Every song cycle presents its audience with a narrative. Such an idea is hardly new; indeed, the idea of an unfolding plot or unifying concept (textual or musical) is central to the concept of the song cycle as opposed to a collection of songs, at least after the mid-nineteenth century. (1) A great deal of musicological literature devoted to song cycles is aimed at demonstrating that they are cohesive wholes, often connected both musically, by means of key relationships, motivic recall, and similar techniques, and textually, by means of the creation of an overarching plot or concept. (2) One type of narrative results from a poetic unit being set to music in its complete state, without alteration. I am interested here, however, in another type of narrative--one that unfolds when a composer chooses to link previously unrelated poems musically, or when a poetic cycle is altered by rearrangement of the songs or omission of one or more poems. This type of song cycle may create its own new narrative, distinct from its literary precursors.
In an effort to demonstrate the internally cohesive qualities of these song cycles, typical analyses assume that the narrative is contained entirely within the musical cycle itself--that is, that it makes no external references. I do not mean to disparage this type of analysis, which often yields insightful results; however, this approach ignores the audience's ability to make intertextual connections beyond the bounds of the song cycle at hand. To turn briefly to the works of Schumann as an example--which in many ways form a nexus in the study of narrative in song cycles--a number of analyses have attempted to clarify or interpret the narrative of the cycles by illustrating (with varying degrees of success) the unifying features of the works. (3) David Neumeyer (1982), for example, has argued for the need to analyze the music and the text jointly in order to fully understand the way in which the narrative functions in Dichterliebe, and Patrick McCreless (1986) has examined the effects of Schumann's song selection and order on the Eichendorff Liederkreis, op. 39. More recently, in responding to the argument that the concept of an immanent narrative in most Schumann cycles is anachronistic--put forth largely by David Ferris (2000) and Beate Perrey (2002)--Berthold Hoeckner (2006) has attempted to establish anew the musical unity of the cycle.
Each of these excellent analyses focuses on the importance of the creation of narrative in these works, yet despite the extreme popularity of Heine and Eichendorff as poets, none of these studies seriously examines the possibility that audiences would be familiar with their works before hearing the cycle, and that such knowledge might affect a listener's understanding of the work. In cases in which well-known literary or poetic works provide the sources for the song texts, surely listeners familiar with the original would be tempted (or even forced) to project that narrative onto the song cycle, despite whatever alterations a composer might have made to the texts. In other words, while composers are indeed free to create new narratives through, for example, the selection of particular poems from a larger poetic cycle, the audience's knowledge of the original will affect their understanding of the work at hand.
This layered understanding may at first seem to limit the ability of composers to create a narrative in song cycles with preexisting texts. On the other hand, there also exists the possibility for composers to make use of the audience's knowledge. One way of doing so is to use the music composed for a song cycle to refer outside itself and beyond the text, suggesting elements of the original narrative that do not appear in the texts selected for a song cycle. Umberto Eco (2004) has explored the possibilities of a similar literary technique that he terms "intertextual irony," in which authors make frequent allusions to …