Synthesizing Performance

Abstract

The author describes synthesizing performance from abstract processes, her exploration of the computational potential of performing algorithms and her development of audio technologies to communicate real-time instructions to performers in her work M.T.Brain (2012).

From Meaning Comes Music

Onto what sort of surface shall “aesthetics” and “consciousness” be mapped?

—Gregory Bateson,
Mind and Nature

Recursively filtered through billions of brains, music is an aural lens, a prism for hearing the patterns of physical, mental and cultural organization. I used to make music in order to discover its meaning, but now I make meaning in order to discover its music.

This perspective shift was an attempt to free my music from the restrictions of tradition and habit. Instead of making music, I explored methods for organizing human interactions through space and time. The people were the performers; their interactions were the performance. The composition was the mechanism of organization.

Music became a tool. Its meaning functioned as the glue for other performance outcomes. Some pieces were made from simple rulesets. Others used technology to communicate organizing information during performance. When the rules could not be explained in advance, I created visual, gestural and/or aural cueing systems. Despite trying to free myself from tradition, I gravitated toward known systems of musical organization, integrating aspects of notation, improvisation, sonification and conducting in order to coordinate performances.

Computers Made from People

In a class I taught at California Institute of the Arts, we organized anti-disciplinary improvisations using the rules of computer algorithms. To perform a Bubble Sort algorithm, the students wore large numbers on their shirts and began by shuffling themselves into a single line. Functioning as an unordered array of people, they systematically stepped down the line, comparing the numbers worn by adjacent performers. When two numbers were out of order, the performers swapped positions in line. After a few iterations, they were standing in ascending order.

This is when I noticed that the performance did not represent the algorithm; instead it was the process itself. The performance computed its own output by physically sorting the performers in space. The performers were the data, or rather, wore the data. The performance ran the process. I call this performatization—the human performance of an abstract system. Performatization is performance synthesized from decontextualized processes.

In another performance of the Bubble Sort algorithm, we replaced the numbers with improvised musical gestures. After comparing their performed gestures, the performers swapped positions in line only if they wanted to. The experience of previous iterations informed the swapping decisions of subsequent iterations. The probability of a performed comparison generating a specific Boolean result correlated with the aesthetic preferences of the performers. Human personality and interaction determined the nature of the sort over time.

Generally, computer science is concerned with algorithmic efficiency, where the quality of the algorithm is judged by the amount of resources required. When algorithms are performed by people, efficiency can be less important than experiential and aesthetic values. However, performers have brains. They can contribute opinions and feelings to a computational process. By introducing human decision-making into a computational process, the result is an algorithm capable of self-determining its outcome.

Performance as a computational medium evades the limitations of deterministic systems by incorporating the indeterminacy of human decision-making. When computer programs run indeterminate processes, the indeterminacy is simulated using a deterministic system. Inversely, when computers made from people run deterministic processes, the determinacy is simulated using an indeterminate system.

Fig. 1.

Kristin Grace Erickson, M.T.Brain, performed by the Improvised Music Theater class, California Institute of the Arts, 2012. Performers respond to computer-distributed audio instructions while wearing channel numbers. (© Kristin Grace Erickson)

Organized Communication Communities Organization

I developed an audio cueing system called M.T.Brain (Music Theater Brain) for distributing discrete channels of real-time audio instructions to multiple performers. In M.T.Brain performances, the performers hear audio instructions combining text-to-speech audio, synthesized tones, audio samples and metronome clicks through headphones connected by very long cables to a multi-channel sound card. To...