Renard (1922). Richard Taruskin summarizes its intended mode of performance as follows: “a troupe of ‘buffoons, ballet dancers, or acrobats’ . . . act the story out in pantomime on a trestle-stage, which they never leave, while the singers remain seated with the instrumentalists in the rear, their voices disembodied after the fashion of Diaghilev’s Coq d’or.”12 The first step would be to discard the dancers and acrobats and to mobilize the singers to enact the characters: Cock, Fox, Cat, and Ram. This effort would be rather challenging, as Stravinsky often uses multiple singers’ voices to represent the voice of a single character. For this reason, the vocal parts are not designated by the characters’ names, as they are in an opera, but as tenor 1, tenor 2, bass 1, and bass 2. Not all vocal music can support singers enacting characters. It requires approximately one-to-one mapping of singers to characters.
One commonplace exception is works that involve characters at different stages of life. In adapting Alison Bechdel’s autobiographical graphic novel Fun Home (2006) into a musical (2013, music by Jeanine Tesori), the librettist Lisa Kron decided to split the central character into Small Alison (age nine), Middle Alison (age nineteen), and (Adult) Alison (age forty-three). Even if the work does not prescribe a character to be divided into multiple roles, a director may decide to divide it between multiple performers. Hans-Jürgen Syberberg’s opera film Parsifal (1982) distributed the singing and acting to different performers, most radically in the case of Parsifal, who was sung by the Heldentenor Reiner Goldberg but acted by a boy (Michael Kutter) and, after Kundry’s kiss, a woman (Karen Krick). Deaf West Theatre’s production of Spring Awakening (2014) involved character doubling of a different sort. Deaf actors were paired with hearing doubles who performed most of their vocal utterances. Wendla’s and Melchior’s singer doubles also appeared onstage for much of the show, often interacting with their signing counterparts (chap. 8 contains a more detailed discussion).
Narration scenes also pose no obstacle to distinguishing between operas and narrative songs. In the course of enacting a character in an opera, a performer may also tell about character, as when a soprano enacts Lakmé in Delibes’s 1883 opera and in so doing tells the legend of the pariah’s daughter. Scenes of narration do not collapse the distinction between enacting character and telling about character because the singer’s act of telling about the pariah’s daughter is embedded within her act of enacting the character Lakmé.
Operas versus Oratorios and Cantatas
Opera is not, of course, the only genre of vocal music capable of presenting stories by means of singers enacting characters. Some examples, incidentally all by Handel, include the oratorios Il trionfo del Tempo e del Disinganno (1707) and Esther (1732), the serenata Aci, Galatea e Polifemo (1708), and the cantatas La Lucrezia (1709) and Clori, Tirsi e Fileno (1707). The difference between these genres and opera hinges on how they are intended to be performed: theatrical performances involving sets, costumes, and singers enacting characters in the case of operas, and concerts involving singers performing in modern dress, standing in place, without backdrops or props in the case of oratorios, cantatas, and serenatas. Determining the intended mode of performance is not always a simple matter. Some composers endorsed both theatrical and concert performances.13 Even if the composer has a definite opinion on the staging of his or her work, performers may decide to pursue an alternative approach. In what follows, I will be speaking about performances that comply with the author’s intentions with regard to performance means. But any work that supports storytelling by means of singers enacting characters could theoretically be staged as an opera, and my account of operatic storytelling would be equally applicable to such performances.
Perhaps surprisingly for those who regard opera as a primarily musical art form, what makes a musical performance with singing an operatic performance, as opposed to a concert, turns on the function of its extramusical components. In operatic performances, what singers look like, what they are wearing, and the movements they perform typically make things true about the characters they play: what they look like and how they move. With dramatic songs, cantatas, oratorios, and serenatas, by contrast, the singers’ appearance and movements do not typically generate facts about the appearance and behavior of the characters they play. It would be highly idiosyncratic for a singer performing La Lucrezia to appear in Roman garb and to pretend to commit suicide at the end. More commonly, the singer will appear in modern formal attire. The inappropriateness of her dress, in the context of the world of the story, does not disturb listeners, since her apparel is not intended to represent that of the character she plays.
To better understand this distinction, Gregory Currie’s category of visual fictions will be useful. “Visual fictions are distinguished from non-visual ones by how content is determined. With a visual fiction, content is determined, in part, by what we see. We see, on stage or screen, a man who moves in a certain way. That man plays a character, and his visible movements determine as part of the content of the play or movie that the character moves in that way.”14 Although Currie is describing cinema, his description also holds true for operatic performances. However, since sound is necessary to opera, while it is not in the case of cinema and other forms of theater, I propose a new category for operas called audiovisual fictions.15 If operas are audiovisual fictions, then songs, oratorios, cantatas, and serenatas (in addition to instrumental works such as Strauss’s Don Quixote [1898]) are aural fictions: their content is determined primarily by what one hears.
The connoisseur of song recitals may protest that I am too hasty in dismissing the importance of seeing to appreciating such performances, asserting that watching their favorite singers contort their faces and gesticulate is an important part of their appreciation of vocal recitals. Indeed, most singers school their facial expressions and body language to be appropriate to the content of the songs they sing. These actions may be similar, even identical to those they might perform in an opera.
Empirical research in the psychology of music perception has shown that even musically trained listeners’ evaluations of the expressivity of instrumental music depend not only on aural information but also on visual information, such as the performers’ gestures and facial expressions.16 In performances of narrative vocal music, gestures and facial expressions may make certain story facts more salient. Take, for example, “Ich grolle nicht” from Schumann’s song cycle Dichterliebe (1840). The speaker has been jilted in love but declares, repeatedly, that he bears no grudge. If a singer were to perform this song with an angry facial expression and a clenched fist, one would be less likely to take his words at face value. Admittedly, the turgid pounding of the piano ought to prompt the astute listener to come to this conclusion even without these visual cues. But I can imagine another scenario in which a singer, through facial expressions and body language alone, marks an utterance as ironic that would not otherwise be interpreted as such. An example, albeit from an opera performance, is Dmitri Tcherniakov’s staging of “Ah, chi mi dice mai” from Don Giovanni (Aix-en-Provence, 2010).
Mozart and Da Ponte intended Elvira’s aria to be a sincere, if exaggerated, expression of the anger and hurt she feels after being betrayed by Don Giovanni. The asides performed by the Don and Leporello contribute levity to this moment for the audience. However, they remain unseen and unheard by Elvira until the end of her aria. In Tcherniakov’s production, all of the characters are part of an extended mafia-like family. Elvira is Don Giovanni’s wife, not a forgotten fling, and she sings this aria as an ironic performance for her wayward husband. The target of her ridicule is the kind of woman Don Giovanni perceives her to be, the kind of woman portrayed in most performances of this aria: one whose entire self-worth is tied up with her ability to attract and retain a man. Since the music and text of the aria have been unchanged, it is primarily through the visual elements of the performance—the singer’s posture, gestures, and facial expressions as well as other staging choices in this aria and the preceding dramatic action—that the character’s ironic intent is conveyed.
Watching singers’ gestures and facial expressions is important to the full appreciation of song performances, even more so than piano recitals, I suspect, because of the rich representational content of vocal music. This observation does not, however,