"The culture wars" have raged furiously in the academy throughout the past decade, demanding our engagement at every scene of institutional power, from scholarly journals and curriculum committees to admissions offices and boards of trustees. After years of struggle, it comes as no surprise that many of us are suffering from battle fatigue: weary of politics and touched by nostalgia, we long for the time when we can return to the tasks of making music—purely, simply, and undisturbed. Yet, just when it seems that "the culture wars" have waned (and can be dismissed as a journalistic invention), they wax again with a vengeance, as if triggered by the explosion of some long-forgotten land mine.1

Why have "the culture wars" persisted for so long, and why should musicians in the academy continue to care about the outcome? The struggles persist, I believe, for two reasons. First and foremost, the underlying social and economic problems that have fueled the conflicts have not been resolved. Inequities in society will always be manifested in the narrower domain of academe, and the days are long gone when the academy could function essentially as a private club. In short, musicians in higher education have no choice but to grapple with issues of race, gender, and class, especially with regard to the policies that determine institutional access, all the way from undergraduate admission to entry into the professorate.

Second, "the culture wars" continue because they touch on moral and philosophical issues of enduring concern to all members of the academy.2 As musicians who teach in colleges and universities, whether as scholars, composers, or performers, we all have a stake in the debate over cultural values and music's role in shaping and expressing those values. All musicians, especially those who teach, deal with questions of meaning and value every day, though we are rarely called upon to articulate our views in explicit detail. Few would deny that music possesses deep cultural significance, that it even can bear philosophical meaning, yet many recoil at the idea that a Beethoven symphony is irreducibly political. These issues are of great importance, as they go to the heart of music's claim to be more than sophisticated entertainment. The language that is used to delineate music's social meanings is thus of interest to all thoughtful musicians, and it is in the light of these larger concerns that the ideological disputes in current music theory and musicology should be examined.

Reading the continuing debates between contextualists3 and their formalist counterparts on the analysis of absolute music, it often seems that the two camps are debating utterly different issues. Susan McClary, Lawrence Kramer, and other contextualists accuse formalists both within music theory and historical musicology of stifling a politically-informed criticism of music,4 only to find themselves accused of methodological sloppiness.5 At the very least, the resulting discourse obscures the necessity of both approaches for music scholarship that seeks to account for music and its social significance. More troubling, however, is that the formalist and contextualist arguments are mutually reinforcing, with scholars from both camps using each other's positions to justify intellectual rigidity and ideological purity.

Formalists hold that absolute music is radically nonverbal and transcends the limits of discursive language; it therefore can only be analyzed in technical terms. Leaving aside contextualist criticisms for the moment, it should be pointed out that there are at least two advantages to the formalist approach. First, formalists protect musical meaning from being reduced to something non-musical. By emphasizing music's syntactical sufficiency, formalists insist that we understand the musical work as aesthetically autonomous: the temptation to substitute a verbal meaning for something that is nonverbal is thus firmly resisted. Second, by maintaining a steadfast focus on music's structural relationships, and by restricting the analytical vocabulary to technical terms as much as possible, formalist analyses offer great specificity of explication. Though this sometimes results in analyses that are so complex as to be comprehensible to only the most advanced students, this should in no way be taken as evidence that formalism is methodologically misguided.

The disadvantage of this extremely narrow approach to analysis is not its complexity, but rather that it allows virtually no room for cultural critique. Indeed, the problem of music's relationship to society is simply ignored, left to other, less "technically precise" disciplines such as music history and sociology. It is precisely this categorical refusal of formalists to engage music in its social context that makes them vulnerable to the charge of being socially repressive: if the social meanings of music cannot be explored, then its ideological functions remain unacknowledged and unchallenged.

Contextualists, on the other hand, begin with the assertion that music possesses extra-musical significance and that this significance can be articulated. Drawing on a wide range of critical strategies—semiotics, deconstruction, narratology—these scholars explore the ideological implications of such narrative structures as functional tonality, sonata design, and the concerto format. Susan McClary's work has drawn attention to the symbolic residue of vocal music in the instrumental repertoire, reminding us that the musical presentation of texts, especially in opera, inevitably affects our understanding of textless music. Other contextualists, such as Lawrence Kramer and Carolyn Abbate, have demonstrated that instrumental music achieves its significance through reference to cultural tropes of all kinds, and that the difference between absolute music and more overtly referential art forms is one of degree, not of kind. Finally, contextualists seek to negate the objectivist bias of traditional analysis, restoring the cognizing listener to something like an equal footing with the musical work.

But therein lies the difficulty. Does this emphasis on the listener's role lead to a subjective projection of meaning? Does musical analysis become simply the occasion for political discourse? What methodological safeguards prevent contexualists from substituting a vague, if ambitious, social interpretation for a precise, if limited, formal one? At the heart of these questions is the issue of metaphor: "understanding and experiencing one kind of thing in terms of another."6 Formalists, such as Charles Rosen, castigate contextualists for their analytical reliance upon metaphor, yet fail to acknowledge the metaphorical dimension of their own work.7 Contextualists, such as Leo Treitler, defend the use of metaphor, but avoid the hard questions concerning its precise function in music analysis.8 While neither formalists nor contextualists have bothered to interrogate the predicative status of metaphors, much attention has been paid in the past decade to the role of metaphor in music analysis by music theorists such as Marion A. Guck, Robert Snarrenberg, and Janna Saslaw. These scholars have succeeded in exposing many of the fallacies of formalism, yet the weaknesses of the contextulist position have received far less attention. In this essay I will argue that by refining our understanding of metaphor and its place in music analysis, we need not feel compelled to make an artificial choice between formalism and more socially inclusive methods. I will begin by assessing the formalist position in the light of scholarship by Guck, Snarrenberg, Saslaw, and others, and I will address previously undiscussed problems on the part of contextualists with regard to their use of metaphor. Finally, I will offer a group of protocols that prescribe the limits of metaphorical interpretation in music analysis.


Formalist suspicion of metaphor

Formalist analysis is inseparable from the rise of absolute music, and is predicated upon absolute music's aesthetic and social autonomy. This affirmation of the autonomy principle is reinforced by the formalist concentration on the analysis of individual works, a practice that implies that it is possible to understand the classic works of composers from Bach through Boulez strictly through reference to the score. It is important to bear in mind, however, that while music theorists have always been concerned with logic of form and clarity of presentation, these functions were rarely isolated from the study of composition until the nineteenth century.9 Tinctoris and Zarlino codified the contrapuntal practice of their predecessors and more conservative contemporaries, and formulated the fruits of their analyses as compositional prescriptions. Albrechtsberger, Beethoven's teacher, taught analysis as a means to understanding the practice of composers, not as the explanation of music's meaning. Even Rameau's treatise on harmony, though vast, was concerned only with the logic of harmonic function, not with the comprehensive explication of a composition's self-referential meaning. Other aspects of the composer's art—the use of compositional rhetoric, the presentation of text, etc.—were likewise taught with a view toward practical implementation by composers. It was only with the rise of a repertoire regarded as fully transcendent from language that the technical explanation of such music became an end in itself.10

Of course, the music now commonly regarded as "absolute" was not understood in quite this way at first. The combination of formal sufficiency and expressive richness found in the late symphonies of Mozart and the instrumental music of Beethoven was at first explained in an almost mystical fashion. Jean Paul, E.T.A. Hoffmann, Wilhelm Wackenroder, and other early romantics made bold claims for this repertoire—that it was capable of conveying a truth beyond language, and that it had become a cipher of the divine. Significantly, these early interpreters of absolute music were not professional musicians, but literary figures, and their foremost concern was with music's expressive powers. Even the more technically sophisticated Hoffmann nonetheless blended thematic analysis with emotionally evocative prose. These early writers on absolute music clearly felt the need to attribute extra-musical significance to this emerging repertoire. It was only much later, with the Wagner-Hanslick controversy, that the idea of a music that was radically autonomous and completely self-referential fully emerged.

Opposing the view of the Wagnerians, who held that music was inextricably tied to emotion, if not to language, Eduard Hanslick stressed that music created its own meaning which was irreducibly musical. In his classic text, On the Beautiful in Music, Hanslick asserted that we may experience certain emotions when listening to music, but these emotions are not to be confused with musical meaning; that is a matter of musical logic:

To the question: What is to be expressed with all this material? the answer will be: Musical Ideas. Now, a musical idea reproduced in its entirety is not only an object of intrinsic beauty but also an end in itself, and not a means for representing feelings and thoughts.11

Music thus became a unique form of pure play: unquestionably significant, yet of uncertain signification.

It is in this context of polemics over the aesthetic status of music that the work of Heinrich Schenker should be understood. Schenker may be regarded as the first theorist to have provided a comprehensive account of musical meaning as completely self-referential. As Pieter van den Toorn points out in Music, Politics, and the Academy, Schenker was not opposed to the metaphysicians of absolute music; rather, he wished to establish a more rigorous method for the investigation of the repertoire's formal achievements. Though his work has been faulted on many grounds—from its metaphysical foundations to its positivist methodology—it has provided the prototype of today's formalist music theory. Schenker's explanation of musical events as the unfolding of a single elegant structure is widely taught by American theorists in our leading universities, and has paradigmatic significance even for those scholars who reject his assumptions regarding the cultural and historical supremacy of functional tonality. Schenker's primary point—that advanced music functions purely according to its own formal principles—is accepted as axiomatic by all formalists. From this viewpoint, the most important feature of absolute music—indeed, its grandest achievement—is its radical aesthetic autonomy.

Music's aesthetic autonomy is inseparable from the larger process of cultural rationalization in the nineteenth and twentieth centuries, and thus carries profound social and philosophical implications. As the formal possibilities of twelve-tone equal temperament and the sonata principle became more apparent, composers became increasingly concerned with the exhaustion of music's formal latency through the elaboration of its own materials. In a manner that is consistent with Weber's theory of rationalization, musical form became a function of internal consistency, not external reference and "expression." Just as music had become less tied to the external demands of patrons of the church and court, so did music become less bound by affective conventions: expression thus came to be understood as a musical by-product, not the cause and motivation of musical form itself. However, if absolute music is entirely self-referential, has its mimetic function been essentially lost? Though contextualists such as McClary argue with some justification that the Affektionslehre remained a vital component of musical production and reception throughout the nineteenth century,12 these mimetic tropes eventually became peripheral to absolute music. The achievement of this repertoire is its structural complexity and formal sufficiency: it does not require verbal concepts to be comprehensible.

Absolute music's self-referentiality and radical autonomy would seem to make it socially useless, yet Adorno has argued that the objective character of all autonomous art is paradoxically the source of its social value, for it provides epistemic resistance to the devouring activity of the subject: "If any social function can be ascribed to art at all, it is the function to have no function. By being different from the ungodly reality, art negatively embodies an order of things in which empirical being would have its rightful place."13 Again, recalling the historical context in which absolute music first appeared, this was the period of the emancipation of the bourgeois subject. The bourgeoisie had attained an unprecedented freedom of movement, both socially and economically, yet this had ironically compromised the objective conditions of freedom: only by remaining blind to the social consequences of freedom and its foundation in class conditions could the ideology of freedom be maintained. This irony was reflected in the initial reception of absolute music: the music was praised for its expressiveness (its subjective power), while its defining aesthetic feature was its structural autonomy (its objective character). The objective structures of absolute music were never meant to become the projection screen for the subjective fantasies of its listeners; rather than existing for the subject in ideological affirmation, absolute music confronts the subject with social truth through absolute faithfulness to its own rational procedures.

Given this emphasis on absolute music's radical self-referentiality, it follows that the analysis of absolute music should focus just as radically on the object. Hence, formalist analyses attempt to explicate musical meaning only through reference to a given composition's use of materials. Knowledge of formal conventions is assumed, but the interest is not in the homology of music and form; rather, what is emphasized is the individuality of each composition, the way each establishes its own universe of meaning. Other vital factors—how music is understood, how much information is processed by the listener, how this affects the aesthetic experience as a whole—are disregarded in the interest of accounting for the structural sufficiency of the whole. Though formalists rarely deny that music provokes emotion, they simply do not see the explication of that emotion as a proper concern of analysis. Just as the composition is aesthetically autonomous, the analysis must preserve and protect that autonomy from the threat of mimesis or extra-musical concerns. Following a lengthy and illuminating application of pitch-class analysis to the second movement of Beethoven's Ninth Symphony, van den Toorn writes:

And there can be no better way of representing this music, it seems to me, of drawing the listener into the web of its detail. For it is precisely by "noticing" that detail, the detail of an individualization, that the experience is given greater depth. Indeed, for those mindful of the context, the objective likely to surpass all others is that of experiencing more of that context. And the greater the depth, the more intense the experience; the greater the potential for intimacy.14

According to van den Toorn, if one accepts the radically self-referential character of absolute music, one will restrict musical understanding to a single context—the autonomous work of music—and refrain from expanding the analytical frame to include seemingly non-musical concerns. (It should come as no surprise that van den Toorn defines the musical experience as one of intimacy, a radically private and unmediated form of communication between composer and listener.)

The objective of formalist analysis—understanding the functioning of a musical composition as a completely self-referential system—results in a severe suspicion of metaphor, for metaphorical language opens the gate to unlimited references beyond the confines of the music's structure. Though metaphors appear in virtually all formalist analyses, they are enlisted rather apologetically and are never the focus of the analysis. They usually occur as vehicles for technical understanding, enlivening otherwise arid prose with a rhetorical flourish.

The avoidance of metaphor helps to explain why the language of formalists has become increasingly obtuse and jargon-laden: if metaphors are disallowed, one has no choice but to describe the object in ever-greater technical detail. Preferably, this approach to analysis yields quantifiable results, and there is no reason why this should disturb a thorough-going formalist: if analysis aspires to the same logical rigor as its object of enquiry, it need not possess obvious humanistic appeal. The point is not to explain the music's relationship to the subject, but rather to probe the musical object itself—in antiseptic isolation. The less analysis relies on prose, on language, the more it will rely on number, the very source of music's formal self-sufficiency.

There are at least three contradictions that follow from the ambitions and assumptions of formalist analysis, each stemming from the inadequacy of propositional language to account fully for musical meaning. First, every form of analysis constitutes a simplification of the phenomenon. Though this may seem to be a trivial observation, it merits detailed examination because the charge of simplification is the one most frequently leveled against contextualists and all others who challenge the formalist methodology. In "Music à la Mode," Charles Rosen criticizes McClary and others for drawing frequently on metaphor, charging that "all metaphors simplify."15 Indeed, they do. Metaphors focus on specific features of disparate domains, calling them into relief with utter disregard for any totalizing objective. Yet, knowledge of any kind depends upon simplification and cognitive selectivity, for particulars only take on significance through reference to concepts. The moment of selectivity in the construction of knowledge is part of a larger dialectic, consisting of the accumulation of information, the assimilation of that information through concepts, followed by the critique of those concepts, which in turn allows for the further accumulation of information. This is as true for formalists as it is for contextualists: in an effort to fully explicate music's technical functioning, formalists disregard all "irrelevant" factors, such as music's emotional, social, and political significance. As was stated earlier, this is not a bad way to start. The problem, however, is that formalist methodology does not allow for a moment in which the analytical frame can be expanded: the relevance of the "irrelevant" is rarely acknowledged, much less integrated through a subsequent analytical step. The categories of formalist analysis—ranging from the simplest elements of rhythm and pitch to the most complex concepts of large-scale form—help control the excess of significance in the musical phenomenon. Though the repertoire of absolute music unquestionably requires such conceptualization, formalist methodology highlights those aspects of music that can be quantified and generally ignores those aspects (such as timbre) that cannot.

Hence, the knowledge offered by formalists is necessarily limited by their methodological assumptions. Metaphors may simplify, but so inescapably do formalist analyses. The issue is not whether or not to simplify, but how to acknowledge simplification as but one step in analysis. Simplification and focus lead to insight, but to insight that must be challenged by an expanded frame of reference. Formalists should begin their analyses knowing that detailed analysis of music's self-referential form must be brought into play with music's social functions.

There is a second contradiction in the formalist objection to metaphor: to speak of music at all is to speak metaphorically since music and language constitute separate domains. We all have a tendency to think of sound in terms of language, and formalist language is designed to conceal the discrepancies that inevitably arise. Even at its most rigorous and precise, the language of formalism is metaphorical, betraying our common reliance on image schemata, which Mark Johnson defines as the "abstract patterns in our experience and understanding that are not propositional in any of the standard senses of that term, and yet. . . are central to meaning and the inferences we make."16 Indeed, the work of Johnson and George Lakoff has made it abundantly clear how thoroughly metaphorical our thought processes actually are.17

Though we may cast our thought in propositional form, most of our thinking requires the application of spatial referents to nonspatial domains. The paucity of aural vocabulary should come as no surprise, nor should the fact that we tend to rely on visual metaphors when describing sound. Sound (especially recorded sound) is sensorially isolated: we cannot see, touch, taste, or smell a sound. As a result, we tend to use metaphors borrowed from these other senses to account for aural stimuli. To describe a pitch as "high" may seem obvious enough, but it is only true metaphorically: a pitch is higher on a staff or higher in the vocal tessitura, but strictly speaking, the pitch is faster, since its perceived "highness" is a function of its vibrational frequency. Our notational system is metaphorical through and through, combining descriptive and prescriptive functions so successfully that most trained musicians think habitually in terms of notational imagery.

If the underlying metaphorical character of the formalist vocabulary is now evident, this in turn points up another way in which formalism is far from objective. Lakoff and Johnson have demonstrated that the metaphors we commonly use are thoroughly intertwined with cultural values. For example, words like "up" and "down" take on positive and negative connotations because of our bodily experience. To be "up" is to be operative, to be superior, to be correct, positive; to be down is generally the reverse. This can be traced to the simple fact that healthy, functioning human beings are able to stand and walk, and are hence upright. Given the pervasiveness of these metaphors, and the unconscious manner in which they are learned and used, it should come as no surprise that we are rarely aware of their cultural connotations.

The spatial metaphors discussed to this point are used primarily to describe facts and the simplest kind of relationship between facts (e.g., "This pitch is higher than that pitch."). Yet, as Lakoff and Johnson have demonstrated, even these common metaphors are inscribed with cultural value. What are we to say then about metaphors of analysis which are deliberately fashioned to describe the complex relationships between musical facts? Janet Levy has explored this question in detail, examining the valuative implications of organicist and naturalist metaphors.18 With regard to the former, she argues that the penchant of some theorists (especially Schenkerians) to prize organic unity in music is revealed by the use of such metaphors as "germ," "prefigured," and "concealed seams."19 Theorists who subscribe to organic unity (or "monogenesis," to use Goethe's term), assume that music should utilize such relationships and hence invoke organicist metaphors to persuade the reader that such relationships exist. Metaphors such as these are so pervasive and so insidious that many students adopt them unconsciously.20 The fact that they are sometimes quite apt for describing aural similarities does not negate the fact that they bear specific values, values that come with their own genealogy of cultural and social practice.21

In Chapter Three of Music, Politics, and the Academy ("Schenker and His Critics"), van den Toorn argues that Schenker's metaphors are less problematical than those of Leo Treitler and other contextualists:

Terms such as prolongation, neighbor note, reaching-over (übergreifen), and initial ascent are openly descriptive of tonal phenomena and were obviously intended to describe such phenomena by means of a physical imagery that was related to Schenker's two central metaphors, those of motion ("directed motion") and organic growth. There is little reason why such a vocabulary should prove inhospitable to anyone with a need for expressive analytical discourse, for an overt metaphorical presence.22

The problem, however, is that not everyone "with a need for expressive analytical discourse" desires "directed motion" or "organic growth," as both of these metaphors are laden with contested cultural value and meaning. Must we all assume that the tonal repertoire is fundamentally based on the idea of directed motion to preordained hierarchical goals? Though van den Toorn is accurate in pointing out the spatial references of other Schenkerian terms, it is unclear why Treitler's "poetical" terms are less analytically acceptable than Schenker's "technical" terms, since both are ultimately metaphorical.23

Beyond these contradictions of formalist analysis (the inadequacy of formalist categories to the complexity of music and the metaphorical nature of formalist language), there is a third problem to be considered: the objectivist character of formalist analysis promotes the ideology of autonomous art by denying all aspects of music that cannot be empirically proven.24 Though it is comforting to believe that music can be studied in a pure, isolated state, music is thoroughly enmeshed in the world. When the analysis of music starts and finishes with the isolated musical work, it promotes the fiction that the knowledge yielded through such analysis is complete. This is the message that has been given to numerous generations of musicians by scholars at our leading institutions of higher learning. Despite the fact that such scholars teach the history of music, they do not teach, in the words of composer Herbert Brün, "the history that music made": they do not teach music as a social and political phenomenon, a phenomenon that only achieves its ultimate significance—indeed its ultimate aesthetic significance—in the context of social and political life, not just cultural life.25

As Jürgen Habermas has taught us, there is no such thing as objective knowledge, knowledge free from human interests.26 The very questions that we bring to the object of inquiry reveal our biases, our interests, and that much more the path that we wish to follow in the pursuit of answers to those questions. The problem, of course, is that formalist analyses are predicated on objectivity, thus making it even more difficult for scholars to acknowledge their interests, let alone interrogate them. Though formalist accounts of the musical object are indispensable, they conceal as much as they reveal. This is not to say that every interpretation is a misinterpretation—that there are no criteria for truth—but rather that formalist interpretation fails to acknowledge the fact that it is incomplete, or, if it does so, it is only incomplete from the standpoint of formalist principles. This situation would be greatly helped by the simple recognition of these limitations on the part of formalists, something that these scholars have been heretofore been reluctant to do.

When whole domains of intellectual enquiry are disallowed, an ideology is being served. Ideologies function less on the basis of what is openly said than on the basis of which questions are permitted. Formalists essentially subscribe to Wittgenstein's dictum, "What we cannot speak about we must pass over in silence,"27 yet they should remember that Wittgenstein himself rejected this prescription in favor of a much more complex approach to language and reality. The first step toward a theory of analysis that is adequate to the complexity of music as an aesthetically autonomous, yet socially contingent phenomenon, is the acknowledgment of formalism's necessity and inadequacy. Though contextualists will dispute the aesthetic autonomy of absolute music, formalists are on solid ground when demonstrating the internal consistency of pure instrumental music. Such demonstration prevents absolute music from becoming the screen for programmatic projection. Only when absolute music's aesthetic autonomy has been affirmed can the next interpretive move be taken: the exploration of absolute music's social contingency.


Contextualist reliance upon metaphor

Because contextualist analysis is patently opposed to objectivism, contextualists give much greater emphasis to the subjective aspect of analysis, the point of reception. Contextualists thus reverse the emphasis in the dialectics of interpretation, starting first with the socially constructed listener. The question McClary and other contextualists ask is not "How does this music make sense?", but rather, "How does the listener make sense of this music?" By calling attention to the listener's cognitive process, contextualists are able to open the door to the social world—the world of symbols, socially constructed meaning: the world of metaphor.

Contextualists gladly enlist metaphor so as to escape the limits of propositional language. A metaphor, whether simple or outrageous, is uttered as a literal falsehood, but one that points to an important analogy with another realm. To cite Paul Ricoeur, "Metaphorical meaning does not merely consist of a semantic clash but of the new predicative meaning which emerges from the collapse of the literal meaning, that is, from the collapse of the meaning which obtains if we rely only on the common or lexical values of our words."28 In a manner that is consistent with the contextualist use of metaphor, Ricoeur suggests that metaphors transform our understanding, achieving a new kind of meaning through the "collapse of meaning." But what kind of meaning is achieved through a literal falsehood? Can such meaning be offered in the form of a truth claim? It would seem that many contextualists have not considered the implications of these questions. In their eagerness to systematically develop metaphors that convey music's analogical relationship to the world, contextualists always risk being seduced by the very power of their own metaphors. This can take at least two different forms: theoretical seduction and methodological seduction. The former results from using metaphors to make truth claims, the latter from simply forgetting that a metaphor has been invoked in the first place.

In a recent essay, "Constructions of Subjectivity in Schubert's Music," McClary writes the following about the second movement of the "Unfinished" Symphony:

Movements such as the second movement of the "Unfinished"—as well as sections of many other compositions—celebrate such differences as utopian: in these, the field is established by pleasurable free play, and the forces that threaten to disrupt are successfully defused. But there was another side to Schubert: a side that produced victim narratives, in which a sinister affective realm sets the stage for the vulnerable lyrical subject, which is doomed to be squashed. This is true, for example, of the first movement of the "Unfinished" Symphony.29

While McClary does not explicitly state that the meaning of Schubert's music corresponds to her reading of it, and thus make an overt truth-claim, her metaphors carry that implication: after all, the movements don't "celebrate," McClary "celebrates" a particular interpretation. What is particularly bothersome in this case is that McClary focuses squarely on the issue of reception early in the essay, describing how her students decided that Schubert was gay all on their own.30 Had she followed that line of thought more strictly, her reading would carry greater force. As it is, she so loads her metaphors that the interpretation is weakened—weakened not because the music cannot justify the interpretation, but because its status as a subjective reading is subtly abandoned through a dubious predication.

The second problem, that of methodological seduction, most often occurs when a metaphor has been elaborated fully into an allegory. Consider McClary's remarks on the second movement of Mozart's Piano Concerto in G Major, K.453:

For it does seem that the piano was an unlikely participant in the celebration of its own submission. The happy ending, the requisite closure, was attained at too high a cost: the lobotomy at the moment of recapitulation. To be sure, the valued quality of inevitability is present here—but in the literal sense of "unavoidable:" regardless of how hard the soloist protested and tried to escape this conventionally predetermined fate, its struggles were futile. This is hardly the sense of "inevitable" we think we are celebrating when we valorize eighteenth-century music, but perhaps this iron fist is always hiding within the velvet glove.31

At this crucial point in her analysis, McClary seems to hope that we have been so persuaded by her fundamental metaphor (the concerto as dialogue) that we will accept all of its extensions. The problem, however, is that her assertions are made in reference to her metaphorical construction. Has she forgotten that she is using a metaphor? The metaphor has become so developed that it has, indeed, started to take over.

In defense of McClary, she should not be faulted for offering strong readings of any music, whether by Schubert, Mozart, or other composers. So long as she does not make truth-claims on the basis of her metaphors, her work can be argued against, but not falsified. Of course, that is what truly bothers the formalists such as van den Toorn and Rosen, that metaphorical interpretations are persuasive without being true. McClary's ultimate truth-claim is about music as a social phenomenon ("A piece of music therefore can be perceived as a dialectic between order and noise, a strategic model of how violence or deviance may be tolerated and channeled within a given social framework."32), and she uses her reading of a particular piece of music to support that truth-claim.

Van den Toorn and other formalists ultimately are arguing against this fundamental premise of contextualist criticism. As we have already seen, however, the problem is that metaphors are so much a part of our mental processes that formalists cannot protect music and culture from their influence. When contextualists are wrong on the music, their work is unconvincing and they should be criticized. However, when metaphorical interpretation is thoroughly grounded in the details of the score, contextualists cannot be faulted for simply invoking metaphor as a rhetorical tool. It is not the use of metaphors in general but the implications of particular metaphors that formalists find undesirable.

The predicative status of metaphor will probably always be a matter of dispute between those who wish to assert that metaphors express what conventional language cannot and those who deny such power of expression. Lakoff and Johnson, for example, view metaphor as "imaginative rationality":

Metaphor is one of our most important tools for trying to comprehend partially what cannot be comprehended totally: our feelings, aesthetic experiences, moral practices, and spiritual awareness. These endeavors of the imagination are not devoid of rationality; since they use metaphor, they employ an imaginative rationality.33

According to this view, the value of metaphor is measured by what it suggests rather than by what it fails to precisely denote. Certainly when one considers the excessive character of absolute music—its syntactical clarity and semantic vacuity, its nestedness in cultural convention and its ability to break free of those conventions—metaphor would seem to be an indispensable analytical tool. In a seminal essay on the role of metaphor in analysis, Marion A. Guck writes, "Metaphor-based analyses, in encompassing the complexities of relations in pieces, capture something of the processive effect of many patterns interacting. They can provide the means to approximate faithfully the complexities of our peceptual manipulation of pieces."34

And yet, what do metaphors actually reveal? According to Donald Davidson, we must never forget that metaphors are literally false: metaphors do not magically say what the literal meanings of words cannot say. "What matters is not actual falsehood but that the sentence be taken to be false."35 This is the maddening mystery of metaphors: the meaning is vivid and seems to be clear, but in actuality is not. Both the creation and deciphering of metaphors require imagination; there are no rules that govern their function. They powerfully suggest relationships between disparate domains, but they in no way prove them. Indeed, what they supposedly prove would be impossible to express: that is why a metaphor is used in the first place.

Does this mean that metaphors are pure fancy? Not even Davidson makes this claim: "Generally it is only when a sentence is taken to be false that we accept it as a metaphor and start to hunt out the hidden implication."36 Note the words, "hidden implication." From this it is clear that Davidson does not intend to dismiss metaphorical significance, but rather, false predication. According to Davidson, metaphors mean exactly what their words say: they create a literal falsehood that suggests a meaning that cannot be paraphrased. Metaphors require that the reader consider relationships between disparate domains that cannot be made using propositional language. The reader must "hunt out" the possible significance of a patent falsehood. This idea is similar to Wayne C. Booth's comment: "Metaphors are seductive because they require the reader to participate in an imaginative process, a kind of dance."37 If an analyst uses a metaphor, he or she should always be aware that the price of metaphorical assertion is that the significance of the metaphor cannot be proven. This is not to say that a metaphor cannot be persuasive, however, and more will be said shortly about that matter.

Given the ambiguous predicative status of metaphor, it is easy to understand why an analyst who casually invokes metaphor might forget that he or she has actually done so. After all, if one is opposed to an objectivist epistemology (as contextualists are by definition), and the objectivist criticism of metaphorical interpretation is based on the inability of metaphors to bear truth claims, one might just push blithely on, insisting that metaphors do in fact, reveal hidden truth. But the "truth" that is supposedly "hidden" is a product of the imagination—albeit a useful, insightful product of the imagination. Metaphors are not to be evaluated on the basis of their truthfulness, only on the basis of the persuasiveness. Contextualists may view this assertion as a denial of metaphorical power, yet I hope to show later in this paper that this is not the case.

Another problem confronts contextualists: Does it matter that the causality of analogical relationships (conveyed through metaphors) cannot be proven? According to Lakoff and Johnson, metaphors have the power "to create a reality rather than simply to give us a way of conceptualizing a preexisting reality."38 Since this description could apply to lies as well as metaphors, it is important to draw a number of distinctions.

Metaphors may not "create a reality" in the objectivist sense, but they create connections in the mind of the reader. The structural relationships between social dynamics and musical structure are not there in the music, somehow waiting to be discovered; this is the false logic of objectivism which affirms facts above the relationships we construct between them. Contextualists draw connections where none "existed" before, citing analogies between disparate realms and arguing for the full consideration of the implications. At their best, contextualists such as McClary, Subotnik, and Kramer argue so persuasively that the reader accepts the analogy as a permanent component of musical understanding. The point is not to demonstrate causality (a relationship that is always difficult to prove, even by the standards of objectivists), but rather to persuade the reader that a relationship is significant. Though this may seem like a modest accomplishment, it is not. One of the primary accomplishments of socially-informed criticism—from Adorno to McClary—is the establishment of new analytical categories. Once one has learned to think in terms of analogical relationships between musical and social structures, it is virtually impossible to think about music without recourse to these categories. Contextualists have made it easier to interrogate the social significance of absolute music, and that is an invaluable contribution.

The real problem is when the issue of causality is not faced squarely. Contextualists all too frequently leave doubts as to their claims, implying a causal relationship where one is undemonstrable. If contextualists are willing to accept that causality is not the issue—that the issue is one of persuasion and argument—then no such claims need be made. This even supports the political objectives of some contextualists, insofar as it places the interpretive emphasis on contemporary, rather than historical, reception. One of our advantages over historical audiences is that we can see outside the frame of the original habitus, we can draw connections between music and society that no historical audience could have imagined. So, falsifying contextualist claims on the basis that no such understanding existed at the time is to (in time-honored fashion) use the status quo as an argument against consciousness. Contextualists need not argue on the basis of causality to alter the modes of musical perception.

There is a final point of contradiction in contextualist analysis. If contextualists simplify, it is usually on not one but two levels: the musical and the social. Just as formalists call attention to certain features of music at the expense of others, so do contextualists through the use of metaphor. The great challenge of the metaphorical interpretation of absolute music is to enhance our understanding of this repertoire's social embeddedness without diminishing our understanding of its formal integrity. A concerto may indeed be structurally analogous to a social dialogue, but it is much more as well. To reduce a complex phenomenon to such an equation is to substitute one meaning for another, a move that denies music its legitimate aesthetic autonomy. Analogies may be drawn between music and society, but such analogies are necessarily incomplete. Only through immanent critique can the subjective and objective moments of analysis be held in tension and not allowed to collapse into false identity.


The limits of metaphorical interpretation

Having examined the strengths and weaknesses of formalists and contextual analysis with regard to the use of metaphor, we are now in a position to consider three guidelines for metaphorical interpretation:

  1. Since absolute music is not denotative, any attempt to attribute extra-musical significance to absolute music will require the use of metaphor.
  2. Since metaphor entails the mapping of analogous features from one domain to another, and the aesthetic autonomy of absolute music results from its formal sufficiency, the metaphorical interpretation of absolute music should be grounded in music's formal structures.
  3. The elaboration of metaphor in critical discourse has only one function: to demonstrate the aptness of the metaphor.


1. Since absolute music is not denotative, any attempt to attribute extra-musical significance to absolute music will require the use of metaphor.

Formalists should acknowledge the metaphorical dimension of their analyses. If it is true that metaphors seduce, and that they often seduce the metaphor maker, then the seductiveness of metaphor is most powerful when it is least acknowledged. As was argued earlier, the formalist prejudice against metaphor is based, in part, on the belief that the technical language of music is free from metaphor, a belief that is false. The value of formalist analysis lies not in its nonmetaphorical character, but rather in the precision and specificity that is offered through its methods.

Formalists should further recognize that this technical precision does not remove music or analysis from the world: by attempting to account for musical meaning (through the use of language, obviously), the cognitive assumptions of our culture are imposed through the subtlest kind of metaphors. Yet the knowledge offered through formalist analysis is not diminished in value because it is contingent, since this is a condition of knowledge generally and does not detract from the contribution of rigorous formal analysis. Formalists should therefore acknowledge the epistemological contingency of their work, knowing that—despite these limits—strong formalist analysis is indispensable for understanding the internal functioning of musical elements. Furthermore, formalist analysis implicitly affirms the aesthetic autonomy of absolute music, and rightfully asserts that knowledge of this repertoire must attempt to account for its claim to formal sufficiency (even if this claim is ultimately revealed as false). This is true despite the fact that no analysis is ever definitive, since our understanding of music changes as the tools of analysis change: this fact alone demonstrates the contingency of musical knowledge. What should be abandoned is not strong formalist analysis, but the claims of such analysis to be objective: formalist analysis is objective because it focuses primarily on the object of study, not because it offers purity of perception or understanding.


2. Since metaphor entails the mapping of analogous features from one domain onto another, and the aesthetic autonomy of absolute music results from its formal sufficiency, the metaphorical interpretation of absolute music should be grounded in music's formal structures.

Contextualists should acknowledge that their analyses are fruitful only if they take formal rigor another step forward. Without the precision of formalist analysis, contextualists run the risk of simply offering programmatic substitutes for musical meaning. By first demonstrating absolute music's aesthetic autonomy, contextualists can retain the social value of this autonomy before expanding the analytical frame to demonstrate absolute music's social contingency. A full understanding of these twin aspects of absolute music—aesthetic autonomy and social contingency—guards against the subjective and objective fallacies in musical analysis.

Theorists and musicologists are not presented with a choice between formalism and contextualism; only when these methods of analysis are followed simplistically are they truly antithetical. If contextualists first attend rigorously to the details of musical meaning before interrogating music's social implications, they are not projecting politics onto music; rather, they are elaborating the significance of music in a variety of social and political contexts. Likewise, if formalists acknowledge that music exists only in social contexts and that formalist analysis is necessarily incomplete, their project need not be understood as an implicit denial of music's social meaning.

If the twin fallacies of subjectivism and objectivism are to be avoided, metaphors must be understood as entailing an epistemic split-reference. We understand metaphors by maintaining two incompatible domains in our minds, not by collapsing them into a third meaning. Every metaphor is a kind of fiction, but a fiction that enlists our imagination to see connections between domains that cannot be articulated through propositional language. As the Ricoeur points out, the Majorca storytellers begin their tales with "it was and it was not."39 Every metaphor functions in precisely this way—suggesting, pointing, but never disclosing a truth that could otherwise be expressed propositionally.


3. The elaboration of metaphor in critical discourse has only one function: to demonstrate the aptness of the metaphor.

This is not the same thing as explaining the meaning of the metaphor. As was just asserted, metaphorical understanding requires a split-reference. Also, to explain the metaphor is to kill it, as Donald Davidson has noted. Metaphorical elaboration must always acknowledge its imaginative aspect and never suggest that a fiction has been magically transformed into propositional truth.

Though it may seem obvious and unremarkable, it is worth remembering that the metaphor is the distinct contribution of the analyst; it is not in the music. Since no connections naturally exist between music and the world, the analyst invokes a metaphor to draw a connection between the domains. The metaphor must be elaborated in order to support its invocation, not to prove its truth; therefore, both domains encompassed by the metaphor must be elaborated in the greatest possible detail to strengthen the analogical connection. Though thick description may not prove causality, that is not its burden: it need only help us imagine "possible worlds." Once the analogical relationship is established between two domains—through the use of metaphor—the stage is set for a different kind of analysis, one that attempts to document empirically the various levels of mediation connecting the musical text with the social context.

Davidson correctly distinguishes "what words mean and what they are used to do," and reminds us that metaphors should be understood according to their purpose.40 Metaphors are used to persuade: the literal falsehood suggests a different meaning through the combination of disparate domains. Yet, according to Davidson, the possible meanings are too vast, too lacking in specificity, to bear truth claims. If they are persuasive, it is because the particular metaphor is strikingly apt or because we have been persuaded of its aptness. (Davidson steadfastly denies that we have been persuaded because the meaning of the metaphor has been "explained," since in his view there is no metaphorical meaning that can be explained.)

All analysis is about the proclamation of meaning and thus reflects particular interests, and metaphor is obviously a tool for the promotion of certain interests. The fact that metaphor remains a rhetorical tool in no way detracts from its value, for it is only through metaphor that music's relationship to the world can be explored. Only when a supposedly objective analysis is offered as an alternative is metaphorical interpretation seen as less valuable.

1See the collection of commentary by Todd Gitlin, Annette Kolodny, et al., in "Have the Culture Wars Ended?" The Chronicle of Higher Education, March 6, 1998: B4-8; see also Wendy Steiner, The Scandal of Pleasure: Art in an Age of Fundamentalism (Chicago: University of Chicago Press, 1995).

2The ideas expressed in this paragraph were largely stimulated by Bethany Bryson's fine paper, "Understanding the Canon Wars: A Comparative Study of Four Universities," presented at the Conference on Social Theory, Politics, and the Arts, Drexel University, Philadelphia, October 9, 1998.

3Though these scholars are frequently identified as the New Musicologists, this term carries the connotation of faddishness. "Contextualists" is the more accurate term, given the insistence of these scholars on always understanding music in its social, political, and philosophical contexts. See Rose Subotnik, "On Grounding Chopin," in Developing Variations: Style and Ideology in Western Music (Minneapolis: University of Minnesota Press, 1991), 142.

4See in particular Susan McClary, Feminine Endings: Music, Gender, and Sexuality (Minneapolis: University of Minnesota Press, 1991) and Afterword to Jacques Attali, Noise, trans. Brian Massumi (Minneapolis: University of Minnesota Press, 1985); Lawrence Kramer, Classical Music and Postmodern Knowledge (Berkeley and Los Angeles: University of California Press, 1995); Ruth Solie, "What Do Feminists Want? A Reply to Peter van den Toorn, Journal of Musicology IX, No. 4 (Fall 1991): 399-410; and Rose Subotnik, "The Role of Ideology in the Study of Western Art Music" and "Musicology and Criticism," in Developing Variations, op. cit.

5See in particular Pieter van den Toorn, Music, Politics, and the Academy (Berkeley and Los Angeles: University of California Press, 1995) and Charles Rosen, "Music à la Mode," New York Review of Books 41 (23 June 1994): 55-62.

6George Lakoff and Mark Johnson, Metaphors We Live By (Chicago and London: University of Chicago Press, 1980), 5.

7Charles Rosen, "Music à la Mode," 55-62. Rosen focuses at length in this essay on such quotes from McClary as the following, in reference to the first movement of Beethoven's Ninth Symphony: "The point of recapitulation...is one of the most horrifying moments in music, as the carefully prepared cadence is frustrated, damming up energy which finally explodes in the throttling, murderous rage of a rapist incapable of attaining release."

8See Leo Treitler,"Language and the Interpretation of Music," in Music and Meaning, ed. Jenefer Robinson (Ithaca and London: Cornell University Press, 1997), 23-56.

9One might argue that medieval theorists such as Boethius were not primarily concerned with compositional practice, but neither were they concerned with the analysis of individual works of music.

10For a fuller discussion of the gradual emergence of formalist methods, see Ian Bent's preface to Music Analysis in the Nineteenth Century (two volumes), Ian Bent, ed. (Cambridge, UK: Cambridge University Press, 1994).

11Eduard Hanslick, On the Beautiful in Music, trans. Gustav Cohen (New York: The Liberal Arts Press, 1957), 48.

12McClary, "Narrative Agendas in `Absolute' Music: Identity and Difference in Brahms's Third Symphony," in Musicology and Difference: Gender and Sexuality in Music Scholarship, ed. Ruth A. Solie (Berkeley: University of California Press, 1993), 326-344.

13Theodor W. Adorno, Aesthetic Theory, ed. Gretel Adorno and Rolf Tiedemann, trans. C. Lenhardt (London: Routledge and Kegan Paul, 1984), 322.

14van den Toorn, op. cit., 119.

15Ibid., 59.

16Mark Johnson, The Body in the Mind: The Bodily Basis of Meaning, Imagination, and Reason (Chicago and London: University of Chicago Press, 1987), 2.

17See Janna Saslaw, "Forces, Containers, and Paths: The Role of Body-Derived Image Schemata in the Conceptualization of Music," Journal of Music Theory (Fall 1996), Vol. 40, No. 2: 217-243.

18Janet M. Levy, "Covert and Casual Values in Recent Writings about Music," Journal of Musicology 5, No. 1 (Winter 1987): 3-27.

19Ibid., 4.

20Ibid., 5.

21For an extensive discussion of organic metaphors in Schenker's work, see Robert Snarrenberg, "Competing Myths: The American Abandonment of Schenker's Organicism," in Anthony Pople, ed., Theory, Analysis, and Meaning in Music (Cambridge, UK: Cambridge University Press, 1994), 29-56.

22van den Toorn, op. cit. 90.

23The focus of van den Toorn's criticism is Treitler's essay, "History, Criticism, and Beethoven's Ninth Symphony," in Music and the Historical Imagination (Cambridge, Mass.: Harvard University Press, 1989), 19-45. Treitler employs a more emotionally suggestive vocabulary in his analysis because he is trying to account for the listener's psychological experience of the music. Though Treitler discusses the music in technical detail, his point is not to explain its formal self-sufficiency, but rather its engagement of the listener's spatial and temporal imagination.

24Charles Rosen, The Frontiers of Meaning (New York: Hill and Wang, 1994), 126: "Even more than literature and the visual arts, music can never be arrested by any system of analysis or interpretation, either formal or historical. It is natural to look outside or beyond the music, to find the ways in which it can temporarily and provisionally assume different kinds of significance. Nevertheless, music will not acknowledge a context greater than itselfsocial, cultural, or biographicalto which it is conveniently subservient."

25Herbert Brün, Professor Emeritus of Music at the University of Illinois, uses this phrase frequently in his lectures and seminars.

26Jürgen Habermas, Knowledge and Human Interests, trans. Jeremy Shapiro (Boston: Beacon Press, 1971).

27Ludwig Wittgenstein, Tractatus logico-philosophicus, trans. D.F. Pears and B.F. McGuinness (London: Routledge and Kegan Paul, 1961), 151.

28Paul Ricoeur, "The Metaphorical Process as Cognition, Imagination, and Feeling." On Metaphor, ed. Sheldon Sacks (Chicago and London: University of Chicago Press, 1978), 144.

29Philip Brett, Elizabeth Wood, and Gary Thomas, eds. Queering the Pitch: The New Gay and Lesbian Musicology (New York and London: Routledge, 1994), 225.

30Ibid., 209.

31Susan McClary, "A Musical Dialectic from the Enlightenment: Mozart's Piano Concerto in G Major, K.453, Movement 2," Cultural Critique, No. 4 (Fall 1986): 158.

32Op. cit., 133.

33Lakoff and Johnson, op. cit., 193.

34Marion A. Guck, "Musical Images as Musical Thoughts: The Contribution of Metaphor to Analysis," In Theory Only 5 (1981): 41.

35Donald Davidson, "What Metaphors Mean," On Metaphor, 40.


37Wayne C. Booth, "Metaphor as Rhetoric: The Problem of Evaluation," On Metaphor, 52.

38Lakoff and Johnson, op. cit., 144.

39Op. cit., 151.

40Op. cit., 30.

3591 Last modified on October 17, 2018