Emptyset - Futuristic sounds between machines, intelligences and bodies

Emptyset is a multidisciplinary project founded in 2005 by James Ginzburg and Paul Purgas, focused on the materiality of sound technologies and the relationship between sound and space. Through performance, sound design and electroacoustic experimentation within the tradition of analogue media, the duo has developed a distinctive sound that sits at the intersection of electronic music, noise and improvisation. Their records, from “Collapsed” (2012) to “Borders” (2017) and “Blossoms” (2019) up to “Dissever” (2025), have been released by Thrill Jockey, Raster Noton and Subtext. Emptyset will perform “Dissever” on Monday 16 February as part of the Inner_Spaces series at the Auditorium San Fedele in Milan. This interview is a condensed extract of the conversation recorded and published for Archive Ricordi's podcast series “The Music Folder”.

Let's talk about your projects. How do you approach a new work? And at what stage of your research does each of them take shape?

JG: There's no linearity to the process, we have several ideas gestating at different levels of maturity. It's also interesting that some ideas give rise to whole networks of outcomes. It can become difficult even to define where one idea ends and another begins. Then a conversation starts about how to implement that idea. I remember when we began working on a project called “Signal” (2014-15), in which we used the ionosphere, or that layer of the atmosphere, as a signal processor via long-distance radio transmissions. At some point Paul mentioned, almost in passing, the idea that the ionosphere was a kind of pulsing magnetic field, moving and changing at the edge of the Earth, and that signals could interact with it, and that it might be possible to sonify the interaction between the waves and that environment. From that idea a process of testing feasibility began. On top of that, my stepfather was an expert in long-distance radio transmission, having worked as a military radio operator and also for NPR in America. In any case, if the result doesn't sound or come across as poetic as the idea itself, then it's not really something we feel we can place within our practice.

When did that happen with “Dissever”? And how is the record structured?

PP: “Dissever” grew out of research stemming from the fact that James was working with Osmium, an improvisation collective exploring the history of instrument-building, web optics and related technologies. At the same time I was researching electronic music in India between the 60s and 70s. We reflected on how the genesis of electronic sound evolved over the course of the twentieth century. It's a cross-dialogue between different musical genres, but also between art and the history of technology. We started thinking about that moment around 1969 when cosmic rock and space rock were exploring a certain kind of futurism; at the same time, synthesis was starting to emerge and commercial synthesizers were becoming available. So that science-fiction narrative language around electronic sound was developing in parallel. On top of that, there was the whole New York minimalist school starting to establish itself. So there were all these practices, both avant-garde and mainstream, trying to shape and sculpt the language of the future. It seemed to us a rich space to step into and think about a certain kind of imagery. It's a kind of sonic fiction-making: it's like thinking of this transmission from 1969 to the present and imagining a slightly messier timeline, bringing different genres and disciplines together into something that could appear as a single unified object.

JG: One of the things we reflected on early in our work together is that, between the late 2000s and the early 2010s when we were focused on the interaction of basic signals with analogue recording equipment, we were still looking at old media. It's interesting. The rise of digital technology is the main driver of most production processes, and it mirrors the cycle of economic systems geared toward accelerating increases in productivity, not necessarily for the sake of qualitative outcomes. Although digital technologies have expanded the range of possible sonic experiences by orders of magnitude, the demise of old media technologies wasn't due to their being dead ends in terms of new possibilities, but to their being dead ends in terms of efficiency. So experimenting with the idea of other speculative futures for music starting from the mid-to-late 1960s ties into that whole set of reflections. I think it's also interesting how all this can be transposed. For example, I recently went back to reading all of Stanislaw Lem's works, and again “Neuromancer”, for the first time in years. It's really interesting to sit with the different moments in which sophisticated minds tried to imagine how every aspect of society would be transformed by technology, and to observe the limits of imagination: what lies beyond the horizon and makes it impossible to speculate about the future, and what instead turns out to be surprisingly accurate. As a curious example, I've noticed that in a lot of texts from the 60s and 70s, no one seemed able to imagine anything beyond the cathode ray tube as the main method for generating a screen.

It's striking how machine-centred the things you mention are, with such a strong fascination for electronics. Listening to your music, I became interested in the processes you set in motion to create your sound. Specifically, for “Dissever”, what machines did you use? How do you work with analogue and digital instruments? How do you create communication between analogue electronic devices, computers and algorithms?

JG: The studio where we recorded the album has a big mixing desk from the early 1980s. The actual core of the music consists of a series of sine waves generated by a computer. They start out as ones and zeros and a bit of noise. Everything else you hear is the way these basic elements trigger and interact with a series of parallel analogue processes. In the live room, the sound coming from the mixing desk was sent to various amplifiers. There were very physical interactions - for instance, Paul was in the live room swinging a microphone between speakers and mics to create a kind of phasing effect. Other pieces of gear generated automatic flanging, phasing and delay. A friend of ours owns a plate reverb that came from Pink Floyd's studio from the late 1970s - it's literally just a large metal plate with a transducer - and we'd send the signal into it. Then there were a number of processes happening in parallel that we could force to interact with one another. In the interaction of these different processes, at different thresholds, unexpected qualities always emerge. Any one of these processes taken on its own would be very familiar and would produce a flat, retro result: the sound of a vintage signal and a vintage piece of gear. But in the complexity that arises from the interaction of all these processes - adjusting thresholds, moving faders, tweaking parameters by hand, triggering patterns manually - you get complex, unpredictable qualities. A large part of what you hear comes from the fact that all these processes happen in parallel, and an emergent organic complexity arises from their interaction.

PP: It's interesting to think about digital modelling in the sense that, at the moment we're in now, the way circuits operate, the way microphones physically shift phase in space as they move, the way physical air vibrations layer on top of each other - that pure spatial materialism is what gives the record its momentum. And it still requires direct, hands-on intervention: going into a space, setting up a series of parallel signal chains and letting them perform. That's what gives the record its organic, live quality. It's also what shaped our approach to presenting the record in a live format. It's always tricky to figure out how to reverse the process when you mix a record and recreate that effect in a live context. But with “Dissever”, since everything was looking back at that period - like the emergence of multitrack recording in the late 60s - reflecting on that technological moment became the backbone both for building the record and for bringing it to life on stage.

In this process, what role does improvisation play?

JG: The actual foundation is a chord, it's simply a series of sine waves, a series of harmonics. Everything dynamic in the music is an improvised performance that interacts with all these processes. It's not an additive compositional process. It's a subtractive compositional process, in the sense that there's something already happening. It's as if the marble were already there and you were carving something out of it. There's this chord, and to create something with formal meaning it has to be gesturally manipulated to give it shape. In the studio we not only created an interesting result, but also a very clear template for the performance. Because in the studio it works precisely because it's actually a performance.

PP: When you work on a project for such a long time there are various paths, different iterations, whether it's the ionosphere as a signal processor, neural networks as compositional processors, or building our own instruments. There have been multiple iterations of events and sets that opened different doors, exploring those spaces. With "Dissever" the idea was to build simple forms. They were sound objects to interact with. It was almost an anti-competitive method, a way of creating a very tactile system, in terms of how it's animated and how it aligns.

I'm very interested in this aspect. How do you handle the relationship between the electricity of analog instruments and the artificial intelligence of digital instruments through your technical setup? Between human gestural operation and automated computerized processes?

JG: I think we pushed the exploration of automated processes to the limit with "Blossoms", because we had access to a generative adversarial neural network with which to create music. We trained it on our catalogue and on a few days of performance, and it was a slow process, but over the course of a month it produced increasingly sophisticated output. It was 2019, and by the end of the process it was clear that in five or ten years you'd be able to synthesize any music trained on a library of music history. There's no other outcome when you push automation to its extremes. Looking at this and realizing a certain inevitability, we understood that, to be meaningful, a sense of futurism has to resonate with what it means to be alive right now. There was no way to generate an anticipatory sense of the future based on that technology, because it felt like an infinite loop of cloning. This pushed us to look backward to find another way forward, even rejecting the forms of automation that have largely driven our lives. The automation we interact with on this record is a circuit that changes a VCA's level from minus 30 to plus 30. It's the dynamism of an electrical signal, not that of a neural network. Obviously computers are involved in our processes, but I don't know anyone who's worked seriously with tape in the studio and would want to go back to it. In some cases it's much more sensible to record onto a computer in the form of a digital file. There's no poetry in wanting to be a purist. The idea isn't a rejection of technology.

PP: With "Dissever" it wasn't just about thinking of technology as a set of technical processes. It's also tied to Atlantic technology and its corporate manifestation. Looking at the '60s and '70s in a more utopian sense meant thinking about a moment when people could pick up instruments and create a language to build the future themselves. That cycle of optimism felt like a powerful basis for looking backward with the aim of looking forward. It was also a reflection on the economic, social and political landscape of the time. There was a genuine sense of experimentation in the arts, in music and in society. The future was being proactively built, rather than assembled as a collage from fragments of already predetermined history. After "Blossoms", which was a direct confrontation with AI music production and which felt to us like a rather dark manifestation of the future, "Dissever" comes across as more optimistic. We haven't talked about "Blossoms" and "Dissever" as yin and yang, but in a certain sense they're part of a duality.

Do you think noise is the point where these two ways of operating converge?

JG: Noise is intrinsic to every system. There's no system without noise. Perhaps the idea of emergence is more interesting than noise itself, because human gesture is an emergent process. Complex AI systems will have unexpected emergent behaviours, but a human being is an infinitely complex object with peculiar sensibilities. Our sonic exploration is possible because, for evolutionary reasons, we're extremely sensitive to timbre. The ability to distinguish minimal differences in sound, in language and in rhythm reveals a dizzying level of complexity. You could call this complexity noise because it's unimaginable and impossible to schematize. And yet within noise there's an ordered complexity, from which formal experiences, ideas and expressions emerge that we can interact with.

Your tracks offer sensory experiences that explore the sounds of machines, ranging from kraut-rock to abstract techno. When you compose, do you think more in terms of looping or layering?

JG: I'd say neither, because our music isn't made up of disparate sound sources, nor is it built that way, especially on "Dissever". There's something happening - in this case a series of sine waves - and the way you interact with what's happening is the music. There are no constituent parts beyond the series of sine waves that trigger processes through electrical impulses. The closest we've come to layered work was with "Borders" (2017) and "Skin" (2017), where we imagined cosmogenesis as a process, reflecting on how complexity emerges from nothing, sonifying that abstract idea and playing with the ideas surrounding it. By playing with recursiveness, the idea of intelligence emerges. We positioned ourselves as gestural agents within that process. For those records we chose reduced, archaic instruments - a very simple stringed instrument and a rudimentary drum kit - to interact with and trigger sonic processes. The goal was to create something perceived as a whole, not as separate parts. This extends to "Dissever" too, and partly to "Blossoms". With "Dissever", much of the dynamism you hear on the record lies in the manual manipulation of microphones, mixers and other equipment.

To listen to the full interview on Spotify.

Ride the wave of great music

Add OndaRock to your favorite sources on Google Discover. Don't miss our reviews, milestones and latest news straight in your feed.

Follow on Google