Complexity and Postmodernism

[Edward theurj Berge]

I came upon this free ebook, Complexity and Postmodernism by Paul Cilliers (Routledge 1998). From the introduction:

“Complexity and Postmodernism explores the notion of complexity in the light of contemporary perspectives from philosophy and science. Paul Cilliers contributes to our general understanding of complex systems, and explores the implications of complexity theory for our understanding of biological and social systems. Postmodern theory is reinterpreted in order to argue that a postmodern perspective does not necessarily imply relativism, but that it could also be viewed as a manifestation of an inherent sensitivity to complexity.

As Cilliers explains, the characterisation of complexity revolves around analyses of the process of self-organisation and a rejection of traditional notions of representation. The model of language developed by Saussure—and expanded by Derrida—is used to develop the notion of distributed representation, which in turn is linked with distributed modelling techniques. Connectionism (implemented in neural networks) serves as an example of these techniques. Cilliers points out that this approach to complexity leads to models of complex systems that avoid the oversimplification that results from rulebased models.

Complexity and Postmodernism integrates insights from complexity and computational theory with the philosophical position of thinkers like Derrida and Lyotard. Cilliers takes a critical stance towards the use of the analytical method as a tool to cope with complexity, and he rejects Searle’s superficial contribution to the debate.

Complexity and Postmodernism is an exciting and an original book that should be read by anyone interested in gaining a fresh understanding of complexity, postmodernism and connectionism.”

[Edward theurj Berge]

Here's something from Byrant's "Time of the object" referenced and linked above consistent with Cilliars:

"Substance is not that which is opposed to temporality and process, nor is it an abiding identity that persists beneath changing qualities, but rather it is temporal through and through. As such, substance must produce itself from moment to moment and perpetually faces the threat of entropy or dissolution from both within and without. Substances are negentropic unities whose identity consists in their operations through which they produce themselves across time. As such, they evolve, change, and mutate in all sorts of ways."

[Edward theurj Berge]

Another free e-book on the topic is DeLanda's Intensive Science and Virtual Philosophy (Continuum, 2002). Therein he explores the rhizomatic complexity of Deleuze.

[Edward theurj Berge]

A sample from ISVP:

"Delueze is...a realist philosoper....[but] not a realist about essences or any other transcendent entity...something else is needed to give objects their identity and what preserves this identity through time. Briefly, this something else is dynamical processes" (2-3).

[Edward theurj Berge]

In the "real and false reason" thread I contended that there was another kind of math at the root of an alternative complexity different from the algebraic sets Commons uses to build his hierarchical complexity. I made reference to Deleuze and his use of differential calculus and indeed DeLanda is going into that in chapter 1. He uses this math to ground Deleuze's notion of multiplicity and the manifold. He says:

"A Deleuzian multiplicity takes as its first defining feature these two traits of a manifold: its variable number of dimensions and, more importantly, the absence of a supplementary (higher) dimension imposing an extrinsic coordinatization, and hence, an extrinsically defined unity. As Deleuze writes: 'Multiplicity must not designate a combination of the many and the one, but rather an organization belonging to the many as such, which has no need whatsoever of unity in order to form a system'" (12-13).

[Edward theurj Berge]

"Unlike essences which are always abstract and general entities, multiplicities are concrete universals....[and] is typically divergent....unlike essences, which as abstract general entities coexist side by side sharply distinguished from one another, concrete universals must be thought as meshed together in a continuum. This further blurs the identities of multiplicities, creating zones of indiscernibility where they blend into each other, forming a continuous immanent space very different from a reservoir of eternal archetypes" (22-3).

[Edward theurj Berge]

"Besides the avoidance of essentialist thinking, Deleuze's speculation about virtuality is guided by the closely related constraint of avoiding typological thinking, that style of thought in which individuation is achieved through the creation of classifications and of formal criteria for membership in those classifications. Although some classifications are essentialist, that is, use transcendent essences as the criterion for membership in a class, this is not always the case. For example, unlike Platonic essences which are transcendent entities, Aristotle's 'natural states,' those states toward which an individual tends, and which would be achieved if there were not interfering forces, are not transcendent but immanent to those individuals. But while Aristotelian philosophy is indeed non-essentialist, it is still completely typological, that is, concerned with defining the criteria which group individuals into species, and species into genera" (41).

[Edward theurj Berge]

Ah, we're getting to a flat mereology like we've seen in Bryant:

"Species are individuals, not kinds...[and] does not represent a higher ontological category than the individual organisms that compose it.... The relations of individual species to individual organisms is one of whole to parts, much as the relation between an organism and the individual cells that compose it. Moreover, the relation of parts to whole is causal; the whole emerges from the causal interactions between the component parts.... While an ontology based on relations between general types and particular instances is hierarchical, each level representing a different ontological category (organism, species, genera), an approach in terms of interacting parts and emerging wholes leads to a flat ontology, one made exclusively of unique, singular individuals, differing in spatio-temporal scale but not in ontological status" (46-7).

[Edward theurj Berge]

A sidebar on DeLanda's writing style: Like Bryant his style is precise and highly intelligible, making very complex scientific and mathematical ideas understandable to the layman like me. Quite refreshing.

[Edward theurj Berge]

As he said earlier, he uses the science of dynamic systems. From this he explores how undifferentiated, intensive capacities give rise to differentiated, extensive forms. As but one example he uses embryogenesis. When an extensive form is completed we get an idea similar to Bryant's withdrawal. He says:

"But the basic idea is that is that once a process of individuation is completed, the intensive factors that defined this process disappear or become hidden underneath the extensive and qualitative properties of the final product" (59).

[Edward theurj Berge]

Here's more on the virtual, similar to the withdrawn. (Bryant discusses the differences between the concepts in TDOO, particularly chapter 3.*)

"An individual may be characterized by a fixed number definite properties (extensive and qualitative) and yet possess an indefinite number of capacities to affect and be affected by other individuals.... Deleuze, in fact, always gives a two-fold definition of the virtual (and the intensive), using both singularities (unactualized tendencies) and and what he calls affects (unactualized capacities to affect and be affected)" (62).

*  For example: "Another way of understanding the concept of virtual singularities or attractors is in terms of Spinoza's concept of affect....[which] links the concept of affect to the capacities of an object.... Tthese affects consist of both an entity's 'receptivity' to other entities and the various capacities an entity has to act" (3.4).

[Edward theurj Berge]

More from TDOO on DeLanda:

"The attractors of a substance....are the generative mechanisms within an object that preside over the events or qualities of which the object is capable. However, while serving as the condition of these events or qualities, these attractors are not themselves qualitative or events. As DeLanda puts it, 'attractors are never actualized, since no point of a trajectory [of an object] ever reaches the attractor itself.' As such, the attractors or singularities inhabiting the endo-structure of an object are radically withdrawn. They are that which serves as the condition for the actual dimension of an object, for the local manifestations of an object, but are never themselves found on the actual side of an object" (3.3).

[Edward theurj Berge]

On 69 DeLanda uses a familiar image, with a twist. Extensive structures are at the bottom, intensive in the middle and the continuous and undifferentiated virtual at the top. However he notes this should not be considered a hierarchical relation. His preferred image is as follows: "A better image here would be a nested set of spaces, with the cascade acting to unfold spaces which are embedded into one another." I'm picturing more the venn diagrams I've used before, since embeddedness in not complete subsumption of one into the other hierarchically but rather in shared spaces, still maintaining their own space(s) apart from such relations.

Also recall TDOO: "This is a variation of Cantor's Paradox. Cantor's paradox demonstrates that there can be no greatest cardinal number precisely because the power set of any cardinal number will necessarily be larger than the cardinal number itself. In a stunning inversion of the ancient thesis that the whole is greater than the sum of its parts, the power set axiom reveals, to the contrary, that the parts are always greater than the whole" (6.2).

And recall Latour's whole being smaller than the parts.