Abstract
This paper discusses the notion of emergence, the result of the collapse of both scientific and artistic barriers which have contributed to the rise of Artificial Life art. A discussion of artists who use biology as model and computers as material will lead into a description of some current Artificial Life art works. The author finishes with conclusions about the work of art in relation to Artificial Life techniques and interactive art in particular.

In our age the primacy of machines - amplifiers of muscle - have been surpassed by the primacy of computers - amplifiers of the mind. The profusion of media that the computer has fostered has collapsed barriers between unique professions, and a postmodern emergence of new forms, artistic and scientific, seems to be blooming forth everywhere. Computers and their use in art are now more than ever creating dynamic relationships between the works themselves and the viewing, interacting public. While the physical manifestation of the machine made distinctions between the self and the machine clear, the computer and its manipulation of information and ideas, make its separation from the self often less obvious. This is evident when it comes to world wide web (web) and the hours one can spend melting into the masses of interesting information. Although the WEB has allowed those able to tap into this dizzying array of databases an unprecedented information wealth, it has also created an abundance that cannot be completely trusted or thoroughly navigated. Still the world wide web has framed a cultural moment that could aptly be called synthesis, where ideas can be shared rapidly across time and great distances.

Emergence is the new paradigm for a global change encompassing this earth. Artists, scientists, theoreticians and researchers alike are no longer solely concerned with fields of pure research to the exclusion of other perspectives; they are instead concerned with the convergence and sharing of knowledge from all fields. The British developmental biologist, Mae-Won Ho, speaks of "a global phase transition" that is circling the earth and touching all disciplines, and can be characterized as:

An emphasis on integration over fragmentation, on cooperation rather than competition, on dynamics and process in place of the static and mechanical, on nonlinear distributed interrelationships and emergent properties of collective wholes instead of linear unidirectional or hierarchical control of incidental parts.

For Mae-Won Ho this is both acknowledgment that we construct reality and that we are moving away from scientific reductionist views that separate humanity and the sciences from the natural.

In the sciences this synthesis includes the joining of quantum electrodynamics and chemistry on one hand and the fusion of the biological sciences with mathematics, physics and computers for the purposes of computational visualization. In the multimedia arts it is difficult to find work that is not interdisciplinary in nature.

Physician-scientist Jonas Salk, in past discussions with Jacob Bronowski, speaks of this synthesis as the next cultural evolutionary step in the ascent of man in the cosmos. Bronowski speaks of the convergence of the cultures of science and humanism:

The evolution of Complexity demonstrates that in the physical sphere, complexity proceeds from elementary particles to atom to molecule. In the biosphere, molecules, cells and organisms replicated themselves. In the Meta-biosphere, the human mind and human culture have evolved...

 For Salk this points to the merging of the cultures of art and science. This Meta-biosphere is evident in the fractured multiple views of postmodern art with new art forms that seamlessly utilize the profusion of scientific research, surplus scientific material and artistic motivation. Walter Benjamin predicted that this profusion of new media would collapse distinctions between genres. This is epitomized by the artist Laurie Anderson and her performances that demystify technology and its use, or Myron Krueger's research at the University of Wisconsin, participating in the creation of Virtual Reality. The world wide web has been a key player in fostering a free exchange of information that has unified the sciences and the arts as they move across ideological barriers. This free exchange has created a mix of forms which, like the primordial soup that allowed the first sparks of carbon-based forms of life, has fostered a strange synthesis know as silicon-based Artificial Life. In one recent web search using the key words of "Artificial Life" I came up with 14,647 hits using the Alta Vista search engine.

Analogs
Technology Recapitulates Phylogeny is both an observation I've made and a play on Ernst Haeckel's notion that "ontogeny recapitulates phylogeny," in which the ontogeny, or growth of the fetus in the womb recaps certain phylogenetic stages through which life has evolved. During gestation, humans progress from a single cell to a time when the fetus has rudiments of gills and a tail. Before Haeckel's ideas surfaced, Aristotle believed there was an analogical relationship between organic history and human development. Pre-Socratic thinkers like Anaximander, Anaximes and Democritus postulated an analogy between cosmic history and human development.

In our time, with human cultural development so inextricably intertwined and dependent on technology, our existence surrounded in a strangely comfortable embryonic sack of chips and wire, it is no wonder that a relationship between technology and phylogeny seems evident. Along analogical lines market forces can be viewed as a strange commercial form of heredity where a particular form of technology will be passed down to a latter form, regardless of the drawbacks. This is evident with the QWERTY keyboard which was better for manual typewriters, but was passed down to computer keyboards regardless of more efficient designs that have surfaced. In biological systems a species may have vestiges of semi-functional systems that served them in a previous environment, but do not serve them in the evolved environment, such as our appendix. Strangely, there are now hybrid forms which one cannot readily identify as being either technological or biological. Neural network computers composed of bacteriorhodopsin, a bacterial-based molecule derived from the chromophores (molecules that allow us to see color) in mammalian eyes, are used for storing images. These bacterial molecules switched on (red) and off (green) are manipulated to store information in binary forms. Dupont has recently created artificial spider silk based on research into real spider silk production, rumored to be twenty times stronger than steel of the same width.

Still, I am not so interested in these specifics as with what seems to be an inevitable and overall evolution toward intelligent systems, both biological and technological. It seems humans are increasingly realizing the exquisite natural models available from carbon-based life forms. This would stand to reason, as biological forms have had a 3.5 billion year head start to explore molecular variation. While genetics have permitted the passing of biological heredity from parent to child, what has truly distinguished humans from our past kin is not our 100,000 year old biological selves (as indeed we reveal little biological difference to our ancient forebears), but instead the cultural memory which has followed human kind.

Computers can certainly be considered as cultural and memory augmentation devices. Hans Morovic is one researcher who has observed: "cultural development proceeds much faster than biological development". Structurally, computers continue to develop, and seem to mimic the successive stages of development that less evolved forms have gone through, progressing from single transistors (cell) to very large scale integrated circuits (neuron) to massively parallel neural networks (brain). Many would say, that what is still lacking in computers, are the body and sensory elements that would allow them to develop a form of consciousness. I believe we represent that body as a remote sense extension, which rejoins with the computer as information processor and integrator. We further act as that body by continually researching, manufacturing and modeling ever-faster hardware/software. Theorists are further citing the web as an expanded form of an emerging consciousness. Roy Ascott calls this a "Noetic network" in which our minds and the information networks come together to create a new space of consciousness.

When looking at Moore's law (that we double the number of transistors on a chip each 18 months) and current human brain power, Hans Morovic believes that there is a genetic takeover underway and that within fifty years computers will:

...take pride when they look back at us and refer to themselves as our descendants.

Artists like Stelarc manifest these notions when they reference the body as "obsolete" and ask when body replacement parts and microrobots will colonize to improve the species:

It is time to question whether a bipedal, breathing body with binocular vision and a 1400cc brain is an adequate biological form. It cannot cope with the quantity, complexity and quality of information it has accumulated...it is only when the body becomes aware of its present position that it can map its post-evolutionary strategies.

Precedents
As a group, artist have always been inspired by natural living systems. On the front end one could relate figurative art and mimesis as a formal variant of natural systems modeling. With the computer and its control possibilities, artists would later look to process and change within biological systems for models. Many of these artists were inspired by Norbert Wiener and his writings, a mathematician out of MIT who looked to create cybernetic systems that attempted to get at the problems of computer control and communication by modeling natural systems. These problems have found some solutions in the related fields of dynamical systems, chaos theory, neural networks, artificial intelligence, autonomous agents and Artificial Life. The artists listed below could be seen as direct precursors to the use of living systems as models, for the creation of Artificial Life art; some looking to the basal systems of the universe.

In 1969 Nicholas Negroponte and the Architecture Machine Group at MIT created a work titled "Seek", which attempted to anticipate gerbils and their objectives by rebuilding metal blocks around the gerbils objectives of nesting. Still, the software and its computer-controlled artificial environment were outsmarted by the gerbils as it seemed the scientist and the computer could not adequately anticipate what the gerbils' objectives really were. British artist Harold Cohen has utilized artificially intelligent programming out of Stanford University's Artificial Intelligence lab, seeing the "machine as analogy for human intellectual processes". Cohen's computer is run by a program called AARON, which gives it the ability to draw two dimensional works on paper based on rules derived from human perceptual behaviors. Myron Krueger has built numerous environments called "responsive environments" which demonstrate computer perception and adaptation. Krueger envisions the aesthetic of the work arising out of the "collaboration between the artist, the computer, and the participant" which come out of a network of response relationships with the environment. James Seawright has created computer-controlled sculptures based on garden themes that are both sensitive to their environments and to each other and offer viewer/participant interaction. His desire is to create a "...patterned personality. Just as a person you know very well can surprise you, so can these machines."

Sculptor Robert Mallary's article "Computer Sculpture: Six levels of Cybernetics" has defined six levels in which computers can be integrated into the sculptural process, and by extension, into the virtual world of the computer. His last level is now upon us with Artificial Life art where the computer "...has achieved some kind of organic, self-replicating mode of existence, ..."

LEARNING At this point in time Artificial Life forms are being produced by artists, scientists and researchers alike, forms which can learn and adapt to environmental changes. They can, as Darwin had defined in 1859, evolve through the processes of natural selection. In this section I will define Artificial Life and discuss a few Artificial Life techniques. I will then discuss some of the works being produced and conclude with some issues raised by these works. The individuals mentioned do not by any means represent a comprehensive list of all the artists working with Artificial Life, but just some of the artist I have had direct exposure to.

Chris Langton, who put together the first Artificial-Life Conference in 1987, has defined Artificial Life as:

...the study of man-made systems that exhibit behaviors characteristic of natural systems. It complements the traditional biological sciences concerned with the analysis of living organisms by attempting to synthesize lifelike behaviors within computers and other artificial media...whereas biology is largely concerned with the material basis of life, Artificial Life is concerned with the formal basis of life.

While biology is the study of carbon chain chemistry, Artificial Life could be thought of as theoretical biology, which looks instead to creating models of biological systems.

Artificial Life artworks could be considered as a subgroup of Artificial Life research in that most artists are more concerned with creation of an aesthetic as opposed to testing theoretical biology. Which is not to say that the techniques utilized by artists are not considered real Artificial Life or that Artificial Life researchers for that matter cannot find a visual or behavioral aesthetic in their research, but that motivations often differ. Artificial Life artworks seem to break into two clear branches along the lines of hard Artificial Life and soft Artificial Life. Hard Artificial Life artists contend that they are really creating life and soft Artificial Life artists submit that they are creating simulations. From here they branch into many subgroups. There are physically manifested works which express themselves in the three dimensional world of form, and the evolved behaviors of the forth dimension of time. These are generally robotic and include senses that unify the systems to act and react in some special behavioral category. These are complex works to realize because they include not only the algorithms of Artificial Life but also the robotic systems and sensors to give them complex and adaptive behaviors within a real-world environment. These environments tend to be very uncontrolled and unpredictable. Works that manifest themselves in virtual space seem to be more prevalent than physically manifested works, though many seem to have some input from the outside world which may start the simulation off or act to change and effect the simulation in some fashion. These works are able to generate virtual creatures within artificial worlds which have more specified environmental conditions and genetic predictability. There are still other artists who have mixed both virtual and physical worlds with Artificial Life programming and environments.

One form of Artificial Life uses genetic algorithms. With genetic algorithms researchers map a set of parameters into 0-1 bit strings and then map the bit strings into a desirable fitness function. The bit strings are then subjected to repeated computer cycles in which the fitness of each bit string is analyzed and evaluated against the fitness function, copies of the more successful bit strings are made in proportion to their fitness, and individual bit strings are altered by random mutations and mating with other bit strings. The most fit bit strings are the ones that win.

Peter Beyles has been producing music using genetic algorithms as an alternative to pure human design as early as 1991. He began with cellular automata and is now employing genetic algorithms to realize original evolved musical compositions. Yves Amu Klein produces sculptural works inspired by Octopus behavior and the growth of Fungi. These works exhibit complex adaptive behaviors triggered by human inputs and other "Octofungi" through evolving genetic algorithms set up by the artist. Karl Sims has used a Thinking Machines Connection Machine and John Holland's genetic algorithms to create "Panspermia" and other Artificial Life works that create evocative and lifelike worlds.

Many Artificial Life programs consist of populations of simple programs, with no specification controlling all the others. These programs react to local situations in an environment, or to each other, with no global behavior controller, allowing the behavior to evolve out of the all the local interactions. Therefore, any behavior which arises out of all these local interactions can be considered emergent. The author and Mark Grossman have produced three robotic arms, an example of a physically manifested work which reacts to viewers and each other with pulsed infrared eyes, and microphone ears. They use telephone tones as an audible language to pass messages about viewer proximity and the movements of the other arms. They display an inorganic flocking behavior toward sound but away from human bodies, and were based on Rodney Brooks' Subsumption Architectures. At this years International Society of Electronic Arts at The Art Institute of Chicago I had the pleasure of experiencing Nick Baginsky's "The Three Sirens" a robot that uses unsupervised neural networks to learn rules about improvisation and instrumental virtuosity. They learn to control their motors and mechanical characteristics in which the feedback of the sound is further analyzed and creates a free improvisation of neural network composed music.

Tom Ray has created "Tierra" a platform for the study of the genomic evolution of artificial organisms at the University of Delaware and the ATR Human Information Processing Research Laboratories in Kyoto. The evolution of Tierra machine-language computer programs proceeds without explicit direction or intervention from a human operator. Tierra represents an bottom-up approach to general Artificial Life simulation and could be considered a virtually manifested work, that is, it exists within the box. Another virtually manifested work is produced by Biota.org a consortium of artist/researchers which have been creating a Virtual Reality Markup Language (VRML) testbed which allows users to plant a seed in cyberspace and begin an Artificial Life simulation. "Nerve Garden 1" uses fractal algorithms and a finite state machine to simulate 3D worlds and "Nerve Garden 2" (in process) will utilize neural network control of evolution in the new world, with a generative engine called nerves. Another virtual Artificial Life work is by Joseph Bates in association with twelve other researchers who produced the work "Edge of Intention" in which animated creatures called "Woggles" interact with each other and with a forth creature which is controlled by a user. These works present an on-screen world in which the autonomous "Woggles" interact, displaying emotions through facial and body language expressions, and develop complex interaction with both the viewer/participant "Woggle", and the others on the interior of this artificial world. Australian artist Jon McCormack has created "Turbulence" a CD-ROM based playback of virtual chimeras and synthesized forms created using Artificial Life algorithms.

Laurent Mignonneau and Christa Sommerer have created "Interactive Plant Growing" in which Five potted plants are used as input transducers, sensing the viewer/participant which changed Artificial Life variables within the program to grow virtual plants on the screen. This work is successful at unifying real-world interaction with a clear virtual manifestation of that interaction. Recently they have worked with Tom Ray in the production of "A-VOLVE". Naoko Tosa has produced an Artificial Life artwork called "Neuro Baby" in which the participant induces a baby's head to react based on the tone of voice the system hears. Neural networks were used to decipher and understand these tones and adapt by eliciting appropriate baby responses like crying or cooing.

Artificial Life programming will certainly have more profound impacts for the arts as hardware becomes more complex and we develop better Artificial Life software and algorithms which can be implemented to utilize this hardware power. Field programmable gate arrays (FPGA) are one promising example of hardware that can be exploited for evolutionary techniques. Researchers have already had some success utilizing Artificial Life techniques to allow FPGA's to alter their connection instructions and allowing them to evolve a desired fitness.

Perhaps the greatest potential for the arts and Artificial Life techniques is that they have presented opportunities for both artists and viewer/participants to develop true relationships with the computer that go beyond the hackneyed replicable paths of "interactivity" which have thus far been presented by the arts community. The word interactivity is often overplayed perhaps because it demonstrates a kind of technical seduction although there may not be a deeper aesthetic realization. In other words, finding a real poetry of interactive form and content is reliant upon simple button activation.

With Artificial Life programming techniques, for the first time interactivity may indeed come into its full splendor, as the computer and its attendant machine will be able to evolve relationships with each viewer individually and the (inter) part of interactivity will really acknowledge the viewer/participant. This may finally be a cybernetic ballet of experience, with the computer/machine and viewer/participant involved in a grand dance of one sensing and responding to the other. It will also allow new sculptural and virtual algorithmic manifestations which will far surpass our wildest imaginations.

Still, Artificial Life artworks raise questions about the originality of the work. Past artistic experience could be seen as a dialogue that an artist would have with the self and the culture that surrounds him or her. In a case like this who is the artist? The participant? The programmer? The hardware designer? Perhaps all of the above. Is Artificial Life work too reliant on easily replicable form, or the excitement that it looks "natural"? Perhaps just because it looks natural should not be enough to classify it as art. Since many of these works find distribution by publishing on the WEB and in so doing tend not to receive the associated criticism that would be evident in a gallery or museum environment this article is a first attempt to raise some of these questions. I believe the community of producing Artificial Life artists would welcome a critical dialogue.

One difficulty with some Artificial Life artworks is that the systems may not seem to be responsive to the changing environment, as the work demonstrates its own internal desires. This can make the work seem unresponsive or uncaring. Natural living systems manifest their complexity through their interconnected relationship to an ever-changing environment. Still, it illustrates a cultural moment in which the computer may not need to be concerned with the participant/viewer. Other Artificial Life artworks can be considered more collaborative in that initial conditions are set up by the artist or participant and then manifested by the system and software. There seems to be an additional element of human selection involved in many works where the artist/researcher personally selects evolved elements which are considered aesthetic. So human interventions further act as a selective pressure in the evolution of the Artificial Life art piece.