On Assemblage: What’s the Problem?

After hearing a few remarks on the equivalence between actor-networks and Deleuze’s agencement, which did not sound right to me, I went back to read one of my favourite commentaries on his concept of “assemblage”. I believe discussing it could start moving me along towards what I might possibly mean by a ‘nomad social studies’. A way in into this discussion will be the ‘target’ of “assemblage” or the problem it sets to address and why, in my view, it cannot simply be imported into social science and theory; especially not in the way actor-networks have been.

In Brent Adkins’ Deleuze and Guattari’s A Thousand Pleatues: A Critical Introduction one can find a fascinating specification of Deleuze’s concept of “assemblage”. It is a highly productive treatment because Adkins’ explains Deleuze’s concept within Deleuze’s own philosophical framework. I believe this is important for attempts at placing the concept in conversation with social science, because Deleuze’s framework is very hard to translate into terms which make sense to knowledge-making practices of the kind practiced by mainstream disciplines (sociology, organisation and management studies, international relations, and perhaps, to a lesser extent geography and anthropology). I will try to explain this point.

It is somewhat customary by now in the social sciences to deploy the concept of “assemblage” as a way of denoting some sort of arrangement between things — perhaps a network-like structure or a fragmented array. There might be disagreements on the kind of shape assemblage is meant to denote and scholars indeed use it variously. However, if one follows Adkins, that is actually not very relevant. Or, rather, it is not what is conceptually innovative about the notion. Regardless of the exact formation which “assemblage” is supposedly meant to represent, the concept of “assemblage” in Deleuze is devised to handle a completely different set of questions; non-representational ones.

As elaborated by Adkins, the concept is not a description of things (a representational/epistemological question), but rather the answer to the question What is a Thing? (an ontological question). To put differently, and in terms that make more sense to social science, “assemblage” is not about an attempt to describe empirics using a theory which more accurately resembles the world, but about how we think the very separation between empirics, theory and the world. Or differently still, the concept is not part of A Theory of Things, but part of A Theory of How to Ask What a Thing Is. Deleuze himself, when asked about the fragmented charter of the book A Thousand Plateaus, commented that the concept gives the book its ‘thematic unity’. Thus, a reading of the concept as representational — that is a kind of shape or form in which the chapters in the book are arranged in, which are then denoted ‘assemblage’ — is quite easy to elicit. However, as mentioned, this is not what “assemblage” is meant to help us with. It is meant to help us think, rather than describe.

The concept of “assemblage” is devised to help us notice and engage the philosophical problem of “things”, namely that a “thing” combines two contradictory properties: stability and change. In Western metaphysics, the projects of Heraclitus and Parmenides, as Adkins reminds us, can be read as about the minimisation of one of these poles, while Plato’s project can be read as about the strict separation between these two and a rendering of the problem into the discontinuity of the sensible and the intelligible (or what is termed hylomorphism).

Largely, and if indeed the history of Western philosophy is footnotes to Plato like Whitehead proclaimed, the concept of “assemblage” was devised to bypass an entire edifice which equates a thing’s stability with its essential, intelligible nature on one hand and a thing’s ability to undergo change with its accidental, sensible nature on the other. Whitehead identified a similar problem at the root of Western metaphysics when he discussed ‘the bifurcation of nature’. This kind of split, due to various historical reasons, is replicated in the social sciences and their image of objectivity. The very existence of a social science (the possibility of a scientific account of social nature), in a way, rests on this separation, or on the position of hylomorphism.

Thus, “assemblage” is not another ‘theory’ (an intelligible) with which to conceptualise ‘empirics’ (a sensible), but a collapse of the image of science that gives social science meaning, to begin with. With the separation between theory and empirics so entrenched in social science (as a modern science; Stengers is helpful here), it is hard to think of a more alien starting point for social science than “assemblage”, as it is a speculative concept meant to assert the continuity of the sensible and the intelligible.

In other words, “assemblage” is a huge commitment. The crux of the matter is that even though there are different experiments and people are doing various things, I am not convinced we know what a social science that beings with the continuity of the sensible and the intelligible even looks like, let alone what sort of practice it requires. Regardless of whether or not we follow Deleuze’s proposed intervention (how he ‘solves’ the philosophical concern), a different image of science is needed if we to uphold the commitment. I see this to be a fundamental issue and a very interesting path to follow. One which leads much further afield than a description of networks, fluidity, or any other representational metaphor meant to provide ‘empirics’ with a more accurate ‘conceptualisation’.



  1. Interesting post!
    Though, ultimately, I’m not sure that I understand why the notion of assemblage should affect the way social scientists work.
    If I understand Deleuze correctly, then the notion of assemblage supplements his theory of difference in itself, that is, of intensity; an assemblage is an assemblage of intensities. In this sense, it is merely another metaphysical theory in the tradition of Plato and Leibnitz, except that it does not begin with substance but with difference.
    Do you think that traditional social theories operate on the assumptions of a metaphysics of substance? And that working with a metaphysics of difference should affect social theory in some way? It is interesting to think of the relation between metaphysics and social theory, but I think some mediation between the two is required. That is, I think that we must first look at the point in which the two encounter each other.


    1. Thanks! Yes, I would say that definitely. Social theory takes its object to be a substance – ‘the social’. I read Latour’s Reassembling the Social in this very sense, moving away from a social theory that deals with substance. But, this is a more profound point, and you are right, what would a social science that rests on Spinoza or Leibnitz look like instead of the one we have that goes back to Kant? As long as we understand science as representational (again – about a particular kind of account of nature) we are in the realm of a bifurcated nature. Deleuze’s argument will be that the natural sciences are not representational (and this is largely the argument of science studies and the STS ethnographies as well). There is a lot of talk about ‘performativity’ in social science, but a performative science will look completely different that the sort of ‘accounts’ found in journals and mainstream research.


  2. The thing is that I’m not sure that Deleuze would deny that the sciences are representational. The whole point is not denying that there are things like individuality, resemblance, identity, etc. but rather finding their conditions of actuality in the realm of pure intensities. I’m less familiar with his later writings, but in Difference and Repetition, for example, the whole point is not to abandon the ordinary and empirical differences (differences that are subordinated to identity and sameness) but rather to find them again at the end of the journey, after going through difference in itself. To me it seems that the social sciences’ point of departure is simply these differences in themselves after they have been synthesized to ordinary objects and individuals, and therefore that the whole metaphysical debate is largely irrelevant to them.

    Liked by 1 person

    1. I agree. That is also my reading, as he does not renounce representation but thinks there is ontologically primary processes to that. However, he would claim (or the texts are allowing a reading in which) the experimental sciences (contemporary physics) do not represent empirics — they are experimental in the sense of acting on the world and forcing particular actualities (rather than describing or representing). There is a good quote by Stengers which I like about this. I will try finding it. For me, you are still reading this thing as about a debate between bodies of knowledge, while I am reading this as about practice – what do we write in texts, what is the status of these texts and how does that stand in relation to what we claim to be objects of description.


  3. Itamar, this could be a possible answer: check out this book which tries to do a ‘deleuzian social science’, or at least ask the question what Deleuze could mean for social science. This blurb from Strathern is key:

    “This remarkable work… creates a compelling radicalism from which to broach issues and problems that turn out to belong to no one discipline.”

    One answer, their answer, is that a nomad social science will need to ask questions that are outside disciplines. If the disciplines are that with which we think about the world, if the disciplines ‘cover’ the world in representation, then it needs to think outside of these actualisations. So now that’s the question, how to even engage that. I think it’s a fascinating question.


  4. I agree: the translation of a philosophical concept to a sociologycal tool is, at least, problematic. As Deleuze and Guattari say in What is philosophy?, a concept always respond to a singular problem, in this case agencement —and I agree with you— responds to the problem of being. I think that assemblage is different to agencement, in this sense Latour has been careful. To him, assemblage allows to trace the associations and relations of actors, a methodological function is linked to it. Deleuze goes beyond a methodological problem: his philosophy is not only a criticism to representation but also a criticism about subject and object, that is the reason why he introduces the notion of haecceity as opposite to subjectivity. Because of that, Latour find a lot of sense in agencement for defining assemblage.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s