Tonight, I’m heading to Brighton Dome to see a performance of Brian Eno’s seminal album Apollo as part of Brighton Festival, which Eno is curating this year.

Two years ago, I wrote an essay about Eno (and the wider theme of generative art) as a piece of coursework for a course called “Generative Creativity” (part of the EASy MSc at the University of Sussex). I thought today might be a good opportunity to put it online for posterity. Below is Part One of the essay, which is essentially a short biography of Eno, concentrating on his contribution to generative art and music.

Part One: Brian Eno

Generative approaches are gaining popularity in the creative arts, and their application to music is one of the most promising and active. The term “generative music” was coined by Brian Eno, and his contributions to the field have been increasingly ambitious since his early experiments in the 1970s.

In the mid-sixties, Eno became interested in the school of musical minimalism, including composers such as Terry Riley, Philip Glass and Steve Reich. In particular, Reich’s 1965 piece It’s Gonna Rain fascinated Eno. It is based on a tape recording of an American preacher saying the eponymous phrase. The recording was looped and played on two identical tape machines simultaneously. Due to slight inconstancies in the playback speed of the machines, the two loops gradually drifted out of sync. After some time, the listener becomes entirely habituated to the speech itself, and the perception of the artefacts created by the slowly shifting phase of the recordings becomes the main focus of the work. This notion of a very simple process being used to create a constantly changing and unpredictable musical texture became a foundation of Eno’s career.

His work in the early 1970s was the beginning of his foray into “ambient music” (another phrase he coined). This genre is usually described in terms of the output itself and its appreciation as a musical “background feature” of the environment [1]. However, Eno saw the underlying process, the “idea that it’s possible to think of a system or a set of rules which once set in motion will create music for you,” as the guiding principle of his work. [2]

The 1975 track Discreet Music used a very similar approach to Reich’s imprecise tape machines. Eno used a long-delay tape echo, fed with two simple melodic lines of different lengths from a synthesiser. The timbre of the lines was occasionally modified using a graphic equaliser, but the main interest of the piece is the interaction of the two lines with themselves and each other as they are shifted in time by the echo unit.

Similar processes underlie many of Eno’s ambient albums, with varying contributions from live performers, and more “standard” studio manipulation and production techniques. On a piece from Ambient 1: Music for Airports he used six simple building blocks, composed of sung notes recorded on tape loops of differing lengths. These elements do not change throughout the piece, but their specific configurations and clusterings are highly unlikely to repeat themselves. On the record, the track is eight minutes long, but this is merely an excerpt from a potentially infinite piece of music.

In the following decades, Eno began to notice parallels between his analog, studio-based techniques and certain concepts from science and computing. Cybernetics, the study of feedback in complex systems such as living organisms, provided a way to study how stable patterns could emerge without external control. Eno saw connections to self-regulating stability in group improvisation, where musicians may have only loose guidance in what to play but environmental factors such as the resonant frequency of the room may influence the outcome of the performance. He wrote several essays on the subject, including one called “Self-Regulation And Autopoiesis In Contemporary Music.” [3]

He became fascinated with John Conway’s Game of Life, in which simple rules applied to a grid of cells produce patterns which persist through time and space. He wanted the dynamics of the simulation to “become intuitive” to him. “I wanted to be able to understand this message that I’d found in the Steve Reich piece [...], in my own work, and now in this. Very, very simple rules, clustering together, can produce very complex and actually rather beautiful results.” [2]

At the same time, his own work was becoming more sophisticated. He began to move from directly interacting with patterns of sounds, to manipulating rules about how sounds were made. Computers presented an obvious tool with which to explore this idea. He became aware of a piece of software called Koan Pro, which could be used to control a hardware synthesiser on a computer’s sound card. Koan follows a set of rules which specify the parameters for around 150 elements of each composition, such as the instruments used, the timbre and attack of each note, and the relative proportion of each interval in the scale used to build the music. These rules are followed probabilistically, and form what Eno calls a musical “seed,” a framework within which the machine improvises to produce unpredictable and infinite pieces of music. [4] In 1996 he released Generative Music 1, an “album” containing twelve of these seeds on a floppy disk. Listeners ran a version of the Koan software on their own computer, which generated the music based on Eno’s rules.

Eno has also worked in the visual arts, often combining projector-based installations with his generative music systems to produce hybrid pieces. However, his recent work conflicts in some ways with his guiding principle of generating the complete piece from simple foundations. Rather than using the computer as a “way of growing little seeds” [2], his 2006 work 77 Million Paintings By Brian Eno comprises a corpus of more than 300 of his own hand-drawn images, and combines and overlays them to produce a continually changing visual experience.

Brian Eno sees generative music as a third alterative to recorded music and live music. “Like live music, it is always different. Like recorded music, it is free of time-and-place limitations.” [5] His contributions to the field have been varied and creative, and his future work (including a generative soundtrack to the forthcoming video game Spore) promises to continue his exploration of this new medium.

References

[1] Brian Eno. Music For Airports, Liner notes, September 1978.

[2] Brian Eno. Generative Music. Transcript of talk from Imagination Conference, San Francisco, June 1996. Available from: http://www.inmotionmagazine.com/eno1.html [cited 16th February 2008].

[3] Brian Eno. Keyboard Wizards, Winter 1985. Available from: http://music.hyperreal.org/artists/brian_eno/interviews/keyb81.html [cited 16th February 2008].

[4] Andy Oldsfield. Brian Eno’s Generation Game [online]. July 1996. Available from: http://music.hyperreal.org/artists/brian_eno/interviews/ind96d.html [cited 16th February 2008].

[5] Richard Williams. The Brain of Brian. The Guardian, May 1996. Available from: http://music.hyperreal.org/artists/brian_eno/interviews/guard96a.html [cited 16th February 2008].

The EASy MSc, which I’ve just finished, was a brilliant experience. It involved lots of hard work and was very stressful at times, but was ultimately extremely rewarding. I feel like I’ve learned more (and been more excited about what I’ve been learning) in the last year than at any other time in my life. The highlight was probably my trip to the Artificial Life XI conference in Winchester, and my publication (PDF) at the conference.

I’ve decided I’m going to stay in Brighton if I can, and I’m currently looking for a job. If anyone reading this wants to employ an enthusiastic and creative programmer, please feel free to get in touch! My CV is available here.

So, hopefully I’ll post again soon with something a bit more substantial.

Artificial Life – An agent-based model of meme transfer
Can global cultural distinctions emerge from local convergence? I will be recreating Robert Axelrod’s model of cultural influence from his 1997 paper “The Dissemination of Culture” (The Journal of Conflict Resolution, Vol. 41, No. 2, pp. 203-226, Apr. 1997). Axelrod’s model was implemented in Pascal, but I intend to use Processing. I’ll try to keep the implementation nice and flexible so hopefully I’ll be able to extend it fairly easily. I’ll be trying to reproduce Axelrod’s results and play with variations of the model to see what happens.

Intelligence in Animals and Machines – Implied aspirations of complexity
Many of the researchers working in the fields of evolutionary robotics and others building control systems for simple insect-like robots seem to present their work with an ever-present implication that the distant end goal is human-level, or at least human-style, intelligence. I will examine how valid this implication is, with specific references to evolutionary approaches and the magnitude and nature of the scaling problem.

Formal Computational Skills - Nash Equilibria and Evolutionarily Stable Strategies
I will study the concept of equilibrium points in Game Theory and how they relate to evolutionary biology.

Programming Techniques – Spellchecker in C
This course is about learning to program in C and assumes complete beginner status – it starts from scratch.  I already have some programming experience, so I decided to choose one of the suggested projects to enable me to concentrate on my other chosen topic, microcontroller programming.

Life and Mind

New Approaches to Modelling Evolution/Ecosystems