Understanding Deleuze

Claire Colebrook writes an overview of Deleuze’s philosophy. Deleuze is in tradition of practical “lively” philosophy. What does it mean for a philosophy to be practical? Colebrook compares Foucault, Freud, and Marx as all practical. Marx’s philosophy is intended to be connected to the world and directly change our understanding with it.

Other, more linguistic philosophers, (eg Wittgenstein) aim to understand language, and use common language. That we will realize that things we say are nonsense. Namely, they pose that theory is in a sense fundamentally disconnected from reality. The comparison here raises the ambiguous question of what does it mean for a philosophy to be practical.

Colebrook poses that Deleuze is a positive thinker: that he saw desire as a positive, constructive force that enables meaning. She seems to lay out a quadrant of some philosophers:

	Negative	Positive
Power	MarxIdeas produce power relations.	FoucaultTheories and actions are modes of power. Concepts are instances of power. The master and slave are conceptually codependent and produce power through each others’ existence.
Desire	FreudDesire is something that occurs outside of a norm. Desire is something that detracts from a person and must be fought against to restore normality.	DeleuzeExistence and identity are created through desire. Desire enables identities and relationships.

Overview

It is important to note that A Thousand Plateaus is the second part to Deleuze and Guttari’s “Capitalism and Schizophrenia” pair, the first part being Anti Oedipus. That said, many of the concepts used here are in fact first defined in the first volume.

On Models

Deleuze on models: models are prescriptive. Claims that Western thought is built on radical (single root) model systems, that ascribing to models limit our world view and limits, to subscriber, what is possible.

I would say that the solution in Western *science* or any other constructive movement is to define NEW models, in abundance. This is something that is heavily studied in linguistics, development, etc. New concept/system development does not account for the limitations of ingrained models. In development and education, there is a bit of investigation of concept reformation and development.

What about meta-models? Deleuze attempts to get underneath models (instead of above). Meta models, as might be imagined mathematically, look to define new structures that can turn and encompass others. When models are used in math, science, and programming (models meaning generally varied approaches to representation within a system or framework), they are used in varying applications. Many times scientists, mathematicians, and programmers all try to force more things than can be accounted for into one system, but this is generally recognized as a poor idea.

Frequently, models are defined to address specific problems, and are intended to be used within a specific domain, or from a specific perspective. Change of these may ask for a change in the model being used. Examples are in looking at human behavior, where sociology, anthropology, linguistics, or statistical methods might be used to explain various about human behavior.

On computer code and rhizomes: computer systems are “tree”-like in that they all can be translated (in Chomskian sense) to equivalent computer instructions. They are all founded on some basic underlying models. So, while they may enable interpretation, representation, and thought, in very different ways, they are still executed through the same turing machine. They do enable different means of cognition, but they must be grounded in some fundamental principles.

When applied to programming and simulation, the situation gets trickier. Computer languages, simulations, and representations are all very capable and abstracted. However: programs all must be reduced to machine code and rendered on some form of hardware, eventually. What this reveals is that all things that are simulatable by a computer (or by a formal simulation that satisfies some programmability requirement) are all possible to reduce to one single, ultimate language. This implies that this simulation root underlies all models expressible within a computer.

However, it might be stated that while all simulations share a root of simulatability, they may share roots with other conceptual models and domains, and thus be rhizomes. So, while the execution level of a simulation might be universally translatable, the other levels may not be translated so easily, especially when the representative level is strongly metaphorically coupled to the simulation. A simulation whose execution is tightly bound with its representation is a rhizomic structure, whereas a simulation whose execution and representation are disjoint may be pulled up easily.

On Territorialization

The concept of deterritorialization is coupled with a reterritorialization. To Deleuze and Guattari, individual things have a territory, but when their systems touch upon one another, their respective territories are upset and then reformed. The example given is a wasp touching an orchid: The orchid is upset and disrupted by the wasp by the contact, and correspondingly, the wasp is turned into a part of the orchid’s reproductive system.

The challenge with this model is that it treats the wasp and the orchid as both totally independent systems until they contact one another. Systems are rarely ever totally independent, and do rely on each other. Frequently this may occur via well defined channels, such as the wasp’s fertilization of the orchid, but the notion that systems are structured in connection with each other seems radically opposed to Deleuzian sense. Further, one may scratch the idea of systems as being independent altogether, and understand that any perceived territory of a system is merely a construct or illusion. If we look back far enough, every system can be seen to be composed of multitudes of subsystems. The plant itself is composed of billions of cells which each impinge on each other as part of the plant’s growth. Blossoming in an orchid is a disruption of the plant’s ordinary sympodial pattern. It bears noting that sympodial growth is a from of rhizome. Go figure.

Principles of the Rhizome:

1. connection
2. heterogeneity
3. multiplicity
4. asignifying rupture : independence of models
5. cartography
6. decalcomania

classical linguistics: Language is built on binary differences, furthermore, differences do not *mean* anything. That is, they are arbitrary. Language becomes interesting in its inability to communicate. D&G trying do deny function of representation in knowledge?

Classical representation romanticises the idea of pure meanings, and that before language things were better. Representation aims to point things back to these pure ideas, and thus emphasizes, and is dependent on the notion of lack. Thus, classical representation constantly is a reminder of the lack of pure meanings. But… doesn’t representation project from one system of meanings to another? Why does there have to exist a system of pure meanings? What if I reject the notion of such a thing?