The modern era in Western music began in 1911 with the publication of Arnold Schoenberg’s Theory of Harmony, which declared that “tonality is no natural law of music, eternally valid.” Schoenberg’s announcement came as a shock to the vast majority of musicians who regarded tonality as the indispensable condition of music. Their distress was understandable: tonality had been the basis of Western music for more than three centuries, and had made possible a body of work of unparalleled depth and expressive power. Schoenberg’s announcement was less surprising, however, to those who knew that tonality had in fact been gradually breaking down for fifty years. This process, initiated largely by Richard Wagner, had by 1911 reached a critical stage in which the very future of tonality appeared doubtful. It was Schoenberg who drew the radical but seemingly inevitable conclusion that tonality itself was no longer valid, and that an entirely new musical system was needed to replace it.
Tonality may be defined as a musical idiom in which all pitches are organized in relation to a given central pitch, or “tonal center.” Music which lacks such a center is commonly called “atonal.” Schoenberg’s greatest contribution to Western music was to provide a rigorous theoretical basis for atonality by inventing the so-called “twelve-tone” method.1
The importance of Schoenberg’s achievement is hard to exaggerate. The contemporary composer and theorist George Perle, for example, claims that the development of atonality by Schoenberg and his students “probably represents the most far-reaching and thoroughgoing revolution the history of music has known since the beginnings of polyphony.” Schoenberg himself clearly recognized the significance of his method, of which he said: “I have made a discovery that will assure the supremacy of German music for the next hundred years.”
Yet while composers of all nationalities have been influenced by Schoenberg’s method, and the basic principles of atonal composition he introduced are still being employed today, the method has met with hardly any appreciation or acceptance from the public at large. Concert audiences continue to prefer the music of the past, and seem uninterested in even a limited exposure to atonal music. As a consequence, only a handful of atonal works are performed by our major musical organizations, which rely largely on the repertoire of earlier centuries to keep their audiences satisfied. Active appreciation of atonal music has thus been confined to a relatively few conductors and performers—and, of course, its composers.
Within the past few years, however, many composers have seemingly undergone a change of heart, and have begun to incorporate elements of traditional tonality in their works. The new trend was certified by the inauguration in 1983 of the New York Philharmonic’s “Horizons” summer festival, whose subtitle, “A New Romanticism?,” was meant to drawn attention to a return to tonality on the part of several of the composers represented. Whatever the merits of the music itself, the very fact that the festival took place testifies to the growing number of composers dissatisfied with the austerities of atonality and searching for a more accessible musical style. As for the decidedly unenthusiastic critical response to the festival, this may have been an indication of just how confused much of that searching remains.
Indeed, serious music today is unsure of itself as it has not been since the early years of the 20th century. Today, moreover, the central issue is the same as it was then: what is the meaning of tonality? Has tonality been exhausted beyond recall, or does it remain a valid means of musical expression?
Properly to understand the meaning of tonality we must first consider a more fundamental question: what are the sources of musical expression?
Since Plato and Aristotle, Western thinkers have recognized that music is unique among the arts in its ability to express human character and emotion. We can begin to understand why this should be so by considering one of the simplest of all musical phenomena: the octave.
The interval of the octave is produced by vibrating strings whose lengths are in the ratio 2:1, a discovery usually attributed to Pythagoras. The mathematical ratio, however, tells us nothing about how the octave sounds. In fact, the sound of an octave is not easy to describe: about the best one can do is to say that the notes of an octave sound the same even though they are different. The sound is so familiar that we seldom pause to reflect on how strange it is that the octave exists at all. For if every musical pitch has a different vibrational frequency, as modern physics tells us, we should expect different pitches to sound completely different. Yet we know from experience that this is not so, and the mathematical conception of pitch must therefore be incomplete.
This rather simple line of reasoning leads us to an important conclusion: music cannot be reduced to mathematics. That is not to deny that music has a mathematical aspect, one which has rightly fascinated musicians and philosophers since ancient times. But it should serve to remind us that the ear hears things that mathematics alone cannot explain.
Why, then, is the octave so important? Because it is the fundamental “unit” in terms of which our sense of hearing measures the quality—the degree of tension or dissonance—of all other intervals. The octave is the least dissonant interval because it exhibits the least possible degree of tension between nonidentical notes (this is in fact what it means to say that the notes of the octave sound the same). Indeed, it is with reference to this state of minimal tension that our sense of hearing implicitly judges the qualities of all other intervals. The octave, by providing the standard against which we perceive qualitative differences in musical sound, may thus be said to be the natural measure of consonance and dissonance.
The concept of a “natural measure” becomes clearer when we recognize that it is practically unique to music. Time, for example, has no natural measure: minutes or millennia are equally good units, and the choice between them is solely a matter of convenience. The same is true of distance. Another way of saying this is that time and distance are pure continua, without any intrinsic structure. A given interval of time or distance may be longer or shorter than another, but qualitatively the two are the same. Musical pitch, by contrast, is a structured continuum in which the existence of a preeminent interval (the octave) makes qualitative differences possible.
If the octave is the least dissonant musical interval, an obvious question arises: can the other musical intervals be ranked in order of increasing dissonance?
Some insight into this question is provided by the phenomenon of the overtone series. A note played on any musical instrument actually contains both a dominant pitch (the “fundamental”) and a series of higher pitches or “overtones” whose frequencies are simple integer multiples of the fundamental frequency. The frequency of the first overtone is twice that of the fundamental; the frequency of the second overtone, three times; and so on. Although these overtones are generally much weaker than the fundamental, their presence in varying proportions is what distinguishes the tone qualities of different musical instruments. The intervals between consecutive notes in the overtone series are defined by the simple integer frequency ratios 2:1, 3:2, 4:3, and so on. The first four intervals in this series are: the octave; the perfect fifth; the perfect fourth; and the major third.
It is tempting to assert that the overtone series is just the hierarchy of dissonance we are looking for. The perfect fifth would then be the least dissonant interval after the octave, followed by the perfect fourth, and so on. We must be careful, however, not to confuse mathematical entities (integer frequency ratios) with qualitative sensations (degrees of dissonance). Whether or not there is a correlation between the two can only be decided on the basis of experience and musical practice.
Still, a survey of world musical traditions suggests that the correlation is a rather strong one. First of all, the octave is universally recognized as the most basic interval, as is indicated by the fact that the most fundamental of all music structures—the scale—is founded upon it. The perfect fifth and the perfect fourth clearly rank next. As the Harvard Dictionary of Music puts it, “. . . from the point of view of musical composition of all eras, these two intervals must be regarded as consonances second only to the unison and the octave.”
The importance of the fifth and the fourth is demonstrated, for example, by the ubiquity of the tonal pentatonic (five-note) scale, which occurs in virtually every musical culture (this scale corresponds to the black keys on the piano). As the ancient Chinese recognized, the notes of this scale are obtained by ascending in fifths (or descending in fourths) from a given pitch.
Secondly, the music of many cultures (such as ancient Greek, Arabic, and Persian) is based on tetrachords, or four-note divisions of the perfect fourth. The function of the perfect fourth in tetrachordal music is to serve as a consonant frame of reference for the complex melody woven around it.
Thirdly, the most common form of primitive polyphony is “parallel organum,” in which the voices move in parallel fifths or fourths. In the more complex types of organum developed in medieval Europe, the octave, fifth, and fourth remained the only acceptable intervals for use at the beginning of a musical piece and in cadences.
Perhaps the clearest illustration of the status of the fifth and the fourth is the common use of these intervals as “drone” accompaniments to a complex melody. This type of drone is familiar to Western listeners from bagpipe music; it also figures prominently in the music of India, where the melody or raga is accompanied by a drone of a perfect fifth (less commonly a fourth). In general, a drone must be a relatively consonant interval to assure that it will not interfere with the melody. The common use of fifths and fourths as drones shows that these intervals occupy an intermediate position between the consonant octave (or unison) and the more dissonant intervals which occur higher in the overtone series.
Why is the hierarchical nature of consonance and dissonance so significant? Perhaps most importantly, it makes possible the phenomenon of “tonal motion,” one of the central expressive characteristics of music. Even the simplest melody conveys the sense that the notes are in motion, that the music is “going somewhere.” Such motion is not merely change of pitch, but rather the variation of musical tension which arises from the intrinsic dissonances of the successive intervals. Tonal motion is therefore directed: it is always felt as motion toward or away from some state of tension or relaxation. Moreover, the phenomenon of tonal motion appears at all levels of musical practice.
Perhaps the simplest example is the musical scale, whose characteristic motion is a departure from the tonic note followed by a return to the same note an octave higher or lower. Melody exhibits more complex patterns of motion. Harmony provides even richer resources, and makes possible such powerful and familiar vehicles of tonal motion as the tonic/dominant polarity of Western music.
The concept of tonal motion leads us to the very source of musical expression: the resemblance of musical motion to human action. Like musical motion, human action is inherently directed: it proceeds through alternations of desire and satisfaction, striving and fulfillment, tension and release. Music and human action thus have the same forms and are characterized by the same polarities. This, I would suggest, is what Aristotle meant when he said (in Book VIII of the Politics) that music “imitates” human character. As he observes elsewhere, “there seems to be in us a certain affinity for musical modes and rhythms, which leads some philosophers to say that the soul is a tuning. . . .” This affinity accounts in turn for music’s extraordinary expressive power, and justifies the critic Walter Pater’s famous dictum that “all art constantly aspires toward the condition of music.”
Now let us return to Arnold Schoenberg and the significance of his renunciation of tonality. Probably the best introduction to this subject is Schoenberg’s 1941 lecture, “Composition with Twelve Tones.” Here Schoenberg discusses the basic principle of atonality, which he called the “emancipation of the dissonance”:
What distinguishes dissonances from consonances is not a greater or lesser degree of beauty, but a greater or lesser degree of comprehensibility. In my Theory of Harmony I presented the theory that dissonant tones appear later among the overtones, for which reason the ear is less intimately acquainted with them. This phenomenon does not justify such sharply contradictory terms as concord and discord. Closer acquaintance with the more remote consonances—the dissonances, that is—gradually eliminated the difficulty of comprehension. . . .
The term emancipation of the dissonance refers to its comprehensibility, which is considered equivalent to the comprehensibility of the consonance. A style based on this premise treats dissonances like consonances and renounces a tonal center. . . .
Schoenberg’s first atonal pieces, composed after 1908, were written in a style which is commonly called “free” atonality. The best-known of these works are the monodrama Erwartung (“Expectation,” 1909) and the cycle Pierrot Lunaire (“Moonstruck Pierrot”) for reciter and chamber orchestra (1912).
By Schoenberg’s admission, free atonality proved most suitable for very short and extremely expressive works, and could not provide a substitute for the organizational principles of tonality. Schoenberg describes how he eventually found a satisfactory solution to this problem:
After many unsuccessful attempts during a period of approximately twelve years, I laid the foundation for a new procedure in musical construction which seemed fitted to replace those structural differentiations provided formerly by total harmonies.
I called this procedure “Method of Composing with Twelve Tones Which are Related only with One Another.”
This method consists primarily of the constant and exclusive use of a set of twelve different tones. This means, of course, that no tone is repeated within the set and that it uses all twelve tones of the chromatic scale, though in a different order. . . . [The basic set] functions in the manner of a motive. This explains why such a basic set must be invented anew for every piece. It has to be the first creative thought.
Beginning with the waltz from Five Piano Pieces (1923), Schoenberg used the twelve-tone method for a body of work encompassing nearly every major musical genre, including the Variations for Orchestra (1928), the opera Moses and Aaron (1930-32), concertos for violin (1936) and piano (1942), and numerous chamber works. The twelve-tone method was further applied by Schoenberg’s students Anton Webern and Alban Berg, and the three are often referred to collectively as the “second Viennese school.”
The greatness of Schoenberg’s achievement is beyond question, and is now properly appreciated by most knowledgeable musicians. The meaning of that achievement, however, is still far from clear. Musicians have generally been reluctant critically to examine the conceptual foundations of atonality, perhaps for fear of lending support to its vocal and often irrational opponents. Yet Schoenberg’s method in fact demands such an investigation, if only because it offers probably the first example in history of “theoretical” music: instead of rationalizing an existing practice which had developed incrementally over time, the theory precedes the practice and determines it.
Consider, then, the basic tenet of the twelve-tone method: “emancipation of the dissonance.” Schoenberg’s explanation of this principle, quoted above, begins by acknowledging the traditional view that the “dissonant” tones or intervals are those which occur later in the overtone series. Concord and discord are therefore not contradictory terms, but differ only in degree. At this point, however, Schoenberg departs radically from the traditional view, claiming that the distinction between consonance and dissonance is due solely to lack of exposure, and disappears upon “closer acquaintance” with the more remote intervals. From this he concludes that it is appropriate to ignore the distinction entirely by “treat[ing] dissonances like consonances and renounc[ing] a tonal center.”
Schoenberg’s explanation clearly reveals his root assumption that the distinction between consonance and dissonance is conventional and therefore arbitrary. But this view is incorrect. The octave, we have seen, is the standard by which we measure the qualities of musical intervals, their relative consonance or dissonance. If the octave is natural, so too are consonance and dissonance. To attempt to overcome this distinction is rather like trying to stop seeing the world in color.
It is striking that Schoenberg provides no evidence whatsoever for his view that consonance and dissonance are interchangeable, but merely asserts it as an axiom of his method. And it is the more striking that Schoenberg’s practice in this matter is not consistent with his principles. The twelve-tone method includes the principle of octave equivalence, whereby tones an octave apart are regarded as the same note. Schoenberg’s method thus implicitly acknowledges—and indeed draws its coherence from—the special natural status of the octave. Schoenberg, however, apparently failed to see the necessary consequences of this status, and his decision to treat all other intervals interchangeably introduced a deep inconsistency into his method.
The fact that there are conceptual difficulties inherent in the twelve-tone method does not mean that it has no musical value. For one thing, the method has produced a body of music of extraordinary density and structural complexity which has proved eminently satisfying to many composers and performers. More importantly, atonality has opened up new possibilities for musical expression by extending the range of human actions that music can imitate. Schoenberg himself regarded this as his major achievement. Speaking of his work in 1949, he said: “Forty years have since proved that the psychological basis of all these changes was correct. Music without a constant reference to a tonic was comprehensible, could produce characters and moods, could provoke emotions, and was not devoid of gaiety or humor.”
Schoenberg did not specify the particular characters, moods, and emotions he had in mind. Like all musical idioms, however, atonality is better suited to express some characters than others. The expressive qualities of a musical style are notoriously difficult to describe, but one useful indication is the nature of the poems or other literary texts a composer chooses to set. An examination of the texts chosen by Schoenberg and his students suggests that atonality is particularly well suited to imitating actions of an extreme or abnormal character.
Erwartung, for example, is about a woman who wanders through a forest at night in search of her lover and comes upon his murdered body; Schoenberg’s musical setting was aptly described by the philosopher Theodor Adorno as a “seismic registration of traumatic shocks.” The macabre and often horrifying character of Pierrot Lunaire is sufficiently indicated by the titles of such poems of the cycle as “Grave Robbery,” “Gallows Song,” and “Beheading.” Alban Berg’s early opera Wozzeck (1925) is the tale of a persecuted soldier who murders his unfaithful mistress and then drowns himself. Berg’s twelve-tone opera Lulu (1937) tells the story of a beautiful woman’s “progress” from an object of desire to a convicted murderess to a prostitute who meets her death at the hands of Jack the Ripper.
Of course, not all the texts set by Schoenberg and his students are as dark as these, but the prominence of such texts in atonal music is nevertheless remarkable. It is fair to say, indeed, that subjects like those of Erwartung or Lulu would never have been chosen by composers before the 20th century. Their liberal appropriation by atonal composers strongly suggests that atonality is most comfortably at home in a spiritual territory where extremes—not to say depths—of human action and feeling are explored.
Is atonal music also capable of successfully expressing the wider range of normal human action and emotion? The experience of nearly eighty years suggests that the answer is no. In spite of Schoenberg’s claim that atonal music is “not devoid of gaiety or humor,” a convincing atonal comic opera has yet to be written, and subjects of a noble or tragic character have not fared much better; most atonal composers, like the rest of us, find it difficult to imagine an atonal Marriage of Figaro or Romeo and Juliet.
In spite of the obvious expressive limitations of atonality, Schoenberg insisted throughout his life that atonal music would eventually be accepted by concert audiences. Needless to say, Schoenberg’s hope is unlikely to be fulfilled. The reason is simple: atonality is not very well suited to expressing things traditionally considered “beautiful.” In fact, the word beauty almost never occurs in discussions of atonal music, even among its most avid partisans. This is not really surprising, for Schoenberg himself spoke of being “cured of the delusion that the artist’s aim is to create beauty.” One may, of course, assert that being cured of such delusions is just what audiences deserve; this was the position of Theodor Adorno, who wrote that “the dissonances which horrify them testify to their own condition.” Such an attitude, however, does scant justice to the traditions of Western music and the experience of those who know it well.
Adorno did speak truly when he characterized atonality as a “musical domination of nature.” What he meant by this was that atonality replaces the traditional, “natural” structures of music with a set of quantitative laws that allows musical sound to be controlled and manipulated mathematically. This new orientation is evident in the twelve-tone method itself, which regards the tones and intervals of the basic set as functionally equivalent elements differing from each other only quantitatively. But the truth of Adorno’s remark has been demonstrated most clearly by the development of musical “serialism” in the decades following World War II.
Serialism generalizes the principles of the twelve-tone method by taking pitch as only one of the several musical “parameters,” including duration, instrumentation, dynamics, and so forth. Each parameter can assume any one of a “set” of discrete values, and musical composition is then regarded as the progressive transformation and permutation of these sets by means of mathematical operations. Given this complete mathematization of musical practice, it is not surprising that serialism has effectively renounced music’s traditional goal of representing human action and emotion.
Adorno’s “musical domination of nature” is reminiscent of Descartes’s famous statement that the purpose of modern science is the “mastery and possession of nature.” In fact, the advent of serialism means nothing less than the triumph of the Cartesian perspective in music. Like Cartesian science, serialism replaces traditional conceptions based on experience and common sense with mathematical abstractions. Each relies on the rigorous application of a quantitative method to assure precise control of its “material.” Most importantly, neither modern science nor musical serialism has much use for human subjectivity.
It has often been remarked that modern science effectively banishes man from nature by denying objective status to his most fundamental concerns and desires. With remarkable consistency, the same can be said of much modern music. Until recently, music afforded one of the few secure refuges from the expanding dominion of scientific abstraction, but it hardly serves this purpose any longer. The price we have paid for making music “scientific” is the demise of musical expression in our time.
If atonality and serialism do not provide adequate substitutes for tonality, does it necessarily follow that tonality remains a valid means of expression for the creative composer? Perhaps Schoenberg and his students were correct on the main point and tonality did die in 1911, never to rise again. Eighty years later, however, we may be forgiven for viewing this contention with some skepticism.
It is surely undeniable that in 1911 tonality had been stretched to the breaking point in the works of Schoenberg, Berg, Richard Strauss, Gustav Mahler, and Claude Debussy, leading Schoenberg to conclude that the demise of tonality was a historical necessity. Note, however, that all but one of the composers just named are German or Austrian. Schoenberg’s assumption is obvious: Austro-German music is supreme, and therefore its fate will define “historical necessity.” It was thus hardly accidental that Schoenberg said, “I have made a discovery that will assure the supremacy of German music for the next hundred years.”
The presumption of Schoenberg’s attitude is extraordinary, and it is surprising that so few commentators have taken issue with it. To refute it definitively one need only list the large number of composers who continued to write tonal music after Schoenberg’s announcement of its demise. A partial enumeration would include Paul Hindemith and Carl Orff in Germany itself, Maurice Ravel and Francis Poulenc in France, Zoltán Kodály in Hungary, Bohuslav Martinu in Czechoslovakia, Sergei Prokofiev and Dmitri Shostakovich in Russia, Ralph Vaughan Williams and Benjamin Britten in England, Manuel de Falla in Spain, Samuel Barber and Roy Harris in the United States. Moreover, the work of Béla Bartók and Igor Stravinsky managed to be highly innovative while remaining within a recognizably tonal context (Stravinsky’s later atonal pieces excepted). As this list shows, tonality still provides a rich fund of resources for original musical expression.
Nevertheless, it would be rash to conclude from all this that Western tonality is the “natural” language of music, as has sometimes been claimed. Western tonality is but one musical system among many, and its unqualified superiority to all others is by no means obvious. What I have tried to argue here is the broader point that tonality itself, as a system which acknowledges hierarchy in the realm of musical sound, is natural. In this sense, the music of all cultures and historical periods is tonal, with one salient exception: our own today. In its explicit denial of nature, this exception merely proves the rule that tonality is the natural language of music.
If tonality is natural, are some tonalities more natural than others? The question is dangerous, because no musical system is natural without qualification. Each was developed by a particular people at a particular time, and for a different purpose. In short, all musical systems are conventional. Nature provides the elements that make musical expression possible; convention shapes these elements into expresssive musical works. The paradoxical relationship between nature and convention in all human affairs is captured by Aristotle’s description of the city-state in the Politics: it is natural, yet “he who invented it was the greatest of benefactors.” The same is true of music.
Aristotle’s remark also implies, however, that we are free to choose our conventions well or badly, with a greater or lesser degree of understanding. Western tonality owes its greatness to a set of conventions which allowed an unprecedented synthesis of expressive richness and formal control, of what Nietzsche called “enthusiasm” and “self-possession.” The tragedy of Schoenberg’s attempt to reestablish this synthesis was that the conventions he chose violated the constraints which nature had provided.
It remains to be seen whether the future of Western music can equal the achievements of its past. The essence of the musical art, however, will remain what it has always been: to discover those conventions which most fully realize the natural expressive possibilities of tone.
1 Although Schoenberg disapproved of the term “atonal” and preferred to call his music “pantonal,” I follow what has come to be the standard usage.