|
|
_____________________
Joel Slayton
|
Introduction This paper presents an analytic perspective on the nature of Art as Network. The network (internet and intranets) is described as having properties of a nonlinear complex system, which serve to enable a network semiotic enabled by entailment meshwork. The network semiotic includes the characteristics of hierarchical structure, tendency towards self-organization and chaotic attractors, resulting in the emergence of self-similar alternates which function as dissimulators of interaction and immediacy. Art as Network is an exploration of the pandemonium of agencies forming entailment meshwork. Art as Network is intended as a subject of discourse, not declaration. Certainly, I do not suppose that the particular theoretical investigations that I have chosen to discuss herein are the only points where the discourse of Art as Network is possible. My emphasis on complexity and semiotics involving entailment mesh, nonlinear dynamics, self-organization, homogeneity, dissimulation and the notion of alternates, is to establish a framework in which are present domains from which the artist might experiment. A Network Semiotic Actions of information immediacy and interactivity concede into a network semiotic of infinite and simultaneous regress. Similar to the properties of hyper-dimensional mathematical matrices, in which meshment frameworks are evident, the network semiotic represents an interrelated set of information structures that emerge as a complex system. I propose that this is accomplished through entailment mesh frameworks (meshwork) coded into patterned actions of data behavior. Entailment being the product and feedback of meshwork conditionals (autocatalytic patterns), which are signified only by the computational behavior of data. Consider the network semiotic as a linguistic system in which signs and symbols function as complex interactions of information agencies. Such a semiotic incorporates a nonlinear dynamic from which entailment meshwork emerge, including all domains for information association and function. An entailment is a set of rules pertaining to the emergence of self-organization by a system in which uncertain conditions and influences are present. Developed by cyberneticist and information theorists Paul Pangaro and Gordon Pask in the early 1980's at MIT, the concept of entailment was postulated as an essential component of their research into conversational and learning systems. They deduced that emergent conversation (brainstorming) was the result of self-organizing entailment processes from which inferencing emerged. An entailment mesh was thought to be the principal semiotic agency (data pattern) responsible for concept inferencing resulting from information filtering, association, referencing and substitution. Their research involving machine/machine conversation demonstrated entailment as a complex autocatalytic feedback loop in which sub-systems influence a tendency for adaptation towards less than optimum goals. That is, there was no specific objective or purpose guiding the conversational system. The notion of entailment can be applied to the network, for nonlinearity, uncertainty, self-organization and inferencing play significant roles in the overall adaptive and evolutionary character of the semiotic as a complex system. The twist, of course, being that network entailment is enabled by linguistic structures, which function as patterned actions of data behavior and not signifiers of symbolized meaning. The nature of entailment meshwork is such that they produce specialized information events (internally or externally) as their exclusive function. Perhaps the most studied type of meshwork is the 'autocatalytic loop', a closed chain of processes that display self-stimulation and self-maintenance. Autocatalytic loops change their internal structure so that they become more or less receptive to other information events. Manuel De Landa describes the effect, "The catalyst provides a meeting of two substances, facilitating (or inhibiting), their reaction and therefore, the accumulation (or decumulation) of the products of that reaction". As he indicates, self-stimulation and maintenance would be central features of entailment meshwork, directly shaping the processes of information filtering, association, referencing, replacement and substitution composing the network semiotic. Entailment meshwork simultaneously function as signifier and signified within a syntactic and procedural linguistic that is unique to the network semiotic. This self-referentiality denotates the consequence of nonlinear association, in that entailment meshwork form catalysts for other entailment reactions (internally or externally), which in turn generate new meshwork conditions as product catalyst for further entailment. Evolution of the network semiotic is the result of a nonlinear dynamics of entailment adaptation. The semiotic exhibits no adaptive tendency towards optimization of resources or efficiency, rather there is continuous shifting among possible phase state alternatives. In essence, the network semiotic moves between different states of equilibrium in response to the multitude of uncertain objectives represented as expressions of network applications. Entailment meshwork formation (or dissolution) serves to modify the network semiotic through self-referential feedback resulting in new agency structures (data actions). The consequence being that the network emerges itself in more or less stable states over time. De Landa speaking on the combinatorial possibilities of meshwork states, "The number of possible hybrids of meshwork and hierarchies--are immense (in a precise technical sense), and so an experimental and empirical attitude toward the problem would seem to be called for. It is surely impossible to determine purely theoretically the relative merits of these diverse combinations. Rather, in our search for viable hybrids we must look for inspiration in as many domains as possible." Evolution of the network is a matter of influences from entailment experimentation, a matter of heuristically testing the vast number of alternatives which may stimulate phase shifts. Inferencing, the ability to pass from one proposition, statement or judgment to another, whose truth is believed to follow from that of the former, is indicative of entailment meshwork functionality. Maturana and Varela describe how an entailment meshwork might be considered as inferencing data structure, "In the simplest autocatalytic loops there are only two reactions, each producing a catalyst for the other. But since this basic two-node network establishes itself, new nodes may insert themselves into the mesh as long as they do not jeopardize its internal consistency. Thus, a new reaction may appear (using previously neglected raw materials or even waste products from the original loop) that catalyzes one of the original reactions and is catalyzed by the other, so that the loop now becomes a three-node network. The meshwork has now grown, but in a direction that is, for all practical purposes, "unplanned." Herbert Simon uses the term "nearly decomposable" to further suggest that within a complex system individual elements composing the system are not fully autonomous, but have negligible relationships which cannot be completely discounted. Perhaps entailment meshwork are also "nearly decomposable" as artificial autocatalytic processes in which interrelations with other entailment meshwork are as Maturana and Varela suggest, linked by the remnants of dysfunction and waste. Complexity Complex systems involve phenomena characterized by interactions of individual agencies (sub-systems, elements, component processes and datum as content), that self-organize at a higher systems level, and exhibit emergent and adaptive properties. Gilles Deleuze and Felix Guattari propose the concept of 'machinic phylum' to reference the overall set of self-organizing processes within a system (including complexity) in which previously disconnected elements suddenly reach a critical point at which they begin to 'cooperate' to form a higher level entity. This 'evolution' is indicative of a set of adaptive alternatives, a systems machinic phylum, through which it may or may not bifurcate. Not simply a taxonomy of evolutionary possibilities, the machinic phylum encompasses the algorithmic design for the features of its current manifestation, as well as, all potential designs and procedures for selecting those designs. Analysis of the network semiotic as a complex system requires a rich interanimation of top down and bottom up theorization. A top down strategy in which general principals illuminate specific instances, enables philosophical characterization of complexity within the system. Bottom up analysis, examines applied instances which serve to illuminate capacity and threshold of the systems potentiality. Developing questions appropriate to these two analysis modes is essential if we are not to lose ourselves in the minutia of detail represented by technological assessment alone. For example: How is complexity reflected in the network and what implications does this have for how we think about describing a network semiotic? If the network semiotic is a nonlinear, self-organizing system capable of bifurcation, how does linguistic coherency within the system flow from its primary agencies? Such questions begin to frame a typology of descriptions serving to typify the network as an artificial system composed of information processes. The network semiotic is necessarily an adaptive system, exhibiting evolutionary processes resulting from the selection pressures of specific entailment conditions. In that information theory explains organized complexity in terms of the reduction of entropy (disorder) that is achieved when systems absorb energy from external sources and convey it to pattern or structure, the network semiotic is such that it emerges only in the coherency of its linguistic functions. The network semiotic procures data from its environment (internal and external via human/machine, machine/machine interaction) and finds regularities in the data (associations), and compresses these into internal signification models (datum as patterned activity) that enable entailment. To suggest that this phenomena is the result of a self guiding system, is to side with Deleuze and Guattari in a corresponding theory in which the network behaves as a virtual machine capable of articulating itself. Deleuze writes, "Simulation remains simulation... The organic machine is driven both by life force (as is affirmed in the movement to dominate the environment) and by machine, insofar as the machine is called on to develop maximum efficiency or to ensure functional movement." The network semiotic is the 'called upon machine', a self-deterministic system simulating a model of itself under the guise of efficiency. In other words, the network evolves as it attempts (appears) to be what a network is. According to Herbert Simon, "Chaotic systems are deterministic dynamic systems that, if their initial conditions are disturbed may alter their paths radically. Although they are deterministic, their detailed behavior over time is unpredictable, for small perturbations cause large changes in path." Chaos theory has been used to explain a wide range of adaptive and evolutionary characteristics in the biological and physical sciences and in the study of social and economic systems in which self-determinism is evidenced. Most, if not all complex systems exhibit deterministic chaos as a principal feature of their adaptive and evolutionary nature. In classical nonlinear theory, a system either arrives at a stable equilibrium or oscillates permanently in a limit cycle. A chaotic system, however may arrive at state in which it would remain permanently unless affected by a strange attractor. Complexity theory identifies attractors (static, periodic and chaotic) as perturbations that influence direction and course of system evolution. Attractors as specific perturbation patterns of linguistic activity can directly affect state transitions of the network semiotic in deterministic and unpredictable manners. Shifts in overall appearance and behavior of the network can be attributed to such perturbations affecting the ability of the system to sustain equilibrium. Chaotic patterns of linguistic activity (datum as autocatalyzed attractors) emerge in specific entailment meshwork. Based on the features of deterministic chaos, the network semiotic is bound to seek new pattern, as well as sustain tendency for adaptation. Debased of a correspondence between the literal subjects of the signified (subject) to those of evolution (chaos), the network, as complex system, lacks in singularity of definition or purpose. Transparency of coherence is depthless, making the network as virtual machine nearly incomprehensible as anything other than a highly decentralized technological infrastructure. Such a perspective tells us little about the network as an artificial system. Datum, Hierarchical Systems and Independent Software Objects Theorists from the physical, biological, social and computer sciences along with philosophers, cultural theorists and literary critics have collectively lacerated traditional conceptions of content, deconstructing its nature into the stored, sorted, distributed, and acted upon. Reasoning for this approach is obvious. Content, meaning, information and data can be segregated conceptually, the properties of each scrutinized with quantification and measurement analysis. Purely structuralist perspectives however, do not suffice to describe a model in which content can be explicated as a self-determinate function within a system - one that is not directly the result or product of a structure. Digital information for example, is not content. 0's and 1's are merely self signifiers and hold no meaning beyond their differentiation. They hold no symbolic content. In terms of computation, content is content only as a model of the content object, a hyper-simulation shaped from the elemental signs of 0's and 1's as autonomous mathematical entities. The lowest level of symbolic data structure as agency are therefore linguistic functions enabling complexity to arise in relationships of 'if, then, and, else, do, while.' It seems apparent that digital content is an aggregate of such linguistic abstractions, the datum. Datum are the fundamental agency (element) enabling algorithmic conceptualization to emerge in entailment meshwork of varying quality, function, capacity and purpose. Information networks are structured hierarchical organizations of datum agencies. Herbert Simon, in his pioneering work, The Science of the Artificial, describes a hierarchical system as, "one that is composed of interrelated subsystems, each of the latter being in turn hierarchic in structure until we reach some lowest level of elementary subsystem." Inferring the properties of the elemental subsystems from observation of the system in question is a most difficult task. Behavior is not necessarily indicative of the nature of information agencies of which a system is a composite. We can only infer about the nature of the agencies as theoretical elements. In the case of information networks, datum conceptualized as the final most elementary subsystem (agency) of the network semiotic theoretically explains the complexities of entailment meshwork which they enable. What then are the dynamics involving datum as an agency of entailment? One answer is that datum are not the output of a procedure, but rather a pattern of digital activity. According to Douglas Hofstadler, "meaning carrying objects carry meaning only by virtue of being active, autonomous agents themselves." Conceived as patterns of action, datum are most likely content intensive in terms of their instructional value with respect to other datum. Acting as an interfacing agency, the datum patterns with other datum to combine (or uncombine) into more (or less) complex propositional relationships. In that the network semiotic operates a linguistic conveyance between lower and higher level subsystems of entailment, the appearance of emergent coherency is an inevitable by-product of datum patterning. Datum processes are those that hold, relate, associate, reference, link, tie, merge, substitute, replace, identify, locate, determine, retrieve, list and index information. Datum are therefore symbolic in the sense of having denotative power to carry meaning as action. A reasoned supposition is that datum are necessarily predisposed to particular configurations of pattern and are therefore delimited in capacity. This perspective, even if true, does not account for all potential organizations spawned by datum within the network semiotic. It simply does not follow to insist that all possible meaningful expressions can be identified prior to the emergence of pattern. It is more likely that datum continue to emerge new capacities for autonomous activity by autocatalyzing into new aggregate relationships. Alternates A new feature of complexity in semantically rich domains can be postulated, that of the alternate. A mimicked substitution of datum which ripples throughout the semiotic influencing entailment meshwork, autocatalyzing self-similar capacities of immediacy and interaction. Alternates, like datum, are capable of self-determination and denotation of meaning through pattern. Unlike datum, alternates contort simulation patterning into stylistic equilibrium, expatiating the self-organizing tendencies of entailment meshwork into replicant and homogenized linguistic performances. The nature of the network is necessarily one of dissimulation. Information as an 'experience event' (hypertextual) on the surface of semiotic appearance is pretense to the functionality of the system. If the 'experience event' is derivative of nonlinear dynamics of entailment meshwork and autocatalytic linguistical functions of alternates, then the network is a simulation of its own functionality. Dissimulation appears as a hypertextual coherency across the semiotic in which the surface behavior of the network is the network. Homogenized diversity, the outcome of a hyperlogical destiny of immediacy and interaction, is present in the effects of repetition in which everything looks and behaves the same regardless of referentiality and meaning. Bifurcating states of permutation and chaos are transposed into self-similarity, in which differentiation is reduced to the continuities of coherence. "A medium is a medium is a medium", states media theorist Friedrich Kittler. In semantically rich domains the alternate is meaningful only as a pretense to the functionality of other alternates. Self-similarity of immediacy and interaction is not an equivalency in the perceived, but rather a coded derivative enabled by the alternate as a dissimulator of network appearance. This countenance of the network semiotic is most interesting in that exposition of self-similar patterning throughout the system encourages homogenized diversity as a stabilizing factor. Immediacy and interactivity attached to all datum contribute to the deception that the network cannot enact its own theme. Art as Network To conclude, having raised more questions than with which this essay began is alas, the goal. Art as Network is a discourse. A discourse that requires the dropping of orthodox positions on the nature of either. How do data structures relate to network semiotics? Is there a taxonomy of Art as Network? What is entailment as an art strategy? How does dissimulation evolve homogeneity? What conditions should be satisfied in order that Art as Network be identified? Is Art as Network self-deterministic, aware of it's emergence, a condition of the pandemonium of agencies (datum and alternates) forming entailment meshwork? Is it an art of correlation, homogenous redundancy and self similarity? The large scale paradigm of complexity is coming to influence both research and artistic enterprise involving networks. It is evident that some deeply held concepts such as authorship and interpretation are fragmenting, giving way to more adequate categories of analysis appropriate to the study of self-organizing systems. Orthodox conceptions of art as expression and network as technology must be replaced by art as system analysis and art as empirocriticism of the paradigms converging the play of forces shaping networks. Art as Network represent a correlative conjunction bridging Art and Network into a hybridized state. I advocate that the role of the artist is one of
theoretical analysis accompanied by large amounts of experimental work.
Designing systems to see how they behave, in my opinion, is art. This
is not to suggest that every system design can or should be considered
art. Issues of contextualization remain. What distinguishes art from
the research sciences and commercial entrepreneurship is a very thin
veil. Relevancy, an important criteria, is tied closely to the collision
of art with the physical, biological and social sciences, as well as
with the discourses of cultural theory and literary criticism. To be
sure, the topics of identity, cognition, virtuality, ubiquity and heuristics
are of looming importance to all disciplines, inlcuding art. In turn,
the investigation of nonlinear dynamics, hierarchical structure, self-organization,
entailment meshwork, autocatalytic datum, alternates, chaos, transparency,
dissimulation and homogeneity will enable an appreciation
of the potentiality of Art as Network. copyright 1998 Footnotes Bibliography
Manuel De Landa, War in the Age of Intelligent Machines, Swerve Editions, Zone Books,
New York 1991 - Distributed by MIT Press |
|
|
|
|
|
||
|
|
|
|[ X ]|
|
|
|