The electro-acoustic chamber composition Répons by Pierre Boulez is more than a technical marvel: it fascinates with its Proustian reflections on time and a contemplative, almost Eastern emphasis on sonority as an end rather than just a means. The work's title refers to the responsorial form of Gregorian chant, with its alternating solo and choral parts, and Boulez has likened its spiral architecture to the corkscrew interior of the Guggenheim Museum in New York. Widely considered one of the signal masterpieces in the electro-acoustic form, Répons debuted as a work-in-progress in 1981, with further evolutions unveiled in 1982 and 1984. Leading the Ensemble InterContemporain, Boulez recorded his latest, 45-minute version of the work in 1996 to inaugurate Deutsche Grammophon's "20/21" modern music series.
Boulez scored this complex, highly textural soundscape for a chamber ensemble plus six soloists
playing two pianos, harp, vibraphone, glockenspiel and cimbalom [a Hungarian
dulcimer]; the soloists' performances are transformed by computer and diffused
electronically throughout the auditorium. The special audio apparatus required
by Répons will be in full effect for a rare performance of the piece
on 22 March 2003 at Carnegie Hall, part of the venue's two-concert SonicFest
with Boulez and the Ensemble InterContemporain.
In 1985, not long after debuting his third version of Répons, Boulez discussed the piece and its gestation, as well as his overall philosophies regarding electronics, with Josef Häusler.
Josef Häusler: In 1955, under the title "À la limite du pays fertile" ["At the edge of the fertile land"], you wrote an essay on the exigencies, possibilities and boundaries of electronic music. At the end of this text, you write that which would attract you most and spur you on to your greatest achievements would be the confrontation of the instrumental with the electronic sound worlds as multidimensional constructions. Unquestionably, this article gives a conceptual underpinning to a work that was presented to the public in its first version three years later and which has yet to find its definitive form: Poésie pour pouvoir. In the meantime, you developed the "pays fertile" even further: you founded the IRCAM institute in Paris and you lead it; you presented a second work that combines the two sound worlds: Répons for six soloists, chamber ensemble, computer sounds and live electronics. Are there direct connections between Poésie pour pouvoir and Répons?
Pierre Boulez: There are no direct connections, because the two works are
completely different in their compositional material. If I wanted to take up
Poésie pour pouvoir again, I would have to rework and modify the
material. I regard the material of the orchestral layer as valid, but not that
of the electronic layer. Let me explain. I find taped music inflexible. When I
was working on Poésie pour pouvoir, I still had little experience
— this work was practically the first. Since that time, I have
conducted pieces of other composers who use tape and have learned that one is at
a disadvantage because one must constantly take the synchronization into
account, leaving no room for the coincidences of interpretation. I don't mean
this only in the sense that one can select structures at will, but that the
gesture of interpretation — as I would like to call it
— is completely paralyzed, because one must pay attention to too many
things that have little to do with the actual performance. The time of a tape
recording is not psychological time but rather chronological time; by
comparison, the time of a performer — a
conductor or an instrumentalist — is psychological, and it is really
almost impossible to interconnect the two. Therefore, in Répons where I likewise incorporate
taped music, the tape layer is a music without exact synchronization: it serves
as a type of 'background' and is, in that sense, not accurately merged into the
mix. Naturally, there are cross-connections — analogies exist in the
structural relations and proportions — but there is no direct
synchronization: I hold this to be almost impossible to realize.
JH: Nevertheless, two things are made clear when one considers Poésie pour pouvoir and Répons next to each other: with Poésie pour pouvoir, there is a large orchestra in the center of the hall, a considerable number of loudspeakers on the hall walls and a rotary loudspeaker on the ceiling; the general concept is modeled on the image of an ascending spiral. With Répons, there is a chamber ensemble composed of 24 performers — not a large formation — again situated in the hall's center; furthermore, there are six soloists in a large circle at the periphery of the hall, and likewise a great number of loudspeakers. Here, the image of the spiral no longer applies; one thinks more readily of circles or rings: in fact, the piece at one time bore the working title "der Ring." Doesn't this imply a certain parallelism with Poésie pour pouvoir?
PB: True. The concept, which concerns the outside framework, the setup, the space, is identical. Thus there is a center first, which is conceived for the eyes as well. Stravinsky said, "One hears the music, but one must also see it." Therefore, one sees here — with the chamber ensemble — the music. It is played by musicians who use no electronics, no amplification, no distortion, no "sound disturbance," so to speak. In the circle, or in the square around it, sits the public; behind the public, two musical circles are formed at the hall's periphery by the soloists, who are modified electronically, and loudspeakers.
JH: I would like to ask about another parallel. The sounds of the chamber ensemble you were just referring to are neither amplified nor distorted; however, the sounds of the soloists are transformed, and furthermore, there is certain sound material on prerecorded tapes. We have thus a threefold sense of the sound events: the normal sounds of the instruments, the distorted sounds of the instruments and the synthetic electronic sound. This threefold perception of sound was also apparent in Poésie pour pouvoir. Is this threefold perception of sound based on a way of thinking that is central to Pierre Boulez?
PB: Yes. But it was not yet possible at that time for Poésie pour pouvoir to transform instrumental sound directly during the performance.
JH: Not this, but on the tape — or the tapes — instrumental sounds were already transformed.
PB: Certainly, as I have already mentioned, the principle is the same. Nevertheless, a great difference exists: in Répons, the sound transformations occur in real time, and I consider this crucial. Having conducted many concerts, I no longer wish to neglect the direct influence of the performer during the performance.
JH: You are halfway to anticipating the response to my next question: what for you, in recent years, has been the most important experience in the field of electronic music?
PB: It has to do with this "real time." The computer makes it possible to transform sounds at the moment that they are produced during a performance. Of course, this occurrence is based on a prearranged plan, but the programs used are determined at the last moment, which makes them accessible in many ways. One is not bound necessarily to a certain order of programs, and that is most important to me, because this allows you to play with them. I believe that in future we will have musical scores that are determined on basic levels — in the structure, for example — but whose parameters can be modified at the last instant.
This is why I am working with my assistant Andrew Gerszo on a concept that is most important for me: a synthetic score in which certain items can be determined in the last instant. In this way, we would have the ability to alter more than just minor things; we could effect complicated conditions. I imagine scores whose elements mix practically only in the real time of the performance, so that an event occurs in a certain form only if one truly wants to have it in such a way at that particular instant. One could also — though it would require a little more time — modify the parameters: the duration, for example, or the volume. So this is what is important to me: that the compositional structure remain the same but without being bound to a prefixed form.
JH: Répons was performed in 1981 and 1982 in Donaueschingen, Paris and London, as a collaboration between IRCAM and the Freiburger Experimental Studio der Heinrich-Strobel-Stiftung des Südwestfunks. Where do you see possible connections — or seams — between the two institutes? Where are the differences situated?
PB: I believe we have in IRCAM — at least at present — the privilege of being a younger organization. Right from the beginning, we did not rely completely on analog technique, but from the outset, we favored the computer. That is an advantage, but also a disadvantage. An advantage, because our devices are the latest. In addition, we were bound to no tradition and so were freer; however, it could also be that compared with the Freiburg studios we have practically no experience with analog technique. But even in Freiburg, the sound distributor, the Halaphone, is now also attached to a small computer. During the collaboration on Répons, everything was connected through the computer, because computer technology allows us to shape sound, to transform it. The space distribution as well as the entire allocation of the individual similar and digital devices came from the Freiburg studio and were likewise connected through a computer.
I am sure that in the future the Freiburg studio will be more strongly inclined toward computer technology — also in the area of the sound transformation. Because, you see, I had the following experience: 13 years before — in 1970 — when I was working on the blueprints for IRCAM, I said to myself: there are computers, there are also other possibilities. But then, in the next 10 years or so, it turned out that the computer triumphed. This technology is simply indispensable: it functions fastest, it is most refined, the means for shaping sound is available, and so one can think the composition through structurally.
What always disturbed me about analog technique is what I would like to call the Mechanische [mechanical aspect]; it disturbed me because it tends to be too simple. For example, it was difficult to modify spontaneously through the delay mechanism alone; with repetitive processes, the repetitions always run in the same way.
For me, I say it again, the structural composition is the beginning, but I also attach great importance to the modifications, the variations. With computers, one can have these structural variations in each moment. Therefore, computer technology is, at least to my compositional nature, closer than analog technique.
JH: This brings us to the question of sound transformation in Répons. The sounds of the six soloists are transformed electronically live during the performance. What means and degrees of transformation do you apply?
PB: There are two types of transformation: the tone quality transformation and the structural transformation — the structural extension.
Tone transformation is not new; little has been invented, but one can do things more precisely with more precise means. There is the ring modulator and the frequency modulator. Those are the two methods of transformation that I used in Répons, and a third form of transformation occurs in the "Wallpaper Music" through phase shifting.
Concerning structural aspects, one can go further with computer technology than with analog technology. For example, I can give a rhythmic profile to a sound through delay. This profile can remain the same, and in addition it can be modified. Now when I gather two, three, four or six such sounds, I also get two, three, four or six different sequences, and I can still modify these sequences in each instance. In Répons, there is a passage in which what the soloists play is delayed according to a certain rhythmic sequence; however, when this passage returns, it is transformed through a new time sequence with a changed timing. I could also couple this whole process with a Klangfarben, a timbral, transformation.
JH: One of your techniques for sound transformation is the gate-switch function: here an instrument or instrumental group is controlled by another instrument or group, which leads to the fact that a certain timbre or certain dynamic is imposed on the controlled group.
PB: This principle works two ways. In the first one, the sound transformation occurs only if a certain instrument or a certain group plays; the transformation then follows the musical curve of this instrument or this group, and projects it onto another group. In the second case, an already existing sound transformation is finalized, as it were, by the entrance of the group or the instrument. Thus, one can use the gate-switch positively and negatively. The procedure offers an enormous wealth of possibilities, if one thinks that the structure of the music and the relation of the instruments to each other constitutes a defining role.
By the way, coming back to the delay: here I also used "aleatoric" structures in Répons. Let's assume I have two sequences, A and B. Between the two I can "flip flop" in any direction — which is controlled by coincidence — periodically. Periodic coincidence, because it is important to me to avoid regularity. Of course, this results in quite unforeseeable results, which are always surprising, but they are also always structural because the field is already there. Of course, I did not predetermine the exact structure; however, I did exactly define the field of possibilities, in which coincidence then intervenes. I enjoy using this technique, because it mixes a quite strict structure with more aleatoric structure, which lets the coincidence break through the rigidity.
JH: Is this procedure applied with all solo instruments?
PB: Yes. Not at the same time, but successively. It is like this: if one has six instruments as in Répons, it is important not to do too much because otherwise the result will be completely muddled. A relationship between order and disorder must always prevail. Disorder can go as far as chaos, but the chaos is to be regarded only as a limited case.
JH: A final means of transformation in Répons is the spatial transformation, the spatial migration of the sound.
PB: Like the transformation created with the gate-switch, the spatial transformation is bound to certain external processes. In the "positive" case, it was a musical curve — at this point, the transformation has geometrical forms: circles and loops. This process, which goes from loudspeaker to loudspeaker, can be predetermined, and it can, so to speak, be mechanical: moving ahead at a certain rate. The rate could also depend on the volume of the sound: if the dynamics are very strong, the motion runs fast and dies gradually with the disappearing sound. I also set the space distributor in relation to certain sonic parameters. A piano, for example, has in its lower register considerable volume; a harp in the high or middle register has a quite low volume. Naturally, that results in quite different rates.
However, I do not want to suggest that one thereby plays as in tennis or car racing, back and forth or in a cycle with everything following exactly. For me, sound mixtures are far more interesting; one feels that they rotate somehow, but one can't have the picture of an exactly comprehensible movement. With these different rates and with these different times, one gets an impression, which is much richer for me than a more normal circle or a loop to the left or right. For me, this is too anecdotal.
JH: We should speak now of the tapes, the so-called "Wallpaper Music." The six soloists have these tapes on reel-to-reel tape recorders next to them and set them off after certain signals during the course of the piece. How were the tapes produced?
PB: All together, on the same machine, the so-called "4X." This little computer not only transforms sounds but generates them, too. All tapes were manufactured and produced according to the same structural principles, containing only synthetic sounds.
JH: Sounds produced through a preset program by the machine and then stored for use as "Wallpaper Music" were then the result of a program designed by Pierre Boulez?
PB: I composed a score containing all the data. Pitches, volumes, duration, overtone spectra, phase shifts — everything, really everything is dictated by me, derived from experiments and tests, which I undertook alone and with Andrew Gerszo. Andrew then translated these data into computer code. That is sometimes very strange: one supplies quite simple raw material, but the result is complicated in such a way that other musicians or assistants have great difficulty when they read the score and compare the resulting sounds. Because the technical processes abstract the original data so completely that it is often very difficult to make the connection between score and sound. One finds oneself in a perfectly untempered world: original data for the pitches are given as a well-tempered tuning, but through the phase shift, one abandons the well-tempered tuning. Sometimes, I work with phase-shifting in very high overtone ranges from frequencies that are inaudible or musically meaningless — at one or two Hz — and if one then hears the 20th or 25th overtone, one gets completely untempered intervals. I find that very interesting, because one operates with imaginary pitches as a basis.
JH: What is the relationship of the sound material of the tapes to the sound material of the instrumentalists — is it the soloists, the chamber ensemble or both?
PB: The starting point is the same. For example, in the phase in which the "Wallpaper Music" is used, all pitches are fixed on one chord; the chord is played by the chamber orchestra, the soloists put melodic lines and ornamentations over it. The chamber orchestra also plays a type of variation, but the variation returns again and again to the chord. The chord forms an ostinato, which lasts for a long time and is "decorated" with melodic lines and rhythmic variations. Then the tapes are added. They are based on the same chord, but with the phase shifts and overtone distortions of which I spoke earlier. In part, these tape sounds make the chord strongly noticeable; in part, they distort it completely, and they occur only if the soloists play at a certain volume: then the "Wallpaper Music" works like a beacon. If the musicians play quietly or not at all, the light reduces itself, goes out completely. Thus one hears the chord very clearly in the center by the chamber orchestra; and if the soloists step out in a certain way, the chord also sounds over the loudspeakers in the synthetic music from the periphery: as "Répons," as in "Responsories," a kind of antiphonal music with the chord in the center.
JH: Let's dwell for a moment on the "4-X" machine. In a text that IRCAM published in 1981, we read: "the '4-X machine' — a computer with extremely short calculation time — is connected to the soloist via contact microphones. A program for pitch detection produces small scores for artificial instruments." How are scores produced through the machine?
PB: It has two functions. First, the machine can store something; therefore, it has scores in reserve, which are pre-planned and prepared. The second function is the detection of pitch, which is done through a filter. One must tread with caution here. The pitch detector does not react so quickly, and if one is too hasty, the machine can become confused. It sometimes made inaccurate substitutions between the basic tone and the first overtone.
The "trigger," which starts the machine, can be connected with one specific pitch or a number of pitches. For example, if one plays a G very loudly, then the machine understands: "OK, now I must react," and then it supplies a sound transformation, a rhythmic delay or some kind of score. Schematically, there are again two possibilities: if one — to stick to our example — plays a G on the piano, then it becomes the "trigger" of a program that can modify or enrich the pitch; or the G releases a program which transforms the tone of the piano — for instance through a ring modulator. If one plays an A — the A of the middle range [440 Hz] — the program will stop. If the A is followed by a B, the program "Time Delay" can begin. If afterward the C or perhaps again the A is played, the time delay is displaced. In other words, the instrument influences through its sound what will be chosen from the memory of the machine.
One also can choose scores. In order to remain with our example, it would be possible with the G to elicit from the machine a score of — let's say — 30 seconds duration. If you want a 30-second score, you only have to play a G, and that would be everything.
JH: What does such a score look like?
PB: It's a synthetic score, and thus a combination of synthetic sounds. With the machine, one can choose either a program for sound transformation or a score. The instrumentalist can choose the sound transformation for his own instrument or for another — that's also possible. And if you choose a score, it is possible to determine the start and end through the way you choose.
JH: "Répons," "responsory," "antiphony." It's your intention in this piece to create different types of responsories between the different sound groups. Did you also compose antiphonal responses within the individual layers?
PB: Yes. The responsive dialogues occur in particular between the soloists. The soloists are not a block that plays "against" another block. I particularly think of a place where the soloists are practically on their own and the chamber orchestra in the center forms only a type of sound background, in contrast to the beginning section, where this center group is the only one heard. Thus sometimes the soloists are quite independent and "speak" to each other by means of electronics. They send each other messages: the first pianist, for example, sends the "Wallpaper Music" to the vibraphonist, he passes it over to the cimbalom player and so on, and they play with this "background music."
JH: You spoke of the initial part, in which the chamber orchestra in the center of the hall plays alone. No soloists there, no sound transformation. In this section, there are relatively long trills, which again and again are interrupted by the insertion of woodwinds and brass. Is this already meant as a "Répons"?
PB: There is a type of "Répons" between the three groups of the chamber orchestra: woodwinds, brass, strings. Later, everything mixes, but in the beginning, the groups present themselves rather purely; one really hears them as groups. Then one understands the importance of these three groups; moreover, one understands the "Répons" principle much better, if one has already heard the pure color. Then the color modification becomes much more obvious.
JH: Let's return to the beginning of our conversation. "À la limite du pays fertile" is one of your essays from 1955. If I try to sum things up, the "pays fertile" of the future is for you the area of computer-oriented live electronics in connection with traditional instruments. Is that correct?
PB: I believe so. Yet one must find the right way to communicate. Pure computer music —derived from an aleatoric structure or in a form that is only determined in the last moment or runs in real-time cooperation of different forms and structures — can have its function for an exhibition, a specific group of listeners, etc. But in a concert, with its closed form, pure computer music has no connection with gesture. A performance experience, and therefore the instrument, must be the means of communication. Yet an instrument has its limits, which can be expanded through sound transformation. But one still would have the impression that it is a matter of interpretation, not something entirely fixed, as something dead.
Music is not like a picture. A picture is finished; you walk through the exhibition, you look at it, and already due to its material you have a quite different time experience than when listening to music. The time it takes to look at the picture is already contained in it. In music, the time experience is linked also with coincidence, with little things which modify a score each time. A score is — at least for me — only a model; in order to stay interesting, this model must become alive, even with errors: something is correct or is not correct, or it is correct in another way according to differences of interpretation: one interpreter stresses here somewhat more strongly than the other, there an interpreter is slower or faster — that is a material one can make flexible. For me, this definition of time, this aliveness of time, is the most important thing in music.
This article was originally published in Teilton,
Schriftenreihe der Heinrich-Strobel-Stiftung des Südwestfunks, vol. IV, Kassel, Bärenreiter, 1985, pp. 7–14.
Translated from the German by Clemens Merkel and Christina Unger. Reprinted by permission of Bärenreiter-Verlag.
