In this post my aim is just to outline the general features of DeLanda’s assemblage theory as presented in the first chapter of A New Philosophy of Society with little in the way of critical commentary or questions. Here I want to get a clear picture of what he sees as the central features of assemblage theory and how it differs from other social theories.

1. Relations of Exteriority versus Relations of Interiority:

DeLanda begins chapter 1 with the declaration that we must removed the entrenched metaphor of society as an organism. Within the organismic metaphor, society is compared to the human body, such that 1) all parts are dependent on one another, and 2) all parts (institutions) work together like organs in an organism to promote the harmony of society as a whole. Here it is notable that this conception of relations between parts is not restricted to organismic conceptions of society, but also to structuralist conceptions of society. The key thesis shared by these orientations is that parts have no existence or being apart from the whole to which they belong. Thus, for example, when we talk about a sound in language, we cannot say that “b” has an existence of its own independent of other sounds in language, but rather that “b” exists only in a phonemic relation with other sounds: b/p. The concept of structure is such that elements have no independent existence apart from their relations. As a consequence, elements are their relations within an organismic or structuralist conception of the social world.

read on!

DeLanda objects to this thesis on the grounds that it prevents us from explaining emergence. As DeLanda writes,

Allowing the possibility of complex interactions between component parts is crucial to define the mechanisms of emergence, but this possibility disappears if the parts are fused together into a seamless web. Thus, what needs to be challenged is the very idea of relations of interiority. We can distinguish, for example, the properties defining a given entity from its capacities to interact with other entities. While its properties are given and may be denumerable as a closed list, its capacities are not given– they may go unexercised if no entity suitable for interaction is around –and form a potentially open list, since there is no way to tell in advance in what way a given entity may affect or be affected by innumerable other entities. (10)

Relations of interiority are thus relations in which the components that are related have no independent existence apart from the relation in which they exist. This concept of relation, DeLanda argues, must be abandoned. While components do indeed interact with one another, they also have an existence that is, in principle, independent of these interactions with one another. Here DeLanda draws a distinction between 1) the properties an entity exemplifies in an interaction or relation with other entities, and 2) the capacities an entity has to interact with other entities. The point of this distinction is that entities always harbor unexercised capacities that might produce very different properties were the entity to enter into different relations with other entities.

DeLanda’s point is thus that we must not confuse the properties of an entity with the capacities of an entity. Properties of an entity are local results of interactions between entities. For example, the water boils because it is heated up. Capacities of an entity are powers that an entity possesses, regardless of whether these powers are exercised or not. The confusion of entities with their powers is what Roy Bhaskar called “actualism”. Actualism reduces the being of an entity to the properties that happen to be actual or occurrent in that entity at a particular point in time. Here my own distinction between virtual proper being and local manifestation closely mirrors DeLanda’s distinction between capacities and properties. If this distinction is so crucial, then this is because it allows us to think the manner in which objects can depart from certain relational networks within which they are enmeshed and produce different properties or qualities.

In contrast to the concept of society and the world as an organic totality defined by relations of interiority, DeLanda proposes a concept of the world and society based on assemblages composed of relations of exteriority. The central feature of relations of exteriority is that the components of an assemblage may be detached from it and plugged into a different assemblage in which its interactions are different (10). DeLanda argues that the properties of the parts of an assemblage can never explain the whole or assemblage they constitute. Those wholes or assemblages are dependent on the component parts (there’s nothing mystical about emergence), but have a structure of their own that is irreducible to these parts.

Where relations of interiority are defined by necessity (part A cannot exist without part B and part B cannot exist without part A), relations of exteriority, contends DeLanda, are only contingently obligatory. Here DeLanda gives the example of co-evolution (12). In Deleuze and Guattari’s famous example of the orchid and the wasp, there is a co-evolution that took place producing orchids that look like female wasps, thereby capturing the wasp as a part of the flower’s reproductive organs, but it would be a mistake to argue that the relation between the orchid and the wasp is a relation of interiority. Wasps could die out and presumably these orchids would either die out themselves or would reproduce in some other way. Likewise, were this species of orchids to die out, wasps wouldn’t cease to exist, but would simply move on to other flowers. In short, these relations are local and contingent. Each of these entities can be situated in new or different relations, wherein, perhaps, they will actualize different properties and behaviors.

DeLanda emphasizes that the components of an assemblage are always heterogeneous (11). Assemblages are never composed of one type of entity, such as, for example, language, but are composed of a variety of different types of entities that exist in relations of exteriority to one another. Part of a good assemblage analysis would consist in thinking the relations between these heterogeneous entities in an assemblage.

An Aside on OOO and Assemblage Theory: It is clear that between OOO and DeLanda’s assemblage theory, there is a great deal of proximity. Like DeLanda, OOO argues that relations between objects are relations of exteriority rather than interiority. Any object can, in principle, be detached from the relations in which it exists. Likewise, DeLanda’s distinction between capacities and properties is very close to my own distinction between virtual proper being and local manifestation within the framework of ontology. Finally, DeLanda’s conception of the relationship between parts and wholes is designed to do something very similar to the meriology proposed by Graham in his object-oriented ontology (which I have wholeheartedly taken over). Graham argues that the universe is such that we have objects wrapped in objects wrapped in objects and so on. Like Harman, DeLanda emphasizes that the components of an assemblage are entities in their own right that have an existence independent of the assemblages to which they belong.

DeLanda deploys the concept of relations of exteriority to, in part, explain relations between the macro- and micro-level of society. He’s interested in maintaining the thesis that the micro- components of a society are, in some important sense, independent of the society to which they belong and therefore can enter into relations with other assemblages. Here OOO is entirely on board. The major difference between OOO and assemblage theory seems to be that DeLanda holds that there are only relations of exteriority, whereas OOO maintains that there are both relations of interiority and exteriority. Each object is composed of relations of interiority or what I call “endo-relations” and what Graham calls “domestic relations”. Does this entail that we’re back to the organic metaphor? No, because within the framework of OOO the parts that make up an object are in their turn objects and are therefore independent of the macro-scale assemblage to which they belong. More on this another day.

2. Matter and Expression:

DeLanda argues that every assemblage exists along two dimensions: Along the one dimension, there are what he calls the material and expressive components of assemblages. Along another dimension, there are what he refers to as the processes which either stabilize the identity of the assemblage by increasing its degree of internal homogeneity, or destabilize it (territorialization and deterritorialization). It is here that we begin to see a real difference between DeLanda’s approach to the social and other reigning approaches to the social, for a good deal of continental social theory has focused on the expressive (language and meaning) to the detriment of the material.

The material refers to those material elements that an assemblage requires in order to exist. As DeLanda writes,

The components of social assemblages playing a material role vary widely, but at the very least involve a set of human bodies properly oriented (physically or psychologically) towards each other. The classic example of these assemblages of bodies is face-to-face conversations, but the interpersonal networks that structure communities, as well as the hierarchical organizations that govern cities or nation-states, can also serve as illustrations. Community networks and institutional organization are assemblages of bodies, but they also possess a variety of other material components, from food and physical labour, to simple tools and complex machines, to the buildings and neighborhoods serving as their physical locales. (12)

Scant attention has been devoted this dimension of social organizations within continental social and political theory. Take the example of Zizek. Zizek focuses all of his attention on the expressive dimension of the social, on content, on meaning, on the signifier, yet ignores this material dimension. As a consequence, the implicit thesis seems to be that the material dimension contributes nothing to why the social as it is. Rather, the social is to be found, according to Zizek, purely at the level of the social. And as a further result, it follows from this that Zizek holds that the only thing relevant to producing political change lies in the domain of the expressive.

This bias for the expressive over the material is not restricted to Zizek, but can be seen in Habermas, Rawls, and Badiou as well. Everywhere we find an emphasis on the expressive, such that the material becomes all but invisible. The point is not that the expressive is not a component of social assemblages, but rather that the expressive is only a component of the expressive. If we wish to understand why the social is organized as it is, then we need to take the material into account as well. This point will become clearer in a moment.

3. Territorialization and Deterritorialization:

Territorialization and deterritorialization refer to those processes within an assemblage that either function to stabilize the assemblage or destabilize the assemblage. DeLanda says that the concept of territorialization is to be understood literally:

The concept of territorialization must be first of all understood literally. Face-to-face conversations always occur in a particular place (a street corner, a pub, a church), and once the participants have ratified one another a conversation acquires well-defined spatial boundaries. Similarly, many interpersonal networks define communities inhabiting spatial territories, whether ethnic neighbourhoods or small towns, with well-defined borders. Organizations, in turn, usually operate in particular buildings, and the jurisdiction of their legitimate authority usually coincides with the physical boundaries of those buildings. (13)

To my thinking, DeLanda’s concept of territorialization is one of the most remarkable features of his thought, for, in a spirit very similar to that of Latour’s actor network theory, DeLanda insists that wherever we talk about social relations, real connections must be involved. Thus, for example, if we speak of “the market”, we speak poorly and about nothing at all, for we haven’t done the work of mapping real connections that enable something like the market to exist. Just as a conversation requires some mode of real connection between the participants, markets need real connections between economic actors to take place.

In failing to map the heterogeneous entities that make up the market (both material and expressive), we turn the market into an omnipotent ghostly entity because it’s everywhere and nowhere, because it’s purely abstract, and thus is indestructible as is the case with any phantasm. How do you target a ghost? No, if we want to target the market we should proceed like a general. The general doesn’t talk abstractly about “the enemy”. No, what the general wants is the locations of the opposing army, the supply lines that feed the army, the location of munitions factories, the technologies used to transport weapons, supplies, and soldiers, the manner in which people in the opposing army communicate with one another, the recent movements of that army, and so on. The general wants to know the territorializing processes of the opposing army. And through this knowledge, the general gradually acquires the knowledge necessary for engaging with the opposing army.

When we speak of “the market” we speak like rubes because we proceed as if we know what we’re talking about, without having any concrete map that would allow us to engage with this entity. Rather it just becomes a ghost that is everywhere and nowhere. A critical theory that practiced assemblage theory or object-oriented ontology would look a lot more like military intelligence reports than ideology critique or deconstructive debunking. It would seek to determine how certain social assemblages are put together, what their gears and pulleys are, so that it might devise ways of strategically targeting these things. With the exception of Marx, our current critical theory knows next to nothing about the social world it is trying to change. It endlessly spins its wheels in the critique of the realm of expression, ignoring all the material components that hold our social assemblage together in the way it’s held together. This is a bit like trying to do surgery without having taken any courses in anatomy or having taken any x-rays of the patient to see what’s wrong. In this regard, it’s difficult to escape the impression that a good deal of political theory talk is more posturing directed at other academics than anything designed really to change things.

The territorializing dimension of assemblages refer to those processes by which an assemblage stabilizes itself, reinforcing both its own identity and the identity of its components. Let’s take DeLanda’s example of a conversation. A conversation can be territorialized in the spatial location of a face-to-face encounter. As the conversation unfolds within this territory, the contingent relation between the two participants become stronger and stronger, their beliefs and attitudes become more and more alike, and so on. This doesn’t entail that the participants in the conversation have to agree with one another. The participants in the conversation might be extremely hostile to one another. However, as their antagonistic discussion unfolds, their reactions to one another become more and more standardized, such that the participants find ways to both provoke one another and such that responses can now be reliably expected. Here the relation between the elements has been stabilized. Yet this stabilization is an open stabilization. All sorts of new interactions can be generated within this framework– one day the discussion can be about Kant, the next about Derrida, the next about Brandom, and so on –yet the basic assemblage and pattern of interactions underlying the assemblage remains the same. The point here is that these stabilizations require real connections in order to take place. In Latour’s terms, they require work.

Deterritorialization, by contrast, refers to the intervention or appearance of components that destabilize an assemblage, either causing it to change or perhaps even causing an entirely new assemblage to emerge. To illustrate this, let’s compare a college philosophy department with the blogosphere. A college philosophy department is a territory. Persons that land within this territory are likely to undergo processes of territorial coding or homogenization as a result of interactions with their colleagues. Not only are the conversations that take place in a department likely to lead to more or less shared frameworks (which might nonetheless be very diverse), but they are also likely to exercise pressure on the members of the department to keep up with a particular set of readings or references and to pursue certain tacit lines of approved research to get tenure. For the person whose academic life completely unfolds within a particular department, the other members of the department more or less come to represent the voices of the entire academic world. If it isn’t articulated by them, it doesn’t really exist for this person at all. Here “das Man” is not some abstract entity, but is something that is right there in the department.

For such a person, the blogosphere can very genuinely be a space of deterritorialization. Here the nature of the person’s intellectual encounters is transformed. The person encounters new references, new texts, enters into new conversations, and so on. Note that entering into the blogosphere does not entail an escape from territory as such, but rather an entrance into a new territory. It is this new territory that leads to new forms of organization. And indeed, while the blogosphere can be a deterritorializing agent for the philosophy department (the academic sets out on new lines of inquiry that, in their turn, impact other members of the department, causing the assemblage to change), other territories emerge within the blogosophere. Thus, in the blogosphere, we see that only certain blogs link to one another. Standardized themes of discussion emerge across these blogs. Norms of conduct and interaction emerge. Different blogs are coded in different ways (“over there are the Deleuzians, over there are the Derrideans, oh look, there’s the Lacanians, etc”), certain concepts become established or coded, etc., etc., etc. In short, a new territory emerges that has its own boundaries and limits.

Setting aside all of these expressive components that belong to the theory blogosphere, the point not to be missed is the role played by all the material components within this territory. Servers, fiber optic cables, high speed internet connections, electricity, computers, wireless, etc., etc., etc., are all necessary for these sorts of connections to be formed. The assemblages formed in the blogosphere were not possible for Hegel or Kant, nor are they possible for people in areas where such technologies are not available or don’t exist. As a consequence, certain forms of change are not possible for assemblages such as this. The point here is that theoretical shifts aren’t simply about expression or ideas, they aren’t free floating abstract entities. Rather, they require real connections in order to come into existence and are sustained by real connections. Consequently, it is important to attend to that which falls outside the expressive but which contributes to the expressive. For example, if we want to overcome the dominance of right wing politics in the United States, perhaps a big part of this project consists not simply in ideological critique, but also in making cheap communications technologies available to rural areas so that the young and bored might have access to communication alternatives besides the conversations that take place in their local newspapers, their churches, and their daily conversations. Internet traffic maps are startling. Traffice tends to come from the coasts of the country, with hardly any traffic from the center of the country. When we correlate this with voting patterns in the United States some troubling questions emerge.

It seems to me that DeLanda’s concept of territorialization and deterritorialization cries out for a very different concept of time and space. Time and space should not be conceived as containers or milieus within which events take place, but rather as meshes of connective relations. There’s a very real sense in which I am closer in space to Graham, despite the fact that he is in Cairo, than I am to my department chair. This is because my real connections with him are more persistent, varied, and ongoing than my connections to my department chair who is on another campus and who I only interact with occasionally over the course of the semester. He is fifteen miles away, Graham is thousands of miles away. Yet in social space it is not geographical distance that defines the connection, but rather connectivity. Similar points should be made about time. Here social time should be thought in terms of speeds of connection or how quickly a message can be disseminated throughout an assemblage. The time of the APA and SPEP is rather slow compared to the time of the blogosphere. These former organizations require annual meetings for new expressive contents to circulate throughout the social field (and even then, this depends on the participants attending certain panels), whereas new expressive contents circulate throughout the blogosphere very quickly, in the blink of an eye. As a consequence, these assemblages have very different temporal structures.

At any rate, this is enough for now. Next up is Alex Reid with chapter 2 over at Digital Digs.