Reviewer biography

The Art of Artificial Evolution: A Handbook on Evolutionary Art and Music

by Juan Romero and Penousal Machado, Editors
Springer, Berlin Heidelberg, New York, Germany/USA, 2008
460 pp., illus. 91 col., 78 b/w. $119.00
ISBN: 978-3-540-72876-4.

Reviewed by Jussi Parikka
Anglia Ruskin University, Cambridge


Digital culture has been characterized by neo-Darwinist tones (at least) since Richard Dawkins’ highly influential texts from the 1970s on. His later Biomorph software summed up some of the ideas by showing how the powers of evolution can be harnessed as part of digital creativity in visual design. However, if one looks at several of the art pieces made with genetic algorithms, one gets quickly a feeling of not “nature at work” but a Designer that after a while starts to repeat himself. There seems to be a teleology anyhow incorporated into the supposed forces of nature expressed in genetic algorithms. Or, as I think Manuel Delanda wrote somewhere, one gets the feeling the Nature ran out of ideas after a while of watching genetic art pieces that branded the imaginary of cyberspace and the emerging digital culture of the 1990s.

The Art of Artificial Evolution, edited by Juan Romero and Penousal Machado is a recent attempt to collect key ideas from this Neo-Darwinist wave around digital design. It promises to be a “handbook on evolutionary art and music” both offering some historical contextualisation and chapters on recent practical work on swarms, evolutionary algorithms and digital design.

The book is divided into five parts: I) Evolutionary art, II) Evolutionary Music, III) Real-World Applications, IV) Artistic Perspectives and V) Future Perspectives. Matthew Lewis’ introductory chapter “Evolutionary Visual Art and Design” starts from the usual suspects of Dawkins, Karl Sims and William Latham and offers a quick overview of different projects until recent times. He concludes the chapter addressing the problem mentioned above: how to make sure the possibility space for evolution in the evolutionary programs remains open and is not reduced to the human being doing the selection of what is the fittest (i.e. most suitable for the present purposes) of the alternatives calculated. What is revealed is a field far from forces of evolution running wild, and instead a handy tool for automating certain patterns in visual design or for example furniture. Indeed, there has been a continuous gap between the utopian promises of cyberculture discourse and the everyday work with evolutionary algorithms and swarm patterns. Through a narrative production, the meticulous design work has been turned into a paradigmatic mode of digital culture as a metaphorical ecology of a kind. In chapters such as Christian Jacob’s and Gerald Hushlak’s the use of ideas of evolution is introduced exactly in the context of creating e.g. chairs and other examples of “blob furniture” – and other interfaces with “real world” interaction, for instance in swarm installations. The chapter is an enjoyable read.

The Art of Artificial Evolution keeps its feet on the ground, and understands “art” mostly as the craftsmanship with digital tools. The chapters deal with evolutionary algorithms in rendering of photographs, The Electric Sheep project commenced in 1999 that works as an online collective and distributed intelligence/design piece (that for participants expresses itself as a screen saver), and for example with swarm granulation (in sonic arts). Chapters such as Tim Blackwell’s on swarm granulation are good examples of the subtle exchanges between various spheres of sensation and relationality. The chapter discusses swarm algorithm interactions with both sonic and non-sonic participation; for example the coordination of users’ bodies with swarm simulations. Such examples from the sonic sphere work best to illustrate the underlying radical temporality of digital culture (the chapter also addresses live algorithms), and flag how necessary for any cultural theory of sound and vision it is to understand the material modes of production of phenomenological events. Popular cultural transpositions of swarms into human phenomenology are familiar from examples such as Batman Returns (1992) and the Lion King (1994).

Embodied interaction with algorithmic swarms is also addressed e.g. in Alan Dorin’s chapter on virtual ecosystems and the already mentioned Jacob and Hushlak chapter which predicts the emergence of “future evolutionary design systems” that “will become hybrids of interactive, human-directed evolution and design ‘ecologies’ where design solutions compete for survival in an ecosystem that implicitly defines the constraints of the design spaces.” (p.164). Jon McCormack extends in his chapter some of the considerations in the book towards “software as a performance instrument” which is a nice way to frame the idea of evolutionary software intertwined with human agency.

What is at times troubling in some of the texts is the vague characterisation of art. It is too often mentioned only as the “subjective element” in the design ecologies or through references to aesthetically pleasing products – as something “interesting” or “beautiful.” In a way, the book demonstrates a very Kantian approach in the sense Kant articulated in his Critique of Judgment: “There is no science of the beautiful, but only critique.” Even though the algorithmic, neo-Darwinian processes are embedded in algorithmic and scientific discourses, the “art” element in the book remains undeveloped. Also the Neo-Darwinian element is very emphasised, and such approaches as Greg Lynn’s endosymbiotic models of parallel evolutionary algorithms are not discussed (for a stimulating critique of Neo-Darwinism from such a perspective, see Luciana Parisi’s chapter in The Spam Book, forthcoming from The Hampton Press.) In addition, some of the references to cultural theory (e.g. Galanter’s discussion of “postmodernism”) rely on generalisations and some troubling inaccuracies.

The collection works well, as the subtitle promises, as a handbook. A nice bonus is the DVD attached with visual and sonic examples as well as some experimental software for trying your own go with a bit of evolution and swarming.