<<

James Harley The Electroacoustic Minnesota State University Moorhead 1104 7th Avenue South Moorhead, Minnesota 56563, USA of [email protected]

Of the close to 150 compositions that Iannis Xe- output can be distinguished (see Table 1). In each, nakis created, only a handful involve electroacous- the technical means change somewhat, as do the tics (sounds directly produced by electronic, digital, aesthetic concerns. There are connections to or other studio means). Those works, however, are the instrumental music he was writing around influential beyond their number. Any history of the same , and there are also connections must place Xenakis as a central to the being written by oth- figure, both for his innovations and for the impact ers (though surprisingly few). Throughout his life, his music has had on successive generations. His as clearly exemplified in his electroacoustic works, involvement in the creation of multimedia ‘‘specta- Xenakis sought to ‘‘extend the limits of musical cles’’ brought him wide exposure, although his un- thought’’ (Robindore´ 1996). compromising aesthetic vision precluded fame and fortune on a popular scale. Nonetheless, through his work, Xenakis (see Fig- Stage One: Musique Concre`te ure 1) presented a bold, charismatic persona: he was a revolutionary, both in politics and in . The res- Xenakis arrived in as a 25-year-old refugee in tive students of the 1960s, in particular, were drawn November 1947. He decided, having barely escaped to him, to his peculiar mixture of forward-reaching (as a condemned insurgent) with his life, to and noisy, pounding . There devote himself to music, a dream he had guarded are many parallels between Xenakis’s work and ex- within himself but which had been sidelined by perimental elements of , particularly turmoil, both political and personal (see Matossian in the embracing of technology and high-density/ 1986 for fuller details of Xenakis’s early life). As a high-amplitude sound set off by disorienting, hallu- trained civil engineer, Xenakis found himself work- cinatory light-shows. The electroacoustic music of ing in the architectural studio of , Xenakis, Concret PH in particular, provided a link where he was able to earn a living while pursuing between the ‘‘academic’’ components of the 2000 In- music in his spare time. He would have known vir- ternational Conference in Berlin tually nothing of contemporary music, but the mi- and -oriented ‘‘off-ICMC’’ celebration, held lieu he worked in was certainly cultured (Le concurrently. Additionally, a new Asphodel release Corbusier was acquainted with Edgard Vare`se, for of his massive electroacoustic work, Persepolis,is example), and he would no doubt have heard vari- accompanied by ‘‘re-mixes’’ by, among others, Japa- ous broadcasts on Radio-. He very likely lis- nese artists. tened to the early broadcasts of , It is my aim to present here a brief overview of who, in 1948, presented his first experiments in Xenakis’s electroacoustic music in the manner of a musique concre`te on the radio and in concert. tutorial. Certain aspects or selected works from Xenakis met in 1951 and began this area of his compositional output have been ex- attending his classes that year, a habit he would amined in some detail (Di Scipio 1998, 2001; Hoff- continue more or less regularly for the next two mann 2000a; Solomos 1993), and a more detailed years. Messiaen was a central figure in the Parisian study of these works within the context of the rest new music world; , who began work- of his output is found in Harley (forthcoming). ing closely with Schaeffer in 1949, had been his While there is a strong unity of aesthetic and com- student between 1944 and 1948 (Boivin 1995). Dur- positional technique running through his entire ing this period (1951–1952), Messiaen himself vis- oeuvre, five stages in the trajectory of his studio ited Schaeffer’s studio, producing a short piece, -dure´es. , who sat Computer Music Journal, 26:1, pp. 33–57, Spring 2002 with Xenakis in Messiaen’s class during 1951– ᭧ 2002 Massachusetts Institute of Technology. 1952, also created a tape study during that time.

Harley 33

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Table 1. Xenakis’s electroacoustic music Year Title Medium Duration Studio Recording Stage I: Musique concre`te 1957 Tape 6:53 GRM EMF CD 003 1958 Concret PH Tape, 2:42 Philips/GRM EMF CD 003, Caipirinha multimedia CAI.2027.2, Bvhaast CD 06/0701 1959 Analogique B Tape 1:30 Gravesano/GRM Vandenburg VAN 003 1960 Orient-Occident Tape 10:56 GRM EMF CD 003 (soundtrack) 1960 Vasarely Soundtrack GRM Withdrawn 1961 Formes rouges Soundtrack GRM Withdrawn 1962 Bohor Tape (8-channel) 21:36 GRM EMF CD 003 Stage II: Mixed 1967 Polytope de Montre´al (pre- c. 6 min Radio-France LP recorded) 1969 Orchestra and 75:00 Radio-France Etcetera KTC 1075, tape (4- Asphodel 0975 channel) 1970 Hibiki-Hana-Ma Tape (12- 17:39 NHK Tokyo EMF CD 003 channel) Stage III: Multimedia 1971 Persepolis Tape (8-channel), 55:06 FractalOX, Asphodel multimedia forthcoming 1972 Polytope de Cluny Tape (7-channel), c. 25 min CEMAMu Mode 98/99 multimedia 1977 La Le´gende d’Eer (Le Tape (7-channel), 46:00 WDR /CEMAMu Montaigne MO 782058 Diatope) multimedia Stage IV: UPIC 1978 Mycenae Alpha Tape (2-channel), 9:36 CEMAMu Neuma Records 450-74, (Polytope de multimedia Mode 98/99 Myce`ne) 1981 Pour la Paix Voices and tape, c. 27 min CEMAMu/Radio-France none radiophonic 1987 Taurhiphanie Tape (2-channel), 10:46 CEMAMu Neuma Records 450-86 multimedia 1989 Voyage Absolu des Tape (2-channel) 15:25 CEMAMu Perspectives of New Music Unari vers PNM 28 Androme`de 1997 Erod Tape (2-channel) c. 5 min Ateliers UPIC Withdrawn Stage V: Synthesis 1991 GENDY301 Tape (2-channel) 14:15 CEMAMu Withdrawn 1991 GENDY3 Tape (2-channel) 18:45 CEMAMu Neuma Records 450-86 1994 S.709 Tape (2-channel) 7:03 CEMAMu EMF CD 003

34 Computer Music Journal

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Figure 1. Xenakis at the mixing console, Le Dia- tope. (Photograph by Mali; used by kind permission of the Xenakis family, along with all subsequent sketch and score examples.)

relatively brief work, but sonically intense and quite unlike other music being produced at GRM at that time (such as Orphe´e, the opera by Schaef- fer and Henry, or even De´serts, by Vare`se). There was a strong concern among many of the compos- ers working in the studio to create sounds from relatively discrete instrumental or real-world sounds. Xenakis, on the other hand, sought to cre- ate dense, noisy textures that bore little direct re- semblance to the world around him. Instead, his sounds derived from the same imagination that was creating complex clouds and masses in the or- chestra—an imagination, it should also be pointed out, strongly affected by the chaotic, traumatic sounds of war. (One might also speculate that the hearing loss, particularly in the higher frequencies, Xenakis suffered from his injuries during that time would have shaped, to some extent, his attraction to ‘‘noisy’’ sounds and to high-decibel playback lev- els.) Indicative of the dichotomy described by the ti- tle, Diamorphoses combines noisy, primarily low- frequency sounds—derived from an earthquake, a jet engine, and a train—with more sharply defined, high-register bell sounds. The natural ef- fect of the jet engine winding up is combined with glissandi of other sounds, produced in the studio by It was a few years later, in 1954, that Xenakis means of tape manipulations. Often, the composite was accepted as one of the first members of the sonority produces something quite new, the Groupe de recherches de musique concre`te (re- sources being submerged. named Groupe de recherches musicales in 1958, The formal organization is quite clear: the two now commonly called GRM). His assimilation of outer sections of more-or-less sustained sounds the concerns of contemporary musical culture had (layered noises, glissandi) are contrasted with a cen- proceeded rapidly. That same year, he completed tral, more discontinuous passage, filled with many his first major orchestral score, , a work shifts of sonority (starting with the bell sounds, that explores novel sonorities built from complex then adding others). Xenakis worked in particular configurations of string glissandi (see Figure 2); it provoked a major scandal at its premiere in Don- to create shifts and continuities of density through aueschingen the following year. Drawn to the new layered of sound objects (Delalande possibilities of the studio, Xenakis expressed his in- 1997). For the discrete sonorities, he used probabili- tention to ‘‘study the evolution of timbres, dynam- ties to calculate attack points, layering the tracks ics, and register . . . to make chromosomes of in the studio to achieve a perceivable range or scale attacks’’ (Matossian 1986). of activity.

Diamorphoses Concret PH Xenakis completed his first electroacoustic compo- Xenakis carried these concerns to his next electro- sition in 1957. Diamorphoses (Greek for ‘‘continu- acoustic piece, Concret PH, produced in 1958. (Ac- ity—discontinuity,’’ ‘‘two aspects of being’’) is a cording to Delalande, it was revised in 1961 to

Harley 35

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Figure 2. Excerpt from Metastaseis, beats 309–314.

create a stereo version, then re-mixed in 1969 onto would be heard by many thousands of people four channels.) This 2-min miniature was produced would have been enticing. In addition, the facilities as an introduction to the Poe`me e´lectronique of Le that Philips had developed in order to project Corbusier and Vare`se for the in sound, light, and images were among the most ad- Brussels. Xenakis had been heavily involved in the vanced in the world. design and of the Pavilion (see Figure Unfortunately, Le Corbusier, who was operating 3), often standing in for Le Corbusier who was also under enormous pressure and seriously behind working on a major project in India (Treib 1996). schedule, would not grant Xenakis leave to work at His music was intended to fill the gap between pre- the Philips studio in Eindhoven, Holland. He sentations of the 8-min of the main needed him to supervise the construction of the pa- multimedia work, which occurred at 10-min inter- vilion, and seemed to consider the composition of vals. In other words, it was intended to fill in the the little interlude of secondary importance. Va- time as the audience filed into and out of the pavil- re`se, on the other hand, was able to take advantage ion. Nonetheless, for a young, relatively unknown of Philips’s technology, including special equip- , the opportunity to create a work for pre- ment for the projection of sounds over multiple sentation at a major international exposition that . (The pavilion ended up housing over

36 Computer Music Journal

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Figure 3. Philips Pavilion, Brussels World Fair, 1958. (Used by kind permission of the Getty Research Institute. ᭧ 2002 Artists Rights Society (ARS), /ADAGP, Paris/ .)

the fledgling composer, encouraging him in partic- ular to work out his ideas of combining mathemat- ics and music so that they could be expressed cogently. Scherchen held a ‘‘retreat’’ every year at his estate in Gravesano, to which he invited lead- ing thinkers of the time to discuss a range of top- ics. (Guests included Werner Meyer-Eppler, , Mathews, and others.) The proceedings were published in the Gravesaner Bla¨ tter. Scherchen’s interests ranged to and new technology, and he had actually developed a rudi- mentary studio for carrying out experiments and producing electroacoustic music. By 1958, Xenakis had become interested in the theory of acoustical quanta (Gabor 1947), the pre- cursor to (Roads, 2002). He had developed a proposition for composing on the basis of ‘‘screens,’’ a representation of clouds of quanta 400 loudspeakers.) Xenakis, in his spare time, was for each ‘‘slice of time’’ (Xenakis 1992). The pro- instead forced to work in the rather primitive fa- gression from one screen to the next is governed by cilities of the Philips offices in Paris. He produced a Markov chains, whereby the settings for one screen monophonic version there, and then completely re- exert an influence on the calculations for the next, composed the piece later at GRM. (Xenakis pro- building a form of ‘‘memory’’ into the temporal duced multiple versions of all his tape pieces, up to organization of the music. La Le´gende d’Eer, mixing for different numbers of That year, Xenakis composed Analogique A,a channels, accounting for different playback situa- short work for strings in which there are nothing tions, and so forth.) but discrete notes (no held pitches, glissandi, tre- The only sound source used in Concret PH is the moli, etc.). The music proceeds as a succession of crackling and hissing of burning charcoal. The changing densities, registers, ranges, and intensities study of density and the application of stochastic (along with timbral changes from plucked notes to functions to the calculation of articulation points struck or short bowed ones). In the summer of for each layer of sound were the main concerns. 1959, during his stay with Scherchen in Gravesano, The crackling texture evolves in a continuous fash- Xenakis produced a complementary composition, ion, much like the central portion of Pithopraktra, Analogique B, for discrete electronic sounds, pro- his second orchestral score, completed in 1956. The duced in this case by oscillators. The procedure sense of spatial movement as the music moved was the same, though the particular parametrical along the trajectories through the loudspeakers of constraints are of course different. The composer the pavilion in Brussels would have been particu- worked out an interlocking scheme whereby the larly effective using these primarily mid- to high- two pieces can be performed as a unit, the strings register grains of sound. Certainly, it was utterly unlike any other electroacoustic music of the time, alternating with the pre-recorded electronic sounds, and remains a gem of the tape music idiom. with some moments of simultaneity. B was premiered in Paris in 1960םAnalogique A along with the first French of Pitho- Analogique B prakta, conducted by Scherchen. From that point, Xenakis first met the renowned Swiss conductor Xenakis’s music began to receive more dedicated and researcher in 1954. The attention in his adopted land. A final break with Le elder musician became something of a mentor to Corbusier provided him with the opportunity to

Harley 37

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 quit his day job (for the most part) and devote him- granular texture of burning charcoal from Concret self at last professionally to music. PH makes an appearance in the latter part of the work, this time mixed with water droplets and other sounds. Orient-Occident The formal outline of Orient-Occident is less sharply defined than is usually the case in Xe- One of the first commissions Xenakis received nakis’s music, no doubt due to its original concep- through his association with GRM was for a sound- tion as a soundtrack. The music proceeds from one track to a documentary film by Enrico Fulchignoni section to another by shifts of sonority, effected at for UNESCO. The film, Orient-Occident (1960), by gradual transition and at other times by tried to express by visual and sonic means the rela- sharp divisions. The liner notes of the Nonesuch tions and resonances of artistic relics from various LP state that ‘‘Xenakis looks upon this work as a ancient cultures. The film was based upon an exhi- solution to the problem of finding many different bition at the Cernuschi Museum in Paris, where means of transition from one kind of material to the director spent three months shooting. Xenakis another’’ (Brody 1970). In drawing upon a relatively was not brought into the project until that phase wide range of sonorities, none recognizably linked was complete. Mr. Fulchignoni intentionally gave to any particular culture but related to each other no instructions or ‘‘interpretation’’ to the com- by various means of transition or juxtaposition, Xe- poser, preferring to allow him the liberty of work- nakis enables the listeners (and presumably the ing from his own reactions and analyses viewers, originally) to create their own associations (Fulchignoni 1981). The film links disparate eras between the sounds and images from ancient cul- and cultures primarily through visual resemblances tures. rather than any sort of chronological or cultural narrative. In effect, this more abstract approach en- abled Xenakis to construct his own ‘‘narrative’’ to Vasarely and Formes rouges underscore the images. The original 22-min soundtrack is not well In the period 1960–1961, Xenakis produced two known. What is available is the concert work of more soundtracks in addition to Orient-Occident. half that length extracted from it. A 4-channel ver- Never again would he work in this domain. (Holly- sion was apparently produced at GRM in 1969 (De- wood decidedly did not beckon!) Vasarely (1960) lalande 1997), and available recordings show some was a similar project to the previous, more well- slight discrepancies of detail. (For example, the ver- known piece, a short film by Peter Kassovitz and E. sion released on a compact disc from the Electronic Szabo based on an exhibition of Op Art paintings Music Foundation stops 16 sec before the version by Victor Vasarely (see Figure 4). The composer that had been released on vinyl by Nonesuch.) was quite taken with this artist, and he ended up Compared to Diamorphoses or Bohor, the sounds working with him a few years later on his ballet, of Orient-Occident are much clearer (i.e., less Kraanerg. noisy). A bow drawn over various objects is the Formes rouges (1961) was a short animated film source of many of the sustained sonorities. The by Piotr Kamler. Little is known of the music for percussive sounds are often presented as regular these films, as they were both quickly withdrawn pulsations or perceptible patterns rather than sta- from the composer’s catalog. Produced at GRM, tistical ‘‘clouds.’’ The music is dramatic, with the which is the reason for noting them here, these more reflective sounds being surprised by sharp at- pieces were primarily instrumental rather than tacks or ‘‘ethnic’’ drumming-like patterns. There electroacoustic, apparently. Regardless, the sounds are some , unusual in Xenakis’s electro- would have been assembled in the studio. acoustic output, but also grating metallic sounds During this same period, however, Xenakis was that resemble sonorities from other pieces. The becoming more and more drawn into completing a

38 Computer Music Journal

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Figure 4. Vega image by Victor Vasarely.

stochastic algorithm for composing music using stitutes a radical exposition of the issues of density computers. These mathematical, abstract concerns and transition that Xenakis had been concerned conflicted both with the dictates of producing mu- with throughout this period. It is conceived as a sic for film and with the general aesthetic direction single, slowly evolving gesture lasting close to 22 of the GRM studio as directed by Pierre Schaeffer. min. The composer describes it as being ‘‘monistic with internal plurality, converging and contracting finally into the piercing angle of the end’’ (Brody Bohor 1970). There are two basic textures in Bohor: one, a In 1962, Xenakis completed his most ambitious concentrated, teeming sonority, constantly in mo- electroacoustic work to date, and subsequently tion, ranging in from bell-like sounds (fil- withdrew from being an associate of GRM. Bohor tered to muffle the resonance) to metallic rattlings, was one of the first 8-channel works created at finer-grained metallic sounds closely resembling GRM ( and had created 8- the charcoal crackles of Concret PH, and noisy, channel works back in 1952/1953). The piece con- clashing and crashing sounds; the other is a low,

Harley 39

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 sustained sonority adapted from a Laotian mouth mand it. His works continued to be performed organ, the characteristic crescendo-decrescendo and there and issued on recordings, and he visited a breaking off for breath being preserved. The low number of times in subsequent years to produce sounds are heard twice, primarily, beginning at the new mixes of his pieces and to make other uses of 5’30’’ mark, and lasting about 8 min, and reappear- the facilities (Delalande 1997). ing after the 17’00’’ mark and lasting 1–2 min. The One of the sources of conflict was Xenakis’s other texture continues throughout, adding layers growing conviction that digital technology was vi- to itself one by one until there are several going on tal to the future of music. His growing stature as a at the same time. On the 8-channel version, these composer added weight to his argument that GRM would be distributed around the different channels. should be the first French institution to equip itself Each layer builds and subsides independently, pro- for carrying out research in computer music. This ducing an opaque, constantly-evolving texture that orientation was anathema to Schaeffer, and Xe- at the same time seems to have no clear linear tra- nakis realized that he would be better off working jectory. When the low sustained sound drops out to establish such a facility elsewhere. the first time, there is a noticeable loss of intensity A rift had also developed between Xenakis and and tension, but the metallic sounds build up some of the other GRM composer-associates. A again, leading at the end to a sublimation of all the ‘‘collective concert’’ had been planned for 1962 in- other layers by the noisiest, crashing sonority, volving nine , the others being Claude which sharply boosts its mass and dynamic level Ballif, Franc¸ois Bayle, Edgardo Canton, , over the final three minutes. (As with Orient- Franc¸ois-Bernard Maˆche, Ivo Malec, Bernard Parme- Occident, the version presented on the Electronic giani, and Michel Philippot. Xenakis took charge of Music Foundation compact disc stops 20 sec sooner organizing the event, and put forward the idea that than the Nonesuch LP recording. This is regret- the succession of electroacoustic fragments be de- table, as this last segment consists of a greatly termined according to a probability matrix. His heightened culmination of the crescendo of dynam- proposal was rejected by the others, who preferred ics and density—the ‘‘piercing angle’’ Xenakis to work more intuitively. Xenakis withdrew from mentions.) the project, and there was evidently a certain Bohor is, perhaps ironically, dedicated to Pierre amount of ill feeling (Schaeffer 1981). Schaeffer. The elder director, who had in fact been quite supportive of Xenakis throughout his seven years of ‘‘apprenticeship’’ at GRM, was horrified by Stage Two: Mixed the piece. Part of his reaction was to the extreme level of amplification at the Radio-France premiere, Having made the decision to distance himself from intensified by the distribution of loudspeakers sur- GRM, Xenakis became preoccupied with other con- rounding the audience. As he put it, ‘‘this was no cerns. He wrote music for the stage (Hiketides, Or- longer tiny embers, each with its own allure, this esteı¨a, Medea), began lecturing, finished writing his was an enormous burst of explosions, an offensive book Musiques Formelles, composed various in- accumulation of lancet jabs to the ear at maximum strumental and vocal pieces, did a residency in Ber- volume level’’ (Schaeffer 1981). Xenakis’s ‘‘war’’ lin, and, in 1967, took up a position at Indiana aesthetic was obviously still present, but also para- University. mount was the formal concern for sustaining ten- One of his concerns through this period (and sion for a long period of time through the layering throughout his career) was the spatialized presenta- and variation of density and other sonic parame- tion of his music. He had worked with Philips on ters. Later electroacoustic pieces would carry this the elaborate sound projection system for the Phil- approach much further. ips Pavilion in 1958, and had produced Bohor with Xenakis left GRM, but he did not cut himself off the 8-channel ‘‘surround’’ diffusion in mind. Xe- from the studio definitively, nor did Schaeffer de- nakis extended this thinking into his instrumental

40 Computer Music Journal

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Figure 5. Sketch for Polytope de Montre´al installation.

and stage music as well. Eonta (1963), for brass (who won the Grand Prix de Rome for this work). and , requires the brass players to In fact, the pavilion is still standing, presently move around the stage, point the bells of their in- housing the Casino de Montre´al. struments in different directions, and so forth. In For the large central space of the pavilion, which Oresteı¨a (1966), the chorus members are asked to rises through several floors and is accessible from create ‘‘clouds’’ of stochastic sounds by playing all of them, Xenakis constructed five networks of small percussion instruments and sirens while intersecting steel cables, each outlining curved geo- moving around the stage. Terretektorh (1966), for metrical shapes (see Figure 5). Onto these cables orchestra, distributes the individual musicians were attached 1,200 flash bulbs (800 white, 400 among the audience, so that each listener has a colored), which were triggered by a control system unique experience of the piece. involving a perforated tape and photosensitive cells. This Poe`me de lumie`re comprised a succes- sion of visual configurations, the aim being to ‘‘cre- Polytope de Montre´al ate a luminous flow analogous to that of music’’ (Fleuret 1988). Xenakis used interconnected tech- In 1966, Xenakis received a prestigious commis- niques to compose the music and the ‘‘poem of sion: to create a multimedia work for the French lights,’’ but for him, ‘‘the link is not between them Pavilion at the 1967 Montre´al World . He but beyond or behind them’’ (Varga 1996). would have preferred to design the pavilion itself, The music for Polytope de Montre´al is scored for but that job was awarded to architect Jean Faugeron four identical instrumental ensembles, each to be

Harley 41

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 placed along one of the four cardinal directions The choreographer was Roland Petit, from with the audience placed in the intervening quad- France. Xenakis was given complete freedom by rants. The score, however, has apparently never the choreographer and the commissioners to create been performed live. It was recorded at Radio- his music, aside from the proviso that it be at least France prior to the presentations in Montreal, and 75 min in length, and that it incorporate, if possi- the music was played by means of loudspeakers. ble, an electronic element, along with the chamber The use of a recording, even broadcast from four orchestra, to highlight the sophisticated playback spaced loudspeakers, would obviously not be the system installed in the new facility (with loud- same as performing the music live, but Xenakis de- speakers all around the hall). It was the composer signed the piece very much in terms of layers of so- who recommended that Victor Vasarely be engaged norities, much as he had composed Bohor. Thus, to create the set (see Figure 6). while not strictly electroacoustic, Polytope de Kraanerg (Greek for ‘‘to perfect, accomplish; [ce- Montre´al exhibits similar concerns. rebral] energy’’) is concerned, according to the com- Instrumental sonorities produced by similar in- poser, with ‘‘the overwhelming fight of man’s brain struments in each are fused primarily and the [un]ending obstacles that exist or that he through register, and dynamic swelling effects are himself creates’’ (Xenakis 1969). Xenakis was also used throughout to produce the sensation of move- thinking, apparently, of the student demonstrations ment as the aural spotlight shifts from one ensem- going on in Paris and elsewhere through the period ble (or channel) to another. The layers are treated he was working on this piece, extending those so- independently, so that the overall spatialization is cial conflicts to global concerns: quite complex. The architecture of the music, though, is clear, In barely three generations, the population of being constructed from three contrasting sections, the globe will have passed 24 billion. 80% will each containing a number of distinct sonic ele- be aged under 25. The result will be fantastic ments. While the music is intense, concentrating transformations in every domain. A biological on registral extremes and being filled with dynamic struggle between generations unfurling all over gestures, it does seem to leave perceptual space for the planet, destroying existing political, social, the visual elements that were intended to occur urban, scientific, artistic and ideological frame- alongside. There are no programmatic links be- works on a scale never before attempted by hu- tween the design of the cables, the succession of manity and unforeseeable. This extraordinary flash bulbs, and the music. But, as Xenakis stated, multiplication of conflict is prefigured by the ‘‘We are capable of speaking two languages at the current youth movements throughout the same time. One is addressed to the eyes, the other world. (quoted in Toop 1989) to the ears. The content of the communication is The music of Kraanerg is primarily instrumental different but sometimes there’s a link between the in conception, similar in its textures to the orches- two’’ (Varga 1996). tral score, Nomos gamma, completed that same year. The tape part, originally conceived in four Kraanerg channels, is built from processed instrumental sounds, apparently based on recordings done with The Polytope de Montre´al made quite an impact in the same 23-member chamber orchestra as the live Montreal during its run there in 1967. So, it was parts. The music needed to be written, produced, perhaps not surprising that Xenakis’s name came and recorded within about six months, in time to up a year later when officials at the National Ballet be ready for the rehearsals with the dancers. It of Canada were seeking to commission a new, seems likely that the orchestra parts were recorded modernist work for the gala inauguration of and then sections of that used to create the electro- the National Centre in Ottawa, the nation’s acoustic part. capital (Harley and Harley 1997). The piece is built from blocks of material alter-

42 Computer Music Journal

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Figure 6. Photo of stage de- man. Used by kind per- sign by Vasarely and per- mission of the National formance of Kraanerg. Ballet of Canada (Photo by Anthony Crick- Archives.)

nating between tape and live ensemble. The obvi- there are 21 such passages of stillness, ranging in ous reference is Edgard Vare`se’s De´serts (1954), but duration from 2 sec to 28 sec. These occur primar- Kraanerg is much more elaborate in the structuring ily in the middle portion of the work, so are not in- of the mosaic-like form. The tape often overlaps tended to delineate large-scale formal divisions. the live parts, to create more of a sense of continu- The proportional relationship between tape and en- ity than is found in De´serts. The overlapping may semble does shift at certain points, and these be slight or may be more extended. The continuity changes create a tripartite form. The first section is enhanced by the similarity of material between presents the two entities with more or less equal the recorded sounds and the instrumental ensem- attention (eight blocks each); the second, beginning ble. There is never any confusing them, though, as at 23’00’’, shifts the focus to the live ensemble; the the produced sounds are always processed, some- third, beginning at 52’00’’, settles on the tape, pri- times quite heavily (primarily through filtering, of- marily. The final 6 min, for example, are entirely ten to obtain degrees of distortion). is also for tape. treated as an important structural element, and Within these blocks of sound, Xenakis plays off

Harley 43

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 the orchestral groups (woodwinds, brass, and to so much that the original sources are unrecog- strings) one against another, sometimes layering, nizable. There is a much wider range of sounds pre- sometimes juxtaposing them. The music is often sented in Hibiki-Hana-Ma than in Kraanerg, which built from global sonorities, or textures, but there is understandable considering that the electro- are also moments of and exposed acoustic element must stand on its own. The possi- chamber-like combinations of instruments. bility of deploying up to 12 channels enabled The ballet as choreographed by Mr. Petit was less Xenakis to build up layers and complex superposi- than successful, though the National Ballet of Can- tions of sonorities. The music is put together from ada did present the work numerous times on tour blocks of material spliced into the different chan- in Europe and North America over the next few nels, in a similar process to the ballet piece, though years. Although there were a few concert perfor- extended from three or four layers to 12. There are mances, no other company took up Kraanerg until many sudden shifts of sonority, density, and inten- , of the , sity, and various layers are brusquely cut in or out. created a new choreography in 1988. The music, Major articulation points serve to loosely divide though, was praised from the beginning, and be- the piece into four sections. The first section, last- came known through the LP recording that was is- ing to 3’00’’, is built from a low, booming, undula- sued from the original recording produced at tory sonority over which orchestral glissandi string Radio-France for the rehearsals in Ottawa. sounds are layered. A sweep up to a sustained high- register cluster signals the start of the second sec- tion, which introduces a layer of tinkling bells, Hibiki-Hana-Ma stochastic clouds of whips and pizzicati, and much else. A sudden drop in dynamic level and in the By 1970, 10 years into his career as a professional number of layers at 6’32’’ signals the third section, composer, Xenakis was known around the world. resembling the previous one in featuring succes- For the World Exposition that year, he was sions of a wide range of sonorities, most of which invited to the studio facilities of NHK (Nippon have been heard already. Here, though, the orches- Hoso Kyokai) in Tokyo to compose a tape work to tral winds are introduced, playing both sustained be presented in the Japanese Steelworkers Federa- sonorities and glissando textures reminiscent of tion Pavilion. No doubt one of the great attractions similar passsages in Nomos gamma and Kraanerg. of this commission was that the music was to be At 11’07’’, another sudden drop in intensity/density projected through a sound system of 800 loud- signals the shift to the final, most substantial, sec- speakers grouped in 250 locations. tion. Through this passage, various noise-based so- Hibiki-Hana-Ma (‘‘reverberation-flower- norities are introduced, of both sliding and interval’’), originally conceived for 12 channels, fixed-band types. This material carries on to the was later mixed down to eight for concert diffusion end, layered with previously introduced sounds. (and then to two for the issuance of the recording). The impact of hearing such a wide range of pro- At the radio studio in Japan, Xenakis had access to duced sonorities based on acoustic sources, both an orchestra, and, like Kraanerg, much of the mate- sustained and percussive, as they were projected rial comes from orchestral sonorities. In this case, through the pavilion’s large sound system must the sources were ‘‘typical’’ textures from existing have been powerful. What made a strong impres- scores with emphasis on the strings, particularly sion on the composer, though, were the dynamic featuring glissandi and natural harmonics. To this light sculptures and laser projections presented in the composer added the biwa, a Japanese lute-like that same pavilion by Japanese artist Keiji Usami. instrument, and some additional percussion Xenakis was particularly interested in the new sounds. technology used to control the lasers and to syn- There are varying degrees of studio manipulation chronize the lights with the sound. In this, he was of these instrumental sounds, from virtually none preparing to take on a series of large multimedia

44 Computer Music Journal

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Figure 7. Archeological site of Persepolis, Iran.

projects for which he alone, finally, would be re- site necessitated a ban on construction or renova- sponsible. tion (see Figure 7). All loudspeakers (59), lights (92 spotlights), and lasers (2) needed to be installed in as non-invasive a way as possible (see Figure 8). Stage Three: Multimedia The lasers and spotlights were ranged throughout the site and projected to create ‘‘luminous patterns In 1969, Xenakis had traveled to the ancient site of evoking the Zoroastrian of light as eter- Persepolis, Iran, for the premiere of his ambitious nal life’’ (. A. Harley 1998). From the central por- work for six percussionists, . The Shi- tion of the site, where the 59 loudspeakers raz Festival was organized under the patronage of projected the eight channels of sound throughout the Empress of Iran, and Xenakis and his music the audience area, the lights swept upward and out created quite a sensation (Matossian 1986). He was toward the hillside tombs of Darius and Artaxer- apparently engaged on the spot to create an even xes. There, in the distance, bonfires were burning, more ambitious work for the 1971 festival. This and parades of children carrying lighted torches time, the composer was to create a multimedia wended their way up the hillsides, creating ever- event, with all the resources of Iranian Radio and changing linear patterns. Television placed at his disposal. The music, with its noisy sonorities and overlap- ping waves of intensity, looks back to Bohor. Per- sepolis, though, is 55 minutes in length, a very Persepolis long span for a continuously evolving form. Ac- cording to his sketches, Xenakis constructed the The one limitation Xenakis was faced with in plan- tape from 11 sonic entities distributed among the ning his first real ‘‘polytope’’ was that the location eight channels (see Figure 9). The sketch shown within the ruins of such an important historical gives precise timings for the first reel of the origi-

Harley 45

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Figure 8. Composer’s sketch for distribution of loudspeakers and lights for Persepolis.

nal tape, which ends at 31’30’’. The performance Persepolis within the desolate beauty of the Iranian would have required two 8-track machines so that desert, with spotlights sweeping the sky and fires the piece could be performed with no breaks. Mul- burning in the distance, would have been an awe- tiple layers of similar material create overall tex- some experience. tural ‘‘zones’’ which serve to delineate the form, though the shifts from one to another are rarely easy to distinguish. Polytope de Cluny The sonic entities range from clarinet multi- phonics (3, as numbered in the sketch), to complex, Polytope de Cluny (1972), Xenakis’s next multime- high sustained sounds derived from string harmon- dia creation, followed fast on the heels of Persepo- ics (2), to low, sliding distortions of timpani rolls lis. It was commissioned for the Festival (9), to (7) and ceramic wind-chimes (11). Oth- d’Automne in Paris, and was set in the historic Ro- ers are harder to identify, but one is derived from man Baths of Cluny, just off the Boulevard Saint- recordings of cardboard being handled (6), and an- Germain-des-Pre`s. It premiered in October 1972, other sounds as if a strong, buffeting wind had been and ran for 16 months, four times daily, with the fed through a distortion module (8). The remaining cumulative audience reaching over 200,000. The entities can be identified as metallic, noisy sonori- experience was popular among young people, who, ties. Entities 1 and 4 are not used at all in the first in their radical political and social orientation, part, occurring only in the final moments of the sought ‘‘music that transcended the limits of tradi- piece. tion and nationalism’’ and rejected the ‘‘formal ap- None of these entities are simple or ‘‘pure’’ so- parel and conventions’’ of the concert ritual. norities, by any means, and the sonic intensity of- Underground in Cluny, they ‘‘sat on the floor, sur- ten seems overwhelming. All of the material is rounded by strange sonorities and subjecting them- developed rather than just being repeated, so that selves to perceptual and aesthetic the music evolves, while remaining unified, over experimentation’’ (M. A. Harley 1998). the course of its journey through this violent, The T-shaped chambers of Cluny were, like the shrouded . Hearing the music in the set- historic site in Iran, not to be altered, so Xenakis’s ting for which it was intended, the dark ruins of installation was erected within the walls by means

46 Computer Music Journal

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Figure 9. Composer’s sketch for the 8-channel tape piece, Persepolis,up to 31’30’’.

of scaffolding and cables. Along with the flash CEMAMu (Centre d’E´ tudes de mathe´matique et bulbs reminiscent of Polytope de Montre´al (600 in automatique musicales), with the goal of pursuing number) were three lasers (colored red, green, and computer music and related research. He had ap- blue) projected along pathways determined by 400 parently attempted to set up a parallel operation at adjustable mirrors. Xenakis was able to control the Indiana University during his tenure there, but he operations of the lights, lasers, and mirrors through was unsuccessful, and he resigned his position in digital means. All of the calculations were pro- 1972, the same year of Polytope de Cluny and the grammed on a computer, and the output data was consolidation of CEMAMu. One of his interests then converted to electromagnetic signals. was the refutation of Fourier analysis as the basis This implementation of digital technology for sound synthesis. As he put it in his chapter on brought to fruition one of Xenakis’s long-held ‘‘New Proposals in Microsound Structure:’’ dreams. In 1966, he had established an organiza- tion, EMAMu (l’E´ quipe de mathe´matique et auto- [S]pectral analysis of even the simplest orches- matique musicales), which in 1972 became tral sounds . . . presents variations of spectral

Harley 47

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 lines in frequency as well as in amplitude . . . hind , where digital synthesis had been These tiny variations . . . would certainly re- launched by in 1957. quire new theories of approach, using another functional basis and a harmonic analysis on a higher level, e.g., stochastic processes, Markov Le Diatope; La Le´gende d’Eer chains, correleated or autocorrelated relations, or theses of pattern and form recognition. (Xe- At first glance, Le Diatope (the title of the installa- nakis 1992) tion) would appear to be a variation of Polytope de Cluny. The electroacoustic music is recorded on The electroacoustic music for the Cluny instal- seven channels, with the final channel reserved for lation was produced on seven channels of the 8- control signals. The multimedia performance in- channel tape, with the last track reserved for volves 1,680 flash bulb lights, 4 colored lasers, and electromagnetic control signals for the lights, la- 400 programmable mirrors. This was a far more sers, and mirrors. Timing between the various ele- ambitious project, however. It was commissioned ments, including the music, could thus be precisely for the inauguration of the acclaimed Centre controlled—although, as before, Xenakis sought Georges Pompidou in Paris, and it also included an parallels rather than unions: ‘‘I wanted to establish original architectural component. At long last, Xe- a contrast: the lights are a multitude of points, nakis was able to design the free-standing shell with stops, starts, etc., and the music is continu- within which the spectacle would take place (see ous, for although the sound changes it does not Figure 10). It was also, at 46 min, about twice the ’’ (Fleuret 1972). length, an ambitious duration for lighting effects The music was presented through 12 loudspeak- that required updating 25 times per second. ers distributed around the site. The event, which In addition, the music was commissioned sepa- ran some 25 min, falls in between the 6-min dura- rately, and was clearly conceived to stand as an in- tion of the Polytope de Montre´al and the hour-long dependent piece. (WDR) length of Persepolis. The music bears some resem- invited Xenakis to produce the piece using the fa- blance to the Iranian piece, even borrowing sonic cilities in Cologne, and it was in fact premiered material. There is much that is new, as well, in- there in February 1978, months prior to the open- cluding a wild, brassy sound that is treated exten- ing of Le Diatope outside of the Centre Pompidou sively throughout. There is a greater prominence in Paris. Its North American premiere took place at given to textures of percussive sounds, such as ce- the 1978 International Computer Music Confer- ramic wind-chimes from the previous piece, but ence in Evanston, Illinois (to much acclaim, and others as well, including a plucked African thumb- some controversy). piano. There is a striking moment at the end, when Xenakis, almost uniquely in his entire output, the focus suddenly turns to the thumb-piano, an expended much effort on the program notes for Le exposed single note being plucked slowly and Diatope, presenting his own thoughts and gather- evenly, the ear zooming in on the rich, rattling, res- ing inspirational texts which, while not forming a onances of this distinctive sound. narrative, resonate in multiple ways with the cos- As one would expect, the various sounds are gen- mic, apocalyptic scope of the sounds and lights. erally layered and distributed across the seven The title of the music, La Le´gende d’Eer, is taken channels, though the density is not as high as the from the concluding passage of Plato’s Republic,in earlier pieces, possibly in deference to the vaulted, which a soldier is killed in battle then brought reverberant performance space. Particularly notable back to life full of images of the afterlife, including in Polytope de Cluny are the synthesized sounds, the famous ‘‘music of the spheres.’’ The medieval created using stochastic algorithms. Xenakis was era is represented by Hermes Trismegistus, an al- proud to have been the first in France to produce chemist, who, in a similar way, is given a vision of computer-generated sounds, even if he lagged be- the boundless darkness and light of immortality.

48 Computer Music Journal

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Figure 10. Design of Le Diatope.

The passage from Blaise Pascal’s Pense´es contem- concentrated, directed formal outline. The overall plates the insignificant place of humanity within trajectory, which proceeds without break, is one of the infinity of nature, and Jean-Paul Richter carries a gradual descent, returning at the end to the high the vision further, writing of the terror of being whistling sounds of the opening. The multi- alone in the universe. The final text is a scientific channel projection of the sounds aids perception of description of a supernova, presenting its awesome the different, often overlapping, entities, and also size and energy with detached precision. It is al- allows rather subtle shifts in emphasis to be ef- ways easier to interpret words than music, and fected. clearly, these texts all share a vision of the vastness There are eight basic sonic entities used in La of the universe, with different images of light and Le´gende d’Eer, present or dominant at different life within that infinity. The composer, in discuss- times (see Figure 11). Each is treated to a great deal ing Le Diatope and his other polytopes, confessed: of studio manipulation, including transposition, fil- ‘‘I want to bring the stars down and move them tering, and reverberation. Each is also varied in around. Don’t you have this kind of dream?’’ (Ma- terms of density. These techniques serve, on one tossian 1981). level, to create links between the different entities. Even if less protracted in duration than Persepo- For example, the high whistling tones of the open- lis, La Le´gende d’Eer is vast in scope, with a more ing, smooth at first, are varied with tremolo/

Harley 49

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Figure 11. Chart showing progression of sonic out- lines and sectional divi- sions of La Le´gende d’Eer.

amplitude-modulation effects to create a more a mixture of the plucked mbira (African thumb pi- striated variant. This developed sonority resembles ano), the rattling ceramics, and the tsuzumi. At high transpositions of the rattling ceramic sonority, 25’00’’, a rich electronic entity enters and gradually itself narrowly filtered to produce a recognizable saturates the texture. The pulsating, wave-like con- identity of pitch. These two entities in turn resem- tours of this sonority strongly resemble the pound- ble the high, narrow-band version of the granulated ing undulations of the second section of Jonchaies, undulating noise, and even the drum (tsuzumi) ele- the orchestral score Xenakis completed around the ment, transposed and mixed to create a texture of same time (Xenakis 1977). After close to 8 min, high density. Thus, while the entities are relatively this entity begins to fade out, and the brassy sonor- distinct, they can also be related along different ity takes over again. A number of other sounds en- parametrical continua. ter too, and the brass is less dominant. A wild La Le´gende d’Eer can be divided into eight sec- metallic sonority, sounding like a cross between tions. The first is the clearest, containing only the the amplified and distorted braying of a donkey and high, whistling sounds that Xenakis calls ‘‘sonic an electric guitar, is prominent throughout this shooting stars.’’ Over 6 min, this layer alternates passage. Up to that point, the general range of the between the smooth sound and the striated variant. sounds had been descending. The lowest transposi- The second section begins when other sounds be- tions of the brass and metallic entities carry the gin to enter and the high sonority starts to fade out. music through to the final section, where they fade This is a transitional passage, as five different enti- out over some 3 min as the high tones of the open- ties are introduced one after another, none of them ing enter and carry on to the end. dominating the sound-field. A rustling noise alter- The pavilion that was designed to house this nates with a brassy synthetic sonority, until, at piece, together with the light show that went along 17’13’’, this entity begins to layer a number of with it, was constructed from red vinyl stretched tracks upon itself, commanding most of the atten- over a metal frame. The design somewhat resem- tion. With the abrupt arrest of the brass voices, the bles the Philips Pavilion, although the point of de- fourth section begins with percussive sounds alone, parture was to create a form that would have the

50 Computer Music Journal

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 maximum volume for the minimum surface of like La Le´gende d’Eer, produced for a multimedia outer shell. Rejecting the obvious solution—the performance, though also conceived to be an inde- sphere—as being acoustically and visually poor, pendent concert piece. The Polytope de Myce`nes Xenakis created a more complex form. As he put it, was an outdoor spectacle mounted at the historic ‘‘the effect of the architectural form has a quasi- site of Mycenae in Greece. Similar in style to Per- tactile influence on the quality of the music or sepolis, this event included torches on the moun- spectacle presented within it. This is beyond any tainsides, searchlights crossing the sky, and also considerations of optimal acoustics or proportions’’ incorporated several of Xenakis’s ‘‘Greek’’ instru- (Xenakis 1978). There is an aesthetic character that mental and works. Mycenae Alpha, under the structure imposed on the performances within 10 min in length, served as an electronic interlude it. In the case of Le Diatope, the point of departure between the other pieces. was the sphere, but with the design’s double curva- The music is noisy and dense, made up primarily ture, its ‘‘flights,’’ it is an architecture ‘‘open to the of massive clusters designed so that the score world.’’ would also be visually compelling (see Figure 12). The basic impetus is to move from complex tex- Stage Four: UPIC tures to more stable ones and back again (or to a new complex sound). Interspersed are moments of Throughout 1972–1978, during which Xenakis was more focused, simpler sonorities. The main limita- heavily involved in overseeing the multifaceted tion of that first version of the UPIC was that it production of Polytope de Cluny and Le Diatope, was not possible to mix the different ‘‘pages’’ of the he was also working with his engineers at graphic score. Xenakis’s piece, then, is a succession CEMAMu on the development of a novel approach of different gestures, one following on the other. In to digital sound synthesis. The UPIC (Unite´ Poly- addition, each of these ‘‘pages’’ could be no longer gogique Informatique de CEMAMu) is a computer than one minute in duration. Mycenae Alpha is music system enabling the user to create sounds by built from 12 such graphic/sonic entities, two be- means of a graphic interface (Lohner 1986). All the ing repeated, to make a total of 14. elements of the sound are designed with an electro- In this piece, a structural distinction can be magnetic pen on a large electromagnetic drawing drawn between complex sonorities created by board. (Recent versions have replaced the pen and means of masses of relatively stable note-segments, drawing board with a mouse.) These elements in- and others created from dynamic arcs. It is easy to clude the , the dynamic envelopes of the design intricate glissandi on the UPIC merely by sounds, and the ‘‘arcs,’’ or notes. Interesting sounds picking up the electromagnetic pen and tracing can be obtained by designing noisy waveforms and them onto the design board—held notes, in fact, are complex envelopes, and also by layering as many as treated as a sub-class of the more generalized ‘‘arc.’’ several hundred notes (Marino et al. 1993). Xenakis The design of the waveforms (a hand-drawn wave- had always sketched much of his music on graph form is written into a look-up table) is also critical paper, so such an approach to sonic composition to the sonorities created. The timbres tend to be was perfectly natural. It has also proven to be an noisy, but also static. Variation is achieved through engaging medium for young children (and many the layering of the notes and the dynamic enve- others, of course) to approach creating sounds on lopes. the computer. Mycenae Alpha is remarkable for demonstrating the innovations of the UPIC system. Clearly, Polytope de Myce`nes; Mycenae Alpha though, this piece is far more limited in its sonic content than La Le´gende d’Eer. But, for the poly- Mycenae Alpha, Xenakis’s first piece created on tope at Mycenae, the harsh sonorities would have the UPIC, was completed in the summer of 1978, matched the savage magic of the landscape, so per- not long after the opening of Le Diatope. It was, meated with ancient myths.

Harley 51

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Figure 12. Page of graphic UPIC score from Mycenae Alpha.

Pour la Paix sage of text. Others are more abstract, and the re- maining fragments are supportive, to accompany In 1981, Xenakis was commissioned by Radio- either the or the recited parts. The choir parts France to produce a radiophonic work for the Prix range from straightforward chanting on a restricted Italia (although apparently it was never entered in set of notes, as in the first sequence, to more com- the competition). In response, he put together a plex sonorities—a compendium, really, of elements rather strange, -like work for reciters, choir, from Serment, an a cappella choral work from ear- and electronic sounds created on the UPIC. The lier that same year. texts are taken from two works by the composer’s In spite of the intensity of the texts, poetic rail- wife (a celebrated author in her own right), Franc¸o- ings on war and death, Pour la Paix is rather disap- ´ ise Xenakis: Ecoute and Les morts pleureront. Part pointing as a radiophonic creation. The sequences of this same selection of texts was used in the of material mostly succeed each other with little work composed just prior to this, Nekuı¨a, for choir overlap, with the exception of the electronic and orchestra. sounds which sometimes appear in conjunction Pour la Paix exists in four forms. The full version both with the spoken and sung parts. The produc- lasts over 26 min, and includes all three elements: tion lacks sonic depth and spatial organization, a the spoken text, the sung parts, and the electronic puzzling weakness considering the level of sophis- fragments. There is an abridged version comprising tication common in all kinds of broadcasts, not to the ten choral sequences on their own. The other mention Xenakis’s other electroacoustic works. two versions present the work in pre-recorded This would be his only foray into the medium of form, either completely on tape or with just the re- radio art, in any case. citers speaking their texts live. The spoken texts constitute roughly half the work, there are ten sung passages, and there are numerous sequences of Taurhiphanie UPIC material. The electronic sounds heard in Pour la Paix In 1987, Xenakis was given his last opportunity to cover a much wider timbral range than those in produce a multimedia event. The Festival de Mont- Mycenae Alpha. Some are cinematic, evoking the pellier commissioned him to create a work for the sounds of war, for example, in response to a pas- ancient Roman arena in the Provenc¸al town of

52 Computer Music Journal

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Arles. Bull fights were still held in that same arena, ety of sonorities making brief entrances. The range so the focus of the event became the incorporation of timbres is relatively wide for a synthesized of bulls into the proceedings (along with the fa- score, though still not as rich as pre-UPIC works mous white stallions of the nearby Camargue re- such as La Le´gende d’Eer. A break at 8’00’’ prepares gion). The plan was to wire the bulls with radio for a return to the bands of sliding sounds from the microphones to capture their snorting breaths. opening section, this time interrupted by dramatic These sounds would then be processed in real-time interjections of more concentrated, sweeping using the new version of the UPIC system, which sounds that heighten the intensity of the passage. would be installed in a tower in the middle of the As the music nears its conclusion, the sustained ring. Spotlights were to project patterns of light sounds begin to diverge, some migrating higher, onto the floor of the arena and the animals would some lower. The piece ends abruptly, as if cut off. create dynamic stochastic patterns as they moved Most of Xenakis’s electroacoustic works end in around the ring. To top it all off, twelve percus- similar fashion. sionists were stationed around the upper seats of While there is a certain roughness to Taurhi- the round arena, and they performed selections phanie that might indicate a lack of finesse in the from Xenakis’s percussion music (Idmen B, Ple´ı¨a- shaping of certain details, there is also a strong des, and ). sense of assurance and cohesion. As with the other In the end, a lack of time and resources forced pieces that had their sources in multimedia crea- Xenakis to drop the amplification of the bulls. In- tions, this music stands on its own and carries on a stead, he incorporated brief samples from this un- vitality that long-past spectacles cannot retain usual source into a tape work created at CEMAMu through mere documentation. on the UPIC. This piece, Taurhiphanie, was pro- jected in the arena along with some live manipula- tion of fragments of the sonic material using the Voyage absolu des Unari vers Androme`de UPIC. The animals did not cooperate in creating interesting stochastic patterns; they tended to hud- In 1989, Xenakis produced his most ambitious dle at one end of the ring or another, apparently UPIC work, a commission for the inauguration of traumatized by the pounding of the amplified per- an International Exposition of Paper Kites in Japan. cussion and the extreme dynamic range of the elec- Voyage absolu des Unari vers Androme`de extends tronic sounds. The performance was not an the image of a kite soaring through the air to a cos- unqualified success. mic voyage through space in the direction of the Taurhiphanie, though, remains as perhaps Xe- Andromeda galaxy. While the music is not exces- nakis’s most successful piece created using the sively programmatic, one may quite easily imagine UPIC. At just under 11 min in duration, it is traveling through space while listening, with vari- slightly more substantial than Mycenae Alpha, ous ‘‘episodes’’ occurring along the way. and, given the technological improvements of the The formal outline of Voyage absolu is con- UPIC, it is clearly conceived as an organic whole ceived on a broad scale, though the 15-min, 25-sec rather than a succession of parts, as in the earlier duration is still nowhere near the breadth of Per- two pieces. sepolis or La Le´gende d’Eer. The piece is structured The first 5 min form a continuous gesture, the in two parts. The first, lasting until 8’40’’, is the layers of sustained sounds gradually rising until a more turbulent, with numerous sonic scrabblings high point is reached at 4’22’’. At that point, the and short dramatic gestures surrounding more on- sounds begin to descend again, but are interrupted going layers of arching contours. One striking tex- at 5’28’’. After a series of shorter, disruptive ges- ture occurs at 4’48’’, where a percussive noise tures, a sustained sonority takes over at 5’57’’ fea- pulsation briefly sets up a regular beat that might turing some stable intervals—a ‘‘still point.’’ By go over well at a techno club. The second part is 6’45’’, fractures appear in the texture, with a vari- much more serene, as the sustained sounds undu-

Harley 53

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 late more slowly. In this section, the added sonori- By 1991, his team had succeeded in developing a ties are less prominent in the mix, though there are computer program that would not only synthesize many discrete gestures heard in the background. sounds stochastically, but would also implement The sonic arcs evolve both higher and lower, and macro-compositional procedures similar to those there are well-defined episodes where timbres and defined for the algorithm implemented in 1962 densities clearly shift. The music at 10’38’’, for ex- (Serra 1993). ample, is quite sparse, though by 11’30’’ it starts to On the synthesis level, the aim was to generate a build up again. The final few minutes encompass a that could be varied continuously ac- gradual ascent into the high register, decorated cording to some stochastic function. In searching with aliasing effects that produce low buzzing for the most efficient way to do this, Xenakis and sounds. his assistants eventually adopted a waveform cycle as the basic unit (rather than, say, a sample), sim- plified as a series of line segments rather than a Erod curve (see Figure 13). The degree of change, both vertical and horizontal, from one point in a cycle to By the time he reached the 1990s, Xenakis was the corresponding one in the next, is calculated on clearly not well. In addition, he was preoccupied the basis of a probability formula. The cycle is var- with developing his stochastic synthesis algo- ied both in its amplitude (vertical extent) and fre- rithms. He did, however, return to the studio to quency (horizontal extent). If each cycle of a sonic produce one last UPIC piece. Erod was created at waveform varies radically, the resulting sound will Les Ateliers UPIC in 1997 for a commission from be very unstable, or noisy. In order to obtain a the Bath Festival in England. Xenakis worked range of timbres, a second-order process needed to closely with Brigitte Robindore´, and they evidently be set in place to control the boundaries, or degree experimented with deriving new sounds from re- of possible change, for each operation. This process cordings of earlier scores, both electroacoustic and could of course be dynamic, effecting gradual shifts instrumental. The piece lasts less than 5 min, and in the rates of change. On the higher level, pro- it was withdrawn soon after its premiere. (The con- cesses were implemented in order to select the tribution of Ms. Robindore´ to the production of the number of ‘‘voices’’ (waveforms) activated at any piece was so integral that Xenakis could not in one time, their points of entry in the time-line of good faith put his name on it.) the piece, and the duration of each of these seg- ments for each voice. With these basic tools it be- came possible to input control data so as to Stage Five: Stochastic Synthesis generate an entire piece digitally on the basis of stochastic functions. As early as the late 1950s and early 1960s, Xenakis speculated about the possibility of synthesizing new sonic waveforms on the same basis as his ST GENDY301; GENDY3 algorithm for stochastic composition. He carried out some experiments to that end at Indiana Uni- Xenakis’s first compositional product from this versity, and continued at CEMAMu in Paris. His GENDYN (GENeral DYNamic stochastic synthe- first results were applied to electronic sounds pro- sis) algorithm was GENDY301, presented at the duced for Polytope de Cluny and La Le´gende d’Eer. 1991 International Computer Music Conference in Until the mid 1980s, though, the main preoccupa- Montreal, Canada. A newly-generated work derived tion at his research center was the development from similar control data was produced later that and promotion of the UPIC system. In 1986, with same year for the Journe´es de Musique Contempo- the loan of a minicomputer from Hewlett Packard, raine in Metz, France. This work, titled GENDY3, Xenakis was at last able to return to this project. is the version that has become known through sub-

54 Computer Music Journal

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Figure 13. Segmented waveform cycle subject to time and amplitude variation.

sequent performances and release on compact disc. What is surprising about GENDY3 is the degree In spite of the close relation of the two pieces in of harmonicity present. Every section contains terms of their genesis, they in fact sound com- some number of sustained voices, cutting in and pletely different. out in stochastic fashion, and there are most often GENDY301 is the shorter of the two, at 14 min consonant relationships between the stable pitches. 15 sec in duration. It also exhibits a wider dynamic In some sections, there are also less stable voices, range than the other better-known piece, with ex- either in terms of frequency (creating glissandi) or tremely loud textures entering suddenly on top of timbre, resulting in noisy or rapidly fluctuating so- narrower-range sonorities. In addition, it contains norities. There are some surprisingly lifelike tim- more breaks of silence. Xenakis never stated his bres at certain points, including vocal-sounding and dissatisfaction with the piece, but he may have de- brassy (though more like organ stops than real cided not to release a ‘‘family’’ of works as he had brass instruments) sounds. In fact, there is a great produced in 1962 with the data from his ST pro- deal of organ-like tone to the held sonorities. It is gram. (Those pieces include ST10 for mixed ensem- also true that the unstable sonorities, because of ble, ST/4 for , ST/48 for orchestra, their novelty, stand out from the other sounds in Amorsima-Morsima for mixed quartet, Morsima- the sections where they occur. Amorsima for mixed ensemble [withdrawn], and Atre´es, for mixed ensemble.) At 18 min 45 sec in duration, GENDY3 is clearly S.709 more substantial than the earlier version. Upon lis- tening, it is apparent that, as in the ST scores, After producing the two GENDY works in 1991, clearly defined sections are integral to the algo- Xenakis continued tweaking the GENDYN algo- rithm. The piece is structured, from the listener’s rithm (Hoffmann 2000b). For a monograph concert perspective, as a succession of eleven short sec- at Radio-France in December 1994, a new piece tions, each lasting from one to two minutes, the was generated. S.709 focuses more on unstable, dy- overall progression expanding these durations namic sonorities. These are produced by allowing slightly. The sections are distinguished by registral the vertical and horizontal points of the waveform placement and scope, and degree of instability in segments to vary more widely and rapidly. With a either frequency or timbre. Each sound is relatively duration of just 7 min, this is a much shorter piece consistent in its settings throughout a section, than GENDY3, but the degree of sonic activity— though there are always mixtures of types, and the the acoustical intensity—justifies the truncated ar- individual layers are often laced with a great deal of chitecture. silence, to avoid oversaturation. There are no clear sections in S.709 by which a

Harley 55

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 formal outline can be perceived. Instead, the music to, finally, digital synthesis. A detailed assessment is highly fragmented. The harmonic, sustained of his achievements in this domain in relation to tones of GENDY3 are entirely absent. By contrast, other work in the field remains to be undertaken. It each of what seem to be no more than three or four is safe to say, however, that Iannis Xenakis fol- ‘‘voices’’ are in constant fluctuation in terms of lowed a singular path, guided by his unique charac- pitch, timbre, or both. Amplitudes, too, are highly ter as a Greek revolutionary, trained in modulated. While there are some consistencies— engineering, influenced by architecture, self-taught certain characteristic sonorities—each voice under- in music. The seeds he sowed, his works, and his goes considerable transformation, both gradually technical or conceptual innovations, have proven, and in sudden shifts. The density of sonic informa- and will continue to be, fruitful for succeeding gen- tion within each voice surely dictated the reduc- erations. tion in the number of layers. In fact, for much of the time, there is only one layer sounding at a time. Acknowledgments Taken together, the GENDY pieces and S.709 ´ demonstrate the wide range of possibilities inher- I am grateful to Radu Stan of Editions Salabert for ent in the General Dynamic Stochastic Synthesis supplying me with archival recordings of the elec- algorithm. troacoustic works of Xenakis otherwise unavailable at the time of my research. I am also grateful to Brigitte Robindore´ for detailing the production of Conclusion Erod, Xenakis’s last computer piece, and for allow- ing me to hear it. Others who have been helpful in various ways in my research include: Cornelia Co- Once he developed a workable compositional algo- lyer, Cort Lippe, Gerard Pape, Jean-Michel Raczin- rithm in 1962, Xenakis integrated his creative pro- ski, , and Malcolm Smith. Thanks also cesses and experimental approach with computer go to Doug Keislar for his help in editing the arti- technology throughout the remainder of his career. cle. I would also like to acknowledge the support of It is worth noting, though, that some of the proce- Iannis Xenakis himself, who permitted me to work dures he developed prior to 1962 could also have with the UPIC system at CEMAMu during 1985– been programmed, had Xenakis had access to com- 1987, and who was generous in many other ways. puting facilities. He applied his stochastic (and other) functions to his instrumental and vocal mu- sic, and worked at CEMAMu to generate new References sounds digitally. Toward the end of his career, he at last found a way to reconcile his higher-order Boivin, J. 1995. La classe de Messiaen. Paris: Christian compositional concerns with the generation of the Bourgois. sounds (waveforms) themselves. Even so, Xenakis Brody, J. M. 1970. ‘‘Iannis Xenakis Electro-acoustic Mu- didn’t drop everything else in order to work with sic.’’ Liner notes. Nonesuch Records H-71246. these long-awaited tools. The acoustical richness of Delalande, F. 1997. ‘‘Il faut eˆtre constamment un immi- live instruments and voices remained compelling, gre´.’’ Entretiens avec Xenakis. Paris: Buchet/Chastel & as the vitality of his late orchestral scores such as INA/GRM. Da¨ mmerschein (1994) or Ioolkos (1996) attests. Di Scipio, A. 1998. ‘‘Compositional Models in Xenakis’ Electroacoustic Music.’’ Perspectives of New Music In the domain of electroacoustics, Xenakis’s mu- 36(2):201–243. sic evolved a great deal, from origins in the mu- Di Scipio, A. 2001. ‘‘Clarification of Xenakis: The sique concre`te style through more abstract Cybernetics of Stochastic Music.’’ In M. Solomos, noise-based sonorities that were fashioned into ed. Presences of/Pre´sences de Iannis Xenakis. Paris: continuously evolving waves of intensity, to the in- Centre de documentation de la musique contempo- corporation of transformed instrumental sounds, raine, pp. 71–84.

56 Computer Music Journal

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021 Fleuret, M. 1972. ‘‘Une musique a` voir.’’ L’Arc 51: Roads, C. 2002. Microsound. Cambridge, Massachusetts: 32–35. The MIT Press. Fleuret, M. 1988. ‘‘Il teatro di Xenakis.’’ In E. Restagno, Robindore´, B. 1996. ‘‘Eskhate´ Ereuna: Extending the Lim- ed. Xenakis. Torino: Edizioni di Torino, pp. 159–187. its of Musical Thought—Comments On and By Iannis Fulchignoni, E. 1981. ‘‘Sur Orient-Occident.’’ In H. Ger- Xenakis.’’ Computer Music Journal 20(4):11–16. hards, ed. Regards sur Iannis Xenakis. Paris: Stock, pp. Schaeffer, P. 1981. ‘‘Chroniques xe´nakiennes.’’ In H. Ger- 257–263. hards, ed. Regards sur Iannis Xenakis.Paris: Stock, pp. Gabor, D. 1947. ‘‘Acoustical Quanta and the Theory of 79–88. Hearing.’’ Nature 159(4044):591–594. Serra, M.-H. 1993. ‘‘Stochastic Composition and Stochas- Harley, J. Forthcoming. The Music of Iannis Xenakis. tic Timbre: GENDY3 by Iannis Xenakis.’’ Perspectives London: Harwood Academic Publishers. of New Music 31(1):236–257. Harley, J. and M. A. Harley. 1997. ‘‘Triumphs of Moder- Solomos, M. 1993. A propos des premie`res oeuvres nity: Xenakis’ Kraanerg at Canada’s National Arts (1953–69) de I. Xenakis. Pour une approche historique Centre.’’ Musicworks 67(Winter):24–31. de l’e´mergence du phe´nome`ne du son. Doctoral disser- Harley, M. A. 1998. ‘‘Music of Sound and Light: Xenakis’ tation. Paris: Universite´ de Paris IV. Polytopes.’’ Leonardo 31(1):55–65. Toop, R. 1989. ‘‘The Music of Iannis Xenakis.’’ Liner Hoffmann, P. 2000a. ‘‘Analysis through Resynthesis: notes. Etcetera Records KTC 1075. Gendy3 by Iannis Xenakis.’’ In M. Solomos, ed. Pres- Treib, M. 1996. Space Calculated in Seconds: The Phil- ences of/Pre´sences de Iannis Xenakis. Paris: Centre de ips Pavilion, Le Corbusier, Edgard Vare`se. Princeton: documentation de la musique contemporaine, pp. 185– Princeton University Press. 194. Varga, B. A. 1996. Conversations with Iannis Xenakis. Hoffmann, P. 2000b. ‘‘The New GENDYN Program.’’ London: Faber and Faber. Computer Music Journal 24(2):31–38. Xenakis, I. 1969. Letter to Wallace Russell, Administra- Lohner, H. 1986. ‘‘Xenakis and the UPIC.’’ Computer tor of the National Ballet of Canada, 6 March 1969. Music Journal 10(4):42–47. Toronto: Archives, National Ballet of Canada. Marino, G., M.-H. Serra, and J.-M. Raczinski. 1993. ‘‘The Xenakis, I. 1977. Foreword to Jonchaies. Paris: E´ ditions UPIC System: Origins and Innovations.’’ Perspectives Salabert. of New Music 31(1):258–269. Xenakis, I. 1978. Le diatope: geste de lumie`re et de son Matossian, N. 1981. ‘‘L’artisan de la nature.’’ In H. Ger- [Program Book]. Paris: Centre Georges Pompidou. hards, ed. Regards sur Iannis Xenakis. Paris: Stock, pp. Xenakis, I. 1992. : Thought and Mathe- 44–51. matics in Music, revised edition. Stuyvesant, New Matossian, N. 1986. Xenakis. London: Kahn & Averill. York: Pendragon Press.

Harley 57

Downloaded from http://www.mitpressjournals.org/doi/pdf/10.1162/014892602753633533 by guest on 02 October 2021