Like Sound, Like Flesh
On January 21, at the Opéra de Lille, the ensemble Le Balcon will premiere Like Flesh, Sivan Eldar's first opera with a libretto by Cordelia Lynn based on a text by Ovid. The Israeli composer continues her work in computer music started at IRCAM on voice and sound diffusion, supported and assisted for the occasion by a team of experts: Florent Derex, responsible for sound projection and founding member of Le Balcon, Augustin Muller, computer music designer at IRCAM and with Le Balcon since its first concerts in 2008, and Koré Préaud, sound engineer associated with IRCAM, who is assisting Augustin Muller on the project During this interview, the three companions give us a look behind the scenes.
From left to right, Augustin Muller, computer music designer at IRCAM and RIM, Koré Préaud, sound engineer at Ircam and Florent Derex, responsible for sound projection and founding member of Le Balcon
Do you think that Le Balcon's sound diffusion work, which is part of the ensemble's DNA, may have played a role in Sivan Eldar's desire to ask you to work on her opera?
Augustin Muller: Of course, but it's also because we had already established a relationship. I had worked with Sivan at IRCAM on Heave, for the Exaudi choir, which premiered at Royaumont in 2018. And Le Balcon is a regular partner of the Opéra de Lille, the sponsor of Like Flesh. The collaboration made sense, given our common work and listening habits.
Florent Derex: I remember a very long discussion with Sivan after a performance of Karlheinz Stockhausen's opera Donnerstag aus Licht that she had come to see at the Opéra Comique. We had discussed the performative challenges of sound, its evolution since the 1980s in the context of written music, as well as the quasi-mystical role of the electronic tapes of Acts I and III, called "invisible choirs", which reminded us of the sacred church music that we hear, that surrounds us, but that we don't see. Perhaps she felt it echoed the work she was doing with Augustin on Heave at the time.
Did Sivan Eldar have a specific idea in mind for the electronics and sound diffusion of her opera?
F.D.: One of the main inspirations of Like Flesh is Ovid's Metamorphoses. Sivan has often spoken to us about her very organic relationship to sound, as if it were living, swarming. But your question actually touches on an essential problem of writing in mixed music, which is the relationship between composer and computer music designer. From my point of view and from the point of view of Le Balcon, the musical ideas prior to the writing of the electronics are developed during exchanges between a composer and a computer music designer (or even a musical director), and are eventually written by several people. I totally reject the point of view of having on one side a global artistic idea, a concept, and on the other side its implementation.
Concerning the sound diffusion, it is dissociated in two parallel systems: the first one relative to the electronics which is Augustin's domain—and the other one, more in connection with the orchestra—which concerns me directly. It so happens that for this opera, these are, in fact, two distinct physical systems. The challenge was to define the relationship between ththeese two- and this reflection will continue right up to the work on the stage with the director Silvia Costa.
A.M.: As far as the diffusion of electronics is concerned, the work on Like Flesh is a natural continuation of the work Sivan and I did for Heave. The difficulty of this piece was that it was to be diffused in the gardens of the Abbaye de Royaumont via a system already in place: a network of buried loudspeakers that had been imagined by Manuel Poletti for Carré Magique, an installation by Jean-Luc Hervé.
Just as in Like Flesh, Heave was set to a libretto by Cordelia Lynn, and one of the main themes of the piece was the biological link between the body and plants, the earth, and in particular mushrooms. We talked a lot with Sivan about images and dynamic behaviors of plants. It quickly became clear to us that a strange and fascinating world could be generated by sounds that would gradually rise up from the ground.
Work session at Ircam between Sivan Eldar and Augustin Muller, November 2021 © IRCAM - Centre Pompidou, photo: Déborah Lopatin
For my part, I had already done some experiments in "spatial synthesis", i.e., generating or modifying an electronic sound by taking into account the parameters of space. We gradually moved towards a "distributed" diffusion system, closely linked to a particular method of sound synthesis: Sivan recorded and selected a multitude of small elementary sound grains that we put into play using a computer system that I developed starting from the Antescofo language (developed at IRCAM by Jean-Louis Giavitto). We can then animate movements, forms, sound surfaces populated with these grains, which weave polyphonic and complex webs, intended for this particular broadcasting system, but we can also focus on one or several points of the system in particular to give the listener access to individuality within the collective.
This system seemed to us to be quite appropriate for Like Flesh, the universes being very close, in their references to the plant world and their transformations of the listening space.
What is the dramaturgical role of these diffusion systems?
A.M.: Without giving away the content of Like Flesh, the electronic part supports large parts of the opera. From material generated together in the studio, Sivan edited electronic tapes to enhance the vocal content. Obviously, electronics, as sound coming from nowhere and everywhere, allows for a change in the acoustic quality of the venue or, at least, its listening. The electronics, as well as the instrumental sounds, are sometimes more of a sound design, in the sense that they modify the scenic perceptions, in the same way as the light, for example. This latest use of electronics has been an area of personal—and with Le Balcon—experimentation for many years!
How did the development of these systems go? What avenues were explored?
A.M.: After the first performance of Heave at Royaumont, we tested the speakers, first at the Théâtre de l'Athénée, then at the Opéra de Lille. For a while we considered the idea of wireless speakers, which could be placed "anywhere", but this idea was not chosen in the end.
Koré Préaud: When I joined the project in the summer of 2021, I was able, on several occasions, to set up a scaled-down system that was as close as possible to what would be installed at the Opéra de Lille; a system that is used in all of Sivan and Augustin's studio experiments. In July 2021, we conducted the first full-scale test, installing about 30 speakers in the parterre of the Lille opera house. This in-situ test had several objectives: first, to make sure that the system works—and discovering the premises of this macroscopic spatialization was in fact a very beautiful experience —then adjusting the synoptic plans, schematics, and rack constructions, and, finally, evaluating the technical constraints of its implementation.
Conclusion: we decided to add more speakers!
In the end, which systems will actually be used?
K.P.: For the diffusion of the electronics, we have 56 loudspeakers distributed in a uniform manner on the floor, in the middle of the audience. 37 small speakers form a grid on the ground floor, reinforced by a "crown" of 8 speakers with more generous bass. 9 speakers are placed in a curve on the balcony to create a sense of height. Hidden from the public's view on the sides of the concert hall, two pre-wired racks complete the ensemble.
F.D.: For the sound system, we work with a system called WFS, which is able to reproduce acoustic fields. These systems are made up of rows of loudspeakers facing or surrounding the audience. Their main interest, from my point of view, is to be able to work on the localization of the sound sources in a very accurate way, whatever the listening position. To give you an idea of what I mean: we have all had the experience of having a bad seat at a concert because a speaker is placed right next to us from which we hear an instrument of a voice that is located in a completely different direction. WFS speaker arrays eliminate these problems, as well as the "feeling" of amplification, creating something more natural and transparent.
Photo : Koré Préaud, sound engineer at Ircam © IRCAM - Centre Pompidou, photo: Déborah Lopatin
What are the strictly technical issues of this installation: how do you install this kind of setup in an opera house like the one in Lille?
K.P.: The challenges are related to safety standards. For example, it is forbidden to install electrical connections in the room, which is why the speakers chosen are passive. On the other hand, the cut and soldered cables that will supply the speakers will be custom-made at IRCAM with black syndex in order to be as discrete as possible. The size of the speakers is also important. They have to be compact so as not to disturb the audience while offering a relevant listening experience.
Nevertheless, the technical constraints are not decisive in the implementation of the project. Everything is decided through discussions between the teams.
What are the issues at stake in the performance? Have you developed new computer tools to manage all this, or simply adapted existing tools for the purpose?
K.P.: As far as the sound tools are concerned, such as amplifiers, digital/analog converters, switches, consoles... we haven't invented anything. Everything already exists. The innovation, for us, is to make these different machines work in this very atypical configuration. Ordering and calibrating the system perfectly is quite daunting! All the more so as it must then go on tour. Everything has to be optimized to the maximum so that it can be set up quickly in theatres while still being sturdy enough for transport.
A.M.: As I explained earlier, the equipment and the synthesis system are intertwined. For the performance, we rely on classical diffusion techniques, but "massively" multichannel (several layers with 64 channels). IRCAM's Spat library is at the heart of the electronic performance system, since it lets us read and decode the source files for the sound diffusion system. We are also curious about the possibilities of spatialization of instruments or voices through this particular system that we have not yet had the opportunity to try out in real life. We are eager to rehearse at the Opéra de Lille to finally hear these sounds in their space!