II
I would now like to look closely at two instances of other-voicedness in music, one provided by "La malinconia," the finale of Beethoven's String Quartet in B
[10] This effect of "narrativization" of apparently nonnarrative figures is widespread; for a discussion, see Paul de Man, Allegories of Reading: Figural Language in Rousseau, Nietzsche, Rilke, and Proust (New Haven, 1979), 13–19.
count of them, will prove to overlap with Beethoven's and Schumann's. The demonstration, however, should also lend weight to the historical claim advanced in the last chapter, namely that Freud's psychoanalysis is in part a codification of nineteenth-century expressive and discursive practices.
Before we proceed, one last theoretical issue must be faced. There is nothing especially problematical in the claim that a text, or even the musical setting of a text, harbors an intrusive narrative effect; textuality in general presumes (even where it conceals) the possibility of narrative. But what kind of narrative effects, intrusive or otherwise, are possible in a piece of textless instrumental music? The answer to this question cannot simply be framed ad hoc to cover Carnaval and "La malinconia"; it clearly depends on the larger question of whether instrumental music presumes narrative possibilities of its own.
It is, of course, commonplace to say that some nineteenth-century instrumental music aspires to the condition of narrative in the sense of telling (encoding, alluding to) a story. In an important series of essays, Anthony Newcomb has taken this idea beyond the commonplace by refining what "telling a story" means in musical terms.[11] Drawing on the work of Paul Ricoeur, Newcomb suggests that following a story and following a work of music entail the same basic activity: the interpretation of a succession of events, possibly quite heterogeneous events, as a meaningful configuration. Such organized successions are historically and culturally bounded. They tend to repeat (vary, transform) a limited number of paradigmatic plots, "standard series of functional events in a prescribed order."[12] What Newcomb succeeds in showing is (1) that paradigmatic plots originating in other discourses can be concretized in textless instrumental music and (2) that the narrative strategies associated with these
[11] Anthony Newcomb, "Once More 'Between Absolute and Program Music': Schumann's Second Symphony," 19th-Century Music 7 (1984): 233–50; "Schumann and Late Eighteenth-Century Narrative Strategies," 19th-Century Music 11 (1987): 164–74; "Narrative Strategies in Nineteenth-Century Instrumental Music," in Music and Text: Critical Inquiries , ed. Steven Paul Scher (Cambridge, forthcoming). See also John Daverio, "Schumann's 'Im Legendenton' and Friedrich Schlegel's Arabeske," 19th-Century Music 11 (1987): 150–63.
[12] Newcomb, "Schumann," 165.
plots—techniques of ordering, interpolating, deferring, and so on—can be concretized along with them.
Without challenging the value of this "narratological" approach, we might wish to ask a question that it leaves out of account. To what extent are the musical concretizations of paradigmatic plots and their techniques actually narrative in character? For of course not all organized successions constitute a narrative in the ordinary sense of the term, the sense associated with literary storytelling: that is, the recounting of certain significant actions by a representative (or antagonist) of the social world in which the actions assume their significance. The traditional genres of storytelling, for instance—epic, novel, tale, and narrative history—produce typical successions that differ markedly from those produced by lyric poetry. One might say crudely that lyric successions are organized with primary reference to gestures of reflection, expression, and interpretation, whereas narrative successions crystallize primarily around patterns of action and consequence. (Primary is a key term here; both kinds of organization occur regularly in both kinds of text.) Or, to invoke a durable concept of Kenneth Burke's, one might say that lyric modes of succession foreground the representation of symbolic action, whereas narrative modes embed symbolic action in a dense matrix of practical, social, and physical actions.[13] Lyric typically foregrounds cyclical, parallel, or graduated patterns; narrative typically foregrounds contingency and causality.[14] Beyond this duality of lyric and narrative, other, messier modes of succession abound in mixed or hybridized texts: Montaigne's Essays , Carlyle's Sartor Resartus , Whitman's "Song of Myself," Nietzsche's Ecce Homo , Woolf's A Room of One's Own —the list could go on.
It is arguable that instrumental music leans more towards lyric than towards narrative in its organization of successions; my Music and Poetry claimed as much for the nineteenth century.[15] I would like to
[13] Kenneth Burke, The Philosophy of Literary Form: Studies in Symbolic Action , 2d ed. (Baton Rouge, La., 1967).
[14] For a further discussion of the lyric-narrative duality in music, and a treatment of musical narrative not tied (like the present one) to specific narrative effects, see my "Dangerous Liaisons: The Literary Text in Musical Criticism," 19th-Century Music 13 (1989): 159–67.
[15] Lawrence Kramer, Music and Poetry: The Nineteenth Century and After (Berkeley and Los Angeles, 1984), 4–23.
take up the thread of that argument here, refining it in terms that may open further the question of music and narrative.
In his book Allegories of Reading , Paul de Man suggests that language is literary to the extent that it acknowledges and confronts its own rhetorical (i.e., inescapably figurative) character.[16] If de Man is right, then it should be characteristic of literary narrative to foreground the process of narration—to tell, in effect, two stories: one referential, the other a story about storytelling. Common experience of canonical literary texts, especially those written since the eighteenth (sixteenth?) century, strongly tends to bear out this claim. The result is a certain dissonance between story and metastory. At its weakest, this dissonance simply enacts the familiar distinction between story and discourse, that is, the distinction between what we construct as the story and the textual process of storytelling. But the dissonance is rarely so tame. So many canonical texts thrive on problematized acts of narration that it is almost fair to say that literary narrative is the art of not telling a story. The reader must wrest a tale from the text by contending against an array of narrative effects that may include limited, biased, or unreliable narrators; multiple narrators who recount or understand events in irreconcilable terms; instabilities and fluctuations in point of view; the narration of events that are imaginary, hypothetical, or indeterminate in relation to the story; patches of exegesis that claim to interpret the narrative, and more.
Broadly speaking, these narrative effects can be sorted into two clusters. The first depends on conflicts among the various fictional persons who are imagined to have a share in producing a given narrative. These narrative agents function in the text as subjects, whether or not they also appear as objects. In principle, at least three distinct types of subject-position may operate within any literary narrative: that of the narrator(s), that of the person(s) whose experience or point of view focusses the narrative, and that of the fictive or projected author, who seeks (not always successfully) to integrate and interpret the others.[17]
[16] De Man, Allegories of Reading , 3–19.
[17] Gerard Genette discusses narrative focusing (which he calls "focalization") in his Narrative Discourse: A Study in Method (Ithaca, N. Y., 1980), 185–211; WayneBooth discusses the authorial subject-position (which he calls the "implied author") in his The Rhetoric of Fiction (Chicago, 1961), 71–76.
This multiplicity of discourse-producing subjects is hard to find in fully composed instrumental music. To recast a well-known suggestion by Edward T. Cone, we do tend to refer musical expressiveness to a presiding authorial subject. Judging from our habits of description, however, we also tend to imagine this subject as autonomous, unencumbered by rivals or collaborators.[18] Texted forms often challenge this attitude; the Lied, for example, often projects a negotiation between independent authorial subjects from the setting and from the text. Most instrumental music, however, seems meant to project a single dominant subject that "surround[s] and include[s]" all others, and in which the authorial, narrating, and focusing activities are merged.[19] Various elements of an instrumental piece may confront this presiding subject with agencies but not with agents, with personifications but not with persons.[20] Characteristic motives, rhythms, or styles may all be highly individualized; so, of course, may instrumental solos. Such elements may even harbor the deconstructive potential to claim or displace an independent subject-position. In their normative role, however, even the most distinctive musical "characters" appear as objects rather than subjects; as products of the musical discourse, not figurative agents for the production of discourse. The subject who supposedly produces the music is not represented directly; in nineteenth-century music, this superiority to representation may act as a sign of the subject's (deep, inner) authenticity.
Following Bakhtin, we might utilize this dominance of a single
[18] Thus Cone: "[The musical persona is] a projection of [the composer's] musical intelligence, constituting the mind, so to speak, of the composition in question"; "[The orchestral conductor] symbolizes both the composer's actual authority over the musical events and the persona's imaginary control" (Edward T. Cone, The Composer's Voice [Berkeley and Los Angeles, 1974], 57, 88; see also 2–40). While I am clearly indebted to Cone's account, my disagreements with it run deep, especially with reference to song (see my Music and Poetry , chap. 5, esp. 125–31). I have avoided Cone's term musical persona precisely in order to avoid its continuous affirmation of a controlling, all-embracing subject.
[19] Cone, Composer's Voice , 22.
[20] A similar point is made by Fred Everett Maus in "Music as Drama," Music Theory Spectrum 10 (1988): 56–73.
subject-position to understand instrumental music on the model of the monologue, in Bakhtin's special sense of the term: a dramatized form of introspection that either represses multiplicity or seeks to "surround and include" it.[21] As a monological form, nondramatic music affiliates itself fundamentally with lyric poetry, which also tends to privilege a single subject-position in which the authorial, narrational, and focusing activities are merged. As Carl Dahlhaus has observed, the assimilation of music to poetry forms the centerpiece of nineteenth-century musical esthetics.[22] It is customary to cite the character piece in this connection, but one might also suggest that large forms like the symphony and string quartet share a common ethos with the extended lyric of English and German Romantic poets, a genre in which the monological unity of the presiding subject is disturbed and—at least pro forma—recovered.[23]
The second cluster of troublesome narrative effects involves clashes and parallels among discrete narrative units. And here, monological or not, instrumental music has found both reason and room to operate. Thus Anthony Newcomb:
Schumann, like Jean Paul, avoids clear linear narrative through a stress on interruption, embedding, digression, and willful reinterpretation of the apparent function of an event (what one might call functional punning). . . . [He also uses a narrative device that] involves what the Romantic novelists called Witz —the faculty by which subtle underlying connections are discovered (or revealed) in a surface of apparent incoherence.[24]
Many of the self-consciously cardinal moments of nineteenth-century music rely on quasi-narrative techniques of interpolation: the interplay of themes, recitative, and Schreckensfanfare that begins the finale of Beethoven's Ninth Symphony; the fleeting recollection of the
[21] Mikhail Bakhtin, The Dialogic Imagination , trans. Caryl Emerson and Michael Holquist (Austin, Tex., 1981), 274–88.
[22] Carl Dahlhaus, 19th-Century Music , trans. J. Bradford Robinson (Berkeley and Los Angeles, 1989), 142–52.
[23] The classic account of the extended Romantic lyric is M. H. Abrams, "Structure and Style in the Greater Romantic Lyric," in From Sensibility to Romanticism: Essays Presented to Frederick A. Pottle , ed. Frederick W. Hilles and Harold Bloom (New York, 1965), 527–57.
[24] Newcomb, "Schumann," 169.
Tristan Prelude in Die Meistersinger ; the materialization of a brass chorale in the introduction and coda of Brahms's finale for his First Symphony. We may find it suggestive that these moments can all be taken to reveal some unsuspected lack within the monological subject.
The narrative effects we have surveyed here follow an imperative to combine storytelling with the continuous representation of an epistemological gap. This imperative is deeply rooted in Western culture, an offshoot of the Platonic distinction between philosophy and rhetoric. Narrative produced under this dispensation dwells in the disparity between knowledge, certainty, and truth on the one hand and belief, experience, and emotion on the other. This principle, what one might call the law of narrative (laws can be broken), may explain why lyric effects generally retard or collapse narrative and why narrative effects are rarely found in lyric poetry, though all of them are possible there. The lyric treats as continuities the epistemological differences on which narrative depends. It is, indeed, one of two art forms that have historically been both celebrated and attacked for doing just that.
The other of those forms is instrumental music.[25] And from this it would seem to follow that narrative effects in such music constitute a critical or disruptive process rather than a normative one. By assuming or revealing a narrative impetus, a lyrical form begins to deconstruct itself. We would therefore expect to find narrative effects in just those compositions that are the most "literary" in that they most explicitly call attention to their own contingent, historical, rhetorical character. We would expect, also, to find the "fissuring" that results both to problematize oppositions of force and structure and to pry apart the inner unity of the monological subject. In short: instrumental music seeks narrative as a strategy of deconstruction. With that claim in mind, I will now turn to "La malinconia," a piece of musical deconstruction if ever there was one.[26]
[25] This was especially true in the nineteenth century. For a brief account on the musical side, see Carl Dahlhaus, Esthetics of Music , trans. William Austin (Cambridge, 1982), 24–51.
[26] Some cautionary remarks may be helpful at this point. First, my claim that narrative dwells in a certain gap between force and structure, a gap that lyric bridges, does not imply that narrative is "deconstructive" in a way that lyric is not. Toarticulate an opposition is not necessarily to dislocate it; to reconcile an opposition is not necessarily to negate it. Lyric and narrative, however, can each be employed as a technique to begin a deconstruction of the other—though other techniques are available, too, and a pat matter/antimatter formula should be avoided here. Second, the value of deconstruction, as I see it, is that it keeps discourse circulating and thaws frozen positions. The purpose of a deconstruction, says Derrida, is to open a reading (Of Grammatology , trans. Gayatri Chakravorty Spivak [Baltimore, 1976], 158). Hence my insistence here, following Derrida, on dislocating rather than reiterating the various oppositions of force and structure. For instance, in saying that Schumann's narrative strategies "keep us wondering where we are in what sort of pattern—in such a way as to stress the process of narrative interpretation," Newcomb ("Schumann," 168) reads these strategies as tropes of reflection or self-consciousness. Historically, this interpretation is unimpeachable. Yet Newcomb's reading itself is also a trope of overturning, a privileging of force against structure. And this is problematical, because his own analyses uncover forces of dislocation that can be confronted only by reading against or across the force/structure polarity.