7#hv]_o{rrrs\ttttt ttttJu1xtu uu*vsquuuuv@uuuuuu--- draft only : not for redistribution --- towards the matrix . . . connectionist culture in theory mitchell whitelaw mitchellw@spin.net.au Since the fractal psychedelia of the 80s, ideas from the so-called complexity sciences have been sliding across into culture and the humanities. Through popular science writing, computer graphics and electronic art, notions of self-organisation, emergence and complexity have begun to circulate in the rhetorical fields of new media and cultural theory. Driving this crossover is the resonance that complex systems ideas have with our conceptions of culture: complex systems are composed of masses of interrelated elements, which seem to organise themselves out of disorder into structure; they are open to changes in their environments, and their own states are sensitive to those changes. The subject, the state, language, or the economy have similar properties - what if these entities could be thought about as interlinked complex systems? What happens when we think about culture as a complex system? This paper focuses on the writing of Sadie Plant and Roy Ascott, who use complex systems or connectionist ideas in talking about culture in general and cyberculture in particular. Rather than accepting a transparent importation of connectionism into cultural theory, I want to pay attention to the particularities of its applications and mis-applications in the work of these two writers. The critique of these applications is not an argument against the importance of connectionist thought in cultural theory - it is an argument for a more flexible and detailed way of thinking about culture and connectionism than that provided by the models Plant and Ascott propose. What is at stake here is the shape of a connectionist conception of culture. Plant and Ascott converge on connectionism from different backgrounds - Plant has defined cyberfeminism almost single-handedly, arguing the merits of an alliance between digital technology and the feminist project, while Ascott is best known as a founding practitioner and theorist of network art. They diverge, in the papers I am discussing here, in their intentions: Plants The Virtual Complexity of Culture centres on an argument for a rethinking of disciplinary knowledge. The Ascott papers I am using are polemics arguing for the involvement of telematic artists in a reconstruction of representation, Nature and subjectivity. The reason for considering them in parallel here is the similarity of their constructions of connectionism: for both it is tinged with a utopian futurism and mixed up with a movement towards the virtual, towards the matrix. This matrix, a rhetorical figure central to the arguments of both Plant and Ascott, provides a way to begin looking at their cultural connectionism. In Ascotts writing the matrix appears around 1990, when he talks about the digital matrix that brings all new electronic and optical media into its telematic embrace. Here it is a familiar Gibsonian cyberspatial construction - an abstract plane of computation which blurs into the imaginary space of the telematic network. Ascott describes the computer as pure system, a universal transformative matrix. It is the agent of the datafield, the constructor of dataspace. This figure also appears as the telematic noosphere and in a later paper as electronic space and telematic space. This matrix becomes more explicitly connectionist as it merges with some other metaphorical fields. In Homo Telematicus in the Garden of A-Life, Ascott declares a structural correlation between the bottom-up, distributed, local determination of behaviour that characterises a-life and the art of homo telematicus... art in the telematic culture. This art is distributed, connective, collaborative, emergent from a multiplicity of interactions in electronic space. The matrix is animated by the agency of the artist as connectivist - the artist armed with a systems view and a connectionist philosophy. Ascott declares: ...growth, spontaneously generated levels of order, and self-organisation constitute the dynamic aspects of our practice. The dataspace has become the garden of a-life - a self-organising idyll with homo telematicus at its centre. Plants matrix, while operational in quite different rhetoric, is similar to Ascotts in shape. Plant describes the early history of computation as a move towards the matrix, through higher degrees of abstraction, culminating in cybernetics and connectionist self-organisation. She aligns the matrix with the formal meta-machines of Turing and Wiener: If all individual computer systems are technical actualisations of such abstract machines, it is this abstraction that guarantees their interconnectedness. Like Ascott, Plant conflates this abstract structural commonality with the consensual hallucination of networked communication. And as in Ascott the computer is an unproblematically transparent point of access to the abstract plane: The virtuality emergent with the computer is the immanent processing and imminent future of every system, the matrix of potentialities which is the abstract functioning of any actual configuration of what we take as reality. Further, It is this virtual reality of actual systems which provides the key to a connectionist or synergetic thinking and gives once separated things and zones the abstract equivalence which allows them to link up. This virtuality underpins an argument for radical interdisciplinarity based on connectionist notions of complexity and self-organisation. Plant describes the matrix as an emergent complexity an evolving intelligence in which all material life is involved the virtuality and the future of every separated thing, individuated organism, disciplined idea and social structure. As such, it founds a sweeping application of connectionism: complex systems are everywhere... across what were once distinct natural, social, human and artificial zones. In line with this common construction of a universalising abstract plane, there is a common movement in this writing towards that plane - an abstraction and intermingling of fields. Plant uses the universality of complex systems to rail against the boundedness of disciplinary knowledge, the distinctions between nature and artifice, human and machine, representation and reality. The matrix appears as a formal commonality that is more valid than these existing distinctions. It absorbs its diverse instantiations (the Net, abstract computation, the common properties of self-organising systems) into a blurry gestalt. Ascott blurs the telematic space of the Net with the universal transformative matrix of the computer, but goes further; invoking a hybrid of radical constructivism and post-structuralist discontinuity, Ascott talks about an explosion of meaning producing a swirling infinity of fragments - a dynamic, multiplicitous semantic field. This field becomes a connectionist milieu, a substrate for self-organising processes - the fragments are in fact seeds of new meaning, this swirling infinity is another burgeoning, generative cloud: another version of the matrix. I want to approach this process of abstraction and intermingling through a critique of connectionist thought from Katherine Hayles. While her focus is specifically on artificial life research, her arguments can be usefully related to the connectionism Ascott and Plant are involved in. Hayles argues that a-life is founded on a Platonistic privileging of form (organisation) over matter, to the extent that for a-life, organisation, rather than material instantiation, defines life. While a-life is often linked with other complexity sciences in participating in a move away from reductionist thought towards a more systems-oriented, holistic approach that deals with the emergence of complexity and variety, Hayles argues that (in simulation-based a-life at least) emergence is a kind of reductionism-in-reverse, where [i]nstead of starting with a complex phenomenal world and reasoning back through chains of inference to what the fundamental elements must be, they start with the elements and complicate them... so that the complex phenomenal world appears on its own. Hayles critique alerts us to a basic complementarity in connectionism, between processes of abstraction and embodiment. The first moves from the particular outwards to the system, from the embodied to the structural, from the concrete to the formal. This is the movement that identifies formal and organisational qualities of systems as essential. Hayles point is that this process of abstraction, the extraction of the axiomatic structural properties of a system, is not transparent - it reduces its object, something is lost in the process. She has some justification in calling simulationist a-life Platonistic in its quest to discover the axiomatic basis of aliveness, but it would be a mistake to extend this critique outwards to a-life and connectionism as a whole. Emergence, which a-life pioneer Christopher Langton calls the fields key concept, fills in the second half of the complement, embodiment: the move out of the abstract into the concrete, from the generality of the system back into the multiplicity of its realisations. Emergent phenomena indicate that the interaction of networks of abstract forces (or rules in a simulation) produces unpredictable results - results that resist formal pre-conception, but appear only with the systems instantiation. The connectionist schemas Plant and Ascott set out have been swept up in the abstraction half of this complement and hang there in the ether, with little indication of their concrete manifestations. Ascotts rhetoric never touches the earth, but sweeps over specificity in a hyperbolic rush: there is a constant supply of endless flux and flow, multiplicity and infinite connectivity and boundless constructivity but its difficult to imagine what an instantiation of this system might look like. Following Hayles critique, we might ask what is left behind in this rush towards the abstract: in both writers, what is lost is the heterogeneity and particularity of the systems subsumed by the matrix. For Plant, neutralising difference becomes the primary agenda; the matrix obliterates all existing disciplinary mechanisms and all conventional distinctions: natural/artificial, human/nonhuman, humanities/sciences, theory/practice, representation/reality. The boundaries of traditional disciplines and didactic process set out allowable lines of connection, tree-structures of filiation and inheritance; Plant unleashes connectivity on them like a membrane-rupturing virus, aiming at an orgiastic commingling of leaking and cross-infection. While enthusiastic, Plants emphasis on the rupture of existing structures and the exhilarating intermingling of the once-separate is a particularly one-sided application of connectionist thought. In a complex system, interconnectedness and interdependence are the preconditions for emergence and evolution. But in as much as they are generative, open, and evolutionary, these processes are also self-stabilising, self-preserving, and autopoietic. Varela and Maturanas notion of autopoiesis gives an account of the way an organism-as-complex-system maintains the organisation of its components despite changes in its environment. Autopoietic processes are interlocking feedback cycles that maintain structure and differentiation; more broadly, any kind of emergent order in a complex system has these characteristics too. Connectionist science, and particularly a-life, tends to focus on the capacity of complex systems to produce these emergent phenomena, but Plant seems reluctant to address this aspect of connectionism. In talking about a connectionist model of knowledge, she discusses the convergence of various disciplines into a nexus. She immediately adds; Not that the particular nexus which emerges is some closed system, and self-contained: these new connections do not regroup to compose another disciplinary zone, and can only ever be contingently defined and momentarily circumscribed. Of course such a nexus is not a closed system, nor is it self-contained, and its circumscription, because of this openness, will always be provisional. However in the same way that we are able to distinguish an organism (as autopoietic system) from its environment, we could expect a knowledge-nexus to have some kind of persistent structure, to be differentiated from its connectionist environs. Plants points about the open character of knowledge systems are well made but she seems unwilling to acknowledge the complementarity of change and stability in such a system, or unwilling to talk about these poles in a more complex way than change=good/stasis=bad. Along with this bias towards instability and rupture, Plants argument shows a lean towards the future which is also a selective re-telling of connectionist thinking. Plant calls the abstract plane (or virtuality) the imminent future of every system, and elsewhere refers to the matrix as the virtuality and the future of every separated thing, individuated organism, disciplined idea and social structure. She projects philosophers who are neither modern nor postmodern, but in proximity to the future and continuous with the tendencies and directions in which they write. They run ahead and in anticipation of themselves, and have to do with surveying, mapping, even realms that are yet to come. In casting the abstract plane as being the future of every system and talking about a connectionist transdisciplinary convergence, a sense emerges of a drive towards a state of higher connectedness. In the precis of the paper she refers to an emergent connectionist thinking...beginning to blur the boundaries between the arts, sciences and humanities. We could tag this as straightforward Modernism - hailing the beginnings of the revolution, locating its fulfilment in the future - and in some important ways it is; its a technological revolution, part of a larger progression. But its also interesting to link this up again with the connectionist ideas Plant makes use of. If the abstract plane is where pansystemic, self-organising processes are located then its possible to see how this virtuality is the imminent future of every system. If it describes the systems self-organising dynamics, then it could be seen as accounting for the systems realisation, moment-by-moment, of those dynamics. But in the same sense as the set towards rupture and intermingling, this set towards the future leaves out some important aspects of connectionist thought. The future of a complex system is in no sense a linear drive towards a distant point; it is a moment-by-moment unfolding of states into each other. Each state is the ground for the next, and each state is a partial accretion of all past states. Connectionist systems are notorious for having memories, that is, retaining traces of past states, adapting and evolving around them. These ideas are central to the work of Ilya Prigogine, one of the major figures in connectionist thought. Prigogine emphasises the irreversibility of thermodynamic processes and the importance of symmetry breaks - bifurcations in the history of a system which constrain its present and future states. Prigogine also makes clear the significance of the shift from a classical conception of reversible time to thermodynamic, irreversible time, characterising it as a reenchantment of nature and a validation of our embodied experience of time. In conflating connectionism with utopian-futurist rhetoric and constructing a sense of a connectionist revolution Plant and Ascott neglect the embeddedness of dynamic systems in their own histories. In some ways they seem to be talking about culture as if it were a computational a-life simulation which can be cleanly and absolutely restarted by a transcendent god-operator (the new paradigm starts...Now!). Neither of them deal with the full implications of adopting a complex-systems approach to culture. If the future is complex and connectionist, how has the past been any different? If the logics of systems, hierarchies, feedback loops and self-organisation will deliver us into the technological ecstasies of the next millennium, havent those same logics brought us to where we are now? If we accept Prigogines notions of temporality, we cant claim any kind of revolutionary moment or break with past conceptual systems. We can expect processes of autopoiesis to continue along with the connectionist dynamism that Plant and Ascott emphasise. The institutions and concepts they would liquefy in advance of a connectionist future are the results of complex dynamics themselves; western individualist humanism and the disciplinary division of knowledge are autopoietic nexii or abstract machines with their own processes of exchange, evolution and self-preservation. The same tendency to neglect the complex systems that condition the present shows up in the way both writers deal with technology. For both Plant and Ascott, computer and telecommunications technology is a privileged point of access to the matrix. However this technology seems unproblematically transparent; the computer starts to look like a mystical portal to the plane of abstract equivalence, rather than an item of consumer electronics engaged as much in virtual networks of commerce, cultural inscription and power as it is in the everyday Net-works of home shopping, webcast advertising and cross-media tie-ins. Plant announces: As everything that was once stored in frames, galleries, books and records converges on cyberspace, an on-line global library of the kind promised by Ted Nelsons Xanadu system is being implemented. Once again there is this sense of a converging rush towards the virtuality of cyberspace; but even if the web was as open and polymorphous as Nelsons Xanadu, its clear that the forces determining the nature of the web are anything but benign, civic-minded library-builders. If anything, the growth of online media is accelerating the commercialisation of information delivery. In Ascott the utopian garden of a-life is premised on the transparency of the telematic connection; he declares ours is an art of electronic networking, of intensive connectivity, mind-to-mind collaboration through computer-mediated telecommunications systems. Through such a system the artist can be widely distributed, experiencing telenoia : a life-affirming sense of mind-at-large. Even leaving aside the streaks of Cartesian dualism and corporeal transcendence that run through Ascotts writing, its hard to imagine Ascotts telematic artist as an empowered agent moving through the networks when every net-users experience is of layer upon layer of mediations, technical glitches, software and hardware filters, plug-ins, crashes and dead connections. I can imagine my telenoic mind-at-large ending up trapped in a half-dead ethernet hub in the back room of some university, unable to connect with anything except an ancient VAX... But the absence of technological mediation from these connectionist schemas indicates a much larger blind spot; degrees of telematic connectedness are an index of economic, institutional and political power. The dusty ethernet hub and venerable VAX belong to an under-resourced government institution; the transparent connective bliss of SGI boxes and fibre-to-the-desktop belongs to the well-to-do corporation or pseudo-corporate research institution. Who will receive high-bandwidth access and the software skills to make use of it? The western, middle-class, tertiary-educated, techno-literates will inherit the matrix. What needs to be added to these connectionist models are additional layers of complication, some internal differentiation, some capacity to address the local and specific, some real complexity. A connectionist conception of culture like this must be able to account for differentiations of connectivity and the autopoietic fields of power that effect them. It should be able to talk about systems with histories (as well as futures), and the articulations of instability and self-organisation that form and de-form the abstract cultural entities that Plant sets out to eliminate. Plant herself tips us off in talking about Deleuze and Guattari: she likens the matrix to their machinic phylum, and later quotes Deleuze in describing theory and practice as a system of relays within a larger sphere. Leaping across the Deleuzean synapse we land on Brian Massumis Users Guide to Capitalism and Schizophrenia, an example of a way of theorising culture which makes extensive use of connectionist ideas but retains the ability to talk about specificities of power and embodiment, and doesnt side wholly with the virtual. Massumis discussion of Deleuzes virtual (which comes in turn from Bergson) is particularly valuable here. Massumi describes the virtual as the future-past of the present and as being real and in reciprocal presupposition with the actual... It subsists in the actual or is immanent to it(italics in original). This is the same circuit between the virtual (as the possible futures of a system) and the actual (an embodiment of one of those future states) and the virtual (as the new set of possible future states) that we find with connectionist thought. Massumi calls the virtual and the actual coresonating systems. He uses a classic self-organising physical system (the Benard instability) as the starting point for a theorisation of the formation of the subject/body as a dynamic system, moving in a continuum from the infantile supermolecule of interacting part-attractors to the whole attractor of socially sanctioned molar personhood. He discusses meaning in terms of force, as the encounter of lines of force, each of which is actually a complex of other forces. This is really a systems approach to meaning, and highly connectionist in the complex interplays it invokes. But the network here never blurs into a cloud, Massumi is firm about its specificity: The processes taking place actually or potentially on all sides could be analysed indefinitely in any direction. We could follow through the lines of the network, we could trace their shape and thickness. Massumi talks about power in this context as a reproductive network of forces, as distinct from the productive network of the initial encounter; so power reproduces, it is self-perpetuating, self-stabilising, autopoietic. The nearest Massumi gets to a single, all-encompassing virtual matrix is his monster fractal attractor; a diagram of tendency and possibility which a real system moves through - by comparison Plants matrix seems conceived in more static, spatial terms, as a place or plane in itself. Its impossible to deny the importance of the goals Plant and Ascott pursue; notions of distributed subjectivity, transdisciplinary notions of knowledge. The connectionist paradigm that they construct in pursuing those points is seriously limited, though, in its reliance on technology and its all-absorbing abstraction. If Plant and Ascott misrepresent connectionist thought in their applications of it, they also miss its most significant attribute, its ability to articulate the abstract and the embodied, to talk about the common attributes of complex systems without eliding the heterogeneity and variety of their instantiations. This is the attribute Massumi picks up on in his hybridisation of Deleuzean ontology with connectionism: it seems particularly important at a point where the disjunction between a massive, unthinkable system (ecology, capital, technology) and the local, embodied implementation of those systems is something we constantly face. Connectionist thought is too interesting to be dismissed as a way of thinking about culture; if it is pressed into service justifying techno-transcendence this is a distinct possibility. This paper is a bid to keep the autopoietic theory-nexus forming around connectionism open and unstable, in the hope that it will develop more subtle and flexible ways of thinking the cultural systems around us. for a popular science account see for example M. Mitchell Waldrop, Complexity (London: Penguin, 1992). For complex systems in cultural and literary theory see for example Katherine Hayles (ed), Chaos and Order (Chicago: University of Chicago Press, 1991).  Both writers use the term connectionist to refer broadly to ideas of complexity and self-organisation. As an umbrella term its validity is questionable, but for the sake of clarity and convenience I will follow Ascott and Plant in using it here.  Sadie Plant, The Virtual Complexity of Culture, in Jon Bird, Barry Curtis, Melinda Mash, Tim Putnam, George Robertson and Lisa Tickner (eds) FutureNatural (London: Routledge, 1996)  Roy Ascott, Homo Telematicus in the Garden of A-Life in Tightrope 1/95 (online at http://www.phil.uni-sb.de/projekte/HBKS/TightRope/issue.1/texte/royascott_eng.html) and Telenoia: Art in the Age of Artificial Life in Leonardo, vol. 26 no. 3, 1993.  ________, Is there love in the telematic embrace? in Theories and Documents of Contemporary Art, Stiles, K. & Selz, P., Eds., (Berkeley: University of California Press, 1996) 497.  ibid., 496.  Ascott, Homo Telematicus  Sadie Plant, The Future Looms: Weaving Women and Cybernetics in Mike Featherstone and Roger Burroughs (eds)Cyberspace/Cyberbodies/Cyberpunk (London: SAGE Publications, 1995)  Plant, Virtual Complexity, 206.  Katherine Hayles, Narratives of Artificial Life, in Futurenatural, op.cit., 153.  quoted in Hayles, ibid., and Ascott, Homo Telematicus  Ascott, Telenoia, 177.  Ascott, Homo Telematicus  Plant, Virtual Complexity , 211.  Humberto Maturana and Francisco Varela, Autopoiesis and Cognition: The Realization of the Living. Boston Studies in the Philosophy of Science, 42. (Dordrecht, Boston: D. Reidel, 1980)  Ilya Prigogine and Stengers, Order out of Chaos (London: Heinemann, 1984) On irreversibility see pp7-9, on symmetry breaks see pp160-170.  Ascott, Homo Telematicus  Brian Massumi, A Users Guide to Capitalism and Schizophrenia (Massachusets: Swerve Editions through MIT Press, 1993), 36.  ibid., 65.  ibid., 11.  ibid., 64.  uwww.khm.de/buttons/englisch.gif5http://www.khm.de/projects/TheGarden.gif5http://switch.sjsu.edu/web/frontpages/stillman.cgi?link2=http://switch.sjsu.edu/web/v3n3/militarytoc.htmlb5http://cadre.sjsu.edu/switch/alife/pdq/switch.gif65http://www.zdnet.com.a+EFex M N[\!"$$''((+b+c--==>{>|HHJ0J1R2R`SST.T7UUWtWuZUZV`o``a2aAaqarbmbnbcc)c*cecsddd(d)dadddddeee}eee@ J  00 [eeeeeff-f<f=fwfxffffffgg;ggggh#h$hAhBhRhhhhhhhhhhhh @ @ J@',-Ffx   ]^>?!u!v&z&{+e+f33|>}CCDgHHL^L_NNQQ[[^^ܼܴܬܤܔ܌܄ XLh XLh XLh XLh XLh XLh XLh XLh XLh XLhLh XLh XLh XL XL XL2^`j`k`l`m`n`o`papaqblbmc(c)d'd(dddde eeeeef;f<fvfwffffffggh"h#h@hAhhhhhhhhhhhhh X  X  XLh XLh6  block quote! . h  [!#&'*b,<I0RTVtYUgqQw$Bi%P^lz}} x gh>&0T:DEMX{_ogJ W+{b    eh56^h78  0`r_ G&p....566677;?HH 09Acxhh d EA.h@@H-:StyleWriter 1200 Palatino---+!F@Towards the MatrixMitchell Whitelaw1.4Mitchell Whitelaw