Xenakis as a sound sculptor
Makis Solomos
First publication in French : “Sculpter le son”, in Portrait(s) de Iannis Xenakis, edited by François-Bernard Mâche, Paris, Bibliothèque Nationale de France, 2001, p. 133-142. First English translation (with the help of Sharon Kanach) in welt@musik — Musik interkulturell, Institut für Neue Musik und Musikerziehung Darmstadt, vol. 44, Mainz, Schott, 2004, p. 161-169
fRENCH ABSTRACT
Xenakis a puissamment contribué à cette révolution majeure qui a marquée le 20e siècle musical : la focalisation sur le son, où la composition du son tend à se substituer à la composition avec des sons. Cet article montre d’abord l’importance du son pour le jeune Xenakis, à partir de la lecture de son article grec de 1955, « Problèmes de composition musicale grecque » : d’une manière implicite, Xenakis y pose le son comme « fondement ». Puis est suggéré le fait que toute la musique de Xenakis peut être interprétée comme composition de sonorités. Enfin, on montre que, dans cette nouvelle conception de la musique, Xenakis prend un chemin particulier, qui peut le définir comme « sculpteur » du sonore.
One of the main revolutions —and maybe the most important one— of twentieth century music is the emergence of sound. From Debussy to recent contemporary music, from rock’n’roll to electronica, the history of music has progressively and to some extent focused on the very foundation of music: sound. During this history —when, in some way, composition of sound takes the place of composition with sounds—, Xenakis plays a major role. Already from the 1950s, with orchestral pieces like Metastaseis (1953-54) or Pithoprakta (1955-56) and with electronic pieces like Diamorphoses (1957) or Concret PH (1958), he develops the idea of composition as composition-of-sound to such an extent that, if the expression was not already used for designating a new interdisciplinary artistic activity, we could characterize him as a “sound sculptor”.
Sound as the foundation of music
Xenakis did not develop a theory for characterising this new idea of composition; at least, he did not treat it in his writings in a direct and systematic way—as it is well known, his theoretical work is mostly devoted to the question of formalizing music. But some important paragraphs of his writings refer to this focus on sound. One of his first articles, “Problems of Greek musical composition” (1955, written in 1954) [1] , which is very important for his early evolution, has such a paragraph. Xenakis wrote this article in a decisive moment of transition, that between the early works, like Procession to clear waters (Procession aux eaux claires, 1953), which belong to his Bartókian period [2] , and the first recognized pieces (such as Metastaseis), works that introduce the new musical questions and that place him in the centre of the musical scene. In the midst of this moment of transition, he is not sure which path to take. Although he is living now in France for six years, he is still torn between traditional Greek music (that he refers to as demotic music, i.e. rural traditions) and western music. But he also has to face the gap between past western music and avant-garde music. In addition, the avant-garde is now divided between the serialist school (Xenakis continues to speak about “dodecaphonism”) and concrète or electronic music. Finally, he seems fascinated by jazz music. In simple and direct words, he wonders: “Which then is the best path? What is real music? Traditional European music, dodecaphony, electronic jazz, demotic music?” [3] .
For a young self-taught (or almost) composer and, moreover, in exile, the choice seems impossible. There is no reason to choose one kind of music over another, to join this tradition rather another. This is why, to answer such a question, Xenakis asks another question: “Is there any link between these different sorts of music or are they, in fact, incompatible to the point that some should be condemned as being outdated or as being monstrous creations from abnormal periods?” [4] . His answer puts the question of sound in the foreground:
“The link does exist. It is the basis, the contents of sound
and of the musical art that uses it. Music is made up of sound messages, of
sound signals.
Acoustic sounds are analysed in physico-mathematical equations (being that
they are elastic vibrations of matter) that are measurable: intensity, colour,
time (duration). When we say colour, we include pitch, harmonics, additive
and subtractive sounds, undulations, etc. Consequently, sound is a quantitative
entity
But when sound goes into the ear, it becomes impression, sense, thus a qualitative entity. The psychophysics of music is not yet a science. The good composer could express the senses he wishes” [5] .
This statement could seem to be a commonplace, but such is not the case. Xenakis does not give a traditional definition based on the idea that music is a “language” and, that therefore, sound is defined as raw material and music as the “art of sounds”. He does not oppose a first level of articulation (sound) to a second one (music). Rather, he replaces this dichotomy with another one: quantitative/qualitative. The qualitative aspect of music is not totally defined —Xenakis talks about “impression, sense, thus a qualitative entity”. On the other hand, his definition of the quantitative aspect is very clear: it is sound. Thus, sound is not defined as a first level of articulation, as raw material. Upheld as the link between all different musical traditions, it constitutes “the basis” of music: we could say then that sound is asserted as the foundation of music.
Composing sound
To assert that sound is the foundation of music means, in a way, that the traditional way of thinking about music is reversed. Sound, henceforth, according to Xenakis, is not treated as the starting point of composition, but as its outcome. We could draw a parallel between twentieth century research of the musical avant-garde and the so-called “crisis of foundations”, which happened in mathematics at the same time [6] . While mathematicians tried to found mathematics, musicians tested several ways to found music on sound. From Schönberg’s Klangfarbenmelodie to Scelsi’s mysticism, composers who focused on sound phenomenon have been countless. Using an epistemologist’s terminology, we could define two ways to focus on sound: “intuitive” and “axiomatic”. In the first case, the question is to “let sounds be”, to paraphrase one of Cage’s famous maxims; in the second case, the question is to compose sound.
Xenakis belongs to the most representative group of composers who choose the second way. An important part of his music, both instrumental and electronic, can by analysed and listened to as composed sound, as sound synthesis transposed into the temporal scale of a work. Some of his compositions are very explicit. Such is the case with his two last electronic pieces, composed with the program GENDYN, Gendy3 (1991) and S.709 (1994) [7] . The program GENDYN, based on stochastic dynamic synthesis, completes the project initiated thirty years earlier with Achorripsis (1956-57) and the ST-program of 1962, a project whose main idea was to create a mechanism that, after introducing some data, could produce a whole composition. An algorithm synthesizes sound as stochastic changes on the time-pressure curve. Consequently, there is no difference between synthesis (of sound) and composition (in the traditional sense): in theory, composition is a direct outcome of synthesis —in Xenakis’ words, “macrocomposition” results from “microcomposition” [8] . Another example is the granular conception of sound that Xenakis developed in the late 1950s, in reference to Gabor [9] . At that time, he could only have the intuition of granular sound synthesis (which was implemented in the 1970s). So, he put forward the hypothesis that we could create a “sonority of second order” [10] and composed Analogique A and B (1958-59) to test it: the nine strings (Analogique A) play very short sounds (short arco, pizzicati or battuto col legno) and the tape (Analogique B) is made with clouds of very short sine waves; these short sounds are like “granular sounds” and Xenakis hopes that, for the listener’s ear, they will merge into a global sound [11] . With this hypothesis, composition is supposed to be like sound synthesis in a meta-level.
In Xenakis’ whole production, the above-mentioned compositions are exceptions. The general case is not one where macrocomposition is directly deduced from microcomposition, the first being a transposed sound synthesis [12] . Nevertheless, an important part of his music is like composed sound. To analyze his music in this perspective, we have to replace the word “sound” with the word “sonority”, in order to suggest that the composed objects are not exactly sounds (in the physical sense), but more complex entities, for instance whole parts of an orchestral work.
Let’s take as example the first section (bars 10-62) of Jonchaies (1977), where the strings (and some percussion) are playing. This is a very long part, almost monolithic, with a clear, inner, progressive evolution, which unravels like the evolution of a single sound. The strings are divided into 18 parts, sometimes playing in partial unisons and sometimes completely independently. It begins with a single high-pitched unison line, which follows a sinuous route to descend in an irregular way. Then other lines enter progressively, following the same route and constructing thus a broad heterophony. In bars 24-29, this heterophony reaches the low register and then begins to climb, always following a sinuous route. In bar 43, one of the lines reaches the highest pitch and is progressively followed by the others. Then the melodic route stays in the high register. Finally, some lines again descend. Despite the fact that the global route stretches over a long period, it is continuous and schematic enough that the listener’s ear can follow it throughout as the eye would do with a graph [13] . In addition, this part of Jonchaies is based on one single sieve (scale), illustrated in figure 8 (a sieve that, according to Xenakis, is near to the pelog scale [14] ). But Xenakis is not using this sieve for itself, as a succession of pitches. Pitch here is not a main feature: the sieve is maintained over a long period, and is explored throughout its whole register in a linear manner (irregular ascending-descending lines) and, finally, it is treated, thanks to heterophony, and constructs a giant sound aura. The sieve is used then to produce a sound color. In a way, this part can be heard as a single sound, which progressively spreads and can be explored, as though with a microscope and in slow motion, its inner composition and its temporal evolution.
Many of Xenakis’ compositions can be analysed as successions of sonorities, i.e. successions of sections that can be heard, metaphorically, as composed sound. Within the limits of this article, it is impossible to give other examples and to develop this type of analysis and of listening [15] . Let’s only add that such an analytical method is not exclusive. To listen to one of Xenakis’ compositions as a succession of composed sounds, of sonorities, does not prevent us from hear its dramatic qualities, for instance.
Sculpting sound
Within the “axiomatic” tradition, which leans towards seeing a musical piece as composed sound, Xenakis plays not only a major role, but also a specific one. With him, to compose sound means to work it like a sculptor. Let’s examine the graph with which he composed bars 52-59 of Pithoprakta [16] (figure 3). Xenakis first calculated, with the help of a Gaussian distribution, around one thousand glissandi [17] . But their distribution in time was made thanks to the graph. Thus, the global result —the sonority— is, according to Xenakis himself, a “plastic configuration of the sonic material” [18] . Now, let’s observe more carefully the graph. We see “holes”, empty regions of a whole portion of the register. We could compare this to filtering in electronic music. But we could also think to the work of a sculptor who carves a statue in a piece of stone.
Xenakis often composed with graphs, at least until the end of the 1970s. Many of the sonorities, with which he was the first to experiment and which make his music so original, have been conceived thanks to graphs. One may think of glissandi, a typical Xenakis signature. In Xenakis’ theories, a glissando is the general case of sounds with defined pitch. But it is important to notice that what allowed him to invent massive glissandi is the possibility to draw straight transversal lines in a graph where the two coordinates represent pitch and time. Syrmos (1959), a piece that is a kind of glissandi étude, was entirely composed with graphs. These graphs (see figure 17 for an example) reveal the visual imagination of Xenakis (and thus, his sonic imagination!), who defines different categories of glissandi (ascending, descending, crossed, convergent, divergent, in skewed surfaces). In the 1970s, it is also thanks to graphs that Xenakis invented a new kind of glissandi, the “Brownian movements”, which are characterized by non-linearity. The technique of “arborescences” is also the result of graphs (see an example in figure 7). It means also that, while analyzing a Xenakis composition, we can use _ graphical transcriptions as a tool for analyzing its sonorities. Figure 18 presents my transcription of bars 291-303 of Terretektorh (1965-66), which shows the extraordinary “braids” that characterise important parts of this piece.
Composed with or without the help of graphs, Xenakis’ music represents the antipode_ of the tradition for which music is an art of time. With him, music tends in a way to become an art of space. In a beautiful text on time, the philosopher Xenakis asks: ““Isn’t time simply an epiphenomenal notion of a deeper reality?” [19] ; after a reference to some paradoxical theories and experiments of modern physics, he states that space, “freed from the tutelage of time”, is a more substantial phenomenon [20] . As it sculpts sound instead of developing in time, his music illustrates perhaps this hypothesis.