In der modernen Prosa sprechen wir eine Sprache, die wir mit dem Gefühle nicht verstehen.… Wir können nach unserer innersten Empfindung in dieser Sprache gewissermassen nicht mitsprechen, denn es ist uns unmöglich, nach dieser Empfindung in ihr zu erfinden; wir können unsere Empfindungen in ihr nur dem Verstande, nicht aber dem zuversichtlich verstehenden Gefühle mittheilen.… In der modernen Sprache kann nicht gedichtet werden, d.h.… eine dichterische Absicht kann in ihr verwirklicht, sondern eben nur als solche ausgesprochen werden.

(In modern prose we speak a language we do not understand with the feeling.… we cannot discourse in this language according to our innermost emotion, for it is impossible to invent in it according to that emotion; in it, we can only impart our emotions to the understanding, but not to the implicitly understood feeling.… In modern speech no poesis is possible, that is to say, poetic aim cannot be realized therein, but only spoken out as such.)

Seitdem nun die modernen europäischen Sprachen … mit immer ersichtlicherer Tendenz ihrer rein Konventionellen Ausbildung folgten, entwickelte sich andererseits die Musik zu einem bisher der Welt unbekannten Vermögen des Ausdruckes. Es ist, als ob das durch die Kompression seitens der konventionellen Civilization gesteigerte rein menschliche Gefühl sich einen Ausweg zur Geltendmachung seiner ihm eigenthümlichen Sprachgesetze gesucht natte, durch welche es, frei vom Zwange der logischen Denkgesetze, sich selbst verständlich sich ausdrüken könnte.

(Now, ever since the modern European languages … have followed this conventional drift to a more and more obvious tendency, music, on the other hand, has been developing a power of expression unknown to the world before. It is as though the purely human feeling, intensified by the pressure of a conventional civilization, has been seeking an outlet for the operation of its own peculiar laws of speech; an outlet through which, unfettered by the laws of logical thought, it might express itself intelligibly to itself.)

In reading recent literature on the history and aesthetics of Western music, one consistently encounters references to the "language" of this music, especially with regard to the practice of eighteenth- and nineteenth-century tonality. Although the word "language" is used metaphorically in such cases, the metaphor seems remarkably apt (and convenient), and this no doubt accounts for its persistence. When applied to twentieth-century music, however, the sense of the term—and thus the nature of the metaphor—requires significant adjustment. For here, un-like the case in earlier Western music, one is unable to find that most characteristic feature of all natural languages, the universal acceptance of an enduring set of formal conventions evident throughout a given linguistic domain. Attempts, as in Donald Mitchell's The Language of Modern Music, to define a twentieth-century musical mainstream (in Mitchell's case, Schoenbergian dodecaphony), elevating its technical and systematic foundation to the status of a uniquely "proper" language for the age, appear seriously misguided and in flagrant opposition to the actual course of twentieth-century musical developments. Musical modernism is marked, above all, by its "linguistic plurality" and by the failure of any one language to assume a dominant position.

This plurality and the significant transformations in musical structure, expression, and intent it reflects form interesting parallels with characteristic features of the modernist movement in general; and it is primarily these connections that I wish to explore here. That such parallels exist is hardly surprising, since music—or perhaps more accurately the idea of music—is intimately tied to certain basic conceptions underlying the modernist revolution. Indeed, musical developments of the critical years around the turn of the century mirror with particular clarity the general intellectual and artistic climate of the period as a whole.

Although considerable controversy persists concerning both the nature and chronology of modernism, there seems to be widespread agreement that it incorporated a wish to turn away from concrete, everyday reality, to break out of the routine of ordinary actions in the hope of attaining a more personal and idealized vision of reality. There were of course precedents for this attitude in romanticism, but its artistic manifestations began to take on uniquely modern colorations toward the end of the century. In particular, there is a prevalent move away from realism and naturalism toward a new and radical abstractionism, evident not only in a turn toward less representational modes in the visual arts but in new attitudes toward language in literature and, as we shall see, in music (by metaphorical extension) as well.

It is frequently noted that a "crisis in language" accompanied the profound changes in human consciousness everywhere evident near the turn of the century. As the nature of reality itself became problematic—or at least suspect, distrusted for its imposition of limits upon individual imagination—so necessarily did the relationship of language to reality. Thus the later nineteenth century increasingly questioned the adequacy of an essentially standardized form of "classical" writing—writing that, even though often in "elevated" form, bore a close connection to ordinary discourse—as an effective vehicle for artistic expression. Indeed, it was precisely the mutually shared, conventional aspects of language that came to be most deeply distrusted for their failure to mirror the more subjective, obscure, and improbable manifestations of a transcendent reality—or rather realities, the plural reflecting an insistence upon the optional and provisional nature of human experience. Language in its normal manifestations, with its conventionalized vocabulary and standardized rules for syntactical combination, proved in-adequate for an artistic sensibility insisting upon, in Nietzsche's words, "a world of abnormally drawn perspectives."

This dissatisfaction with "normal" language received its classic statement through Hofmannsthal's Lord Chandos. Writing in 1902, Hofmannsthal conveys through the figure of the aristocratic Chandos the loss of an encompassing framework within which the various objects of external reality are connected with one another and integrated with the internal reality of human feelings. Chandos's world has become one of disparate, disconnected fragments, resistant to the abstractions of ordinary language. It is a world characterized by "a sort of feverish thought, but thought in a material that is more immediate, more fluid and more intense than that of language." Chandos longs for a new language in which "not a single word is known to me, a language in which mute objects speak to me and in which perhaps one day, in the grave, I will give account of myself before an unknown judge." The content and forms of art thus shift away from exterior reality, which no longer provides a stable, "given" material, toward language itself—to "pure" language in a sense closely related to the symbolists' "pure" poetry. "No artist tolerates reality," Nietzsche proclaimed; and, according to Flaubert, he should write "a book about nothing, a book without external attachments, which would hold itself together by itself through the internal force of its style."

It is more than coincidental, I think, that both Nietzsche and Hofmannsthal were intensely musical and intimately involved with music. For both, music provided a sort of idealized model for the reformulation of art and language. Indeed, music acquired the status of a central symbolic image for many of the principal artistic concerns of the years immediately preceding and following the turn of the century. Walter Pater provided perhaps the strongest statement (certainly the most famous) in asserting (in 1873) that "all art constantly aspires towards the condition of music." Removed from ordinary reality by its nonsubstantive and nondesignative nature, music offered the age an ideal embodiment of the notion that art is pure form, and thus pure language. Pater's is only one of a series of such pronouncements appearing in the aesthetic literature of the period, e.g., in Verlaine's "De la musique avant toute chose," or in Valéry's "Reprendre à la musique leur bien." Music, with its apparent indifference to external reality, comes to be viewed as the purest manifestation of human thought—as a "language" capable of producing the sort of "immediacy, fluidity and intensity" that Hofmannsthal found missing in ordinary words.

The tendency to propose music as a model for artistic intentions and aspirations is equally evident among painters. Delacroix, for example, stressed "the music of a picture"; and Gauguin, when questioned concerning the meaning of one of his paintings ("Where are we going…"), said that it should be understood as "music without a libretto." But perhaps the most fully developed argument for a musical "basis" for painting appeared in Kandinsky's writings from the early years of the present century, in which he calls for the creation of a "pure painting" independent of external reality. Kandinsky repeatedly evokes music as an ideal for a more abstract, "object-free" art: "After music, painting will be the second of the arts … [it] will attain to the higher level of pure art, upon which music has already stood for several centuries." Similarly: "Music, which externally is completely emancipated from nature, does not need to borrow external forms from anywhere in order to create its language. Painting today is still almost entirely dependent upon natural forms, upon forms borrowed from nature. And its task today is to examine its forces and its materials, to become acquainted with them, as music has long since done, and to attempt to use these materials and forces in a purely painterly way for the purpose of creation." Indeed, Kandinsky goes so far as to envision the eventual development of a malerische Generalbass and a Harmonielehre der Malerei (i.e., theories of figured bass and of harmony for painting).

The idea of music as a uniquely privileged medium able to penetrate to the essence of reality and thus express things inaccessible to language as such has a history extending back at least to the turn of the nineteenth century. Its definitive philosophical statement was supplied by Schopenhauer, in whose formulation it became a cornerstone of the aesthetics of romanticism. Writing in 1819, in The World as Will and Idea, he praises music above all other arts as a "universal language" capable of expressing, "in a homogeneous material, mere tone, and with the greatest determinateness and truth, the inner nature, the in-itself of the world." Unlike the other arts, it is not a "copy of the phenomenon, or more accurately, the adequate objectivity of will, but is the direct copy of the will itself, and therefore exhibits itself as the metaphysical to everything physical in the world." The composer thus becomes in Schopenhauer's eyes a sort of clairvoyant, privy to truths hidden from ordinary beings: he "reveals the inner nature of the world, and expresses the deepest wisdom in a language which his reason does not understand; as a person under the influence of mesmerism tells things of which he has no conception when he awakes."

This view reflects, and presupposes, a uniquely modern and Western conception of music as an autonomous art, freed from the verbal texts to which it had traditionally been attached and upon which its meaning and significance had always depended. Schopenhauer is explicit on the point that only instrumental music enjoys the special powers he ascribes to the art: "It is precisely this universality, which belongs exclusively to it, together with the greatest determinateness, that give music the high worth which it has as the panacea for all our woes. Thus, if music is too closely united with the words, and tries to form itself according to the events, it is striving to speak a language which is not its own." It was not until the eighteenth century, however, that instrumental music gradually began to emerge as an equal partner to vocal music; and thus only then could such a "pure" music be taken seriously, and questions arise as to what this textless music might be "saying." Already by the early years of the nineteenth century the prevailing attitude toward instrumental music had completely changed. For many it had become the only true music, the only form in which music could attain its highest and purest expression. Wilhelm Wackenroder, writing at the end of the eighteenth century, praises music above all other arts, for "it speaks a language we do not know in ordinary life, which we have learned, we know not where or how, and which one can only take to be the language of angels."

Yet though many thus viewed the insubstantiality of musical material—its "purity"—as sufficient to justify its role as a model for artistic regeneration, for the composer matters were by no means so simple. Indeed, by the end of the century a crisis had developed in musical language as shattering as that in the language of literature. To the composer, the idea that music offered a "pure material" must have seemed grotesquely naive. Far from supplying a sort of tabula rasa on which could be inscribed, free from all external interference, the "hidden hieroglyphics" of uninhibited fantasy, music in fact came tied to a remarkably fixed system of built-in conventions and constraints. Not by chance, this system began to be theoretically codified at just the time that instrumental music began to break away from its vocal-linguistic heritage. It was as if music, suddenly removed from the semantic and syntactic foundation previously supplied by language, had to discover its own grammar. With Rameau's Traité de l'harmonie of 1722 as the most conspicuous initiator, the history of modern Western music theory represents a concerted effort to map out the coordinates of a new and autonomous musical system capable of matching the logical coherence and expressive power of language itself. If music was to be a world removed from ordinary reality (in Wackenroder's phrase, eine abgesonderte Welt), it was nevertheless to be a world of reason, logic, and systemization.

This increasingly systematic conception of musical structure was bound to take its toll. As the nineteenth century progressed, a growing number of composers felt that musical language was becoming frozen in the conventions of an overly standardized harmonic vocabulary and a formal framework too heavily bound to empty symmetrical regularities. By the middle of the century, Wagner was already acutely conscious of the delimiting nature of the inherited style. His inclination to dissolve tonality through chromatic saturation of the triadic sub-structure, producing almost constant harmonic ambiguity, is one well-known symptom of this concern, as is his dissatisfaction with what he had come to view as the meaningless periodicities of "quadratic compositional construction." Wagner wanted music to become "endless melody," free to develop continuously according to its own inner impulses rather than to the "outward forms" of an imposed convention.

Intensifying the growing discontent with a musical language that, flattened out under the weight of its own habits, seemed to be rapidly losing its former expressive power was the dramatic growth of "lighter" music during this period. The nineteenth century gave birth to a veritable industry for the production of music for instruction and household entertainment—not popular music, but so-called "salon" music pretending to a degree of technical complexity and emotional depth designed to satisfy the cultural amibitions of a growing middle class. Such music was turned out in increasing volume throughout the nineteenth century as part of the burgeoning publishing and printing business. Compositions were often offered in periodic series on a subscription basis, and many of the better-known composers of the day provided songs, piano pieces, etc., for such purposes on commission. The degree of banality and sentimentality in these pieces, suitable for unsophisticated yet "aspiring" music lovers, was necessarily high. Hanslick, writing in the 1860s, commented on the phenomenon: "By far the largest portion of the music published here [in Vienna] consists of little dances, practice pieces, and the basest kind of brilliant piano music, which makes no secret of its spiritual and technical poverty."

A sense of malaise thus developed in the musical world paralleling that found in the other arts of the period. For the composer committed to a similar quest for "spirituality," the inherited language of music seemed no "purer" than the languages of such "representational" artistic modes as painting and literature. It was equally burdened with a system of conventions that, trivialized through overuse and exploitation, had been rendered unresponsive to the more immediate and intuitive dimensions of human experience. Trapped under syntactical and formal constraints rooted in the past, the composer was as much the prisoner of an "external" reality as was the poet or painter. He might well have echoed Nietzsche's famous remark: "I fear we shall never be rid of God, so long as we still believe in grammar."

Debussy, writing in the early years of the new century, expresses the dilemma in a typically witty, yet revealing, manner in ridiculing the ossified formal prescriptions of the classical-romantic symphony (a genre generally held to be the highest manifestation of absolute music):

The first section is the customary presentation of a theme on which the composer proposes to work; then begins the necessary dismemberment; the second section seems to take place in an experimental laboratory; the third section cheers up a little in a quite childish way interspersed with deeply sentimental phrases during which the chant withdraws as is more seemly; but it reappears and the dismemberment goes on; the professional gentlemen, obviously interested, mop their brows and the audience calls for the composer. But the composer does not appear. He is engaged in listening modestly to the voice of tradition which prevents him, it seems to me, from hearing the voice that speaks within him.

The inner voice has become the important one for Debussy, as well as for many others of his generation. One can already recognize the condition in Wagner, who praises Liszt's symphonic poems, for example, precisely for "those individual peculiarities of view that made their creation possible." It is what is individual and unique, rather than general and conventional, that now matters.

Yet even Wagner, certainly among the most radical composers of the later nineteenth century, remained faithful to a latent foundation of traditional tonal and formal principles. The triad remains for him an always implicit, and usually explicit, structural norm, even when the underlying diatonic basis is obscured by his richly chromatic textures; and so does the dominant-to-tonic harmonic progression, the main key-defining agent in the classical canon. Moreover, the same is true of all of his contemporaries, and even of the earlier Debussy. Thus Ferruccio Busoni, writing in 1906, can look back over the entire nineteenth century (and specifically to late Beethoven, which he takes as representative of the extremes of musical freedom attained during the century) and comment (in his New Aesthetic of Music, perhaps the first conscious—or self-conscious—manifesto of musical modernism) on the ultimate failure of even its most progressive figures to achieve a radical break with the past:

Such lust of liberation filled Beethoven, the romantic revolutionary, that he ascended one short step on the way leading music back to its loftier self—a short step in the great task, a wide step in his own path. He did not quite reach absolute music, but in certain moments he divined it, as in the introduction to the fugue of the Sonata for Hammerklavier. Indeed, all composers have drawn nearest the true nature of music in preparatory and intermediate passages (preludes and transitions), where they felt at liberty to disregard symmetrical proportions, and unconsciously drew free breath.

Busoni's words again recall Nietzsche's aphorism about God and grammar: the apparent order and logical precision of standardized language is distrusted as bearing false witness to an increasingly unstable world of "degrees and many refinements of relationships." As if in response to this view, which I take to be fundamental to all the main currents of modernism (for "grammar" can be replaced by "conventional tonal structure," by "traditional modes of visual representation," etc.), the major progressive composers of the first decade of the new century undertook a radical dismantling of the established syntax of Western music. This move "beyond tonality" was remarkably widespread (although it assumed very different forms in different composers). It profoundly altered the face of music and supplied the technical foundation for musical modernism.

Although the technical consequences of this musical revolution are, I believe, ultimately comprehensible only within the context of the broader cultural crisis I have focused upon up to now, they are themselves of considerable interest and significance. I will thus turn now to consider some of the more specialized developments in musical language that occurred during the first decade of the century. It will be useful to treat these in rather general terms, for they are thus applicable to a wide range of composers (including Debussy, Scriabin, Stravinsky, Schoenberg, and Bartók) who in other respects might seem to have relatively little in common. Of course these technical developments did not come about instantaneously; they had an extended history. But the final step was taken only after the turn of the century; and in this instance this step produced a difference in kind rather than simply another one of degree.

One way to view the revolution in musical language during these years is as a transformation in the relationship between compositional foreground and compositional background—that is, between the musical surface and its formal substructure. (I have already made tacit use of this distinction in discussing the music of Wagner.) Music is above all an art of ornament and elaboration; and it must maintain a subtle, and often fragile, relationship between its variegated embellishments and the simpler, stricter, and more solid supporting framework that holds these embellishments together and supplies their foundation. Indeed, a striking feature of the foreground-background relation is the mutual dependence of the two. The under-lying framework is often not sounded at all, but must be deduced from the implications of the foreground; while the foreground, though actually sounded, owes its "grammatical" meaning solely to its connection with a "virtual" background. The history of Western music theory can be read as an attempt to codify a set of rules for, on the one hand, approved background relationships and, on the other, permissible foreground divergences. To take a few relatively simple examples applying mainly to "local" levels of structure, such theoretical concepts as consonance, diatonicism, triad, and fundamental progression belong to background phenomena, while those of dissonance, chromaticism, and auxiliary tones belong to the foreground.

Since at any given moment the background elements are not necessarily present on the surface, their proper apperception must depend upon strong conventions concerning what is "normal" and thus structural, as opposed to what is "abnormal" and thus superficial and ornamental. All Western music, at least since the Renaissance, displays a more or less complex interaction between foreground and background structures. Although the degree to which these levels can depart from one another has varied considerably from style to style, it is characteristic of the post-Renaissance period as a whole that a sufficient balance is maintained to ensure that the underlying structure is never seriously threatened. During the nineteenth century, however, this balance begins noticeably to waver. Since the background represents what is essentially fixed and unchanging, while the foreground contains what is unique, individual, and characteristic in a composition, it is not surprising that an age of such marked individualism should produce a radical shift in the foreground-background dialectic, tilting the balance heavily toward the surface. The growth of chromaticism, an emphasis on novel dissonances, an ever-greater exploitation of motivic and thematic elements at the expense of architectural ones—all this reflects a significant structural realignment. By the latter part of the century such technical innovations often make it extremely difficult to hear an implied background at all through the heavy accumulations of wayward foreground detail. The latter becomes so complex, so laden with multiple, entangled, and often contradictory layers of implication that the underlying structure (to the extent that one can still be inferred) is brought to the edge of collapse.

The more adventurous composers of the nineteenth century countered the problem largely by structuring the foreground features of their compositions at the expense of background ones. The various techniques of thematic transformation evident in Liszt, Wagner, and other composers of the period serve to hold together through surface correspondences extended spans of music whose background structures have been seriously weakened. Similarly, lengthy symphonic movements are often organized according to shifting and opposed key areas that, according to conventional background criteria, form dissonant relationships applicable only to local formal contexts (e.g., the C/B dichotomy in Strauss's Also Sprach Zarathustra).

Yet even in such extreme instances, traditional background structures continue to exert a strong influence. Despite the often exotic surface peculiarities, the music maintains at least a latent reference to the standardized grammar of Western tonality. The triad still represents the sole harmonic norm (no matter how rarely a pure triad may appear); and the traditional dominant-to-tonic progression still retains its key-defining function (though it may now appear more by implication than by actual statement).

Nevertheless, the growing strain brought on by the conflicting claims of foreground and background in complex European music reached a crisis point by the end of the century. If, on the one hand, the substructure became too obscure, the "meaning" of the foreground was apt to seem unclear; whereas if the substructure was rendered too clearly audible, the luxurious surface detail so typical of fin-de-siècle textures tended to sound like nothing more than "junk"—i.e., decoration in the worst sense of the term. One notes the latter problem, it seems to me, to some degree in even the greatest composers of the turn of the century. It is especially evident in such figures as Reger and Franck, who attempted to reconcile a classicizing tendency with a penchant for the most progressive technical procedures of the day. Thus Reger's complex modulations and intensely chromatic voice-leading are contained within a highly regular phrase structure with cadential points defined by blatantly unambiguous dominant-tonic progressions. The heightened chromatic motion on the surface seems to have no influence upon the substructure, which sounds through with schematic clarity. Both surface and background take on the aspect of cliché: the surface, because it acquires the attributes of a momentary decoration without wider repercussions; the background, because it provides a too "easy" (because too obviously conventional) resolution for the entangled interrelationships suggested by the surface. But it is not only in Reger that one hears the problem: the specter of kitsch looms over even the greatest achievements of an age in which music threatened literally to become pure ornament.

A solution demanded a major restructuring of the received musical language. In the broadest terms, it involved a projection of musical phenomena previously considered to belong solely to the foreground—elements that are ephemeral, passing, structurally unessential, and thus, in a sense, accidental (the "chance" results of voice leading, etc.)—onto the structural background. I have already noted a tendency in this direction in nineteenth-century music, in the increasing emphasis on individual foreground features. Nevertheless, the moment when agreed-upon background relationships no longer supplied even an implicit matrix for controlling the confusion of surface detail marked a fundamental turn in the history of compositional thought. A fixed and conventional conception of musical structure gave way to one that was variable, contingent, and contextual—dependent upon the specific attributes of the particular composition. Those qualities of uniqueness and individuality, of the ephemeral and accidental, that had previously marked the fore-ground alone now characterized the background as well.

The final, conclusive break occurred in the first decade of the century, in Scriabin, Debussy, Schoenberg, and Stravinsky, as well as others. Significantly, the leaders in this musical revolution were themselves all nurtured within the tonal tradition and produced in their earlier careers compositions written according to more or less traditional tonal assumptions. The rift with the past led to different responses, but the particular solution of each composer can be largely understood as a direct outgrowth of the stylistic evolution of his earlier music and thus of a particular orientation toward tonality.

It will be helpful to consider briefly two composers who arrived at radically different solutions to this common problem: Scriabin and Schoenberg. Scriabin's early music seems remarkable mainly for its conservatism, its unequivocal harmonic relationships ordered within an essentially conventional larger tonal context. Yet as in much music of the later nineteenth century (Scriabin was born in 1872 and his first mature works began to appear in the 1890s), the harmonic structure is covered with a dense network of auxiliary tones that, although clearly subordinate (there is never any doubt about the triadic background), resolve only with the greatest reluctance. The obvious disparity between the rich accretions of surface detail and the all-too-apparent harmonic underpinnings produces formal-expressive problems similar to the ones noted in Reger. Only in Scriabin the dissonances are prolonged over such long spans that the whole structure seems to float precariously over the delicately maintained chordal foundation; and in the later 1890s and early 1900s, the complex surface sonorities are increasingly emphasized at the expense of their background supports. When triadic resolutions do occur, they sound more and more like perfunctory nods to tradition, dictated solely by protocol. The heart of the music has been displaced from the substructure to the surface, so that the resolutions sound like a breach of faith.

Scriabin's own particular development of extended chromaticism and delayed resolution is closely tied to his use of elaborate dominant type sonorities. The dominant seventh is the one chordal type within the traditional vocabulary whose tonal function is, at least under normal circumstances, unambiguous, and which is thus able to define a key area entirely by itself. By focusing upon elaborations of such harmonies in his earlier music, Scriabin was able to preserve at least some degree of tonal definition; despite the increasing avoidance of resolution, one is usually able to infer what the resolution should be. Moreover, up until about 1907 tonal resolutions do ultimately occur, although they may be delayed right up to the final measure. The moment arrives in Scriabin's evolution, however, when the dominant-type sonorities completely lose their functional subordination to an inferred background tonic. The dominant, one might say, has moved deeper into the structural background to become an "absolute" sonority in its own right, with a meaning no longer dependent upon its relationship to a simpler, more stable structure. The dominant-type harmony, in fact, assumes the role of a center, or tonic, itself; but it is a new kind of unstable tonic, whose priority must be contextually defined within each composition.

Significantly, Scriabin referred to this new tonic sonority as the "mystic chord," for to him it was the source of previously unimagined musical power. Moreover, he conceived of it as built up of intervals of a fourth, thus distinguishing it from previous harmonic norms. Yet the chord can just as readily be viewed as a series of thirds, in which case it conforms to traditional conceptions of triadic extension. What was actually novel about the chord, then, was not so much its internal structure, or even the way it sounds in isolation, but its functional location in the background. There it shed its traditional grammatical meaning, acquiring a new and seemingly inscrutable one more in keeping with Scriabin's growing mystical orientation. Only through such drastic structural means could music become more responsive to those transcendent and visionary claims that increasingly occupied the composer from about 1908 to 1915, the final years of his brief life.

Schoenberg's development, though different in many ways, reveals significant parallels with Scriabin's. In his earlier works, too, the surface elaborations of a still basically tonal language are stressed to a point that eventually brings about the latter's dissolution. But Schoenberg's chromaticism is the product of rich webs of thematic and motivic development that bury the structural background under a complex, thickly woven contrapuntal overlay. Whereas in Scriabin the harmonic background moves slowly and projects its triadic nature with relative clarity, in Schoenberg's music of the early 1900s the density and speed of the counterpoint produce a constantly shifting harmonic basis that at every moment appears ready to dissolve the argument into complete tonal uncertainty. Dominant-type harmonies, though still present, are increasingly deemphasized as too suggestive of unwanted conventional resolutions. The stress is on highly varied dissonant complexes, which sound like opaque, heavily refracted distortions of the traditional harmonic functions that were fast becoming grammatical impossibilities, or at least embarrassments, to Schoenberg's ears. The final resolutions in the op. 8 orchestral songs, for example, or those of the Second String Quartet and Kammersymphonie, are still triadic; but they seem like reluctant tributes to a remote and distrusted authority.

Schoenberg's theoretical writings also reflect his new conception of foreground-background relationships. In a famous passage in his Harmonielehre, first published in 1911, he points to several momentary vertical structures cut out of compositions by Bach and Mozart, claiming to show that the sort of complex and highly differentiated dissonant harmonic structures found in his own work were already present in music of the eighteenth century. What for Bach and Mozart were passing "accidents," the result of surface contrapuntal elaborations firmly tied to an unmistakably inferrable triadic background, have become for Schoenberg absolute entities warranting theoretical investigation and explanation in their own right.

In Schoenberg's music, as in Scriabin's, the moment at which the latent background completely receded, leaving virtually no trace, is approached gradually, almost imperceptibly; but sometime around 1907-8 a final margin was irreparably traversed. Despite this step-by-step evolution, the consequences were fundamental. Schoenberg's own awareness of having made a critical turn is apparent in the preface to his song cycle Das Buck der hängenden Garten (generally considered to be the first major composition in the new style): "For the first time I have been successful in coming near an ideal of expression and form which I had had in mind for years.… Now that I have finally embarked upon this path I am conscious that I have broken all barriers of a past aesthetic." And later he remarked of the last two movements of his Second String Quartet, a work briefly predating the cycle: "No longer could the great variety of dissonant sonorities be balanced out through occasional insertion of such tonal chords as one normally uses to express a tonality"; it was no longer "appropriate to force the motion into the Procrustean bed of tonality."

Schoenberg thus sacrificed a traditional background in order to allow the compositional foreground to speak more freely, unencumbered by the constraints of a conventional syntax. Here, finally, was a music that could communicate directly, unmediated by external controls, and that was thus actually able to approach that "purity" of language so indiscriminately attributed to music in general by those working in the other arts. Yet the price to be paid was severe: Schoenberg's newly liberated foreground projected a "language" that no one, not even the composer himself, could understand, at least in the sense that one had always been able to "understand" traditional tonal music. As the composer himself remarked in his Harmonielehre, referring to the advanced harmonic constructions found in music of the century's first decade: "Why it is as it is, and why it is correct, I am at the moment unable to say."

There can be no coincidence, certainly, in the fact that Schoenberg's final break with traditional tonality initiated the most productive period of his creative life. Within a two-year span from 1907 to 1909 he completed seven major compositions, including such extended works as the Second String Quartet, op. 10, Das Buck der hängenden Garten, op. 15, the Five Orchestral Pieces, op. 16, and Erwartung, op. 17. The sense of a release, of a newly won freedom suddenly available beyond the "barriers of a past aesthetic," is evident in both the quantity and character of this music. Yet Schoenberg, working at the outer edges of what then seemed musically possible (at least to one committed to the notion of a continuously evolving tradition), increasingly felt the strain of operating at such disorienting heights, where only his unconscious, intuitive feeling for what was musically valid could serve him as guide. After the brief period of unprecedented productivity coinciding with the first explorations of the atonal terrain, Schoenberg's output decreased dramatically, coming to a virtual halt by 1916. For seven years thereafter no new compositions were published; and when new works began to appear again in 1923, they revealed a composer embarked upon a radically different course. In the intervening years Schoenberg had evolved a new musical system intended to replace tonality, one that—like tonality—would provide a method for consciously determining compositional choices.

This was the twelve-tone system, which Schoenberg envisioned as supplying the basis for a new "musical language," the lingua franca of a new stage in musical his-tory. For despite the revolutionary character of many aspects of his thought, Schoenberg remained committed to the idea that this next stage would share with past ones a dependence upon a set of widely accepted compositional conventions, within whose terms all composers could shape their own personal statements. As a consequence, he came to view his own earlier atonal works as representatives of an essentially transitional phase of music history. Writing in 1932 on the historical necessity of the twelve-tone system, he commented upon his atonal work:

The first compositions in this new style were written by me around 1908.… From the very beginnings such compositions differed from all preceding music, not only harmonically but also melodically, thematically, and motivically. But the foremost characteristic of these pieces in statu nascendi were their extreme expressiveness and their extraordinary brevity.… Thus, subconsciously, consequences were drawn from an innovation which, like every innovation, destroys while it produces. New colorful harmony was offered; but much was lost.… Fulfillment of all formal functions—comparable to the effect of punctuation in the construction of sentences, of subdivision into paragraphs, and of fusion into chapters—could scarcely be assured with chords whose constructive values had not as yet been explored. Hence it seemed at first impossible to compose pieces of complicated organization or of great length.… the conviction that these new sounds obey the laws of nature and of our manner of thinking—the conviction that order, logic, comprehensibility and form cannot be present without obedience to such laws—forces the com-poser along the road of exploration. He must find, if not laws or values, at least ways to justify the dissonant character of these harmonies and their successions.

Schoenberg's change of attitude was by no means exceptional. Following World War I, Western composers generally tended to pull back from the heady, more experimental atmosphere of the prewar years. Manifestations of a new point of view were everywhere evident: e.g., in the simpler, more objective and more "everyday" type of music fostered by Satie and Les Six, and in the various moves toward a "new classicism" by such otherwise diverse figures as Stravinsky, Bartok, and Hindemith. Yet what seems in retrospect most telling about all of these developments was their failure to produce a new set of musical procedures even remotely comparable—in terms of commonality, of reflecting a consensus—to those of traditional tonality. Thus the twelve-tone system, though indisputably one of the most remarkable and influential technical achievements of twentieth-century music, has remained an essentially provisional method, occasionally employed by many composers but consistently used by relatively few. Nor did the widespread neoclassical turn of the between-the-war years produce an even marginally unified technical orientation; rather, it gave rise to a series of strongly personal and thus divergent and idiosyncratic reformulations of technical and stylistic traits drawn from virtually the entire range of Western music history. Cocteau's famous "call to order," which reverberated throughout the postwar period, remained in this respect largely unanswered.

From the present perspective, then, it would appear that the most important historical moment in defining the main coordinates of twentieth-century music was the widespread break from traditional tonality that occurred during the first decade of the century. From this moment springs the unprecedented stylistic, technical, and expressive variety of the music of the modern age—in short, what I have previously referred to as its linguistic plurality. Despite the numerous attempts that have been—and continue to be—made to offer a systematic account of Schoenberg's prewar music, the true force and significance of this music lies, it seems to me, precisely in its determination to speak in an unknown and enigmatic tongue that largely defies rational comprehension.

This may help explain the unique position this music continues to occupy in our consciousness. Along with other composers of the time (one thinks also, inevitably, of the Stravinsky of the Rite of Spring), Schoenberg set the essential tone of music in the modern age. He attempted to transform musical language from a public vehicle, susceptible to comprehension by ordinary people (but thereby also limited to more or less ordinary statement), to a private one capable of speaking the unspeakable. Music became an incantation, a language of ritual that, just because of its inscrutability, revealed secrets hidden from normal understanding.

The fifteen songs of Das Buch der hängenden Gärten are settings of poems drawn from the volume of that title by Stefan George, who himself favored "a language inaccessible to the profane multitude." George's distinct elitism equally colors Schoenberg's aesthetic. (The composer once commented: "If it is art, it is not for all; and if it is for all, it is not art.") But Schoenberg's elitism can be understood in part as an understandable reaction against a musical language that had lost its fundamental expressive core and thus its capacity to challenge, to illuminate, and to astonish. The composers of the first decade of the century undertook to revive musical language by reinventing it. They tried to disengage musical sounds from their inherited attachments, to set them free from conventional associations in pursuit of what Schoenberg (along with Kandinsky) called the "spiritual." In sober retrospect, they may seem to have failed; yet theirs was a brave and exhilarating effort that fundamentally altered the nature of musical discourse.