The new artistic research project “Spirits in Complexity”1 is a cooperative effort together with the Johannes Kepler University (JKU) in Linz, the KTH Royal Institute of Technology in Stockholm, the University of the Arts in Reykjavik, and several artistic and scientific experts. It is supported by the Austrian Science Fund (FWF) through its PEEK and AI Mission Austria programmes.
Being at a university of music, we are keenly aware of how musical practice entails close practical and affective relations with involved objects—meaning acoustic instruments and other technological equipment. Over time, these objects become trusted professional partners that require but also resist bodily and intellectual interaction. Such non-human entities are attributed thing-power, with the ability to exceed the status of mere objects and generate encounters with an out-side of our daily experience. Working relationships with them can be of reciprocal kinship or of a more negotiatory or even confrontational nature or even take on ritualistic and spiritual forms. In our project, the notion of spirits is used metaphorically to indicate an opaque complexity and stands for a self-willedness of technological systems. Such tools depend on various kinds of measurability, which often contrasts with the motivation and practice of artistic expression—especially in music. So how can the uses of technology in artistic practices change if they shift from an analytical perspective to a relationship of negotiation, confrontation, or trickery? What forms of knowledge are thereby implied? And furthermore, what is the practical and theoretical impact of contemporary technologies’ black box-like nature? How can artistic practice approach, navigate, and populate unknown complexity and its associated context? The project began in April with a symposium at the mdw that included internal sessions involving all of the project partners as well as public talks and artistic performances. As an artistic research project, we explore our topic within the arts and mainly by means of the arts, employing a diverse set of methods that are often interdisciplinary and of an experimental or speculative nature.
At the mdw, the project is headed by Thomas Grill at the Artistic Research Center with artistic researchers Angélica Castelló, Marco Döttlinger, Patrik Lechner, and Hui Ye, cultural sociologist Tasos Zembylas, and student assistant Miriam Jochmann. Machine learning expert Arthur Flexer is our partner at JKU. Each of the involved researchers is responsible for a specific perspective on our topic:
Angélica Castelló explores the personal lives of music machines such as tape machines and synthesizers, how they have been shaped by use and acquired patina and special characteristics, be they fully functional or “out of order.” Disintegratory processes in which a machine begins to disclose its materiality and structure seem especially revealing.
Patrik Lechner seeks to challenge the physicality of materiality and the environment by modeling them in virtualization, which makes it possible to tweak fundamental parameters. Procedural modelling permits the introduction of unnatural shifts into the characteristics of music systems and their environments, generating new insights and creative possibilities.
Thomas Grill studies audio technology and its environment within a framework of feedback strategies, where the initial audio information gradually fades into the background in favour of systemic properties, dependent on the relationship between material/structural and contextual components—with only a minimal touch of human intervention.
AI (music) systems abstract the farthest from physical and sensory phenomena with the promise of a certain realism learned from massive corpora of examples. The complex nature of this technology and the fact that its functionality is not fully understood make it a true “black box” example. In this context, Hui Ye investigates the mechanisms of AI voice synthesis by exploring the possibilities of non-linguistic “vocalisation” by the machine while Arthur Flexer subverts generative music systems via so-called adversarial attacks. Marco Döttlinger, in turn, works on AI-based compositions that involve human and non-human participants as co-creators.
Zembylas and Jochmann observe, document, comment on, and analyse enfolding research processes that are often non-linear or intertwined and hence call for reflection, feeding back into our work as a team.
Overall, we argue that moving away from a controlling attitude towards technology in the arts and towards a discursive, exploratory practice of “making kin” opens up crucial and otherwise inaccessible potentials.
Recent public activities have included Marco Döttlinger’s project points of no return at the festival Klangspuren Schwaz on 10 September,2 the performance My voice my void by Marino Formenti, Thomas Grill, and Irena Tomažin at the Musiktheatertage Wien on 18 September,3 and a quartet performance by Castelló, Döttlinger, Grill, and Lechner at the Speculative Synthesis Symposium on 27 September in Graz.4 AI-based compositions by Döttlinger for “The Morricone Project” with the Black Page Orchestra will be presented at Musikverein Vienna on 8 May 2025.5
