Example 1.3. Stravinsky, Octet, movement II, variation B

      Within this framework, trombone 2 may be cast in the role of a private with a chip on his shoulder. Fed up with taking orders, he cracks a rude joke. The vulgar sound of his glissando might even suggest the target of his joke: the corporal’s sexual potency. Noticing the ensuing behavior exhibited by the other members of the platoon—the circus-like bassoon ostinato accompanying the flute, who lackadaisically mocks trumpet 2’s dotted rhythms—we may conclude that the private’s prank succeeded in putting a halt to the march, undermining the corporal’s authority and causing a ruckus.

      By now, my remarks on the Octet do constitute a narrative, based on my earlier stipulations, but a narrative of my making, not Stravinsky’s.³³ One might argue that the same is true of my description of Don Quixote, but that would be to elide crucial differences between these two cases. Although my discussion of Strauss’s work did involve creative extrapolations on my part, these were invited by Strauss and guided by features of his work—namely, its title and program.

      Stravinsky’s work may be an apt prop for imaginings that constitute narratives, but there is no indication that Stravinsky invited such imaginings. Its title as well as the titles of individual movements (Sinfonia, Tema con variazioni, Finale) do not recommend extramusical associations. Stravinsky’s statements about his work provide further evidence that he did not intend it to present a story. The first sentence of an article he published shortly after the work’s premiere is “My Octuor is a musical object.” He elaborates that it “is not an ‘emotive’ work but a musical composition based on objective elements which are sufficient in themselves.”³⁴ In his ghostwritten autobiography, Stravinsky sanctioned Walter Nouvel to publish the following aesthetic credo on his behalf: “I consider that music is, by its very nature, essentially powerless to express anything at all, whether a feeling, an attitude of mind, a psychological mood, a phenomenon of nature, etc. . . . Expression has never been an inherent property of music.”³⁵ Such a claim is, of course, dubious in the extreme, as it is not even supported by the composer’s own works, including the passage under consideration. Nevertheless, it does suggest that Stravinsky did not intend the Octet to represent a story.

      At this point, one might object that I have chosen decidedly cut-and-dried examples. A more challenging case would be Mahler’s Todtenfeier, which originated as a symphonic poem but eventually became the first movement of the composer’s Second Symphony (1895). Adam Mickiewicz’s verse drama Dziady (Todtenfeier in Siegfried Lipiner’s 1887 German translation) served as inspiration for Mahler’s compositional activities, but, unlike Strauss, he never made this fact public.

      There were two actions Strauss performed that contributed to Don Quixote being a narrative. First, he modeled his composition on the characters and events of Cervantes’s novel such that appropriately informed listeners could hear those characters and events in his work. Second, he took steps to ensure that listeners would be appropriately informed by gesturing to Cervantes’s work in his title and indications in his score and by sanctioning Arthur Hahn to publish a more detailed program in his guidebook to the symphony.³⁶

      Mahler’s decision not to publicize the story on which his Todtenfeier was based suggests that he did not intend it to be taken as a representation of Mickiewicz’s story. Due to the substantial formal changes that occurred during the compositional process, there is also reason to doubt that Mahler’s compositional activities were guided by an intention to achieve a high degree of correspondence between the musical form of his composition and the plot of Dziady.³⁷

      Even if Mahler did not intend for listeners to think about Dziady while listening to his Todtenfeier, one could still argue that he intended them to invent their own stories while listening to his work. Such an argument would find support in Mahler’s decision to categorize his Todtenfeier as a symphonic poem, a genre of program music often used to tell stories. Furthermore, the musical features of the work all but demand extramusical explanations, and narrative listening was a part of the culture of music listening and criticism in Mahler’s sociocultural context. Nevertheless, for my purposes, I will reserve the category of narrative to works that tell or present particular stories.³⁸ Since particularity admits of degrees, the question of the requisite degree of particularity arises. As sketchy as Strauss’s musical representations are in comparison with Cervantes’s novel, I suggest that they are sufficiently particular to qualify as a narrative. But to preserve a distinction between Strauss’s Don Quixote and Stravinsky’s Octet, I am disinclined to lower the bar such that any work that is conducive to narrative imaginings is able to clear it.

      What about works that fall in between these two extremes, such as Beethoven’s Fifth Symphony? In a famous scene from E. M. Forster’s novel Howard’s End (1910), the Schlegel siblings attend a performance of the symphony. While Margaret and Tibby concern themselves with the “music itself,” Helen imagines “a goblin walking quietly over the universe” who is dispelled in the final movement with “gusts of splendour, gods and demigods contending with vast swords, colour and fragrance broadcast on the field of battle, magnificent victory, magnificent death!” Her more sober siblings dismiss the legitimacy of her response to the symphony. In Margaret’s words, “she labels it with meanings from start to finish; turns it into literature. I wonder if the day will ever return when music will be treated as music.”³⁹

      In response to the interpretive flights of fancy that characterized much music criticism of the Romantic era, many twentieth-century critics harbored similar doubts about the appropriateness of responses like Helen’s. In philosophy, Peter Kivy has been one of the most outspoken opponents of narrative-based interpretations of instrumental music.⁴⁰ I take a more moderate view. The struggle-to-victory narrative Helen and many nonfictional interpreters have heard in Beethoven’s Fifth Symphony is insufficiently particular to ground it as a narrative, in my view. Who is struggling? What is the nature of his struggle? With Don Quixote, there are answers to such questions. With Beethoven’s Fifth, there are not.

      Yet, like Forster himself, I am disinclined to agree with Margaret that her sister failed to treat Beethoven’s music “as music.”⁴¹ The impulse to imagine fictional scenarios while listening to music is a natural one, born of our tendency to anthropomorphize objects and make sense of events in our lives with narratives. Narrative listening is not merely a way to make works of instrumental music accessible to children or members of the laity but can also benefit musicians and music scholars. Narrative-based analysis can be a powerful tool for understanding music because it draws attention to conflicts and discontinuities that may be overlooked by other analytical methods.

      Almén is right to be skeptical of many of the stipulations about the necessary ingredients of narratives put forth by literary theorists. However, if they can be faulted for their failure to consider dramatic or musical narratives, an analogous complaint can be laid at Almén’s feet, since his definition is plausible only in the context of instrumental music. Music scholars concerned that our field is a perennial outsider within broader discourses in the humanities ought to consider whether idiosyncratic definitions of shared concepts are apt to bring us closer to or further away from colleagues tackling similar questions in other fields. My motive for proposing a more circumscribed definition of narrative is to facilitate dialogues between music and other disciplines and between music scholars, performers, and members of the general public.

      If works are more akin to processes than products alone, determining whether a work is a narrative involves more than merely studying its structural features. One must also consider the way it was made (was it modeled on a preexisting or composer-authored story?), its intended mode of reception (did its composer intend for listeners to hear this story’s characters and events in the work?), and whether the composer was successful in achieving the desired response (are listeners able to hear the music as presenting this story?). Thus, I propose that a narrative