also an increasing number of works not generally considered to be operas, such as Handel oratorios and Broadway musicals. The Komische Oper Berlin has even staged a production of Mozart’s Requiem. In response to the increasing diversity of items attracting operatic billing, Monika Hennemann has remarked that “there seem to be no limits to what falls under that category.”1
In light of this trend, an attempt to define what differentiates operatic storytelling from that in related forms such as dramatic songs, oratorios, and cantatas may seem like a futile endeavor. In opera studies, there is a long history of skepticism that there is anything particular about the way operas tell stories, from Edward T. Cone’s suggestion that “every song is to a certain extent a little opera, every opera is no less an expanded song” to Hennemann’s more recent questioning of the distinction between opera and oratorio.2 While recognizing the many connections among operas, songs, cantatas, and oratorios as well as the potential to stage many works in the latter categories as operas, this chapter also points to some key differences between these art forms. One distinction my account of operatic storytelling will not make is that between operas and musicals (for reasons given in the overture). In what follows, references to opera and operatic storytelling should be understood to encompass musicals.
My philosophical readers may harbor skepticism of a different sort, arising from the still controversial concept of medium specificity. Following David Davies and Berys Gaut, I understand medium in art as referring not merely to the materials (physical or otherwise) with which artists work but also to the practices governing how they use these materials.3 Skeptics of medium specificity, such as Noël Carroll, target an extreme version of the thesis, by which each medium is believed to possess properties that are unique to it and artists are instructed to exploit only those unique properties.4 My aim is merely to identify some features that differentiate opera from related media, such as song and oratorio, not to suggest that composers and librettists ought to focus their attention on only these features.
I also reject some of the evaluative claims endorsed by other moderate supporters of medium specificity. Gaut, for instance, suggests that the effective exploitation of the medium-specific features of cinema makes visual extravaganzas such as Cloud Atlas (2012)—chock-full of montage sequences and special effects—cinematically better than static, dialogue-driven films such as My Dinner with André (1981).5 Although commonsensical, such a position would seem to limit the possibilities that are legitimate to pursue in any given medium, a situation I am keen to avoid. In my view, successfully exploiting the properties particular to opera may or may not generate operatic works that are superior in any respect to those that are less characteristically operatic. Nevertheless, knowing what medium the artists were working in, and its particular strengths and weaknesses, plays an important role in explaining and evaluating works of art. For example, understanding the challenges of conveying on the stage the kind of deep psychological investigations at which novels excel renders Britten and Myfanwy Piper’s success at doing so in their 1973 operatic adaptation of Thomas Mann’s Der Tod in Venedig (1912) all the more virtuosic.
As I did in my exploration of narrative in the previous chapter, I will begin by stating the obvious and then venture to more adventurous propositions. Operas require music, specifically song. Jerrold Levinson has defined “paradigm song” as “a melodically and rhythmically distinctive arch of fully-fledged tones of definite pitch, produced in the form of vocables coalescing into words and sentences, and typically with support, primarily harmonic, from some cohort of instruments.”6 Given that this definition would seem to exclude many commonplace components of opera—recitative, vocalise, Sprechstimme, and unaccompanied singing—I am inclined to be more generous. However, that is not to suggest that anything goes. Take, for instance, the Montréal-based composer Luna Pearl Woolf’s “voiceless opera” Mélange à trois (2014) for violin, cello, and percussion. In this work, the performers do not merely play music; they also use their bodies to act out characters in a fictional narrative.7 But at no point do they generate sounds with their vocal cords. Their “singing” is made up solely of the sounds they create with their instruments. As such, I regard this work as less an opera than a work of instrumental theater that is inspired by opera, particularly its penchant for melodramatic plots.
Another work that calls into question the nature of singing is Christine Sun Kim’s Face Opera II (2013) for nine prelingually deaf performers. Although it concludes with the performers generating vocal utterances, it is primarily their facial expressions (from the American Sign Language lexicon) that Sun Kim is inviting her audiences to regard as singing.8
While acknowledging recent explorations of the limits of song and, accordingly, of opera in contemporary music and performance art, this study concerns opera of a more traditional sort. As such, I will be using the verb to sing in its more conventional or literal sense to refer to utterances produced by the performers’ own vocal apparatus that possess a discernable pitch or pitches with at least enough sustainment to allow for pitch discernment. Given that I will be discussing modernist operas and works of musical theater, the bar on tunefulness will need to be sufficiently low to include not only Sprechstimme but also Rex Harrison–esque “talking on pitch,” immortalized in the film musical My Fair Lady (1964).
Many genres of music tell stories through song. In this chapter, I isolate the medium-specific features of operatic storytelling by comparing it first to narrative songs; then to cantatas, oratorios, and serenatas; and finally to plays and films that contain songs but are not generally considered to be musical dramas.9
Operas versus Narrative Songs
One of the remarkable features of opera, in contrast to most other musical genres, is its practitioners’ commitment to storytelling. Since its origins around 1600, opera has been understood as a medium for presenting stories. By far the most common generic label applied to the first operas was favola in musica, story presented through music. And so it has remained, even throughout the aesthetic upheavals of the previous century. While practitioners in virtually all other art forms in the West were abandoning narrative, even representation, librettists and composers kept on representing stories through their works. There are nonnarrative operas, such as Einstein on the Beach (1976), but these are few and far between, and they remain on the periphery of the opera canon, in part because of their denial of our expectation for storytelling. Significantly, we are not similarly disturbed by nonnarrative songs or cantatas. With oratorios and serenatas, the expectation of a story is higher. Notably both were used as opera substitutes during Lent and papal bans. But nonnarrative examples exist, including the most famous oratorio of all, Handel’s Messiah (1742).
It would appear that operatic storytelling can be distinguished from that in songs through Plato’s distinction between mimesis and diegesis or, in the terminology I will employ, enacting character and telling about character.10 Operatic storytelling involves singers enacting characters: singers’ utterances represent characters’ utterances, and singers’ actions represent characters’ actions. In most songs, by contrast, the singer takes on the role of a narrator who summarizes or paraphrases characters’ speeches and merely describes their activities.
As Cone has noted, there is not always a clear distinction between enacting a character and telling about a character. Schubert’s Lied “Der Erlkönig” begins and ends with a narrator telling about the characters, but in the internal stanzas, the singer impersonates the utterances of the father, son, and elf king, thereby vocally enacting these characters.11 Furthermore, many operas contain scenes of narration (a situation that I will discuss in more depth in the following chapter). From this evidence, Cone suggests that the difference between song and opera is primarily a matter of duration.
Cone is too hasty in his conclusions. One key difference between operas and songs is that one-to-one mapping between singers and characters is standard in the former but not in the latter. In an opera, typically all of the utterances made by a given singer are to be understood as representing the fictional utterances of the character the singer is playing. The less this is true, the more challenging it will be to perform the work as an opera.
Suppose