In 1812, sixty-two years after the death of Johann Sebastian Bach, E.T.A. Hoffmann wrote:

There are moments—above all, when I have been reading in the works of the great Sebastian Bach—in which the mathematical relationships of music, the mystical rules of counterpoint, in fact, awaken in me an inner terror.

(“Kreisleriana,” in Fantasy Pieces in Callot’s Style)

This picturesque assertion is less extravagant, less romantic than it may appear. To most lovers of music who have an intimate experience of the works of Bach—most of all, those who play them—his music seems to reveal, with a direct physical impact, the most profound hidden aspects of the language of tonality. It is generally beside the point to try to distinguish Bach’s imagination from his knowledge or craft, his powers of expression from his manipulation of “the mystical rules of counterpoint,” but his craft was always acknowledged as formidable, even by those who did not appreciate his style. Several of his contemporaries testified to his ability, when given two different themes, to see at once all the different ways that they could be played together, with all the possible contrapuntal combinations. He knew immediately what could be done with a single theme: his son, Carl Phillip Emmanuel, showed him a fugue he had been writing and asked if any other variations were possible; after a moment’s glance, he replied firmly, “None.”

The expressive power of Bach’s music has several sources. One is his ability to sustain the tension of an ornamental line or arabesque at greater length than any other composer before Chopin. Even more important is his exploitation of the dissonance of chromatic harmony: no composer of his time came near him in daring and originality. Most important of all, perhaps, is his understanding of the emotional possibilities inherent in counterpoint—that is, in the way one voice can imitate another, in the dramatic effects that can be drawn from a melody inverted or played against itself, and also from the subtle way the harmonic implications of a motif can be radically altered when it is played with another motif of different character.

A new book by Laurence Dreyfus, Bach and the Patterns of Invention, takes up Bach’s creative process. It is the first study in some time to deal above all with the reasons that music lovers might want to listen to Bach or play him. Most of the recent studies of Bach have concerned themselves with matters of authenticity and chronology, with detailed questions of performance—how large was Bach’s orchestra and his chorus, how did he execute the trills, did he expect the performer to add to the decoration—with the religious symbolism of the music, or with the influence of rhetorical treatises on his aesthetic outlook. There have been studies of numerology in his music, most of them very dubious as are all such studies, but we may rejoice that these have now fallen out of fashion. A more recent development is sociological interpretation of Bach’s music. However interesting many of these studies are, none of them can explain the way his work became the foundation of all Western musical education from the early nineteenth century to the present.

There is an odd popular legend that Bach was an obscure composer during his lifetime (1685-1750), was forgotten for half a century after his death, and was suddenly rediscovered and exalted by the Romantic movement. There is only a grain or two of truth in this, just enough for the myth to survive. By itself a study of the reception of his music will give us very little hope of comprehending what actually took place: only an understanding of the effectiveness of his work for performer and listener can illuminate the extraordinary history of Bach’s music during his lifetime and for two centuries afterward.

“Invention” was a key word for Bach, as it was for many contemporary German musicians. What was most important for a young student to learn, Bach himself insisted, was how to have “good inventions.” The word is sometimes taken to mean the initial theme, motif, or melody, but Dreyfus shows that a much larger sense must be assumed. “Invention” in a fugue covers not merely the opening motif, but also all the interesting things that can be done with it: Can it be inverted (with all the intervals that originally went up now going down and vice versa), augmented (played twice as slow), diminished (twice as fast), and played in stretto (where a second voice plays the theme before the first voice has finished, so that the theme is performed simultaneously by two or more voices but out of phase)? Can it be played in stretto in its original and inverted forms together? Can the augmented form be played in stretto with the original one? “Good inventions” means, of course, that all of this must be done while producing a beautiful harmony, and without infringing the simple rules of counterpoint. “Invention” is what makes the fugue interesting. Dreyfus, following some eighteenth-century writers on music, distinguishes invention from decoration and disposition (or ordering): they make the work agreeable and acceptable—the fascination must come from the invention.


When the genre is a concerto grosso (which contrasts an orchestra with a small group of soloists), the “invention” concerns the opening orchestral section (called the ritornello) but is not confined to that: it also deals with the way the ritornello can be related to the solo episodes, how it can be segmented, transposed into different keys, and reshaped in the course of the movement.

Bach inherited the standard form of the concerto grosso from Vivaldi, whose work swept throughout Europe like wildfire in 1712. Vivaldi established the ritornello in its ideal form: an opening dramatic gesture, a few bars of spinning out a motif and a rhythm, and finally a short conclusion. Bach never improved on (although he often equaled) Vivaldi’s openings, which are various, generally arresting, and often very original. Stravinsky was wicked when he said all too famously that Vivaldi did not write five hundred concertos, but the same concerto five hundred times. It would be more accurate, if a little unfair, to say that Vivaldi wrote five hundred remarkable beginnings of a concerto grosso, none of which is realized as powerfully as their ritornelli suggest. Dreyfus studies the invention that Bach imaginatively deployed in the successive returns of the ritornello and their relation to the solo episodes. In addition, he examines the way Bach simulates concerto form with ritornello technique in solo pieces without orchestra, in which a solo harpsichord or even a solo violin or cello produces a standard ritornello and then mimics a relation between orchestral sections and solo episodes by dynamics and texture as well as by the reappearance of the opening “orchestral” section.

An “invention” is a project, an intention. Dreyfus’s sense of Bach’s intentions does not start with biographical speculation—although it does significantly arrive at biography in the last analysis. He reads the intention directly from the text of the music and from historical reflection on the genre of the work in question. Each genre imposes constraints: a minuet ought not to be written without elegance and it must be in two parts, each repeated; a fugue must display the composer’s contrapuntal skills.

Dreyfus is interested in Bach’s reaction to these constraints, his attempts not only to carry out the demands of the genre but also to escape from them, or even in many cases deliberately to oppose them. Above all, to understand how a project or an intention was carried out, we must have a sense of what will work: not all aspects of a project can be realized, not all the details of an invention will be satisfactory. We need to see the adjustments that the composer was forced to make in order to appreciate the final result, and we need to make the effort to deduce from the final text that part of the original invention which could not succeed, and which stimulated not only compromise but further and more inspired invention.1

A relatively simple example can be given of the inspiration provoked by the impossibility of realizing an original intention. Partita no. 4 in D major from the first volume of Keyboard Exercises was written before some of the other partitas, and Bach’s harpsichord at that time must have extended only two octaves above middle C: it did not yet have the top D that we find in other works. In the first part of the allemande from this partita, a beautiful A minor passage in the first part goes to the top C: Bach repeats this passage in the second part transposed now to B minor. Since the D that would have enabled an exact repetition was not yet available to him, he was forced to think up a new and even more intense version in order to use the passage at its new place.

My example is one of mechanical constraint. Dreyfus prefers more subtle ones, and concentrates on the limitations imposed by musical technique and genre. A simple case of this kind of constraint is what is called double invertible counterpoint, which sounds impressive but is actually extremely simple. In almost every fugue by Bach (and in most of his other pieces as well), the principal theme is eventually accompanied by another theme or motif played or sung in another voice. It was natural and inevitable for an eighteenth-century composer to ask himself, can I invert the position of the two themes, and play the one which is below on top? This does not always work, and the necessary adjustments can give rise to interesting variations. Bach certainly expected these variations to be perceived and admired by performers.


More complex problems may arise when the theme itself is inverted (that is, played upside down), a very common contrapuntal technique, and one that evidently had a particular appeal for Bach. It also produces an effect easy to appreciate, even by listeners unschooled in counterpoint: if a theme has any idiosyncratic character, its inversion is heard as a new and dramatically different form, but clearly related to the original. Sometimes, however, the inversion only produces an awkward melodic line, as in the A minor fugue from the first book of The Well-Tempered Keyboard. In fact, Donald Francis Tovey, one of the most appreciative analysts of Bach, thought the inversion in this fugue detestable. I think, however, that Bach must have enjoyed the awkwardness of this inversion for its characteristic force. At times Bach clearly cultivated a rough and even brutal turn of phrase.

A more admirable example is given by the two fugues with three themes in The Art of Fugue, the last of Bach’s immense encyclopedic projects at the end of his life; this one remained unfinished when he died. The second triple fugue inverts the themes of the first fugue. The inversion of one of the themes is extremely ungraceful, and Tovey pointed out that Bach was careful not to introduce it on its own, but always covers it with another theme in a higher voice until there is a full thick four-voice texture.2 This sophisticated treatment is a genuine “invention.”

Dreyfus remarks on an even more inspired “invention” in the Fugue in F sharp minor from the first book of The Well-Tempered Keyboard, one of the most striking pieces of the collection. The principal theme has a memorable chromatic contour. Bach writes a second theme, or countertheme, which already surfaces in another voice when the first theme has only sounded its three opening notes: the new theme sounds like a decorated free inversion of the first, falling and rising as the original rises and falls. Dreyfus observes that the original theme and its strict inversion cannot be played together for more than a few beats, and then the combination becomes unworkable. It was second nature for Bach to see whether a theme and its inversion would combine, and he was forced to reject the possibility for this fugue; but the attempt to make it work inspired the invention in which the free decorated form makes the initial subject doubly effective, more agitated, and even more expressive.


The advantage of Dreyfus’s approach to Bach is that it makes us listen to his work as he himself listened to the music of his contemporaries, and as they would have listened to his. It does not claim to read the composer’s mind, but it reconstructs some of the processes through which he had to go to compose in each case, and it does so by referring to aural experience, leaving questions of ideology and doctrine temporarily on the side. They eventually move back to the center, however.

For Dreyfus, we must reconsider Bach’s relation to his age, to the composers and critics who were his contemporaries. The traditional view of Bach—found in the famous nineteenth-century biography by Philip Spitter, and in Albert Schweitzer’s once-influential book on Bach—pictured him as a reactionary, out of step with his time, intent upon conserving the great Baroque contrapuntal tradition that was rapidly becoming out of date. This was, to some extent, what some contemporaries of the composer thought, although their opinion was nuanced: they praised his craft and admired his virtuosity both as performer and composer, but thought that much of his work had the swollen, turgid character of old-fashioned German Baroque rhetoric, poetry, and drama of the late seventeenth and very early eighteenth centuries. One critic objected to Bach’s writing out of the elaborate decoration that would normally have been left to the inspiration of the performer or singer, but Bach was not the first to do this: François Couperin had also filled in all the ornamental details of his suites for harpsichord, called Leçons. Most of the graces and trills of a Bach aria are fully indicated, while a Handel aria provides only the scaffold that must be finished by the singer.

However, Bach was working in the small, backwater town of Leipzig for a church with a number of local singers who made up a choir so insufficient in numbers that he complained (in vain) to the church authorities; Handel wrote in London for singers of international reputation, famous for their ability to improvise ornamentation, and who would have resented any elaborate prescriptions from the composer. Furthermore, the detailed ornaments of Bach’s published keyboard works are at least partially explained by their function: the published texts were without exception educational models that told the performer and composer (there was not much difference then) how to play the entire work, including ornaments. It is true that Bach must have relished an expressive, full ornamentation and made it an essential part of the original conception of many pieces; and his decoration generally had a richness and complexity that went against the growing fashion for a simpler, leaner line. By the end of his life, indeed, there had developed a preference for brilliant, less expressive decorative detail, and in this matter he remained attached to an older tradition. He even looked further back into the past by studying the great late-medieval Flemish masters of more than two centuries before him and incorporating some of their art into his own.

This view of Bach as a reactionary figure, at least partially accurate, lasted for more than two centuries until a now famous and provocative article of 1976 by Robert Marshall called “Bach the Progressive.” Marshall emphasized the evident influence on Bach during the last two decades of his life of the most advanced modern stylistic experiments, and showed that there were even clear anticipations in Bach of the stylistic developments that would appear only with the era of Mozart. This was a sensible approach. Bach lived into the period we think of as the Enlightenment, and it is unlikely that he would have remained untouched by some of the radical changes in modern aesthetics; indeed, he himself explicitly insisted on the importance of keeping abreast in Leipzig of all the different stylistic possibilities of his time—although Dreyfus points out that for Bach that meant being aware not so much of the music of other composers, as of the way he himself incorporated different styles in his own work.

Dreyfus takes issue both with the accepted view of Bach as reactionary and with Bach as progressive, but he largely builds on Marshall to carry the argument further. For him, the music of Bach does not sit comfortably either with Baroque or Enlightenment aesthetics. Dreyfus is, I feel, correct but not entirely convincing on Bach’s incompatibility with the older aesthetic of the Baroque: he observes that this aesthetic required an attention to decorum, a demand that each artistic genre be carried out according to rules that were agreed upon by connoisseurs; nevertheless Bach continuously violated decorum and inventively and brilliantly broke the established rules of the different genres in which he wrote. But then so did Handel and Rameau, both Bach’s exact contemporaries, born at the same time; it is what has kept their music, and Bach’s, alive after more than two centuries. Dreyfus is right to combat the idea of Bach as an old-fashioned representative of the past, and he observes that nothing that Bach took from his predecessors was not radically transformed under his hands.

Nevertheless, the sense that Bach’s contemporaries had of the reactionary cast of much of his work will not go away. They accused his music of being monotonously thick, overloaded with learned and decorative detail—bombastic, in short. We must admit this to be true in many instances, but it does not, in fact, elucidate Bach’s relation to the past. He himself described his own compositions in a memorandum of 1736 to the Leipzig town council as “incomparably harder and more intricate” than the older liturgical works still performed. His music did not exactly resemble the music of his predecessors in its bombast: on the contrary, it could be more bombastic than anything ever written before, more complex, more artfully worked out, more oppressive, more ostentatious in its display of esoteric skill.

On the Enlightenment aesthetics which arose during the last decades of Bach’s life between 1720 and 1750, and its irrelevance for his music, Dreyfus is more provocative as well as convincing:

The new Enlightened aesthetic was also a theory that desired the representation of nature to be as effortless as possible. That is, art was to be produced and judged according to how well it offered direct access to the mental representations that lay behind the work, its signifying ideas, if you will. Words, music, gestures, pictures—all were there not to draw attention to themselves as signs but to be as transparent as possible…. Georg Friedrich Meier (1718-1777) …put it this way:… “One must avoid everything through which the attention is diverted from the object itself and is steered principally toward the contemplation of the signs and images in which the object is clothed.”

Any such theory spells big trouble for music, particularly music that is dedicated to thinking through its own materials. For one thing, the principle of transparency is more problematic for music than for any of the other arts. Moreover, the sheer breadth and detail within a musical experience are always far out of proportion to any underlying affect that can be proposed for it. Whereas a painting can be seen to resemble an observed view, or the plot of a novel can seem to replicate a realistic experience, the phenomenon of music itself does not by and large resemble the experience it may wish to convey…. Given, then, Bach’s continual propensity—indeed, obsessional need—to “divert attention” precisely “to the signs and images in which the object is clothed,” he must be seen as one of the leading musical thinkers to take a decisive step in refuting this damaging view of signs, though he may have lacked an adequately explicit theory to do so.

Dreyfus does not wish to make Bach into a Romantic, but he suggests that it is only in the late Enlightenment with the appearance of an expressive theory of aesthetics that overthrew the representational theories of the earlier eighteenth century and dominated Romantic criticism that Bach’s music could come into its own. At the end of his book, he writes about Bach’s disciples and admirers in the last years of his life:

Even Bach’s followers…were not prepared to admit that the progressive musical thought of the day, for all its elegance and charm, had signaled a regression in technique and a demeaning of the divine task that Bach saw as entrusted to the composer. As it stands, only his music (at least in Germany) opposed these regressions, if only by doggedly pursuing a path at odds with the dominant aesthetic.


The later canonization of Bach, however, is dependent on more than a change of aesthetic doctrine. Dreyfus is right to say that all of Bach’s contemporaries, friends as well as enemies, were surely aware “that Bach’s music by and large went out of its way not to curry favor with anyone’s taste, that the composer wrote music which fundamentally ignored a concern with public approbation.” But this is not as straightforward as it might appear. What does “public approbation” mean between 1720 and 1750 in a town like Leipzig with no opera house and a niggardly financed tradition of ecclesiastical music?

On a larger European scale, what does public approbation mean for a culture in which public concerts are not yet a fully established institution? Bach’s fame as a virtuoso performer on the organ is well attested, and was international in spite of the fact that he traveled very little. Fame as a composer, on the other hand, was largely dependent upon a position at an important court and on the composition of serious operas, paid for at that time by a court. Bach did try to free himself from the obscurity of Leipzig by applying for a court position in the important city of Danzig: there he would certainly have had to compose operas. He was unsuccessful: perhaps his well-known formidable contrapuntal learning put him at a disadvantage.

In any case the hundreds of Protestant church cantatas that Bach composed at Leipzig did not travel. Within a few years of his death, in fact, they were not produceable anywhere. They are not even a significant part of the repertory today, in spite of the commanding presence that Bach enjoys in the history of music. The Mass in B minor was half composed and half compiled from previous works, but it is unlikely that a complete performance was envisaged. It was intended as exemplary. The Passion According to Saint Matthew is another matter: one of the greatest masterpieces of the eighteenth century, it has a dramatic power equal to any opera or oratorio.3 It was performed by Bach himself several times during the last three decades of his life. Historians often date the “Bach revival” from Mendelssohn’s partial performance of this work in Berlin in 1827, but this is absurdly misleading. By 1827 Bach had already been universally recognized for almost thirty years as one of the greatest figures of the eighteenth century. And the rarity of performances of his music in the late eighteenth century is no mystery. The works of all early eighteenth-century composers disappeared from the repertory—Vivaldi, Telemann, Alessandro Scarlatti—with the single exception of the oratorios of Handel. (The oratorio was the only musical genre of the first half of the century to remain economically viable while all the contemporary operas quickly became out of date.) Above all, the institution of public concerts, still in its infancy, was not structured to allow, or even conceive, the performance of the works of Bach then available (although with some difficulty) to professional musicians in Central Europe.

Throughout the late eighteenth century and the early nineteenth, Bach’s reputation was founded almost entirely on his educational works for keyboard: the four volumes of Keyboard Exercises, the two- and three-part Inventions, the two volumes of The Well-Tempered Keyboard, and The Art of Fugue. (The idea that the latter work and the great six-voice ricercar fugue from The Musical Offering were intended for anything other than two hands at whatever keyboard was handy—harpsichord, clavichord, organ, or early pianoforte—is an aberration due to German musicologists in the early twentieth century.) All these great works were written to be studied at the keyboard by those aspiring to the knowledge of composition and performance.

To understand the history of Bach’s reputation, it is important to see that none of these basic works could have been performed in public during the eighteenth century—or for most of the nineteenth, for that matter. A complete public performance of any of them was out of the question: no one then would have wanted to hear twenty-four preludes and fugues in twenty-four different keys at one go. (It is attested that Bach himself once played through an entire book of The Well-Tempered Keyboard for an audience of one, a student who was certainly looking at the score over his shoulder). Even today a complete performance is more a religious ritual than an ordinary recital. Partial performance at a truly public occasion was also not workable, although this could take place in private or for a select group of friends and invited guests.

The first attested such occasion is significant: some thirty years after Bach’s death, Mozart arranged several fugues from The Well-Tempered Keyboard and from The Art of Fugue for string quartet and string trio, adding new preludes of his own devising. It must be emphasized that the fugues from these encyclopedic works are not fully intelligible to anyone not playing them. The principal theme or fugue subject is often hidden in an inner voice, sandwiched between the more easily perceptible soprano and bass, and neither harpsichord or organ is capable of setting it in relief except with stylistic shenanigans that would not have been indulged during the eighteenth century. It is possible to bring out the main theme of an inner voice on the piano, but emphasizing the principal subject of a fugue at the expense of everything else (the way students are taught to play Bach at most music schools) is to distort the whole purpose of the work.

Arranging the fugues for several instruments, as Mozart did, makes it somewhat easier to distinguish the different voices, although it is dubious that Bach would have thought this an improvement. (For Mozart, everything in music had to be audible, but this does not correspond to Bach’s way of thinking about music.) Public or semi-public performance of Bach after his death begins, therefore, with Mozart’s arrangements, and it continues in the nineteenth and twentieth centuries with transcriptions by Franz Liszt, Ferrucio Busoni, Leopold Stokowski, Anton von Webern, electronically with the Moog synthesizer, and finally the Swingle Singers, who turned the Contrapunctus of double invertible counterpoint at the twelfth into a real pop piece.

The ideal works of Bach for public performance are above all his early virtuoso works for organ—preludes and fugues, toccatas and the passacaglia. They have a brilliance, a rhythmic swing, and a verve rare in the later music. In these works, Bach took care to make each appearance of the principal subject audible and effective for the listener. They are not intended, like the educational works, for private playing and meditation. Interest in organ recitals, however, dwindled in the second half of the eighteenth century, never to return as an important element of musical life (except perhaps in England). These are the works that Liszt, Busoni, and Stokowski seized for their transcriptions, and they became a permanent part of the concert repertory. To these may be added a few works for harpsichord, above all the Italian Concerto and the Chromatic Fantasy and Fugue, both popular in the nineteenth century.

Nevertheless, the term “Bach Revival,” as applied to nineteenth-century performance, is a misnomer if it signifies making his music available again to the public: it was not a revival since most of his music had never really been available to the public. Musical life before 1800 had not developed the kind of concert program in which the important keyboard works of Bach could be presented. The piano recital, invented by Liszt in the 1830s, finally made it possible—and even then many works had to wait for the twentieth century before they could find a place in the repertory.

It was the stylistic revolution of the 1780s, the revolution in which the principal figures were Gluck, Haydn, and Mozart, however, that created a new understanding of Bach. Dreyfus is right to call attention to the radical change of aesthetic, but in fact it was something very much more specific that won Bach his supremacy. It is true that the music of the 1750s and 1760s—of such composers as Hasse and Sammartini—is largely insipid: it had lost the contrapuntal richness of the Renaissance tradition. Even in pure instrumental music, composers now employed only the simple accompanying figures required by opera, the most prestigious genre of the time. In the late 1770s, however, the more ambitious composers, Joseph and Michael Haydn, Mozart, Florian Gassman, and others, attempted to recapture the earlier complexity and absorb it into the elegant, so-called galant style. At this point the work of Bach was seen in a new light. By the early 1780s the twelve-year-old Beethoven received his first press notice for “playing chiefly the Well-Tempered Keyboard,” as yet unpublished but recognized by the reporter as the “non plus ultra of our art.” Whether Beethoven played a whole book of the work at one sitting and for whom is unknown. He certainly did not perform it at a public concert.

With the revival of professional interest in counterpoint, basically an attempt to recapture the nobility and prestige of the art of music, the new aesthetic set in relief by Dreyfus became crucial. For this Romantic aesthetic, the means of expression does not efface itself humbly before the meaning, but thrusts itself forward ostentatiously. Bach’s extraordinary craft had been generally recognized internationally by professionals: the expressive force of this craft had been perceived only dimly. The virtuosity of his constructions literally obscured their affective power. Dazzled by his skill, most of his contemporaries could not recognize the emotional intensity.

As Matthew Arnold has remarked: “Many things are not seen in their true nature and as they really are, unless they are seen as beautiful. Behavior is not intelligible, does not account for itself to the mind and show the reason for its existing, unless it is beautiful.”4 Working with the new chromaticism made possible by well-tempered tuning, Bach had transformed and magnified the latent expressive capacities of the most esoteric elements of the ancient contrapuntal art. Recognition of his achievement now became official. It was, however, only the educational keyboard works of the final three decades of his life that were the basis at first for the new evaluation. At last the extraordinary beauty of these works was perceived as well as the tremendous skill—or, rather, the emotion and the beauty of the craft itself were gradually recognized in the end.

The greatest influence on the new aesthetic of the late eighteenth century was music itself, above all instrumental music. It was because music did not appear to refer outside itself, but had a meaning simply constructed with, and derived from, the very elements of expression, that it became a model. The meaning was not what the notes said, it was in the notes themselves: they “called attention to themselves.” This is true of contrapuntal music more than of any other. That is why philosophers like Rousseau who believed that the means of expression should be transparent detested counterpoint more than any other aspect of music. (Rousseau’s ideal, indeed, was absolutely unaccompanied melody.)

Not until this philosophy of language and signs changed could Bach’s music meet with full understanding. But it is a mistake to think that his music passively waited for a change of aesthetic. The gradual increase of professional and amateur interest in Bach in the latter half of the eighteenth century contributed to that change and made it possible.5 In this sense it is not the Romantic movement that salvaged the work of Bach, but his work that helped form Romantic aesthetics, as musicians who wanted to assert the independent dignity, and justify the complexity, of their art gradually realized throughout Europe that his work could be a model and an inspiration. By 1800, the work of Bach was the foundation of all teaching of music throughout Western culture, and has remained so for two centuries.

The difficulty of assessing the growing importance of Bach between 1750 and 1800 comes from the impossibility of using the most common yardsticks: frequency of public performance and the quantity of publication. Public performance in the late eighteenth century was almost everywhere an infinitesimal part of musical life. Private performance at home for oneself and for friends played an even greater role than recordings do today. Publication was less essential when so much music—and literature—was available in handwritten copies. (We would seriously underestimate the influence of Diderot in the 1780s, for example, if we based our estimate on publication; his most important works, like Le Neveu de Rameau and Le Rêve de d’Alembert, circulated only in manuscript copies, and were published posthumously.) The most famous work of Bach, The Well-Tempered Keyboard, circulated throughout Europe among professional musicians in manuscript copies made by Bach’s pupils and friends (and their pupils and friends). When it was finally published in 1801, three publishers independently and simultaneously printed it from three different manuscript sources: this testifies to the growing interest in and pressure for publication of a work that had still an interest largely for the professional.

Even today few of Bach’s works can be intelligibly presented to the public as he would have presented them for the simple reason that few of them were intended for any kind of public audition. Some of them, however, have achieved a spectacular public success. They have been anachronistically salvaged by the twentieth century. The nature of the paradox will appear if we briefly consider the Goldberg Variations.

This is in some ways the most popular of all Bach’s works with the public, in spite of the fact that it lasts between forty-five minutes and an hour and a half, depending on whether one plays the repeats or not. There are dozens of recordings available today, although it was rarely performed before the twentieth century. Its status has been established by Wanda Landowska, by Rudolf Serkin and other German pianists, and by Glenn Gould. The paradox is that the structure is perfectly designed for success in public presentation, and yet public performance cannot have been imagined by Bach. We may go further with Robert Marshall’s thesis and remark that Bach anticipated not only the style of many composers to come but even the format of the modern recital program.

The form of the Goldberg Variations, specifically designed for two-keyboard harpsichord, has many details which are abstract, mathematical, or purely symbolic. Every third variation is a canon6 starting with variation no. 3, and they are arranged in an order purely mathematical; canon on the unison, second, third, fourth, up to the ninth. Starting with variation no. 1, every third variation is a brilliant virtuoso two-part invention (the influence of the sonatas of Domenico Scarlatti is likely). The remaining variations are all different characteristic forms: dances, a fugue, accompanied solos (with the right hand on one keyboard to be played louder than the left on the other). The sixteenth variation which begins the second half of the work is a French overture: unless Bach expected a pause in the work just before it (and there is no reason to think he did), the overture is a symbolic gesture. The systematic alternation of these three kinds of texture—canon, virtuoso invention, and free form—gives an unparalleled variety of style to the work.

The great success for the public, however, is largely owing to the way it ends. Variation no. 25 is an accompanied solo in the minor mode, deeply affecting and always played very slowly: it is a dramatic moment. In the following four variations, the virtuosity and play with sonority reach new levels of brilliance, unmatched even by Scarlatti. The final variation, no. 30, is a comic apotheosis: over the bass of the theme, Bach superimposes two folk songs: a slightly indecent love song and a jolly complaint (“My mother only cooked beets and cabbage, if she had made some meat, I might have stayed home a little longer”). At the end, the original theme, a stately and lyrical sarabande, is repeated.

No one has ever devised a more dramatically effective concert ending. Yet this work could not have been performed at any really public function during the composer’s lifetime, and even at semi-private musicales it is doubtful that it would have ever been played all the way through.7 It was published as another volume of Keyboard Exercises. Each variation is a model of composition. You can learn how to write a canon over a common descending bass on every degree of the scale. Played through for oneself and perhaps a friend, it would have been an occasion for instruction and delight, the classical criteria for great art. In this sense, Bach conformed absolutely to the Enlightenment aesthetic, and realized it on a scale that no other musician had ever attempted. At the same time, he devised the most effective work of his time for an institution as yet undreamt of, the modern recital program.

As far as I know, the first report of even a semi-public performance of the Goldberg Variations is in E.T.A. Hoffmann’s Fantasy Pieces, in the section called “Kreisleriana.” The young musician Johannes Kreisler—a character based on Hoffmann himself—comes to a salon with a copy of the music; someone discovers it under his hat in the next room and demands a performance, thinking it a simple, popular piece. Kreisler begins to play. “All right, I said to myself, listen and explode with boredom.” Several ladies leave at variation no. 3. A couple of Kreisler’s pupils hold out until variation no. 12. Others leave at no. 15. Only one lasts until no. 30. The salon was evidently not the right place for this work.

Beethoven knew and imitated the Goldberg Variations, and Busoni fixed them up with a vulgarly inflated finale. But it is only in the second half of the twentieth century that the greatest instrumental work of the first half of the eighteenth century finally came into its own.

This Issue

October 9, 1997