info I contact
advertising
editorial schedule
acknowledgements
join the realtime email list
become a friend of realtime on facebook
follow realtime on twitter
donate

magazine  archive  features  rt profiler  realtimedance  mediaartarchive

contents

  

Meta-Crash


A-life's mysteries

Stuart Bunt on Mitchell Whitelaw's Metacreation

Trained as a developmental neurobiologist, Stuart Bunt is the Scientific Director and a co-founder of SymbioticA at the School of Anatomy and Human Biology, University of Western Australia.


Mitchell Whitelaw, Metacreation: Art and Artificial Life, MIT Press, Mass, 2004 ISBN 0-262-23234-0

Mitchell Whitelaw has produced a detailed and wide ranging review of artificial life (a-life) in art. Clear descriptions and analyses are presented of many major works in the field, which are divided into “breeders”, “cybernatures” (where “breeder” programs are placed in a more complex ecology), hardware (which overlaps considerably with conventional robotics) and abstract machines or cellular automata. These make interesting and contestable groupings in themselves.

Both “breeder” and “cybernature” programs often draw on evolutionary theory and language. Attempts are made to produce “small worlds” where a-life breeds and competes for resources, but all too often the basic algorithm is that of conventional Darwinian selection. As humans inevitably select interactively or provide the program’s simple rules for selection, these pressures are far less complex and subtle than those found in true ecosystems. When copies of life’s rules are written in, is it any surprise that life-like work results? The predictable outcome is all too often a trivial attempt to make pale, albeit aesthetically pleasing, imitations of the real world.

The major omission in Metacreation is any real coverage of biological works, with Jeremijenko’s One Trees one of the few examples described. In contrast, while little attempt is made to integrate artistic a-life work with the larger fields of biological science, computer engineering and gaming, Whitelaw does acknowledge the overlap. As a biologist I felt the relationship to current debates in evolutionary theory could have been better explored. Much of the artistic a-life work which produces aesthetic outcomes involves little critical evaluation of the science on which its theory is founded.

There has been an odd transposition occurring in the last few decades, with science leading debate while art avoids or ignores the wider implications of a-life work. Is it time for art to regain its creative and pioneering leadership rather than shrinking from confrontation? It would have been interesting to compare a-life art with the politically loaded and controversial work of a biologist like Richard Dawkins, who has also dabbled in the production of a-life.

As Whitelaw notes, “a-life art is under theorised”, a criticism often aimed at many branches of emerging ‘sci-art’, but one that many traditional art practices might pray for! This left me frustrated as I read account after account of works described in isolation, often with no analysis beyond that of the artists themselves. However it is worth the wait as the deeper analysis of works in 2 chapters towards the end of the book covers the relationships between a-life art and society, science and other art practices.

Whitelaw really gets into his stride when discussing the artistic context of a-life. He addresses the essential tension between those who see a-life as rich with creative potential and those who see it as a “value laden technoscientific practice” embedded in conventional scientific and information technology practice. Why is there a need in these artificial worlds to stick to a conventional view of competition based on the macho struggle to breed, kill or be killed? Whitelaw describes the gender issues involved in a-life creation, including womb envy and virgin birth. Does the competitive nature of these artificial worlds have any relationship to the fact that computer programming and robotic engineering are still bastions of male dominance in research and artistic practice?

In his final chapter Whitelaw tackles the issue of ‘emergence’ as related to a-life, asking whether such art works can actually produce effects unforeseen by the author and not determined directly by the computer code. All too often a-life works seem little more than toys; are the “breeders” actually as complex or socially interactive as Tamagotchi? Do the hardware “robots” truly have any emergent properties? Can a-life really be the “life of the future” when so much of it simply mimics the living world? I have to declare a vested interest but I do think a comparison with the cyborgs of Fish and Chips or MEART (www.fishandchips.uwa.edu.au/) would have been worthwhile, since in these works biological material not under complete human control has a major effect on the output. Works where humans interact with living organisms, such as Eduardo Kac’s Genesis, or where tissues are left to their own “creative-semi-living” devices, such as the Worry Dolls of the Tissue Culture and Art group, are serious omissions from Whitelaw’s review.
Kenneth Rinaldo, Autopoiesis, 2000, Ars Electronica-Cyberarts Kenneth Rinaldo, Autopoiesis, 2000, Ars Electronica-Cyberarts
When Ken Rinaldo says he is waiting for the day when one of his art pieces greets him with a spontaneous “hello!”, is he talking about robotics rather than art? Are these works just an extension of the programmer and engineer’s skill? If the creators are more interested in results and discovery than creation, are they more engineers than artists? Whitelaw worries that in many cases a-life may be nothing more than a fancy spirograph reflecting, as suggested by Cariani, the skills and rules introduced by the programmer.

While Whitelaw describes Adrian Thompson’s production by artificial evolution of mysterious, incomprehensible, functional electronic circuits, he does not describe similar work in neural network studies, where due to the almost infinite possibilities, the way the final program operates may be impossible to predict.

Can these works of a-life truly show us “life as it could be”, given that many employ complex culturally and biologically contextualised algorithms? With the exception of bio-art, Mitchell Whitelaw has provided an excellent review of attempts at a-life, enabling the interested reader to decide for themselves.

Trained as a developmental neurobiologist, Stuart Bunt is the Scientific Director and a co-founder of SymbioticA at the School of Anatomy and Human Biology, University of Western Australia.

RealTime issue #60 April-May 2004 pg. 7

© Stuart Bunt; for permission to reproduce apply to [email protected]

Back to top