Observation, Interference and Evolutionary Relationships.
(An overview of the Phumox project)

(originally published in Aminima 14)

- Phumox? What's that?
The Phumox project (established February 2005 by Andy Gracie (hostprods) and Brian Lee Yung Rowe (muxspace) has as its aim the exploration of collective intelligence, organic development and interplay with the dual concepts of symbiosis and parasitism. Phumox examines the notions of data and information systems inherent in the living organic environment and how Artificial Intelligence can be used as a tool to explore them, intervene in them, establish relationships with them and ultimately to become a part of them. As we move deeper into an age where the line between the natural and the technological becomes increasingly blurred and easily bent, Phumox poses questions such as: What constitutes our notion of 'being' and how are awareness and consciousness embodied through information networks? Invoking the notion of the cyborg is inevitable, though Phumox should be seen as a discussion on the wider issues of the junctions between organic and inorganic, natural and artificial. Phumox is more interested in how complex multiple relationships are formed via the exchange of data than in how a single agent can be augmented or "improved".

Phumox attempts to circumvent the limitations of traditional silicon-based von Neumann machines by exploring new forms of computing using organic systems but not to be confused with techniques such as DNA computing or using Zhabotinsky fluid style chemistry as the vehicle for computation. Rather, our "organic intelligence" is a self-contained system that behaves in a predictable fashion to perform some functional, artistic (and possibly computational) task.

A crucial part of Phumox is the recognition of emergent behaviour in organic and artificial systems and how these organic and artificial systems can be juxtaposed to produce emergent behaviour through symbiotic or parasitic activity. Exploring these boundary conditions is the focus, as the most dramatic events occur not in equilibrium but in change. For example, the traditional version of collective intelligence can be interpreted as a form of symbiosis. The Phumox perspective is to ask what are the consequences of creating an environment that promotes a collective intelligence that embodies a relationship closer to a traditional view of parasitism?

By employing the ideas behind Phumox in a range of installation works, presentations and papers we intend to create an epistemological landscape for reflection, contemplation and discussion of the many issues around our complex and changing relationships with the organic. In particular we take issue with our access to the organic, our ability to literally deconstruct, to appropriate, to re-imagine and to reconstruct. We see it as necessary to explore our visions of what new evolutionary processes and procedures may be revealed by the probings of bioscience and the demands of bio-industry. Phumox embraces the notion that a considered cultural response is vital to the successful navigation of the 21st century and its reinterpretations of the nature of life.

- the sliding scale of Parasitism and Symbiosis and their effects on evolutionary change
Parasitism and symbiosis are two simple words that hide myriad complex relationships between various organisms. It is commonly believed that 'parasitism = bad/negative/destructive' whereas 'symbiosis = good/positive/constructive' though much evidence points to the fact that things are far more complex or much less black and white than that - as they so often are in matters of life, death and reproduction. Maybe we need a few choice definitions to get going;

Symbiosis, as proposed by Heinrich Anton de Bary in his text 'The Phenomenon of Symbiosis' (1879), simply means a close association between organisms of different species. At the 'friendliest', most harmonic end of the scale we find Mutualism, which as the name would suggest is an association in which both organisms derive mutual benefit, or that each of the interacting species functions as both a host and a parasite. Phoresis is a fairly loose association and is a specialised form of commensalism in which one organism (the phoront) is usually smaller than the other and uses the larger organism as a transport host. Next up is Commensalism, an association in which one, usually smaller member, benefits from the association, whereas for the other member the association is neither beneficial nor harmful.

Parasitism is characterised by a range of definitions including the following:Parasitism is an intimate and obligatory relationship between two heterospecific organisms during which the parasite, usually the smaller of the two partners, is metabolically dependent on the host. [T.C. Cheng; General Parasitology. Academic Press 1986.] and my personal favourite (and most appropriate bearing in mind the concept of the Phumoxian host) Parasites are those animals which use other living animals as their environment and source of food, at the same time relinquishing to their hosts, partially or completely, the task of regulating their relationships with the external environment. [V.A. Dogiel; General Parasitology. Oliver & Boyd 1964.

Parasites and symbionts are normally considered as the two extremes of an interaction between organisms in which one is termed host, and the other, according to the degree of injury it inflicts or benefit it bestows, is parasite or symbiont. Therefore the concepts of parasitism and symbiosis may be employed in the same sense as both refer to the same general phenomenon. In fact, the classical definitions of mutualism, commensalism and symbiosis do not contain any specific cut-off points that distinguish them from parasitism.
'The parasite-host-environment system is dynamic, with several points of equilibrium. This makes it difficult to trace the thresholds between benefit and damage, and therefore, the definitions of commensalism, mutualism, and symbiosis become worthless.' [A. Ara²jo et al; Parasitism, the Diversity of Life, and Paleoparasitology]

These shifting relationships have been a driving force for evolution since the first vague traces of organic matter appeared on Earth. If we follow the Serial Endosymbiosis Theory of Lyn Margulis we see that complex eukaryotic cells (cells which have their genetic material in a nucleus) came about from endosymbiotic relationships between various microbial life forms. In fact, in may instances the theory goes that many relationships began parasitic, became symbiotic and later evolved into a whole discreet organism. Symbiosis was, and continues to be, a force for evolutionary change - life just finds it convenient. We also find evidence that parasites have had their fair share of influence in evolutionary matters as the efforts by their hosts to be rid of them, their escalating arms race, is a major cause of biological diversity. 'Not only are parasites the most successful life forms on Earth, they triggered the development of sex, shape, ecosystems, and have driven the engine of evolution'. [Carl Zimmer; Parasite Rex]

So symbiosis, parasitism and all the forms of inter-organism relationship along the line between the two are driving forces for evolutionary change. I like to extrapolate from this that the very act of an agent, organic or otherwise, existing in close biological or ecological proximity with another is a force for evolutionary change. And here is where the Phumox project wants to butt in. Maybe the Phumox project, via some form of artistic activity can provide such an agent as a player in a bio-artificial ecosystem. By inserting some form of intelligent device into an organic environment and modulating its behaviours and modes of interaction between more symbiotic or more parasitic what kind of emergent effects can we observe? Would the resulting system be an evolved life form or would the changes and adaptations of the natural organisms involved exhibit some form of Lamarckian evolution due to the new and strange stimulus in its environment?

- the observer as a force of change
One of the fundamental concepts of quantum theory, or more accurately an aspect of it known as the Heisenberg uncertainty principle is that an event is determined or altered by the act of observing it. Or more accurately again, that everything exists in a probability cloud and the act of observing an instant 'fixes' reality at that moment. In the famous 'Schrˆdinger's Cat' thought experiment the probability cloud is pretty small - the cat is either alive or dead - though in more complex situations the probability cloud can grow to enormous proportions.

The observation of nature virtually excludes the possibility of complete objectivity and detachment. As Heisenberg himself said "What we observe is not nature itself, but nature exposed to our method of questioning." So is reality more matter-like or more idea-like? Such a view of nature is obviously speculative, and raises questions about what kind of observer is necessary... could it be anything?
So could it be some kind of Artificial Intelligence enhanced robotic agent inserted into a bio-artificial ecosystem? (Bearing in mind that as we approach the sub-atomic level the distinction between organic and non-organic begins to break down). Could the observing of the organic system alongside the symbiotic/parasitic nature of its relationship be a strong evolutionary force? The Phumox project would like to think so. Perhaps, we have two ways now of stimulating responses and forcing evolutionary behaviour in our ecosystem.

- natural genetic mutation in bacteria and other cells
'Evolutionary genomic change occurs largely by a process of Natural Genetic Engineering. Systemic genome organization means that new functions arise by the cut-and-splice rearrangement of genetic modules. Living cells possess mobile genetic elements and other biochemical functions which carry out the underlying DNA rearrangements. Cells regulate the activation of natural genetic engineering functions. Thus, cells have a capacity for major genome reorganization in response to evolutionary crisis.' [J A Shapiro; 21st century evolution]

It is becoming apparent that natural genetic engineering is a relatively common activity in cells in general and also in small whole organisms from bacteria to the fruit fly. They are 'consciously' altering their own DNA, and turning on and off of genes according to environmental conditions. The necessary conditions can range from stress (lack of food or biological attack) to the need to perform some other activity (colonise a plant or other creature), in fact it seems like it can be anything that the colony wasn't already happily doing. It has been demonstrated, for example, that if E. coli bacteria are plated on media containing high concentrations of a nutrient they cannot use, they begin to mutate much faster than the normal 'spontaneous' or 'random' mutation rate, but only in the genes that will then enable them to use the nutrient and to grow.

Genes and genomes need to be fluid and adaptable to maintain stability on the one hand and to respond to environmental challenges on the other. This is the essence of organic stability, as opposed to mechanical stability. It is also becoming clear that the 'fluid genome' processes are a complex regulatory system for carrying out the 'natural genetic engineering' on which life depends. [Mae-Wan Ho]
The popular wisdom is that genes and the organisms containing them that 'respond' successfully to environmental challenges are not acting randomly but purposefully, pointing to an instinctive intelligence that is beyond purely physical mechanisms and processes. Whether this process has any evolutionary direction is still open to question, though it is likely that over a period of time useful genes could remain switched on and the bacteria would remain in the environment to exploit their new niche. This is getting particularly close to a new species.

We ask ourselves at this point: Can Phumox create an environment of persuasion, stimulation or threat that would lead to the spontaneous mutation of bacterial DNA? Can it attempt to quantify genetic manipulation without getting its hands dirty? And if so, how would our reality changing, quantum observing, parasitic/symbiotic force for evolutionary change, natural genetic engineering stimulator recognise the effects it was having? Once embodied as an 'idea' in the 'mind' of our artificially intelligent agent this notion of evolution would need to be broadcast, replicated, disseminated.


- memes and genes
The term 'meme' was made popular in Richard Dawkins' 1976 book 'The Selfish Gene' and was defined as "a unit of cultural transmission, or a unit of imitation". Although in the emerging science of memetics there are differing interpretations of the word it is generally agreed that a meme consists of some sort of a self-propagating unit of cultural evolution that resembles the gene. Memes have become increasingly popular amongst cultural commentators as they are particularly useful in explaining human behaviour and cultural evolution whilst maintaining connections and metaphorical links with the highly fashionable science of biotechnology.

The notion of a meme as an idea or a thought that is contained in a word, or a collection of words as a sentence forms an analogy to the idea of a gene as a self-replicating set of code. The gene can be viewed as a collection of nucleotides that replicate together and code for some set of behaviours or body parts. DNA strings seen as code, the code of life, is not a new idea, but a well established meme.
Mutation is necessary for memetic evolution just as it is for genetic evolution. Via the process of mutation variations are produced, and the variations that prove 'better' at replication will become more common and therefore have a greater chance at replication again. In memetics the meme is mutated by its environment on a constant basis and changes in an early form will accumulate and get passed on as they replicate. In this way memetics behaves much more according to the Lamarckian view, which pits it directly against the many who are devoted to proving that genetic evolution does not function in a Lamarckian manner.

Would it then be possible for our Phumox-agent, with all its evolutionary powers and ability to not only instigate a process but also to realise what was actually happening, to propagate, mutate and disperse a few choice memes? Maybe the act of viewing, participating in and discussing culture is inherent in all artistic practice - whether it fancies itself as a (pseudo)science or not. If memes are being generated by the audience viewing a Phumox project are there any being generated amongst the other organic participants? Do bacteria even do cultural transmission, or are they happy enough just expressing their genes?

- the phumoxian response
The initial response of the Phumox project involves an Artificial Intelligence system designed by Brian Lee Yung Rowe called Generalized Cellular Signalling, the theory of which is described below in his own words.

'On the surface, there doesn't seem to be much of a relationship between bacteria and brains, though recent research on bacteria implies a profound connection between the two, particularly when viewing bacteria as a collective. Both are systems that consist of hundreds of billions of cells; both are highly connected and can send multiple types of signals to other members of the collective; both exhibit a number of emergent behaviour, including intelligence. It is widely known that the brain is the source of intelligence in humans, but to most people, the idea of intelligent bacteria is more dubious than the idea of intelligent design. And yet, there are numerous observable traits that point to collective intelligence, such as specialization and cooperation within the colony and democratic voting. Given these parallels between bacterial systems and neuronal systems, a number of questions can be raised: 1) What exactly is intelligence? 2) Are bacteria the evolutionary predecessor of brain systems? 3) What other parallels exist between these systems and can we use one system to inform us about the other (and vice versa)?'

The Generalized Cellular Signalling model is a platform for exploration into emergent behaviour and intelligence using cellular systems. Within GCS, a complete virtual environment exists where individual cells act independently and can communicate with other cells either in a neural fashion using relatively fixed connections, or bacterially, where signals are propagated as molecules through a medium. GCS is inspired by two independent paradigms, and like the systems it models, is (hopefully) greater than the sum of its parts.The basic process of early Phumox works is to implement GCS at some point in a distributed natural system and for it to take one or more of the positions of observer, parasite, symbiont or collector and broadcaster of memes. Any one of these activities can be a force for evolutionary, physical or cultural change. The obvious relationship we wish to explore is that between the GCS and living bacterial colonies, to allow the two intelligences to learn about each other, to allow technology to enter into dialog with bacteria on their own terms. We will be working initially with cyanobacteria, species such as Anabaena, which already have well established symbiotic relationships with aquatic plants and are major players in complex ecological cycles. By inserting the GCS at strategic points in bio-artificial ecosystems and allowing a give-and-take, push-and-pull, listen-and-learn relationship to emerge with the bacteria we will begin to gently influence symbiotic relationships to become more parasitic and to push the bacteria to a point where their response is to begin to naturally genetically modify themselves. As the environment adapts to the GCS and its stimuli, we also look to the GCS to adapt to the environment.

With the continuous introduction of the man-made, artificial and chimeric into the natural environment and the gradual convergence of wetware, software and hardware it is worthwhile to explore the adaptive reactions of ecosystems and the possibility of new data types suggested by nature/machine interactions. The work must provoke discussion about how new technologies are changing the way both humans and nature live and evolve, and how many vital processes that shape the world around us are adapting to accommodate new bioartificial ecosystems. Through the discovery or evolution of novel forms of machine and nature communication and examination of its inherent data and intelligence we hope to facilitate new modes of approaching and understanding the techno-organic environment. Phumox frames our interest in exploring connections between machine and nature that are outside the typical areas covered by cybernetics or purely scientific study whilst maintaining the ability to engage or critique.