The first point to raise regards DeLanda’s realism and his definition of this in sociological terms as ‘conception-independence’ (3). In response to Levi’s review of the book’s introduction I wrote:
“‘Mind independence’ is in no way a new idea, it is the mainstream view for social scientists of all stripes. DeLanda’s innovation, it seems to me, is nowhere to be seen in the first few pages (though it abounds immediately after).”I more or less stand by this argument, although I must now make some adjustments. Upon reviewing my notes, it becomes clear that DeLanda does not say that society is ‘mind independent’ – i.e. it has an existence independent of human minds – but that it is ‘conception independent’ – i.e. it has an existence independent of the analyst’s conceptions of it. This distinction is introduced to account for the fact that if human minds disappeared so would society. This is a fair and appropriate distinction but I still don’t accept that it is particularly innovative. It’s still ‘off the shelf’ realism that, to someone trained in social science rather than philosophy, is frankly de rigeur.
To my first comment Michael at Archive Fire argues that:
“While, for the most part I agree with that line of thinking, it must be acknowledged that DeLanda’s audience are not those people [social scientists]. DeLanda is, first and foremost, a philosopher – and specifically a Deleuzian philosopher drawing extensively on what has come to be known as the “continental” tradition. So DeLanda’s project must primarily been seen as philosophical - as an attempt to reach out to those thinkers who, having learned from the intensities of critical theory of the 80’s and 90’s that focused on language and interpretation, may, again, be seeking out a way to supplement their thought with a new concern for material life and the more tangible dimensions of human experience.”I see his point but I don’t think that I can accept his reasoning. DeLanda makes extensive use of various sociologists’ works throughout the book and declares that his intention is to “elucidate the proper ontological status of the entities that are invoked by sociologists and other social scientists” (8). Far from aiming his book primarily at ‘continental’ philosophers he clearly wishes it to be valuable to social scientists too – he’s working at the edge of both traditions and as such should be evaluated as much as possible by the standards of both. This is consistent with his statement in an interview that “a philosopher cannot take … artificial [disciplinary] limits into account, and … should push multidisciplinary approaches to the limit” (Deleuzian Interrogations, 14). It is not that I find his definitions of realism wrong as such, it is just that they are a little simplistic – ‘clunky’ would be my preferred adjective.
However, this is not the main problem. The greater problem, besides his realism’s ‘clunkyness,’ is that both ‘mind independence’ and ‘conception independence’ appear to be attacks on a position of subjective idealism, whereas poststructuralist inspired orthodoxies (which I would assume are primarily in his crosshairs) must be characterised as intersubjective idealism. This is not a particularly massive difference in the grand scheme of things but it does make a difference in this case. An intersubjective idealist may well agree that society has a reality independent of their individual conceptions of it because their views, they would add, are themselves drawn from (or even produced by) the wider socio-linguistic or discursive field. Of course, because they define society in linguistic terms they remain anti-realist, yet they would still dodge DeLanda’s realist haymaker.
In other words, mind or conception independence is clearly a necessary condition for realism but it is not a sufficient condition.
So, moving swiftly on, what do we already know about assemblages?
“[A]ssemblages [are] wholes whose properties emerge from the interactions between parts” (5). “[A]t each scale one must show that the properties of the whole emerge from the interactions between parts” (32). In keeping with the realist temper of the ontology, assemblage based analysis is causal and is “concerned with the discovery of the actual mechanisms operating at a given spatial scale” (31). ‘Micro’ and ‘macro’ are relative terms, with any given assemblage having micro or macro aspects and there existing an effectively unlimited multitude of gradations between the two throughout the cumulative emergence of levels (32). Importantly, “although a whole emerges from the interactions among its parts, once it comes into existence it can affect those parts” (34). Assemblages thus possess powers of ‘downward causation’ – indeed this property may be what distinguishes them from mere aggregations of parts.
Assemblages are thus characterised by ‘levels’ which are continuous insofar as there is no definite dividing line between them and each higher level comprises all lower levels as parts; however, levels are discrete insofar as one can identify a number of parts which form a larger whole, the properties of which are irreducible to the simple aggregation of the properties of the parts.
Insofar as each level has its own properties and dynamics different analytical resources will be required. For this reason we could say that discrete assemblages are characterised by distinct modalities (L. modus “measure, rhythm, manner”) of existence. The sense of harmonic, rhythmic or musical distinction is a useful one because these notions are easy to imagine in terms of continua, yet at the same time it is easy to imagine distinct aural objects emerging from that milieu. In other words, the notion of modes as rhythms or harmonies allows for both the mixing and imbrication of elements and their distinction from each other.
Analytically, assemblages are analysed through a fourfold structure – that is, two intersecting axes, each of which are envisioned as continua. The first division is between material and expressive properties and the second division is between processes of territorialization and deterritorialization. (I will assume that readers are familiar with what these terms mean.) Each level thus requires an analysis of the following factors:
1) Material components of the assemblage.
2) Expressive components of the assemblage.
3) Territorializing processes at work in the assemblage.
4) Deterritorializing processes at work in the assemblage.
This divides up the empirical analyses quite neatly and intuitively. The analysis as a whole, however, should also indicate what causal mechanisms create, sustain and/or threaten the assemblage as a whole and must, therefore, give some indication of how all these elements work together. With this in place we can get down to business…