Christos Hatzis

Monogram written during August - September 1998. Posted on the www: November 1999.

Between 1985, the year of J. S. Bach’s 300th birthday, and 1998 I worked on a number of compositions which are palimpsests of works by Bach. Four of them are based on fugues from Bach’s swan song, The Art of the Fugue. This paper describes my philosophical approach to composing works based on existing music in general, and attempts to outline the techniques used in each one of the works in this series.

My first two works based on the music of Johann Sebastian Bach date back to 1985, Bach’s 300th birthday. One of them, Spring Equinox, was commissioned for the Bach300 Festival in Toronto and was performed two days after our mutual birthday, March 21, the day of the Spring Equinox. Beyond the birthday coincidence, and although it was not until much later that I began to view music history as a procession of shifting paradigms, Bach was already in my mind the focal point, the most eligible representative of eurocentric music history, which encompasses a period of nearly 800 years, from the advent of polyphony to the aleatory of the 1950s. If I were to make a musical commentary about this span of western music, Bach’s was certainly the most appropriate music to make it with.

Prior to 1985, I had composed very little music. Having received my training in composition in the United States at a time when academic institutions were dominated by the ideology and politics of serialism and the avant garde, I had developed in the late seventies a system of composing which consisted of fractals based on the overtone series1. After a few years of not composing at all, I wrote a series of neo-romantic works of which only the song cycle Arcana is still extant. By the middle eighties I was developing as a composer on two separate fronts. One was concerned with expression and the deeper aspects of composition. It was a slow and painful growth process which did not reach maturity until the mid-nineties. The other front was the development of technique. My work on this front can be best described as of a series of experiments which were meant to provide me with an armory of problem-solving mechanisms in composition. Unlike most of my music nowadays, the questions asked and the answers extracted from these pieces were very precise and purposely limited in scope. I always started with non-original material and treated it in the way a photographer treats his subject matter. I applied distortion lenses, changed the magnification, stretched or compressed the frequency or time element, but the original material was always there and always discernible as something different than the manipulations applied to it.

Almost invariably, the subject matter of these compositions was Bach, particularly The Art of the Fugue, Bach’s last composition, left incomplete and with no indication as to the intended instrumentation. It felt appropriate for this sort of treatment, not only because of its non-specificity, but especially because it was Bach’s treatise on technique, his final and most important inquiry into what was possible contrapuntally. Beyond that it has been described by many scholars as ‘paper music’ never intended to be performed, but only studied. The other works of Bach used as source material in this series are the Goldberg Variations and the Concerto for Flute in G minor.

At the time of their composition, I thought of these works as ‘musical photography’, for reasons which I have already explained. I applied to each of the works a different photographic technique and then followed it through vigorously. This means that they can be easily taken apart, and be anatomically exposed, particularly the earlier ones, since they have been put together not so much by a process of composition, but by one of ‘inverse’ analysis. The later works in this series go beyond technical considerations and the investigation of surface aspects in Bach’s music. These later pieces have little to do with ‘musical photography’, or with technique in general, so I have chosen to describe them as a palimpsests, that is something which is written on top of something else. It is a more appropriate description of postmodern composition in general, and in this specific instance, it means an original way of combining unoriginal material, a juxtaposition of various layers of information whose composite meaning is radically different than the original meaning of each separate layer. In Farewell to Bach, the last piece of my Bach series, this process can also be described as ‘cultural counterpoint’. I have used this term to describe a great deal of my music during the late eighties and most of the nineties, particularly those works which are influenced by world music. The term would not be appropriate for any of the other compositions in this series based on Bach, because in all the other works the cultural reference is singular and very specific.

This paper has been written for a number of reasons: to document the compositional techniques used in conjunction with Bach’s music for the benefit of musicians and listeners who might be interested in them; to understand these works better myself, by approaching them in an anatomical manner, something I do not often do; finally, to make this information available to my composition students at the Faculty of Music, University of Toronto, so that they may apply the manner of thinking about music which lies behind these techniques, if not the techniques themselves, in their own search for new compositional tools. Given the plurality of purposes, this paper will occasionally veer into areas which may not be of particular interest to some of its target readers, as for example the specific circumstances which have brought these works about—commissioning, premieres, and subsequent editions. For the benefit of these readers, I have enclosed such reference information in editors brackets: [.....]. A more extended report on how each work was put together will follow such reference material with musical examples and/or other illustrations. This paper will not attempt a detailed analysis of these works. Even when an aspect of a composition is dealt in detail, it is solely for the purpose of outlining the compositional technique used and showing briefly how this technique is implemented in the music. Occasionally, as circumstances require it, I will discuss larger musical issues which may arise from the use of these techniques.


[Composed in 1985, Equivoque was originally written for accordion and two computer controlled synthesizers which bounce accordion samples at sixteenth note intervals across two stereo speakers. The piece was requested by Canadian accordionist Joseph Petric who gave the premiere performance of the work at Toronto’s Music Gallery in 1985 and several performances in the ensuing years. In most subsequent performances, the synthesizer parts were recorded on a digital audio tape and played back during the performance. In 1998, a new version was prepared for harpsichord and tape for performance by Charlotte Nediger in a concert with Tafelmusik on June 6, 1998 at St. Andrew’s United Church in Halifax, Nova Scotia. The harpsichord part is identical with the original for accordion, but the tape part was remixed with harpsichord sound samples and is now available in two different tunings, A=440 and A=415, to accommodate both modern and baroque harpsichords. A further version for piano and tape was also created later in 1998.]

The idea for Equivoque came from experiments with graphic media I was engaged in at the time. I took two similar photos of identical size, placed them on top of one another, and sliced them into small pieces using a ruler and razor blade. Then I proceeded pasting the pieces back together again, but interchanging them in a way that the end result was two checker-board-type collages. Each of the resultant collages had an equal number of pieces from Photo A and B. Looking at each one of the collages to determine if the missing pieces from each photo could be guessed at by virtue of the information provided by those present, I found that, if enough pieces with high information elements were present, like the important elements of one’s face, the mind could easily reconstruct the missing elements from each of the two photos. I changed, therefore, the mechanical checker-board-like arrangement of the pieces with an arrangement whereby most of the high information elements from both photos (eyes, nose, mouth) were maintained and the lower information pieces were discarded, while maintaining at the same time a high degree of articulation between the pieces of Photo A and B. By maximizing the areas of information and discarding the areas of redundancy, both photos became clearly visible in the space of one (Figure 1.)

Figure 1

The next step was to see if this principle could be applied to music. As source material I picked one of the two-part contrapuncti from The Art of the Fugue which seemed most appropriate for this type of manipulation for the following reasons: (1) it is simple—only two parts—which means that it is appropriate for an instrument like the accordion, (2) the original will be recognizable even after extensive manipulation, (3) it has an eighth-note motor rhythm throughout which serves as a quick orienting feature for the ear, after the original is sliced into small pieces with other pieces inserted between them, and (4) it has an unusually high degree of redundancy with directional stepwise melodic writing and sequential melodic and harmonic movement. The compositional elements which make this composition rather predictable in its original form and are normally undesirable serve as perceptual anchors in the derivant work and are, therefore, highly desirable (Figure2).

Figure 2

As in the self portrait of Figure1, I used two different ‘snapshots’ of the Bach original to create Equivoque. The first snapshot is the original piece itself which is in D minor. For maximum contrast, I chose as the second snapshot the most distant transposition from the original which would be A flat minor, with the two keys (snapshots) interchanging at sixteenth note intervals. This means that for the most part you can hear the original D minor material on the first sixteenth note of every eight note beat, while on the second sixteenth note of every beat you can hear the same material transposed a tritone higher. Theoretically, one should be able to hear two separate pieces (tonalities) in the space of one; not in the manner of polytonal music, for the two tonalities here are interposed, not superimposed, but in a manner analogous to switching repeatedly and rapidly between two preset stations on the radio and trying to follow the contents of both at the same time. I say theoretically, because in reality the acoustic interposition behaves differently and not as precisely as the visual one. This imprecision is due to two reasons. The first has to do with the inherent time inconsistencies in the performance of compositions incorporating live instruments and tape. The bandwidth of these inconsistencies can be significantly reduced by good players, but can never be completely eradicated, nor is it desirable that it should. The second reason has to do with how sound is transmitted. Our understanding of what constitutes a ‘musical’ sound consists of not only the original makeup (timbre) of that sound at its source, but also the transformations it undergoes along its trajectory from the source to our ears. Even sounds whose impact is instantaneous, as for example with earphones, have at least some of these after-the-source attributes, acquired at some stage of the recording/transmission process. Of these attributes, reverberation is the most problematic for our particular experiment. When the piece is performed live, there is significant bleed-through from one snapshot to the next, unlike the visual collage of Figure 1, where the borderlines between elements belonging to separate originals have zero width and are only contextually defined. To counter this problem I assigned the two musical snapshots in Equivoque, the D and A flat tonalities, to hard right and hard left positions in the stereo system, so that physical space may be used as an additional means of segregating the two signals (it goes without saying that this applies to the tape part only.) There is a lot of ping-pong-like bouncing in Equivoque, and notes or chords forming tritones between them—the most abundant interval—rarely emanate from the same physical location (Figure 3.)

Figure 3

Even with this separation, the result was rather confusing. My intention was to come up with a situation where the brain perceives all the on-the-beat time slices as aspects of one musical picture and all the off-the-beat slices as aspects of another. Furthermore, I wanted these bytes of information to be kept perceptually segregated, so that the composite snapshots could be properly reconstructed afterwards. That was not what I was getting at all. Instead, the acoustical result was that of bitonality, as if the two keys were superimposed on each other. What I was hearing was the minor mode version of Stravinsky's ‘Petrushka’ chord. Here too, as in my original visual experiment with the two photos, the problem was that there was no differentiation between high an low information in my slicing and selecting process. The solution of keeping high information segments and discarding redundant ones used in Self Portrait was not applicable to Equivoque, simply because Self Portrait is based on two completely different photos, while Equivoque consists of two identical copies of Contrapunctus XII a tritone and a thirty second note apart. I devised instead a system of shifting weights of emphasis between the two tonalities, so that at certain points the overall harmony is predominately D minor, at other times it is predominately A flat, but most of the time it’s a combination of the two. I call the gradual shifting of these weights quantitative modulation to differentiate it from the system of qualitative modulation which Bach uses to move to secondary key areas in the original.

Quantitative modulation finally solved the perceptual discrepancy. With shifting emphasis between the two tonal areas and spatial distribution, a listener with some exposure to western common practice harmonic language can follow the process quite clearly. Once quantitative modulation had been perceptually established on relatively stable harmonic areas of the music, I began to use it in conjunction with qualitative modulation. The aural effect of combining the two processes is quite intriguing. It involves starting the qualitative modulation process with a maximum weight in one of the two key areas and gradually changing the balance until the weights have been completely reversed by the end of the qualitative modulation. At the start of the segment, one hears the original contrapunctus beginning to move towards a secondary key area, just as Bach wrote it with minimal interference from the opposite key area. However at some point during this process the incremental shifting of the weights causes an attention shift to the concurrent qualitative modulation which is taking place one tritone higher. In Figure 4, the original music modulates from A to D a perfect fourth higher, but the listener hears a modulation from A to A flat (the fourth of E flat, the ‘prarallel’ harmony at the beginning of this segment) and, while according to one way of hearing it sounds like a natural arrival, according to the harmonic language of Baroque music, it is anything but.

Figure 4

This dual way of perceiving the same phenomenon is reminiscent of perceptual ambiguities in M. C. Escher’s work, particularly his etching Up and Down (Figure 5) which is conceptually very similar to Equivoque. In both works there is a semantic gray area somewhere in the middle where objects have potentially two opposite and mutually exclusive meanings (up vs. down; lower key vs. higher key.) At the outer edges of the composition, this duality collapses into a single reality (either up or down; D minor or A flat minor). However, as the eye/ear passes through that gray area, which is initially interpreted as belonging to the realm already visited, and enters into the realm of the anti-universe this gray area is mentally revisited and re-interpreted according to the new context. At that point the perceiving mechanism is engaged in two distinct activities: recording and interpreting the incoming information on one hand, while re-interpreting the gray area information on the other.

Figure 5

The perceptual delight or unsettling feeling one experiences from this process is caused by the realization that our empirical sense of reality may be nothing more than a ‘conspiracy’ of our senses2. The subjective forms of this perceptual ‘conspiracy’ are perspective, and its acoustical counterpart, tonality. While both perspective and tonality can be extended and challenged from within with no long-lasting perceptual crisis resulting from this challenge, as chromaticism has constantly challenged and redefined tonality for the better part of two centuries, it is the assault from without which presents the most serious challenge to our sense of reality. In this sense, Escher provides a more fundamental logical disturbance than any of the late nineteenth and early twentieth century painters who abolished perspective altogether in their work, because he continues to engage the viewer in a discourse on perspective—and, by extension, on the nature of perception and reality—even though he uses perspective as a means of demonstrating the futility of holding any objective view of the world through sole reliance on our sensory mechanisms.

Similarly, Equivoque is a discourse on the ‘conspiracy’ of tonality and uses tonality to expose this conspiracy. By the term ‘tonality’ I am referring to the organically evolving, goal oriented, cause and effect dominated harmonic system used throughout the era we refer to as common practice music. By establishing the longer possible travel distance within this system, the tritone, and making instantaneous connections between distant points which are only possible by traveling outside, not within this universe, the implication is that something must exist beyond this universe which makes these connections possible. Think for example of the concept of non-local connections in physics and how this has necessarily affected our understanding of relativity and the nature of the space time continuum.3 In Equivoque, the mirror image of the contrapunctus a tritone higher is, perceptually speaking, not a transposition. To early eighteenth century tonal perception, it represents ‘the other side’, the most distant point which cannot be reached except by means of a long journey through the circle of fifths. To get there instantly raises serious questions about the nature of the tonal continuum based on this circle of fifths and its claim that it is the only possible route for harmonic travel. Soon after Bach’s time, these questions of course would no longer be asked. As the understanding about equal temperament and its compositional implications increased, composers realized that they could in fact travel at warp speed outside the circle of fifths by simply redefining the harmonic function of certain chords via enharmonic respellings—dominant seventh chords becoming ‘German’ augmented sixth chords, and diminished or augmented chords resolving readily to more than one key. By Bach’s time, however, even though the equal temperament system was already in use, this type of harmonic space travel was not yet possible.4

Other than its doubling a tritone higher and a sixteenth note later and the process of shifting weights between the two tonal centers, Equivoque follows faithfully the progress of the Bach original. However, at a point exactly three quarters into the piece, I created a climactic moment by doubling the music of the tape part in octaves and by introducing an added contrapuntal line on the accordion which does not exist in the original. This adds direction and focus to the formal outline of the piece, although it does not change in any way the operating principles established at the beginning of the work. There is no resolution at the end. Like in most of the other works in this series, the music ends with Bach’s signature, the notes B, A, C, H (Bb, A, C, B natural in English notation), The last note of the signature is supported by a diminished chord which contains the two ‘tonics’ of the piece (D and A flat). As the chord decreases in intensity, the note D is sustained after the other the notes have stopped (Figure 6), but to little avail. Even though D is the tonic of the original, it lacks the harmonic preparation which would have established it as a tonic; the lingering impression continues to be that of the ambivalent diminished harmony preceding it. If listeners hear this D as a tonic, by perceptually appraising its role in the background structure of the work, then they essentially encounter a paradox, for they become aware of the antiphatical roles this note plays at the foreground and background levels. Thus Equivoque remains equivocal to the very end.

Figure 6


Figure 7

[Stylus (1987) for recorder, viola an accordion was commissioned by recorder virtuoso Peter Hannan, violist Douglas Perry and accordionist Joseph Petric with funds from the Ontario Arts Council, for a proposed concert involving the three of them which actually never took place. Soon after I composed the work, I made a different version for string quartet and left the top instrumental part of the trio version open for various instruments, primarily upper strings and woodwinds. Even with these adjustments, it was not until ten years later that Stylus was recorded at the Glenn Gould Studio of the CBC Broadcast Centre in Toronto by violinist Marie Berard, Douglas Perry and Joseph Petric for a broadcast on the new music radio program Two New Hours. The first public performance of the work was the string quartet version played on Baroque instruments at the June 6, 1998 Tafelmusik concert in Halifax, where the harpsichord version of Equivoque was also first performed. The performers on that occasion were Tafelmusik members: Stephen Marvin and Christopher Verrette, violins, Patrick G. Jordan, viola and Sergei Istomin, violoncello. The first CBC broadcast of Stylus was of this performance and it took place on July 13, 1998.]

The work is based on Contrapunctus VII, one of the four part fugues in which the main theme from The Art of the Fugue is presented in the dotted eighth note followed by sixteenth note configuration commonly known as the French overture rhythm. The Bach original is included in its entirety in the new work no other material has been added with the exception the closing music signatures, which I will get to later on. Like Equivoque, Stylus is an one idea piece. As the name implies—at least to those who were of age during the vinyl record era—the predominant compositional idea is a method of causing disturbances to the music similar to those caused by a skipping stylus of a phonograph record. I strategically placed repeat marks at various points in the music and then wrote out the results of the interference caused by these repetitions. Most of the time, these repeat marks are not stationary. Their start and end markers slide to the right or the left by a sixteenth note after one or more repetitions, sometimes in opposite directions. At other times, one of the markers moves while the other remains stationary and at other times still, there are repetitions within repetitions which cause further distortion of the original music.

The start and end points in each loop are chosen so that there is melodic continuity between the end of one repetition and the beginning of the next. In many instances, the continuity is such that the listener is not aware of the exact position of the repeat sign, even though he is aware of repetition taking place (Figure 8b.) The melodic fragments perceived by the listener are thus new, not present in the linear unfolding of the original work, although they are virtually there and audible as soon as this linear presentation is disturbed by my interference.

Figures 8a and 8b

This melodic reconfiguration is one aspect of the perceptual disturbance experienced in this piece. The other, of course, is repetition itself and the way it affects our sense of organically evolving time. Even though the first impression of these repetitions is that it sounds like a minimalist piece (in jest, I have often described it as "Bach meets Philip Glass"), the directional and goal oriented nature of the original makes the repetitions feel more like ‘time traps’. Because of this, they produce a type of musical tension and information which is foreign or undesirable in minimalism. The trapping of a linearly and dynamically evolving thought and its inevitable disintegration information-wise within the time trap, is very much at the heart of the perceptual interest of Stylus.

To make this concept of trapping time even more apparent to the listener, it is better to perform the Bach original in its entirety first, followed immediately by Stylus. In the Tafelmusik performance in Halifax on June 6, 1998, a group of Baroque woodwinds performed the original at the left side of the stage followed immediately by the Baroque string quartet version of Stylus at the right side. I had not heard a performance of the original contrapunctus for sometime and neither had the players of the string quartet, who by the time of the dress rehearsal had been quite ‘acclimatized’ to the convoluted time world of Stylus. Upon hearing the woodwinds play the original during the dress rehearsal, and at the points where Stylus deviated from it by lingering on—trapping—selected material, the facial reaction of the Stylus performers was one of disturbance and surprise expressed in the form of repressed laughter. Difficult as this may be to believe, the original sounded elliptical and incomplete ("the Readers’ Digest version" according to one of the Stylus performers), even though the fugal framework which held the piece together was only operational in the original. This framework had been destroyed, or was not immediately evident after my interference with its normal progress. For most of the listeners, who had equally fresh ears for both works, it was naturally Stylus that caused all the chuckles. But for both the performers and the audience, Contrapunctus VII would never be the same again after hearing both works in such close proximity to one another.

What is destroyed in this context is a deep seated belief among western music listeners in the inevitability and finality of the compositional product. Traditional analytical approaches and western cultural norms which give rise to these approaches nearly always treat a piece of music as a final and complete object, not as one possible realization of nearly infinite possibilities. In reality, however, conventional composition is a navigational process within a complex, and consensually established, maze. It resembles more the process of playing a video game than that of creating one.

Imagine yourself playing such a game. At every step of the way, you are confronted with a decision: Go left or go right? If you choose left, then you have to decide which of the seven closed doors in the corridor you just entered should you open and enter. Based on the decision you just made, you will end up in completely different places and in each of these places you will be asked to make further decisions. Like in composition, those decisions are goal oriented. You want to get somewhere at the end, you want to achieve the goal that you set for yourself in the beginning. In western music, or in what I call the music of the Renaissance Paradigm,5 we call this journey composition. Increasingly, however, we also understand as composition the work which the programmer of the video game has done, like the Brain Opera6 of Tod Machover, instead of only the navigational strategies of the end user. The programmer’s composition is virtual: he works in a multithreaded manner which structurally resembles a branching tree; he has to compose every possible path at every step of the way, regardless of whether these paths are followed or not by the person playing the game. In a sense, the programmer is like an instrument maker, except that in this case, and unlike a conventional instrument, the range of navigational choices and operating parameters are hidden in the background, so that the end users will not feel limited in the exercise of their decision making process.

Every time the end user makes a decision, the rich possibilities of virtual composition collapse into actual composition, in the conventional sense of the word. Bach’s works, like those by other composers, are such actualizations of virtual works which Bach never has, nor could have, composed. At the thematic level, one could argue that The Art of the Fugue is a series of actualizations of a virtual theme, the main theme of the work, but at the level of composition, each fugue is an actualization of a virtual composition, which is the common practice harmonic system of the early eighteenth century. Everything composed during that time can be viewed as a different strategy of collapsing the probabilities of this seemingly limitless harmonic system into actual musical events. At the level of an individual work, these strategies become the actual composition, whereas at the level of several such compositions by a single composer, the sum total of these strategies is understood as the composer’s personal style.

In a combined performance of Contrapunctus VII and Stylus, the audience is exposed to the dichotomy between ‘actual’ and ‘virtual’. By its very nature, Stylus undermines our inherited sense of finality and inevitability in a work of art. It resurrects the ‘what if’ reality which most of us abandon after childhood, and which is intimately associated with the concept of game. Stylus is a game, an instructional game, to be sure, but a game nonetheless. In performance, this game is played not only between the audience and the composer, with Bach’s music as the operating language, but also between the composer and the performers.

Stylus challenges the performers’ assumptions about subdivided time by shifting the barlines and the downbeats of the original against a rhythmic grid of sixteenth note resolution. In the early twentieth century, with Stravinsky and Bartok in particular, this grid had also doubled its resolution: to the eighth note from the quarter note previously in effect. By resolution, here I am referring to the psychoacoustic ability of a performer to easily and intuitively measure time in terms of a certain subdivision of the whole note. We all know of the trouble, for example, many musicians had during the early days of this century counting the shifting time signatures of the Rite of Spring, or those of the Bartok string quartets. By comparison, now these problems seems easily manageable, compared to some of the rhythmic complexities abundant in music since the 1950’s. There is of course nothing particularly novel about sixteenth note resolution in present day music. It is not the resolution itself which is causing the problem of rhythmic coordination in Stylus, but the fact that the Bach material comes in a package with a weight system of rhythmic accents which are heavier on the beat and lighter on the off-beats. My further manipulation of the material adds yet a third weight, even lighter, to the off-the-eighth-note sixteenth notes. All this is easily managed when the music is not trapped in a loop. But in a loop, particularly if the start point of the loop happens to fall on an off-the-beat sixteenth note, a metrically weak note in the measure, the musical weight of the note and the weight of its position within the measure will contradict each other. The second half of measure 210 in Figure 9b is identical to the second half of measure 33 of Figure 9a. However three sixteenth notes into the next measure, Figure 9b loops back to the first violin E flat of the previous measure and continues to do so four times before it exits the trap in measure 215. The loop point is an off-the-beat sixteenth note and furthermore is inside the 9/16 measure, not at the beginning of it. It is a difficult passage to play accurately and apply to it the proper emphasis of strong and weak beats.

Figures 9a and 9b

The performers could always choose, of course, to shift the pattern of accents to reflect the new position of the notes within the measure, but that would destroy the intended effect of the phonograph stylus, which does not make adjustments to the accents of the music, as it begins to reiterate segments of the recording. The ideal performance of Stylus would treat measures only as simple metric containers, and in the case of loops, as the dimensions of a time trap. Such a performance would draw its system of accents from the music itself, no matter how unconventionally it happens to be framed in notation at any given point. To accomplish this, one would have to do a comparative study of Contrapunctus VII and Stylus, to determine how the music of the former shifts in relation to the notational mechanism of the latter. From experience, when performers take the time to do this in advance, they spare themselves a lot of difficulty afterwards. This comparative work is not as difficult as one might imagine, because (1) there is no music in Stylus that does not derive from the Bach original, and (2) there is no internal manipulation of the music of Bach, that is to say that there are vertical interferences imposed upon the whole, but no horizontal ones, where—in the latter case—one instrumental line might shift in relation to the rest. For these two reasons, one could actually make a studio recording of Contrapunctus VII, and then create Stylus in the digital audio domain by editing the recording accordingly, instead of having performers deal directly with the complicated and idiosyncratic rhythmic peculiarities of the piece. This, however, would take away one of the most enjoyable aspects of the work, which is its performance component and the satisfaction both the performers and the audience experience when those technical difficulties are mastered.

As with Equivoque, Stylus ends with musical signatures, first Bach’s (Bb, A, C, B natural, in English musical notation) followed by my own initials in the viola (C. H. or C, B natural.) I do this right before the final cadence taking advantage of the fact that the melody pauses momentarily on Bb, the first note of the signatures sequence. In the end, like the beginning, Stylus merges with Contrapunctus VII, thus confirming itself as a deviation, an anomaly in the course of Bach’s own work (Figure 10.)

Figure 10


[Commissioned by the Nouvel Ensemble Moderne for the inauguration of the Glenn Gould Studio at the CBC Broadcast Centre in Toronto in 1992 with funds provided by the Canada Council, The Go(u)ldberg Variations is a work for MIDI piano—a Yamaha Disklavier—and chamber orchestra. The premiere performance was given at the Glenn Gould Studio on September 26, 1992, by the le Nouvel Ensemble Moderne under the direction of Lorraine Vaillancourt. ]

The piece was very much inspired by the circumstances of its commission and by Glenn Gould in particular (1992 was the year of Gould’s 60th birthday, and it also marked ten years since his death.) Even though I had written a number of pieces which paid homage to other composers, paying homage to a performer was not something I had ever thought much about. I had no idea how to make a creative response to a performer’s legacy, since this legacy is one of interpretation, and as such, it evades the notational system which composers depend upon for conveying their ideas. I knew that Gould had a life-long interest in Bach, and so did I, so my first decision was that the piece should connect to Gould by connecting to Bach. It seemed appropriate in this context to use The Goldberg Variations as my departure point, since this was the work which began and ended Glenn Gould’s illustrious recording career. I was particularly interested in his revisited rendering of the work in 1981; it was more defiant, more uncompromising than anything by him I have heard on record. The Goldberg Variations is a rather long work, nearly an hour long, and, if I were to do a creative response to even half of the variations, I would have ended up with a piece of several hours in length. I decided that I would compose this work intermittently over a period of several years and that le Nouvel Ensemble Modernee commission would be Volume I of a multi-volume opus, the rest to follow at a later time.

At the pre-compositional stage, I began to think about new techniques of ‘musical photography’ for use in this piece. Each of the Bach variations on the Goldberg Aria was to be treated differently: my Go(u)ldberg Variations would be a set of meta-variations as it were, each meta-variation using a different approach to the treatment of the original (for purposes of clarity, I will refer to the variations of the Bach original as Variation I, II, III..., and the corresponding movements in my composition as Meta-variation I, II, III....) While thinking about such approaches, I attended a lecture by composer Denys Bouliane at the Goethe Institute in Toronto, where he talked about ways of changing existing and recognizable music by simply ‘stretching’ the chromatic scale and then mapping the notes of the music to the positions in the stretched scale. This changes completely the harmonic character of the music, while the melodic contours of the original material remain intact and recognizable. His approach was very much along the lines of my own thinking at the time, and of the thinking behind Equivoque and Stylus, but it was Bouliane’s presentation which put it all together for me. I took this idea a step further by associating these intervallic stretchings with contractions in time, so that the frequency/time domain behaved like a substance made of rubber: if you stretch it perpendicularly it compensates by contracting horizontally and vice versa.

Certain practical considerations entered into deciding how to stretch the pitch domain. To begin with, since the resultant music from this process had to be within the playing range of actual instruments, the stretchings had to be such that the highest and lowest notes of the Bach material were still possible to obtain from the instruments in their new assignments. This meant that, since Bach’s music did not reach the extreme high and low registers of the piano, it would be better to do an outward stretch from a certain central note. I arbitrarily chose G, just below middle C as that central point. Given the limitations of equal temperament, the stretching method could not be uniform throughout. The smallest uniform stretching would convert every semitone in the chromatic scale into a whole tone, which in turn means that a G two octaves above central G would be stretched to a G two octaves higher still, an impossible or impractical note on most instruments. I had to resort, therefore, to resultant mappings of the chromatic scale which were intervalically uneven. Such mappings would follow a pattern of two semitones followed by a tone (stretching 1, Figure 11a), or three semitones followed by a tone (stretching 2, Figure 11b.)

Figures 11a and 11b

Although they display characteristics of certain scales, such as uneven intervallic content, these resultant mappings are not scales, but the approximation of stretchings of the twelve tone equal temperament. Scales replicate themselves at some interval, usually the octave, whereas these stretchings do not. The pattern of one, two or three semitones followed by a tone does replicate itself, but this is the result of mapping the pitches of the stretched chromatic scale onto the closest positions on the piano keyboard. The notes missing from these stretchings will also be missing from any music which is similarly stretched. Within this new ‘chromatic scale’ there are subsets which correspond to scales and tonalities in common practice music; it may take some getting used to, but it is possible to recognize them after some listening, in the same way one recognizes the major or minor modes in conventional music.

Once the tables of vertical stretching had been determined for the entire usable register of the available instruments (Figure 11a and 11b), I ‘translated’ into the stretched temperament the music of the variations which were to employ this technique at the meta-variation stage. It is possible to do this process manually, by consulting the tables of Figures 11a and 11b, but I used MIDI sequencing software—Cubase by Steinberg—to enter the original music into the computer, and the software’s logical editor to massively affect the transpositions of each pitch to its new position. At the end of this process, there were three forms of each variation in computer memory: the original and two stretchings of it.

Next, I started thinking about the time element and various ways of temporally contracting the intervalically stretched music. The most logical way of doing it, given the limitations of our notational system, would be to calculate the degree of stretching in terms of ratios and then express the inverse ratio as a tempo change in the music. There are two problems with this approach. The first problem is that in performance tempo changes are approximate, even with the most accurate of performers, unless they are introduced as metric modulation and are, therefore, expressed as a simple ratio relationship with the previous tempo. The second problem is that tempo change is a linear time change and would not allow for simultaneous or overlapping presentation of two or more different types of stretched music. For the purposes of what I wanted to do in The Go(u)ldberg Variations, I decided to express these ratios as rhythmic ratios, more specifically as simple tuplet configurations. If the original music, for example, was unfolding as a stream of sixteenth notes, four per beat, the first rhythmic contraction associated with the first interval stretching would be expressed as a stream of sixteenth notes in sextuplet configurations (six per beat.) The ‘beats’ of the compressed music would still be every four sixteenth notes, which means that they would be in a six against four—or three against two—phase relationship with the original beats. The second rhythmic contraction associated with the second interval stretching would be expressed as a stream of thirty second notes , a two against one relationship with the original beat. Playing the Bach music this way would have an aural effect of playing the music at three different tempi: the original tempo, one which is 3/2 times faster and one which is twice as fast as the original (Figures 12a, 12b, 12c.)

Figure 12a, 12b and 12c

The difference between this notation and the notationally simpler option of different tempo markings is that (1) the tempi are now ‘locked’ in relation to each other—in both time-compression schemes, the performers count beats in the original tempo, but subdivide them differently—and (2) two or more time compression schemes can be played simultaneously by two or more groups of players. The latter would have been very difficult to notate and control in performance if it was notated with multiple simultaneous tempi. There are works by a number of composers written that way—Charles Ives7 and Karlheinz Stockhausen8 come to mind—but such works are either intended to be approximate and collage-like, or else require multiple conductors, each conducting a separate ensemble in a different tempo. Besides the fact that there was only one conductor available for the performance of The Go(u)ldberg Variations, Lorraine Vaillancourt, I was personally in favor of a notational system whereby the tempo relationships between various instrumental groups are conventionally notated in the score as rhythmic relations. Although difficult to decipher at first, this method of notation encapsulates best the desired musical effect and allows for a certain degree of analytical clarity on paper as well as in performance.

In addition to the techniques already described, there are additional ones used which will be described briefly, simply because they are not as pervasive as the ones already mentioned Each meta-variation employs these techniques in various combinations:

Meta-variation I (Variation I) uses interval stretching 2 (Figure11b) and time compression 1 (notated as sextuplets). The material is presented as alternating ‘slices’ taken from the original and the stretched versions of the music. The Aria and each of The Goldberg Variations are in two sections, each section repeating twice. These repetitions at the meta-variation stage are never identical and are usually complementary. At each repetition in Meta-variation I, I use the material which was left out the first time around (Figures 13a and 13b.) Look at the Self Portrait (Figure 1) and think of its alter ego, a portrait consisting of all the discarded pieces from the two original photos which were not used in the first assembly. These discarded pieces properly placed could form a composite Self Portrait II, which would have the same structure as the assembly in Figure 1, but entirely different content. The relationship between Figure 1 and its hypothetical double would be identical to the relationship between Figures 13a and 13b.

Figures 13a and 13b

Meta-variation II (Variation III) employs a stretto technique, whereby the original music is also presented in superimposed double and quadruple augmentations of itself. These three different magnifications of the music are assigned to three different instrumental groups: each group plays its own segment and then waits until the slowest presentation of the material is completed; then all three groups start the next segment and so on. Progressively, the vertical alignment of the segments becomes more complex and less clearly defined. Like drawings on separate overhead transparency films, the music of the three instrumental groups was shifted and repositioned horizontally during composition, until maximum clarity was achieved (Figure 14.) An additional technique used in Meta-variation II is the simulation of the ‘Doppler effect’. It involves the sliding downwards in quarter tones of a short segment of music accompanied by a corresponding reduction in dynamics (Figure 15.)

Figure 14

Figure 15

Meta-variation III (Variation IV) is mostly a conversation between the two pianos, one played by a sequencer and the other by a pianist. Here too, the music has been put together by using the musical equivalent of horizontally sliding overhead transparencies. The live piano plays the original music broken down to small segments in order to accommodate the time stretched music of the MIDI piano. The music for the MIDI piano employs interval stretching 2 (Figure 11b) and, in contrast to the ‘rubber band’ principle described earlier whereby interval stretching is accompanied by rhythmic contraction, here the time element stretches as well. All attack positions and durations are multiplied by 1.5 which results in dotted note values replacing normal ones—dotted eighth notes instead of normal eighth notes, etc. A number of notes are ‘erased’ from both piano parts, so that the combined sonic result is more transparent and harmonically more meaningful. There is no a priori rationale for which notes are erased and which are retained; the selection is purely intuitive (Figure 16.)

Figure 16

Meta-variation IV (Variation V) is dominated by the skipping record stylus effect already described in the Stylus section of this paper. Driven by the MIDI piano in a fast sixteenth note motor rhythm, the music alternates between the original and interval stretching 2 (Figure 11b), but does not use any stretching or contraction of the time element. The technique of moving the right hand barline to the right or to the left so that the time trap expands or contracts by an eighth note after each repetition is used extensively throughout this meta-variation. Like in Stylus, sometimes the time trap is reduced to a single eighth note. At a rate of 294 eighth notes per second, this is the most difficult section of the piece to perform. For the premiere performance in 1992 this particular meta-variation was left out, due to inadequate preparation time.

Like in Meta-variation III, Meta-variation V (Variation VII) starts off as a conversation between the two pianos. This time the music for the live piano is the inversion of the original music played by the sequencer on the second piano. The center point for this vertical mirroring is the D sharp above middle C. The reason for this choice is that the two most important notes in this variation, the tonic G and its third B will thus mirror onto each other—G inverting to B and vice versa (Figure 17)—while the tonic chord of G major will invert into an E minor chord, which in turn is a secondary chord in the key of G major. Initially the root music and the inverted music alternate at one measure intervals, but progressively this rate of interchange becomes freer and more intuitive. Overall, the points of change between the original and the inversion are those where the music leads naturally and intuitively into its mirror image. During the repetition of each half of the variation, each piano is joined by a duet of other instruments—an oboe and bassoon with piano I and a violin and cello with piano II—and there is inevitably a more complex relationship between the reflected parts as a result of this addition.

Figure 17

To the extend that Meta-variation IV could be called the ‘Stylus’ variation, Meta-variation VI (Variation IX) could be called the ‘Equivoque’ variation. It uses a technique similar to the quantitative modulation in Equivoque, albeit slightly more complex. Like in Equivoque the two predominant ‘tonalities’ lie a tritone apart from each other, in this case G and D flat. Very often, however, these tonalities ‘slide’ around their center by a semitone, so G major can temporarily become A flat major or G flat major before recovering its original center. Due to the close positioning of these three keys, the aural effect is that of a ‘wobbly’ gramophone record where the tonal center constantly fluctuates upwards or downwards giving you a feeling of acoustical ‘seasickness’. Time wise, the music of the two keys alternates at sixteenth note intervals. Unlike Equivoque, where this alternating relationship persists throughout, here the quantitative modulations are brief: for the most part you hear clear definitions of either one key area or its ‘corruption’ a half step higher or lower. To make the music more readable, when a quantitative transference between two keys take place, a measure of 1/16 is inserted in the music, so that the new key, normally in an off-sixteenth note position, will fall on the downbeat of the next measure (Figure 18.) Except for these sixteenth note insertions, the only other rhythmic treatment in this meta-variation concerns the woodwinds which often repeat material presented by other instruments in rhythmically compressed triplet configurations.

Figure 18

The final meta-variation of Volume I is Meta-variation VII (Variation XII.) Here the ‘rubber band’ idea of combining interval stretching with rhythmic contraction is implemented in its fullest. In Figure 19 we have three statements of the same music, first in its original form by the brass at measure 4, then in its first distortion by the oboe, bassoon, French horn and strings at measure 5, and finally a further stretching/contraction by the flute, clarinet, bass clarinet and MIDI piano at the last eighth note of measure 5 and the downbeat of measure 6 (check this example against the table of intervallic stretchings (Figure 11a and 11b) and against the table of rhythmic compressions (Figure 12a, 12b and 12c.)) Each half of the variation is presented in this manner with small phrases of the original material followed by two successive distortions of it, but as the music progresses, there is increasing overlapping of these fragments, although never to the point where the formal clarity of the music is obscured. When this formal process reaches the half point, there is a repetition of the first half of the Bach material scored for the whole ensemble. This time the variation is played in its original form, uninterrupted and not tampered with, except for the heterophonic treatment of some of the lines in the orchestration. The same scenario is followed for the second half of the meta-variation, ending in a rather grand manner by utilizing all the forces of the orchestra.

Figure 19

So far, all the discussion has concentrated on the compositional aspects of The Go(u)ldberg Variations. This work, however, was conceived as a tribute to Glenn Gould and, as such, it was meant to incorporate aspects of interpretation—Gould’s interpretation of the Goldberg Variations—into the compositional process. Some aspects of interpretation, like his approach to ornamentation, can be conventionally notated. The trills of the Aria and of Variation VII (Meta-variation V) were written out in the score just as Gould played them in his 1981 recording. But that was clearly not enough. I wanted to incorporate into the composition all the aspects of Gould’s interpretation which could be extrapolated from the recording, not just the ornamentation. Furthermore, this enterprise would be necessarily limited to the music for the MIDI piano, where not only the composition but also aspects of interpretation could be programmed in advance. For all other instruments one would have to instruct the players to play their parts in the ‘style’ of Glenn Gould and hope for the best.

During the early 1990’s, Ernest Cholakis, a Toronto based DSP programmer, was developing what commercial music software developers call "groove templates". Those familiar with music sequencers, our present day music word processors, are probably also familiar with quantization templates. Used mostly in making commercial MIDI drum tracks sound more life-like, the principle of quantization templates is that you apply such a template on a ‘square’ MIDI pattern, which in turn causes the individual events of this pattern to be shifted slightly from their original positions and end up sounding more like the way they would sound if they were produced by a live player. Incorporating common performing mannerisms into a lifeless MIDI drum pattern, is like injecting life into it: it begins to "groove", perhaps by having the kick drum anticipate the beat somewhat and the snare drum miss the second and fourth beats by a few milliseconds. Even though Ernest was busy developing such templates for the commercial market, he was keenly interested in exploring the application of quantization templates on other instruments than drum sets and for longer durations than cycles of two, four or eight measures which was then the commercial standard. When I approached him with the idea of applying his quantization templates on Glenn Gould’s recording of The Goldberg Variations, he became quite excited and went to work right away.

First he worked on the Aria. He tracked the deviations of Gould’s performance against a reference tempo for the entire Aria and created a tempo map with input values for every sixteenth note of the music. For purposes of illustration the tempo map of Figure 20 is the result of sampling Gould’s performance at quarter note intervals. One can see not only the minute tempo fluctuations of the music, but more importantly an interpretive ‘plan’ which has to do with starting each half of the Aria at about a tempo of quarter note = 35 and then progressively slowing down to as low as quarter note = 25 at the end of the first half and considerably lower at the end of the second. This process of slowing down the music may have a lot to do with the sense of ‘timelessness’ one experiences while listening to this recording.

Figure 20

While Ernest was working on the tempo map, I created a mechanical MIDI version of the Aria on my computer. I entered Gould’s ornamentation by simply listening to the recording and, as soon as Ernest handed over the tempo data, I created a tempo track similar to the one of Figure 20, but with sixteenth note resolution. Next, Ernest created a velocity map of the recording of the Aria. To do that was trickier than it sounds particularly back in 1992, because one would need a velocity for each separate note in a chord, not just the velocity of each composite vertical event which is considerably easier to extrapolate, but would be of very little use in reproducing the voicing of individual lines that Gould applied to his performance. Ernest sampled the Aria several times, each time playing the music through a digital band pass filter which would only allow through frequencies in a band one semitone wide. He did this for each semitone of the piano range used in the Aria, that is from the low cello D to the B nearly two octaves above middle C. He ended up with a velocity map for every single semitone within this range. By combining the information of these maps, it was possible to reconstruct the individual voicing of chords in the recording and come up with something which was fairly true to the finger action applied by Gould on the piano. Figure 21 shows a composite dynamic map of Gould’s recording of the Aria. Here too, for purposes of illustration, samples are taken at quarter note intervals. Each cycle of the saw tooth wave at the bottom of the table represents one measure of music, the peaks representing the downbeats. Like the tempo map of Figure 20, this table too shows the outlines of a plan, clearly conceived and executed. Both halves of the Aria are treated as dynamic arches, but the second one is overall considerably louder than the first. What is quite astonishing, is that the dynamic peaks of the music increase in intensity along an almost straight line throughout the Aria, the result of incredible control over finger and hand musculature. It would have been impossible for a pianist of lesser caliber to precisely execute so complex an interpretive strategy, whereby you have step-like dynamic ‘transposition’ from the first to the second half of the music, and a continuous linear volumetric increase of the emphasis points throughout the entire Aria.

Figure 21

The combined information of Figures 20 and 21 reveals an interesting insight about Gould’s approach to the Aria. There is an overall tendency in the music to slow down and at the same time get louder. Psychoacoustically, these ‘inverse’ processes—we tend to associate slower with softer and louder with faster—balance each other in an admirable way: the ‘energy’ lost by slowing down is compensated for by energy resulting from a corresponding increase in amplitude. This way Gould manages to keep the performance energy of the Aria consistent throughout, and at the same time create a sense of timelessness by causing a sympathetic slowing of the biorhythms of the listener. As I already said, this happens without any perceivable loss of energy which would have made the whole process anti-climactic. I wanted to incorporate these and similar performance strategies developed by Gould for his interpretation of the Goldberg Variations into the composition of The Go(u)ldberg Variations. The process, of course could be infinitely refined. Had Ernest, for example, not only recorded velocities, but also charted the events on the time domain individually in each of these half-step-bandwidth passes with a high degree of accuracy, we would have been able to know the articulation of each chord—how was the chord arpeggiated in microtime—beyond the point where such knowledge is empirically possible. This in turn would have created a completely different ‘phrasing’ for these chords, as velocities are very much dependent on the sequence of articulation of the chord notes in microtime. However, we were much more interested at the time in exploring the idea of ‘resurrecting’ Glenn Gould as a performer on a Yamaha Disklavier for the inauguration ceremonies of the Glenn Gould Studio in CBC’s Broadcast Centre in Toronto, than in implementing this process to its logical conclusion. The pressures of the project deadline made the latter impractical.

Once the tempo and velocity maps were in place we spent several hours at Yamaha’s Canadian headquarters testing the particular Disklavier piano which would be used in the performance. It’s all well to know how Glenn Gould played the specific Steinway used in the 1981 recording, but to be able to extract the same response from a completely different instrument was the only way we could create the illusion of a ‘resurrected’ performance on stage. We sent MIDI velocity commands to each key of the piano with values ranging from 1 to 127 and measured the decibel output of each string for each MIDI velocity value. With this information we calculated what velocity values would solicit on that piano the dynamics of the Glenn Gould recording and then proceeded to adjust the velocities on the MIDI file of the Aria accordingly. I then made similar modifications to the velocities of all the meta-variations which involved the Disklavier. I did not use tempo maps in the meta-variations, because that would have created insurmountable problems for the orchestra. The way I synchronized the Disklavier with the live orchestra was to have the sequencer send a click track through an earpiece to the conductor’s ear. She then made sure that the orchestra stayed together with the sequencer. To incorporate Glenn Gould’s rubati into this click track arrangement would simply have spelled trouble during the live performance. Synchronizing an orchestra to a click track meant that the conductor would have to conduct ahead of the beat to allow for the normal human reaction time of the performers. This would have been a difficult thing to do, if she was not absolutely sure about the exact placement in time of the beat she was trying to anticipate.

The effect of the combined tempo and velocity maps during the premiere performance was quite eerie: The Aria was played by the Disklavier piano alone in its original form. Hearing rather clearly Glenn Gould’s interpretation while watching the piano keys being depressed ‘on their own’ as it were, gave one the sensation that Glenn Gould was on stage performing, ten years after his death and several decades after he abandoned the concert stage. Unfortunately, this sensation did not last past the mechanical performance of the Aria on the Disklavier. In the meta-variations, the subtleties of interpretation on the Disklavier were overshadowed by the live pianist and the other instrumentalists in the orchestra who focused on the daunting task of addressing the formidable technical performance requirements of the work, before they could give any thought to questions of interpretation and dynamic control of individual notes. A superb ensemble of contemporary music specialists fell short of mastering the rather unrealistic requirements of the work. It would come as no surprise to anyone who knows the piece that to date The Go(u)ldberg Variations has not been performed again, in spite the—genuine, I think—admiration for it by conductor Lorraine Vaillancourt and the performers of Le Nouvel Ensemble Moderne. Had Glenn Gould been alive and sitting in the audience, he would had been delighted, no doubt, by how things turned out that night. The fact that a mechanical, ‘premeditated’ rendering had outdone a stage performance, one by superb musicians no less, would have been considered by him a more appropriate tribute to his memory than a composition written to his honor.

It was sometime later, as I was becoming more familiar with the writings of Glenn Gould and his reasons for abandoning the concert stage, when it occurred to me that here we have a classic case for the defense of a ‘technological’ performance over a conventional one. This is a defense which Glenn Gould would have most likely espoused. By ‘technological performance’ I mean one in which the sounds of the ensemble are sampled individually and are performed through MIDI. This isolation of individual notes as audible performance quanta which would be problematic for the rendering of most kinds of music is actually very appropriate for the mind set behind The Go(u)ldberg Variations, and also quite appropriate for Glenn Gould’s approach to Bach in general. His interpretive tendency was to isolate individual notes and then form continuous lines, not by blurring the boundaries between notes, but by assigning to every note its appropriate, perfectly calculated dynamic. Most of his contemporaries would bundle a group of notes into gestalts or gestures by legato playing and the use of the pedal. Gould hardly ever used the pedal. His preference for distinctive dynamic assignment to each note enables the ear to establish relationships between notes thus articulated, even if they happen to belong to aurally difficult to discern inner voices of the music. His resentment of what he has called the "non-take-twoness"9 of live performance stems from this almost compulsive desire to achieve the right dynamic profile of individual notes and sequences thereof, which can not be achieved consistently from beginning to end during a live performance.

Glenn Gould died a few years before the MIDI protocol was widely implemented by electronic instrument manufacturers. MIDI (Musical Instrument Digital Interface) allows for 127 different dynamic values on each note of the keyboard, a considerably finer gradation than the dynamic system of conventional notation. It also allows performers to experiment with various approaches to interpretation and hear how they would sound before investing a great deal of time and energy on any one approach. For example, one may experiment on how leaving ‘gaps’ between the notes of the outer voices may bring into attention the inner voices of the music, or how applying a specific rhythmic and velocity profile on a short motif may bring out this motif from the midst of a texture with no need to antagonistically ‘push the volume’. One is tempted to speculate on how Gould might have reacted to the MIDI protocol or to the more recent digital audio editing systems which have revolutionized our traditional concepts of recording. Would Glenn Gould, the producer, take over Glenn Gould, the pianist and totally devote himself to a new form of synthetic interpretation, dispensing altogether with the concept of real-time performance in a studio? I would venture that the answer to this question might possibly be in the affirmative. Throughout his professional career, he had been disillusioned with the inability of pianos to deliver his ideas about the music he was playing. His notorious search for the right Steinway to perform on while concertizing, his obsession with getting an instrument to produce the kind of sounds even the piano manufacturers did not think appropriate for their instruments—the case of the infamous Steinway One Seventy-four10 comes to mind—point to a person who would have seen MIDI, digital sampling and digital audio as the answer to his prayers. Gould had prophetically anticipated the interactive nature of listening in a home environment. He saw the listener actively involved in the interpretive aspects of the music by using gadgets which affected the listening experience. Listeners with a personal computer today can do much more than that. Not only they can be involved in interpretive interference, but in meta-compositional11 interference as well: they can change essential aspects of interpretation at home—equalization, tempo, etc.—but they can also reassemble the building blocks of the composition—themes, motives, whole sections of the music—in completely different configurations, thus implementing on their own different realizations of the virtual compositions I mentioned earlier.

Given these conclusions, and in view of the inherent problems of performing The Go(u)ldberg Variations live in concert, I have decided that the definitive performance medium for this work is a MIDI sequence playing high quality digital samples of real instruments. The intriguing aspect of such an approach is that, with such a scenario, one could apply Glenn Gould’s interpretation of the Goldberg Variations to instruments on which he has never performed. An electronically performed segment from The Go(u)ldberg Variations for flute which consists of material from one of the variations, but compressed in time could, for example, be combined with a similarly compressed tempo map extracted from the same section of Gould’s recording. If the flute segment also happens to be intervallically stretched, its ‘stretched’ notes can be assigned the velocities which Gould used for the recording of the original segment. The resultant music would sound related to, but significantly different than its source in the Goldberg Variations, yet both segments, the original and the derived, would bear equally the mark of Glenn Gould’s interpretation. Some global overall adjustments would be necessary to account for the fact that it will be a flute, an oboe, or a violin performing the new material, but this notwithstanding, one would still be able to discern Gould’s interpretation permeating the entire orchestra, not just the piano. Purists may argue that interpretation is instrument-specific and that these indiscriminate transpositions of interpretation from one instrument to another have only curiosity value. I think Gould would have disagreed. His interpretive approach was composition-specific, not instrument-specific, and he has sometimes gone to great lengths to prove this point12. I believe he would have found the prospect of a mechanical reproduction of his ‘style’ quite fascinating and, what is more, appropriate for Bach’s music. He says of The Art of the Fugue: "this magnificent indifference to specific sonority is not least among those attractions which emphasize the universality of Bach"13, and: "despite its monumental proportions, and aura of withdrawal pervades the entire work. Bach was, in fact, withdrawing from the pragmatic concerns of music-making into an ideal world of uncompromised invention."14

Beyond interpretation, Gould’s insistence on technological performance had a moral foundation: He said it best in an interview with Robert Hurwitz which appeared in the New York Times:

I happen to find all of the live arts immoral because one should not voyeuristically watch one’s fellow human beings in testing situations that do not pragmatically need to be tested. Those who stand in opposition to technology do so because they think that there is something inherently wrong with making life simpler than it ought to be, that if you can take a series of problems and reduce these problems by a splice, in some way you are performing an anti-human act15

If Gould was alive today and stood by this position, he would have found himself in a situation where it could be used against the very art which he perfected from childhood and which established him internationally as a twentieth century genius: the art of playing the piano. If the only reason for learning an instrument is to be able to implement one’s ideas about music which lie beyond sound, which he has repeatedly stated on many occasions as his only reason for playing the piano, one can accomplish this today by less cruel means than spending countless hours practicing scales on the piano or on any other instrument. The compositional ideas used in The Go(u)ldberg Variations were made possible in the first place by technology. If it were not for the ‘cut’ and ‘paste’, ‘stretch’ and ‘compress’ macro-algorithms of modern day sequencers, I wonder if I might have not followed a completely different path with this piece at the composing stage. To test, then, with a composition thus derived the limits of human endurance and concentration as it will inevitably be the case every time The Go(u)ldberg Variations is performed on stage by live musicians, is indeed a perverse act of voyeurism. I wonder if Glenn Gould might not even find my application of his ‘style’ on my mechanical rendering of the music as an equally perverse act, preferring instead the de facto anonymity of a ‘square’ MIDI rendering which guarantees the absence of any trace of subjectivity in the interpretation, even the possible traces which he himself was not able to eradicate from his performance. Maybe a ‘square’ MIDI performance would be the best homage to the man who raised musical objectivity to a supreme aesthetic and moral value. Then again, it may be that his often controversial statements about music interpretation should be read against the backdrop of excesses by his contemporaries and not at face value, certainly not the face value these statements may have today.

Whatever Gould’s reaction to incorporating him into my music might be, he would have probably approved of my compositional approach in The Go(u)ldberg Variations or, for that matter, the other Bach-based works discussed here. Whether he might have seen my own role in these works as that of a composer or that of a super-interpreter, in the grand manner of Leopold Stokowski, is less important than the fact that he would have considered these works the result of a legitimate vision of Bach’s music. The following homage to Stokowski’s conducting is in many respects more telling of Glenn Gould’s view of interpreters than necessarily Stokowski’s view of music:

Stokowski is involved with the notes, the tempo marks, the dynamic indications of the score, to the same extent that a film-maker is involved with the original book or source which supplies the impetus, the idea, of his film. So Stokowski’s performances, then, stand or fall by the degree to which he can infuse them with a sense of his own commitment to the project at hand16

Similarly, I wish The Go(u)ldberg Variations be judged by no lesser a standard. Inasmuch as the work communicates an intellectually distilled vision of Bach’s music or, beyond Bach, of virtuality in the compositional process, then it will succeed in its objective. In The Go(u)ldberg Variations, intellectual clarity is raised to a supreme value. As with Gould, subjectivity is rejected in favor of objective and of unambiguous focus on the pure compositional process. This work was a culminating point in my writing: with the subsequent works in the Bach series I turned away from this purely intellectual approach and focused on areas which Gould might have found objectionable: eclecticism, expression, and an approach to sound as a phenomenon inseparable from composition. 


Written in 1992 immediately after The Go(u)ldberg Variations, the Concerto for Flute and Chamber Orchestra is stylistically related to the former in several respects. The first and third movements are very much similar to it and to many of the works in the Bach series discussed so far. However the middle movement with its nineteenth century approach to melodic, harmonic and formal construction is quite different than any of the works discussed so far, and it points to a direction away from the purely cerebral and technical concerns of the previous pieces in the Bach series.

[The work was commissioned by ConAccord Canada with funds from the Ontario Arts Council for a program of my music which was to involve flutist Robert Aitken in a number of my works. It seemed appropriate that, since a great performer of Aitken’s stature and talent was to be involved in this concert, a new chamber work focusing primarily on him should be created for the occasion. Since that concert would have also included the performance of Equivoque and the rescheduled premiere of Stylus, it seemed appropriate that the new piece for Robert would be related to those two and be part of the same Bach series. As it turned out, due to unforeseen circumstances, this concert never took place. Partly due to the fact that there was no longer any specific venue for the performance of this work, and partly due to my own desire to write a substantial composition for such a great soloist, the original concept of the piece was significantly expanded. What started as a relatively brief work for flute and small chamber ensemble ended up as a lengthy concerto—346 pages of score, and over 35 minutes in length—in three movements for solo flute, oboe, bassoon, harpsichord and string orchestra. The work has not yet been performed, mainly because neither Robert, nor I have been able to find an orchestra which can afford enough rehearsal time for the formidably difficult orchestral part. For reasons which will become evident further on, a technological performance like the one proposed for The Go(u)ldberg Variations would simply not be a consideration for this piece.]

As most of the pieces in the Bach series delight in musical signatures, principally Bach’s and mine, the dedication page of the concerto is a signature acronym:

The Bach work of which my composition is a palimpsest is the Concerto for Flute and Strings in G minor, itself a compilation of movements from earlier concerti, the manuscripts of which no longer exist. Strictly speaking, my Concerto for Flute and Chamber Orchestra is a case of ‘double’ palimpsest, the first performed by Bach himself. Judging from the melodic writing and the instrumental range of the solo part, it seems that the two outer movements are probably from a violin concerto, while the middle movement must have been originally conceived for a solo wind instrument. It has been handed down to us as the Concerto for Harpsichord and Strings in F minor, an arrangement Bach probably made for his weekly Collegium Musicum concerts in Leipzig. In my own composition, I took into account the published harpsichord version of the original and the flute version recorded on compact disc by William Bennett and the Academy of St. Martin-in-the-Fields under Neville Marriner (London Jubilee 417 715 LM.)17 As in Equivoque and Stylus, here too, the entire borrowed work is embedded, however fragmented, within my own composition. In fact the outer movements of my concerto comprise entirely and exclusively of the manipulated Bach material. Not only has the original been preserved in its entirety, but the overall structure has been retained intact as well: The linear harmonic and thematic evolution is identical in both the original and the derivant works.

A system of intervallic stretchings and rhythmic contractions similar to The Go(u)ldberg Variations was implemented in the first movement of the concerto. Unlike The Go(u)ldberg Variations, however, where the various altered segments are superimposed on top of each other or overlapped freely, in the concerto they are always presented in sequence, so that the listener may easily become aware of the distortion lenses being applied to alter Bach’s music. In Figure 22, the flute and the orchestra play the first intervallic stretching of an one-measure-long segment from the Bach concerto at measure 35, and the second intervallic stretching of the same segment at measure 36. Finally the original segment is presented at measure 37 with the oboe playing the solo part. No rhythmic contractions have been introduced yet, in order to help the listeners’ focus on the technique of intervallic stretching before distracting them with any other compositional interference. In spite of the fact that its harmonic language is radically different from the music which precedes it, the material of measure 37 sounds like a natural consequence of the previous music. Our sense of the melodic and harmonic kinship of these three segments takes precedence over our sense of the presence or absence of conventionally defined tonality in each of them. When this tonality finally arrives at measure 37, its presence is not felt to be the result of collage, but of the linear unfolding of a musical idea which inhabits equally comfortably both tonal and non-tonal realms. In fact, in the tonally amphibian existence of this movement, conventional tonality functions as an anchor. It imparts to the music a distinctive sense of arrival, not unlike the sense imparted by the tonic chord in the common practice harmonic system.

Figure 22

Rhythmically, the compressions in the first movement do not adhere to the ‘rubber band’ simile which I described earlier; they occur independently of the intervallic stretchings. There is only one rhythmic compression scheme in this movement, a 3/2 compression. If I were to notate this compression as tuplets—in this case triplets or sextuplets—as I did in the first and last meta-variations of The Go(u)ldberg Variations, I would have needed to resort to ‘nested’ tuplets, because a great deal of the music is already in triplets in its unaltered form. As I have already explained, the reason I opted for the more complex notation in the former work, was that the rhythmically compressed segments were layered on top of one another. In the concerto, the rhythmically compressed segments are placed in sequence and, furthermore, they are contained within separate measures, so the rhythmic compression scheme can be easily expressed as metric modulation. Two tempi are in use: eighth note = 140 and its compression, 3/2 faster, which is eighth note = 210. Quite often when the tempo changes, the music is simply repeated: the listener’s ear is guided contextually to either rhythmic compression by keeping all other musical information redundant, or to intervallic stretching by keeping the rhythmic element constant.

The notation of rhythmic compression as metric modulation with constantly changing tempo markings is actually a compromise necessitated by the limitations of the music software used for the notation. The ideal notation would have been one which I developed in the late seventies, where it is possible to write metric modulation as non-conventional time signatures. Time signatures in western music are expressed in terms of fractions, the denominators of which are duple subdivisions of the whole note: two, four, eighth, sixteen, and so on. We do not use three, five, six, seven, or nine in the denominator, preferring instead to notate non duple subdivisions of the whole note as tuplets in the music. But, if, for instance, we wanted a group of two eighth notes from an eighth note triplet, and then we wanted to move immediately to another rhythmic structure, without the third eighth note of the triplet or a rest in its place, we would be up against a notational impossibility. Now, this eighth note within the eighth note triplet is in fact a ‘twelfth note’, because there are twelve of them contained within a whole note. Therefore, if we decided to call them twelfth notes instead of eighth note triplets, then we could have a measure containing two twelfth notes. This would be notated as two eighth notes in a measure with a time signature of 2/12. Let’s assume that the measures immediately preceding and following this 2/12 measure are 2/8 measures at eighth note = 140. There are two ways to notate this three measure sequence. One, the familiar method, would have the time signature remain constant throughout (2/8) and apply tempo markings (eighth note = 140, eighth note = 210, eighth note = 140) to each measure of the sequence. The other, a notationally more elegant way, would have the tempo remain constant (eighth note = 140) and change the time signatures for each measure (2/8, 2/12, 2/8.) Both notational solutions would sound identical in performance. I have used the latter notation on a number of earlier works which exist as pencil scores, and I feel quite disappointed that most music software developers do not offer these unconventional time signatures as an option in their programs. Think, for example, how much easier the passage of Figure 23 would have been to read, if metric modulation was notated as a succession of time signatures. With a constant tempo of eighth note = 140, the sequence of time signatures would read: 1/12, 4/8, 2/12, 2/8, 4/12, 4/8, 2/12.

Figure 23

With the exception of the interpolation of stretched ‘clones’ between segments of the original material and various repetitions of material thus interpolated, the first movement of the concerto follows closely Bach’s formal outline. In fact, the interplay between the original and the derivant material, in essence between baroque tonality and its absence, and the degree of preference given to each serves to delineate Bach’s intention within a contemporary context to an even greater degree than Bach’s semantics alone would be able to do. This seemingly arrogant statement certainly requires explanation, but the explanation is intimately entangled with a view on how contemporary audiences listen to the music of the past, so allow me to divert from my course momentarily to tackle this first.

The context of listening to music in the eighteenth century was radically different than what it is today. Since for eighteenth century listeners the common practice harmonic system represented the only possible harmonic universe, they engaged into the composer’s travel plan in a way that modern listeners may or may not do, depending on the listening circumstances and on their attentiveness. For present day listeners the music of the common practice era is like the Star Trek ‘holodeck’; it is virtual reality, not inescapable reality. Captain Picard and the crew of Enterprise may, for their entertainment, engage in a ‘Wild West’ American frontier adventure, for example, but the adrenaline rush is not as intense as it would have been, had they actually lived in the time and place they choose to temporarily inhabit, and where the "life-threatening" situations they entertain themselves with are real. The intensity of any experience is inversely proportional to the number of ‘escape clauses’ it allows: this is the qualifying difference between virtual and real, unless you want to get into the gray area of more esoteric science fiction writing18.

There were no escape clauses in the listening experience of eighteenth century music lovers. The common practice harmonic system, the only possible universe at the time, was a musical metaphor for their reality. Learning how to navigate in it was for them cultural survival practice. We, on the other hand, easily can, and often do, disengage. We may choose to suspend disbelief, as we do when we attentively listen to the music of the past, and allow ourselves to follow the internal progress of the composition. However, we are aware at all times that what takes so long for the composer to get to can now be arrived at instantaneously, at ‘warp speed’, as it were. This awareness, however repressed it might be during the listening process, sabotages any incentive provided to us by the composer to stay tuned to the end.

To counter this problem, the music of the past can be offered to contemporary listeners as a musical ‘guided tour’, a compositional commentary which engages the listener in a musicological discourse during the listening process. This is an outgrowth of a paradigm that has produced works which are so much ‘about themselves’, they ultimately become self-referential. The latter is particularly true of the outer movements of the Concerto for Flute and Chamber Orchestra, and of all the works discussed so far, in that, instead of being outwardly dependent through selective quotation, they contain the entire reference work within them and proceed to anatomically expose it with tools which are included in the package. It is in a sense like music theory through sound. If some listeners find the first movement of my flute concerto a bit too didactic, it is because the music carries with it—within it—the ‘user’s guide’ to its own vocabulary and syntax.

I am now ready to proceed with the explanation of my previous, seemingly preposterous statement about Bach. In the first movement of the Bach concerto, the main theme is associated with important harmonic arrival points in the music. The three most important such points in my rendering are the tonic area of the beginning (measures 1 - 20 in the Bach original), the tonicized subdominant exactly two thirds into the movement (measures 71 - 78), and the final return to the tonic at the very end (measures 109 - 116.) Each of these three thematic occurrences are left intact in my rendering without any further compositional interference, so that they stand out as three extended tonal ‘clearings’ in the surrounding harmonic forest. Throughout my treatment of the Bach material, the relationship between tonality and its absence is roughly equivalent to the relationship between the tonic and secondary harmonies in the Bach concerto. The sense of ‘comfort’ or ‘familiarity’ one experiences with the tonic chord or a tonicized secondary harmony in common practice music is translated into the analogous sense of ‘comfort’ with our present day experience of tonality as a whole in the midst of other harmonic and textural possibilities. To put it another way, the anchoring role that the cadential tonic chord plays for eighteenth century perception is performed today by tonality as a whole. In the post-serial, post-aleatory perceptual milieu of today, tonality is a comfort zone, an identification with protected infancy, or with prenatal, womb-like protection. Similarly, for the eighteenth century, the tonic had a similar archetypal identification. It too, stood for a time in the past: the pre-tonal, perhaps pre-polyphonic past, before the tonal adventurism of the seventeenth and subsequent centuries made harmonic travel possible. In the Concerto for Flute and Chamber Orchestra, these archetypal associations of ‘then’ and ‘now’ are intentionally played out. When the main theme is restated in the tonicized subdominant at measure179 of the first movement, it follows a tonally and rhythmically unstable section. It’s instability increases exponentially as we approach the restatement of the theme; there are rapid metric modulations and erratic interplay between the quoted tonal material and its non-tonal stretchings. When we finally get to the theme, it is presented in a context of extended tonal and rhythmic stability, that is the stability of the Bach original. The semantic significance of this moment is of equal magnitude as the exact same moment in the Bach concerto must have been for eighteenth century listeners: my rendering is designed to solicit the same response from present day listeners that Bach must have been able to solicit from his contemporaries.

The second movement is a radical departure from the premises of the first. The melody, one of Bach’s loveliest slow melodies, begged to be treated in a manner which acknowledged its beauty. Unlike the opening melodies of the two outer movements, which are essentially the product of the harmonic movement beneath them, this one has a life of its own. Its linear profile is so pronounced, that it could be conceivably played by a solo woodwind instrument without any orchestral accompaniment. It contains within it all the essential characteristics of its accompaniment and it looses very little of its inherent beauty when played alone.

My musical treatment of this movement was the first symptom of a change in my approach to Bach. Up until then, I had looked at Bach through the eyes of Glenn Gould, even before I was consciously aware of what that meant exactly. I saw spiritual greatness in the asceticism with which Bach applies himself to solving difficult contrapuntal problems especially in his large scale fugal works, the Well Tempered Clavier and The Art of the Fugue. I have recently come to believe that the display of intellectual prowess in one’s work is artistically and spiritually arrogant, and according to this belief it would have been easy to see Bach’s engagement in the cerebral pursuit of contrapuntal problem solving as a theologically ‘proud’ act, far removed from the puritan ethic which Gould so much exults in Bach and identifies with. The fact that even today I do not hear Bach’s music this way has a lot to do with the composer’s humility and the higher musical and spiritual purpose to which he has subjugated his urge for intellectual display (I will have more to say about that later, when I discuss the last fugue from The Art of the Fugue in conjunction with my Farewell to Bach.)

Bach’s music points towards a sense of humanity shared by all of us; it does not merely come off as a testimony of personal contrapuntal accomplishment. It was this sense I wanted to depict in the slow movement of the concerto. To do so meant that I would have to disengage from the controlling objectivity of discipline and intellectual rigor and abandon myself in the subjective world of emotional engagement. This approach was also consistent with a subplot in the Concerto for Flute and Chamber Orchestra which was to demonstrate Bach’s relevance to the development of musical ideas since his death all the way to the present. The unmistakably nineteenth century concerto approach of the second movement does necessitate the presence of a great deal of original material of my own, but this material always rides on top of larger expanses of material from Bach. The movement follows the progress of the original up until the diminished seventh chord halfway through measure 18. Instead of resolving to the cadential 6/4 chord, as is the case in the original, my music moves into a development section of over 200 measures in length. This long development eventually works its way back into the diminished seventh chord which gave rise to it in the first place, and the second time around it cadences as it should and concludes with a slightly extended rendering of the remaining three and a half measures of Bach’s music, which in the new context sounds like a recapitulation. At the level of macrostructure, the listener has the sense that the entire development section is like a door which suddenly opens within Bach’s music. Walking through this door takes you into a completely different era: the nineteenth and early twentieth centuries. In accord with the thinking of this era, Bach’s baroque material becomes the stuff of symphonic development. I will not get into the particulars of how the development section unfolds organically; this would normally be the focus of analysis, and our focus here is on compositional techniques and how these techniques are applied to the music of J. S. Bach. It should suffice to say that the predominant technique used throughout the development section is stylistic transformation. The same melodic material is quite often repeated intact, while the harmonic and rhythmic content changes giving the material a new stylistic identification. Further to my earlier discussion about the role of tonality in these works and in contemporary tonal perception as a whole, this journey through various ‘styles’, chronologically and/or geographically defined, is a contemporary transposition of the concept of harmonic travel in baroque and early classical music. The semantic significance of movement through key areas is here supplanted by that of movement through defining and recognizable moments in western music history, from the baroque to the twentieth century. The sonic result is not a stylistic collage, but a continuous transformation, as seamless as the harmonic progressions of common practice music which in essence it replaces as the operating compositional language.

At the opening of the second movement, each phrase of the original melody on the flute is repeated by the oboe. Besides being rhythmically compressed, the repeated melody is also fragmented to smaller motivic units. At the fault lines between these units, the music shifts abruptly by transposition either up or down. The entire music shifts this way—both melody and harmony—creating harmonic relationships at either side of the fault line which are not indigenous to baroque music. In the ‘exposition’ section of the this movement this seismic displacement in the melody and harmony is quite common . Figure 24 illustrates this technique in some detail. Measures 10 to 16 contain the original Bach material. At measure 17 the oboe plays the same melody, only now each motif is transposed at a different level and the harmony follows suit. While the repetition is a 3/2 rhythmic compression, the viola continues playing in the original tempo by playing longer note values. The viola part continues undisturbed at measure 21, while the rest of the orchestra returns to normal—uncompressed—rhythm.

Figure 24

The third movement of the concerto is in 3/8, an ideal meter for 3/2 rhythmic compression, which is again the only rhythmic compression used. The compressed content of a 3/8 measure can easily be expressed as a sixteenth note sextuplet or an eighth note triplet in a 2/8 measure with no need for metric modulation whatsoever. This movement utilizes extensively most of the techniques already discussed in this paper: The intervallic stretching and rhythmic compression of The Go(u)ldberg Variations, and the ‘stutter’ effect of Stylus. Also from Meta-variations II and VI of The Go(u)ldberg Variations, I borrowed the ‘Doppler effect’ and the ‘wobbly turntable’ techniques respectively. Here, these two techniques are combined to produce an even stronger sense of tonal disorientation. For example at the start of the movement, just as the flute soloist and the orchestra end the first half of the first phrase of the A minor theme, the last five sixteenth notes of the music are caught into a 5/16 time trap in the strings. At each repetition the string texture is transposed downwards by a quarter tone. At the other side of these repetitions, the soloist and the orchestra pick up where they left off before the time trap, but this time they are in A flat minor as a result of the quarter tone detuning. A few measures later, after the music undergoes some mechanical adjustments of the ‘faulty stylus’ type, they re-emerge in A minor as in the beginning. This happens several times during the opening part of this movement. What begins as ‘accidental’ bitonality, seemingly resulting from a playback malfunction of the performing apparatus, becomes before too long a tonally dual personality with each of its components making distinct claims to the music. At the first cadence, both keys fight for harmonic conclusion in their own domain. At measures 47 - 50 of Figure 25 the soloist and the orchestra examine together both cadential possibilities. At measure 51, however, the orchestra makes a unilateral attempt to cadence in A flat minor; it is stopped from doing so by the soloist at measure 52 who insists in a cadence in A minor and by measure 53 convinces the orchestra to follow suit.

Figure 25

Figure 25 is a good illustration of not only the harmonic language of this movement, but, more importantly, the way known harmonic processes from the common practice period are used as metaphors for larger processes, subliminally familiar and of interest to the listener. At the risk of investing too much importance to a single cadence, I find that this particular example epitomizes in musical terms a conflict situation between an individual and the group. The individual is in a position of strength: he/she defends the ‘home’ key, therefore their argument is a historical one, the defense of status quo. Musically—for the listeners, that is—this argument carries significant weight: the listeners expect the music to end in the key it started, and they unwittingly identify with the soloist’s plea. The question of course is: can the listeners impart all this directly from the music without resource to extra musical explanation? Well, their instinctive translation of musical processes into something larger and outside the music altogether is the only means they have for making sense out of contemporary music. In an era where there is no common musical practice, cognitive listening is by necessity semiotic. Every time musical processes begin to ‘mean’ something semiotically, they begin to make sense to non musically educated listeners. In many instances, particularly during the middle of the twentieth century, you have composers who expect the listeners to follow the composition on the composer’s terms, that is according to the logic which organized the music in the first place, but, instead, the listeners end up with a semiotic understanding of the music, which perhaps was not foremost in the composer’s mind at the time of the writing. Most often the average listeners’ aversion for serial or aleatoric music is the result of their semiotic identification of this music with psychologically dysfunctional states of mind. Yet, the presence of such semiotically ‘dysfunctional’ music in a context where it is part of a larger struggle between tonality and atonality, to name just one example, might become a metaphor for the kind of struggle a less-than-perfect individual experiences personally and, therefore, identifies with during the listening process. If perfect harmony is too good to be true, a dialectic between tonality and its absence may be a truer reflection of the listener’s inner struggle between aspirations and everyday reality.

In the Concerto for Flute and Chamber Orchestra, even more so than in my previous works in the Bach series, the semiotic content of musical processes is the prime building tool of musical structures. It is active not only at the scale of individual events or gestures, like in the previous example, but also at the larger level of overall structure. Some of these larger semiotic associations are, of course, already imbedded in the traditional notion of the concerto, especially the nineteenth century development of the form. The individual standing defiant against the impersonal group, an anti-industrial metaphor, is very much the psychological undercurrent of the greatest examples of this genre during the romantic era, and is to a certain extend explored in the middle movement of my concerto. There are other, less obvious associations which are explored here as well. In the outer movements, the matter-of-factness approach to the relationship between the soloist and the orchestra is neo-baroque in character, not as confrontational as in the classical or romantic concertos. The possible differences between the soloist and the orchestra are ‘mediated’ by two other woodwind instruments, the bassoon and especially the oboe, which throughout the course of the concerto ‘explains’ the flute material, and establishes a relation relationship to its source. At the flute cadenza of the last movement, the soloist acknowledges the special contribution of the oboe and the bassoon by playing in a trio fashion with them at sections of the cadenza re-introducing material from the first two movements. There is an attempt here to make sense out of three movements which in the original derive from more than one concerto and have dubious musical relationship to one another. This step-sibling genetic difference between them has been further amplified by my dual compositional treatment of the middle versus the outer movements, but, at this point, the soloist and the other two woodwinds are attempting to build a coherent family out of this adopted material. A further such reference at the very end is also a characteristically Beethovenesque ending: just before the final tonic chord, the music is left hanging in the dominant. What follows (Figure 26) is a version of the main theme of the first movement in the major mode, the tonicized dominant, again by the flute, oboe and bassoon. The introduction of such radically different material at this point implies that the music is moving into a new section of the background structure: different key, slower tempo, different material following a long section where the key relationships, the tempo and the material have remained constant. Yet, at measure 707, where the flute melody moves—as it would normally—to the E flat, the third of the subdominant chord, this E flat suddenly becomes the third of the tonic chord—the previously tonicized dominant becoming again a functional dominant. This gets us back to the beginning of measure 701 and to a surprise final cadence.

Figure 26

This is not unlike the type of endings Beethoven favors in many of his works. Take a look, for instance, at Figure 27 which shows the closing measures of Beethoven’s "Pathétique" sonata. The 24 measures preceding this example are essentially in the C minor key area signaling the impending conclusion of this straight forward rondo movement. Then, at measure 198, a ‘Neapolitan’ sixth chord is contextually reinterpreted as the subdominant of A flat major. What follows is the first presentation of the main theme in the major mode. Up until this point, the main theme has been presented each time in its entirety. The expectation is that this should be the case here too, more so, in fact, because the change of mode establishes this presentation as particularly important. This expectation is compositionally calculated: the dynamics are very low, and most performers would probably play measures 201 - 207 at a slower tempo, playing along with the deception that we are at the beginning of a new large structural unit in the piece. At measure 209, however, we become aware of the parenthetical—and misleading—nature of the preceding material and the music reaches immediately a conclusion, before the listener has enough time to rethink the prior course of events.

Figure 27

In Figures 26 and 27 the musical strategies are identical: they entail a temporary disorientation of the listener’s navigational apparatus, so that the final conclusion comes as a complete surprise; they also require a quick conclusion after the deceptive, parenthetical statement, so that the listener does not have enough time to catch up with the composer’s bluff until it is too late, a compositionally desirable situation which has both information and entertainment value.

Of course, to Beethoven’s detriment, the overexposure of his music has rendered many of these informational strategies ineffective. To listen to music in an information-rich manner, one needs to have a basic understanding of the composer’s style, but preferably no prior knowledge of the actual composition. This is the ideal setting for the communication of information-based structures, and it is an almost impossible condition to be met with the music of Beethoven or that of most major classical composers of the eighteenth and nineteenth centuries. When that music was written, the composers had no way of predicting the electronic mass dissemination of their work from the middle of the twentieth century on, and how the familiarity bred by this dissemination would destroy one of the most important aspects of their communication language. It is no wonder that performers most sensitive to the new communication technologies, like Glenn Gould, became instinctively attracted to music whose language is not threatened as much by this type of familiarity, that is music whose linear information content is of less importance, such as the music of Bach and other baroque composers. The more musical value is attached to the harmonic progression and its information content—for example, the more the musical meaning of a composition depends on the deceptiveness of a deceptive cadence—the less information will be there after the tenth or twentieth hearing. During such a high numbered hearing, one may get the impression that a point is being made over and over, and disengage from the music on that account. The re-emergence of baroque and earlier music during the past thirty years with audiences drawn primarily from the communications economic sector is a symptom of this inherent problem which classical and romantic music has been facing during this same period of time. A listener in an information saturated urban environment, is more prone to be attracted to low-information music such as minimalism, baroque, or Gregorian chant, than to high-information works whose information content is already known. The latter can no longer engage one’s attentive listening, since, for reasons already explained, their engaging mechanisms are by now ineffective; they cannot be listened to as background music either, because their composers’ attempt to consciously engage the listener’s attention makes them distracting and ineffective in that capacity too.


This brings me to the last piece in this series. Farewell to Bach was composed in 1998, five years after the concerto—five long years, as both my music and my ideas about music have undergone significant changes in the meanwhile. What precipitated these changes was an increasing awareness of a paradigm shift rapidly taking place in western culture of a magnitude comparable to the paradigm shift from the Middle Ages to the Renaissance. I have discussed this paradigm shift in an earlier paper19, so I will concentrate on those aspects pertinent to the present study.

I believe that the external reason for this paradigm shift—in music at least—has something to do with a radical drop in the value of musical information, which in turn has been engendered by the overproduction of musical products of the classical variety. At this scale of production—recordings, live concerts, videos, television specials, etc.—the finite, non renewable classical repertory can be drained of consumer value, in spite of its temporary blood infusions by ever improving recording technologies and more radical interpretations by a newer breed of performers on rediscovered period instruments. While these cosmetic operations give classical music some temporary lease on life, its long term future is bleak, unless the emphasis shifts soon on new creation. New creation alone is not a long term answer either, unless a deeper understanding of the role which art is called to perform within a rapidly changing social and cultural context is shared by a critical mass of creators and re-creators in the performing arts. There is evidence to suggest that this is happening already, but a great deal of this change is market driven and, as a result, essentially indistinguishable from the similarly driven popular culture. Once the avant garde lost its exclusive hold on the academic and new music scene, the pendulum swung to the other extreme. Elitism was transformed overnight to populism to such a degree that many questioned whether pop music and new classical music might not become one and the same in the near future.

The answer to this question is, I think, both yes and no. Yes, inasmuch as contemporary creation decides to cater to market pressures alone, as it is already doing in some quarters. No, inasmuch as this paradigm shift is also driven by an internal cause which, although ultra-sensitive and responsive to the needs of every listener, is not market driven, in fact it draws a clear distinction between real human need and fabricated market demand. This cause is elitist, because it seeks to address the highest common denominator among listeners, and it is also egalitarian because it also seeks to address the highest common denominator, unlike commercial music which, for reasons of convenience and profit expediency, only seeks to address the lowest common denominator. I have identified this cause elsewhere20 as an urgent need for sanctuary among urban dwellers besieged by our infoculture. A sanctuary can be understood as different things by different people, but most people would associate it nowadays with an environment —sonic or otherwise—protected from unsolicited information. Music which revels in information silence and information quietness, is a musical sanctuary, a sheltering sonic environment, within which the listeners can lower their defense shields, recharge their battered sensory apparatus, and allow a healing process to take place.

Actually, the psychological identification with a sanctuary has to do with more than just information content. It is also associated with cultural conditioning and, not least, with the nature of the sound itself. By cultural conditioning, I am referring to certain spaces, like a church, which one associates with prayer, meditation and abstention from mundane concerns. Whether we are conscious of it or not, when we enter a place of worship we leave a great deal of our information overload—our mundane everyday pressures—at the gate. Churches are usually large, ambient spaces in which sound behaves differently than in a recording studio or a small concert hall. In a church, most of the sound does not reach us from the source, but through reflection. In a non-ambient environment, our listening mode is instinctively that of the primitive hunter. To identify the source of the sound and classify it as coming from a predator or a prey was a question of survival for primitive man. Even though we have stopped being hunters or hunted long ago, subliminal race memory comes into play when we engage our attention upon a sound source, and with it comes the inevitable tension and stress of the primitive hunter. In contrast to this stress, ambient sound is associated with womb-like safety, because it articulates the architecture which surrounds us and protects us from the outside environment. Given these subliminal associations of ambiance, Farewell to Bach was composed in a way that it will sound best in a church, making ambiance an integral part of the orchestration.

[Farewell to Bach was composed during the Spring of 1998.] Even though it was to be the concluding composition to my Bach series of works which in their majority are driven by rather cerebral interests, its compositional approach is quite different from those earlier pieces and similar to my more recent music, where I am concerned with the spiritual and experiental aspects of musical communication. [It was commissioned by the Scotia Festival of Music, an annual festival in Halifax, Nova Scotia, for Tafelmusik, Canada’s preeminent baroque orchestra with funds provided by the Canadian Broadcasting Corporation and the Nova Scotia Arts Council. Farewell to Bach was originally scored for a baroque ensemble consisting of two solo violins, flute, recorder, oboe, oboe d’ amore, bassoon, a small string orchestra and harpsichord and was premiered in this configuration by Tafelmusik on June 6, 1998 at St. Andrew’s United Church in Halifax along with the harpsichord version of Equivoque and the string quartet version of Stylus. A subsequent version was made for modern instruments: two solo violins, flute, recorder, oboe, English horn, bassoon and strings, so that the piece may have a life beyond the comparatively small world of period instruments.]

Farewell to Bach is based on Contrapunctus XXIII, the last fugue from the Art of the Fugue. This is a fugue in three parts—three separate fugues on three different subjects not encountered elsewhere in the larger work—which follow each other with no interruption. Beyond the third fugue, there are a few measures where the subjects of the three fugues are presented together in counterpoint and then the work is abandoned, presumably because Bach died at that point in the composition. This is a fascinating piece of contrapuntal work and there is a lot written about it including a number of proposed conclusions for this last of the master’s works. As an undergraduate student I had studied this particular fugue thoroughly, as well as one of the conclusions proposed by Douglas Green, my counterpoint professor at the Eastman School of Music. Later on, while working on my doctorate at State University of New York at Buffalo, I wrote a term paper investigating some contrapuntal properties—particularly inversion possibilities—of some of the subjects of this triple fugue, enticed, like everyone else, by its incompleteness. When I first embarked on the Bach series in the mid-1980’s, I had planned to attempt my own conclusion to this open ended puzzle when the time was right. That time finally did come thirteen years later, but the prospect of completing the last fugue no longer had any artistic appeal for me. I was getting ready to abandon the Renaissance Paradigm, best epitomized by the music of Bach, and, while it felt appropriate to do so with reference to a composition which Bach likewise abandoned, the possibility of completing this ambitious architecture seemed to be an exercise in vanity rather than artistic judgment.

At that time, I was less interested in this work’s inner structural workings, its contrapuntal framework, both present and implied, and more keen in gaining an intuitive understanding of what must have this work’s importance been for its author at that particular point in his life. I mentioned earlier that, according to my present belief system, I consider the display of intellectual prowess in one’s work as a ‘proud’ act, inconsistent with the puritanical ethic of Bach’s music. These latter qualities are indeed more evident in the two works I mentioned earlier, The Well Tempered Clavier and The Art of the Fugue, which must have been intended by the composer as studies in craft rather than expression, as in fact all of the earlier works in my own Bach series have been. What is problematic with this view is the ‘lateness’ of The Art of the Fugue in the composer’s opus of works. It is still unclear from the historical evidence whether Bach died during the writing of the last fugue, as his son Philip Emanuel notes in the surviving copy of the manuscript, or whether he abandoned the work just at the point where one of his greatest contrapuntal ambitions, the vertical presentation of the preceding three fugal subjects combined with the main theme of The Art of the Fugue, was to have taken place. I would rather think that it was the latter; that at this particular stage of his spiritual and artistic development, Bach saw the futility of this type of accomplishment. It would be theologically more consistent with his self-professed religious devotion that he would choose to end his life with the piety and trepidation of the chorale "Bor seinen Thron tret ich hiermit" than conclude it with a superhuman contrapuntal summation, such as the last fugue would have been, had it been completed. A number or scholars think that this chorale has been set to music after The Art of The Fugue had been abandoned. Even if this might be historically inaccurate, it would be a psychologically more fitting conclusion to the life of this great man: the very act of abandoning his greatest intellectual achievement so close to completing it, would be a crowning spiritual achievement in itself.

It was through this line of questioning that I, a composer of far smaller accomplishment, sought to creatively approach this work and complete a series of compositions which, in a deeper spiritual sense, were guilty of pride and artistic impiety. The majority of these pieces, more so than any other works of mine, are an empty display of technical virtuosity which is subjugated to no higher purpose than the ephemeral pleasure and ego gratification one experiences with mastering a technical task. That this concluding work would not be able to escape some of these drawbacks on account of the very premise upon which it would be built, was clear enough to me even before I embarked on its composition. At the technical level, the choice of J. S. Bach as a model for the ‘photographic’ manipulation of the earlier works was a correct one: a clearly delineated structure, such as his, lends itself to external distortion, because its elements are more likely to be still recognizable after such distortion. At a spiritual level, however, the choice of Bach, no matter what the compositional approach, can never be disassociated from some degree of imprudence. In Farewell to Bach, I addressed this problem by creating a true palimpsest, a composition written on top of another composition, which, otherwise, does not alter the original in any way. I felt that, by leaving the Bach material alone, and composing music which can exist alongside with it, I was not imposing as much on the music of a great master as I have been in the past with my other works in this series.

The pre-compositional decisions about this piece had a different starting point than in the rest. The starting point here was context. I had to compose a work which would be performed in Nova Scotia by Tafelmusik, a period orchestra which had never performed a work by a living composer before, or even one who had not been dead for at least 200 years. Nova Scotia has been attracting a lot of media attention in Canada and in the United States in recent years on account of a new crop of fiddlers who have revived the Maritimes Celtic fiddling tradition and have brought it to the rest of the world. Christopher Wilcox, the Artistic Director of the Scotia Festival of Music suggested half jokingly one day on the telephone that I might consider incorporating some Cape Breton fiddle tunes into the piece, and, although I originally dismissed the idea as not relevant, I kept twisting it in my mind, trying to find some visceral connection—certainly there was no logical one—between Bach and Cape Breton fiddling. Finally, during the months preceding the composition of Farewell to Bach, I was visited by my parents who live in Greece and whom I had not seen for a number of years prior to their visit. I wanted to write a piece dedicated to my father who is getting quite old and who has been a great influence to my own thinking since my childhood. All of these unrelated items wanted to be incorporated somehow into this composition.

The first connection I managed to establish was between J. S. Bach and my father. What my father represented in my physical lineage, Bach represented in my artistic one. This farewell piece for my father is also a farewell to Bach, because, just as I am preparing myself for a life which my parents may not be a part of for much longer, I am also preparing for an artistic life away from a musical paradigm which encompasses Bach and all other composers of the past, from the advent of polyphony to the middle of the twentieth century.

Farewell to Bach chronicles my own coming of age against the background and the structure of my genetic and spiritual heritage. Throughout the piece, the string orchestra plays the Bach original with only minor modifications. In the first section the fugue is preserved intact, but played at an uncharacteristically slow tempo. After the exposition, the woodwinds enter discreetly by threading together melodies and motivic material out of existing notes of the fugue: the new music is born out from the fugue itself and learns ‘how to walk’ by selectively sampling existing information from the original material (Figure 28.)

Figure 28

Gradually my music becomes more independent, introducing additional musical information, but always within the strict guidelines of common practice harmony and counterpoint. While my music sings along with the fugue, it does not actually participate in its structure, imposing instead a motivic structure of its own, unrelated to the thematic content of the original. There is a ‘generation gap’ between the two musics. On one hand, they are in harmony, the younger music obeying the momentary—harmonic—rules of the older, but on the other hand, their structure and outlook are different and remain insular throughout. The young music does not seem to understand the linear makeup of the older music, its wisdom as it were. The haphazard structure of the former does not even acknowledge the entries of the subject in the fugue or the inherent contrapuntal symmetry of these entries. It does however, walk in a circumspect manner around the older music and in this sense, it is never disrespectful. As the woodwinds, particularly the flute, stretch their melodic, harmonic, and rhythmic meaning of their music in relation to the fugue, the style of the superimposed music fluctuates between the baroque and later musical eras, playing and dancing around the canonical stipulations of the original.

The second section is dominated by two solo violins who play extremely virtuosic parts juxtaposed on the Bach material. Melodically, rhythmically and stylistically, they are far more independent from the Bach original than any of the material of the first section. The pervading principle here is cultural counterpoint. I developed the concept of cultural counterpoint in a number of works for instruments and tape which I composed during the late eighties. In these heavily eclectic works the concept of cultural counterpoint was applied not only to the materials, but to the structure as well, and also to the way of thinking about composition in general. In the second section of Farewell to Bach, the two solo violins draw their material mainly from non-western cultures: Celtic fiddle music, middle eastern folk, klezmer and even jazz. In many instances, this material is not in the major/minor mode of western music, but it still works against the original fugue which is in D minor. The two are always vertically compatible, while horizontally, they follow their separate paths, and speak their different, sometimes incompatible languages. Figure 29 shows an example of this. The solo violin line is mostly in the Dorian mode with G natural, instead of the G sharp leading tone of the Bach fugue. In the descending figure of measure 143, the solo violin introduces the G natural in the first half of the measure taking advantage of the passing G natural in the alto voice of the fugue. When, however, the Bach G natural turns into G sharp, the leading tone, for the next four beats, the violin melody finds itself in a difficult predicament: it can either acknowledge the harmonically foreign G sharp—foreign within its own modal context, that is—or avoid the leading tone association, but pitting against it an A flat—the flat fourth of the dominant E. This way there is an vertical convergence between the worlds of the solo violin and the string orchestra, even though the G sharp and A flat mean completely different things within the horizontal contexts of their respective lines.

Figure 29

On the surface, the relationship between the new and the old music is tenuous and occasionally contradictory, but at a deeper level it is symbiotic and one of interdependence. If the first section of Farewell to Bach was structurally a metaphor for childhood, this section is a metaphor for adolescence. In the face of the onslaught by the solo instruments, the string orchestra is gradually persuaded to play along, again not by altering the fugue, but by applying different types of articulation to selected blocks of music. Seen from a certain perspective, the breaking of the music into textural blocks conspires against linear cohesiveness. In reality however, the fugue is the bedrock against which all harmonic relations between the soloists and the orchestra are negotiated, and is thus the only true authority in the music. Furthermore, it owes its authority not to its intransigence, but to the fact that it acknowledges the cultural integrity of the solo—albeit foreign—material and participates willingly in its development.

The third section is a signature fugue whose subject is dominated by the B, A, C, H motif. The music owes its pronounced chromaticism to the fact that the subject is tonally ambiguous, starting deceptively in G minor, and not revealing its D minor tonality until the cadence at the end. Because of the tonal ambiguity of the B, A, C, H motif, the fugal entries are irregular and in most cases recognizable only in retrospect; they often overlap cadences and appear at unsuspecting moments in the music. Even though most fugal entries are in primary key areas, the "nomadic meandering" of the music proclaims, according to Glenn Gould, "a spiritual descent from the ambivalent chromaticism of Cipriano do Rore or Don Carlo Gesualdo."21 It is a retrospective look at earlier music away from the vogue of new simplicity propagated by a new generation of composers which included Bach’s own sons. Bach sees himself creatively as the final summation of an entire era of contrapuntal thinking. He signs this era off with his musical signature and then falls silent forever.

This was the poetic imagery I had in mind when I attempted the setting of the third fugue. As in the first section, I left the exposition alone, starting my music immediately after the exposition with the first woodwind entry to the notes of my own musical signature (C, H, or C, B natural) and introducing the rest of the woodwinds in a manner similar to that used in the opening section. At times, as in Figure 30, the overall texture and harmony is reminiscent of late romantic music, even thought the string orchestra plays the Bach fugue in its original unaltered form. As in the first section, the thematic structure of the new music is unrelated to that of the fugue. However, since the woodwind music is stylistically more ‘recent’ here than in the beginning of the piece, there is less thematic coherence and more emphasis on expression and expansiveness of melodic line, hence the ‘romantic’ character of this section. Consistent with my descriptions of the previous sections of Farewell to Bach, that is as metaphors for specific human ages, this section is the equivalent of the age of maturity and introspection. In this age, the relationship with the ‘father figure’ is least confrontational and more appreciative and understanding. This resignation of the will to be different manifests itself musically towards the end of the third fugue, where the woodwinds gradually merge with the fugal material, literally duplicating it during the last four measures.

Figure 30

Since, as I have already stated earlier, I had no intention of attempting my own conclusion to this contrapunctus, I decided to end the third fugue seven measures earlier—that is at the true end of the third fugue which is measure 233—just before Bach embarked on the task of combining the three subjects in a single contrapuntal statement. This was both musically and intuitively a more appropriate break off point. Still for the purposes of my own composition I needed a conclusion. This conclusion, or coda, is essentially a number of statements of the B, A, C, H motif. The coda starts by reintroducing the two solo violins which have not been active at all since the end of the second section. At measure 336 the string orchestra plays a stretto consisting of two prime and two inverted statements of the signature motif (Figure 31.) This stretto is followed by other interlocking statements of the same motif by the entire orchestra, while the first solo violin re-introduces fiddle material from the second section of the piece. Even though the coda provides a conclusion to the work as a whole, this conclusion still sounds incomplete. This is partly due to the plagal character of the final cadence, but mostly due to the fact that the coda does not provide enough of an answer to the contrapuntal expectations created by Bach during the preceding three fugues. Even though in the foreground the actual progress of these fugues becomes obscured by my own music, Bach’s fugal writing is so rigorous, that an attentive listener focusing on the background structure will still be aware that the contrapuntal summation of the music has yet to take place. No coda of mine—or anyone else’s, for that matter—would have been able to obscure this. I accepted the fact that, at a deeper musical level, this composition would remain incomplete, and perhaps raise in this manner the theological question which I discussed earlier, and which—I choose to believe—Bach also raised and, as a result, abandoned the fugue where he did.

Figure 31


This paper has focused for the most part on compositional technique. The method of presenting these techniques has been personal, rather than objective: I first outlined the questions which were raised during the act of composing and then offered my answers to these questions. This way it is possible for the reader to focus on the questions rather than my answers, and perhaps come up with different—better—answers. As I stated at the outset, one of the original intentions of this paper was to serve as a guide for my graduate composition students, particularly those who have an interest in creatively approaching the music of other composers. However, in the course of writing it, I realized that technique and its application is inseparable from a deeper artistic intention and from the necessity to understand this intention within a wider cultural context. This is the context which gives legitimacy to our work and to our existence as creative individuals. A great deal of the commentary which accompanies the more matter-of-fact description of the techniques is meant to ensure that they are not viewed as abstract templates which composers may liberally apply to any music that catches their fancy. The latter practice may produce immediate results, but would be of little value to the cause of one’s artistic development.

In using these resources, I start by asking the question a photographer asks of a photographic subject. Does my subject need additional commentary beyond what the camera can capture by default? As a matter of course, I do not start using distortion lenses, unless I already see in my subject a hidden quality which a specific distortion can bring to the surface for everyone to see. In music as well as in the other arts, anything less than this would constitute cultural appropriation, a practice both morally and artistically questionable. Inasmuch as Bach can still speak to a contemporary listener as he spoke to a listener of his time, Bach’s music remains the domain of the performer, not the composer. However, if one is able to see in his music certain hidden qualities hitherto unexplored, or if a new artistic statement may be articulated by distorting Bach’s original musical commentary—moreover a statement which can tell us something about our place in history as opposed to his—then compositional distortion is a legitimate artistic practice, very much in keeping with the postulates of post-modernity, and very much a composer’s concern.

Since the 1960’s or so, we have been experiencing an artistic present whose bandwidth is much wider than any of the artistic presents which preceded it. In Mozart’s time Bach was virtually unknown, certainly not performed, most certainly not part of their musical present. Today, Bach’s music—and Mozart’s and Mahler’s and Bartok’s and, for that matter, Perotin’s—is back with us again, very much part of our present, as in fact are the instruments and instrumental/vocal ensembles by which this music was originally presented. Postmodernity has made us aware that our musical present encompasses not only the music of today or that of last month or last year, but the music of an entire paradigm which started with the advent of polyphony and ended as a dynamically evolving phenomenon with the entropy of the 1950’s. A great deal of the music written today is either attempting to make creative and cognitive sense of this era we usually call western music—I call it the ‘Renaissance paradigm’—or is looking beyond that into a new paradigm, one which heralds a new age for humankind and a new global culture. My recent music is already gearing for the latter pursuit, and in many ways is distancing itself from the experiments which I have described in this paper. It is difficult to describe in just what respect this new pursuit is different from my previous music, for there are no clearly identifiable external—formal—aspects of the music one can actually talk about. It is rather a different attitude, an ethic if you like, in one’s relationship to the world and its inhabitants, and in one’s approach to the music of other composers. The difference in my approach to Bach in most of the works in this series as opposed to my approach in Farewell to Bach is very much indicative of this old/new paradigm polarity. To say more than that would not be within the scope of this paper.

In closing, I should mention that the term ‘palimpsest’—to write over something else—does not necessarily imply a relationship between the pre-existing script and the one written on top of it. The traditional practice which the term ‘palimpsest’ describes is in favor of discarding the existing artwork for the benefit of using the same material—the wood of an icon, or the page of a manuscript—for a new creation. This is clearly not the case with my approach to Bach, but I will leave it for the linguists to decide what the appropriate term is for this practice. In the meanwhile, my interest in the music of J. S. Bach has enabled me to engage creatively with, and develop an intuitive understanding of an entire era of artistic endeavor whose time has finally passed. We now live at the time of the Spring Equinox of a new age, the cusp between two great eras of human endeavor. As we are crossing the threshold into a new century, a new millennium and a new eon, it is perhaps the right time to briefly look back to the rich musical tradition which has led us to this threshold and bid farewell to Bach, the composer who over the centuries has emerged as the most venerable icon of our musical heritage.


I am indebted to Ernest Cholakis and Prof. Gustav Ciamaga for their assistance and advice in writing this paper.

  1. The extant works of mine from the late seventies utilizing this technique are: Aztlan (1978) for oboe and harp, Cain (1978) for flute and percussion, and The Law of One (1980), my doctoral composition for harp and two identical chamber orchestrals antiphonally placed. Copies of The Law of One may be obtained through UMI 300 N. Zeeb Road, Ann Arbor, MI 48106 (order number: DA8214158) . For a detailed description of the fractal organization of sound in The Law of One see my paper, "The Law of One: Recursive Structures in Composition", Organized Sound, Volume 3, Number 1, 1998. Cambridge University Press.
  2. Bertrand Russell describes this ‘conspiracy’ of our sensory apparatus in the following passage: "Physics assures us that the occurrences which we call "perceiving" objects, are not likely to resemble the objects except, at best, in certain very abstract ways. We all start from "naive realism", i.e., the doctrine that things are what they seem. We think that the grass is green, that stones are hard, that the snow is cold. But physics assures us that the greenness of grass, the hardness of stones, and the coldness of snow, are not the greenness, hardness, and coldness that we know in our own experience, but something very different. The observer, when he seems to himself to be observing a stone, is really, if physics is to be believed, observing the effects of a stone upon himself. Thus science seems to be at war with itself: when it most means to be objective, it finds itself plunged into subjectivity against its will. Naive realism leads to physics, and physics, if true, shows that naive realism is false. Therefore naive realism, if true is false; therefore it is false". Bertrand Russell, "An Inquiry into Meaning and Truth" p. 15, 1950. Unwin Paperbacks, London.
  3. Best known as the Bell Theorem. J. S. Bell, "Speakable and Unspeakable in Quantum Mechanics; Collected Papers in Quantum Mechanics." 1988. Cambridge University Press. This book includes the entire collection of published and unpublished papers on the conceptual and philosophical problems of quantum mechanics written by John Bell, the leading expositor and interpreter of the modern quantum theory.

  4. There are countless examples of extended chromaticism before Bach, to be sure, and there are passages in Bach’s own music which are extremely chromatic. Bach’s chromaticism is mostly a foreground phenomenon. His background harmonic motion adheres to the principles I mentioned. Earlier chromaticism particularly in the music of composers such as Carlo Gesualdo is pre-tonal and has no bearing to the organic, dynamically evolving, goal oriented tonal system of the common practice era.

  5. Christos Hatzis, "Towards a New Musical Paradigm." Published in Mikro-polyphonie on-line journal, La Trobe University, Australia. This paper looks at the history of western music as a series of shifting paradigms: the Medieval Paradigm followed by the Renaissance Paradigm, spanning a historical era from the advent of polyphony to the middle of the twentieth century, followed by the New Age Paradigm, which is just beginning to musically manifest itself in our time.

  6. The Brain Opera produced by a pool of creative talent associated with MIT is the brainchild of Tod Machover. It involves a complex user interface which consists of a number of terminals resembling video arcade games. From these terminals the end users may guide the melodic, rhythmic, harmonic and structural aspects of the opera in real time, therefore, the actual course of the composition cannot be determined beforehand.

  7. Charles Ives’ interest in multiple tempi played simultaneously by different ensembles stems from experiments his father, a bandmaster in a small Connecticut town, did during special civil celebrations. On such occasions he had several bands approach the town centre from various directions playing different pieces at different tempi. Ives delighted in this superimposition of musical materials, sowing thus the seeds for the development of the collage technique by subsequent composers.

  8. The origins of Stockhousen’s interest in polyrhythmic structures expressed as multiple tempi can be traced back to his article "How Time Passes" (Die Reihe, Volume 3, 1957, pp. 10 - 40; Theodore Presser Company, Pennsylvania.) In that article he discusses the compositional use of polyrhythm as an extension of applying to music structural relationships derived from the overtone series. His work Carré for four orchestras and four conductors was his first application of this principle to music.

  9. Geoffrey Payzant, "Glenn Gould: Music and Mind" pp. 28 and 75, 1978. Van Nostrand Reinhold Ltd., Toronto.

  10. Joseph Roddy, "Apollonian". From "Glenn Gould: Variations" edited John McGreevy, pp. 107-109, 1983. Doubleday Canada Limited, Toronto. Reprinted from The New Yorker Magazine, 1960.

  11. John Oswald’s plunderphonic, a technique of deconstructing and reassembling recorded music from various sources is such a meta-compositional technique. My own approach to the music of Bach as outlined in this article is another.

  12. In a discussion with Curtis Davis, for example, Glenn Gould mentions an experience in a New York studio where he played two Bach fugues on a multitrack tape recorder, one voice at a time, and the thrill he experienced from the result:

  13. the extraordinary thing was that having known these pieces at least twenty years, I saw them and heard them afresh in the most extraordinary way. I had never realized things about the inner voice aspects of those fugues which are manifestly clear; they are simply there to be apprehended by anyone who takes a close look at that score, and yet I had never discerned them in that fashion.

    Glenn Gould & Curtis Davis, "The Well-Tempered Listener." From "Glenn Gould Variations" (ibid.) p. 277.

  14.   Glenn Gould, "Art of the Fugue". From "Glenn Gould Variations" (ibid.) p. 213.

  15.   Glenn Gould, Introduction, "Bach’s Well-Tempered Clavier 1. 1972. Amsco Music Publishing Company, New York. Also in "Glenn Gould: Music and Mind" (ibid.) p. 83.

  16.   Rober Hurwitz, "Towards a Contrapuntal Radio". From "Glenn Gould: Variations", p. 260.

  17.   Glenn Gould, "Stokowski: a Portrait for Radio" (broadcast), CBC (February 2, 1971.) Geoffrey Payzant, "Glenn Gould; Music and Mind" p. 7.

  18.   Although not clear in the liner notes, the continuo part which I have used as a model, particularly in the second movement, must have been created by Christopher Hogwood, to whom I am thus indebted.

  19.   In Paul Verhoven’s film "Total Recall" (screenplay by Ronald Shusett, Dan O’ Bannion and Gary Goldman inspired by the short story "We Can Remember it for You Wholesale" by Philip K. Dick) the difference between virtual and inescapable realities is blurred to such a degree that the two become indistinguishable from each other. The key differentiating factor is memory: on Star Trek’s holodeck, or while listening to classical music, we are aware at all time that we are engaged in a "make believe" reality of some kind. If, however this knowledge—which is intimately connected with the understanding of who we are—was temporarily or permanently suspended by chemical and/or other means, then the virtual reality, might become very ‘real’; from such reality we might not be able—or have any incentive—to ‘snap out’. The main character in "Total Recall", for example, ends up feeling more comfortable with his virtual, fabricated self and decides to continue his existence in the virtual mode; in fact at the very end of the film, we feel uncertain as to whether these two characters, the virtual and the ‘real’, exist at all and are not simply a hallucination of yet a third character, who at the beginning of the film went into a virtual reality ‘vacation’ machine. This crisis of identity, even though still the stuff of science fiction, might turn out to be one of the most fertile grounds for artistic inquiry in the future.

  20. "Towards a New Musical Paradigm." ibid.

  21. Christos Hatzis, "Ritual versus Performance: The Future of Concert Music". Harmony, Number 7, October 1998. A publication of the Symphony Orchestra Institute. 

  22. Glenn Gould, "Art of the Fugue." From "Glenn Gould: Variations", (ibid.) p. 206.

LE FastCounter

Return to Writings Click here
If you have not already visited Christos Hatzis' Home Page click here: arrow.gif