Assemblage,materiality,race,capital

It is with pleasure that I respond to this potentially groundbreaking paper. Though some will decry the coming of yet another buzzword, the term assemblage, as the authors insist, is a refining and consolidation of theoretical trends in human geography since the later 1990s. More than other recent developments such as actor-network theory, non-representational theory and affect theory, this term can capture the breadth and critical impetus of the main ontologist of our time, Gilles Deleuze, who inspires all these conversations. Here I will offer my thoughts on some trajectories for improving the concept of assemblage which I believe are broadly in line with Deleuze’s intentions.

First I would like to suggest not relationality but materiality is the epistemological priority of the geography of assemblages. After all, the era of signs, discourse, statements and representations that assemblage theory seeks to surpass is/was already profoundly relational. For Saussure, famously, words mean absolutely nothing without their interrelatedness with other words. Identity derives only from difference. Likewise, geographers argued that places real and imaginary cannot be understood as given but irrevocably define each other through webs of reference and influence. But what reached its limit in the mid-1990s was precisely this insistence on relation, which had completely obscured the internal composition and force of the terms related. Anti-essentialism could not account for why it was this particular term (the word ‘ether’, the place Moscow) that could become interconnected. Identity also derives from what Anderson et al. (2012) refer to as a holding-together independent from the relations an entity enters into, which in fact partially explains its material agency (better sensed in the French word for assemblage agencement). The turn towards materiality did not do away with interconnectedness of course – to the contrary, connections were now seen to consist of far more than signification – but showed how it was both made possible and thwarted by the physical characteristics of the places and bodies being studied.

If I had to sum up my reflections it is to state assemblage theory has to be not simply realist, as its main proponent Manuel DeLanda avers, but materialist in the sense best known to geographers, i.e. Marxist. The book that is the cornerstone of assemblage theory, Deleuze and Guattari’s (1988) A Thousand Plateaus: Capitalism and Schizophrenia, is throughout indisputably clear that the materiality of present-day assemblages consist of flows of money, commodities, legislation, land ownership, nationalism, science and a global division of wage-labour, all classic Marxist themes. Anderson et al. note that Deleuze and Guattari were in close conversation with the autonomist Marxist tradition, especially Italian communist thinkers like Antonio Negri. This point can be pushed further and innovate geographical research. In particular, how is class assembled and held together? How is Bourdieu enriched through assemblage theory? How does the state as legal and military assemblage manage class, race and sexual domination? Assemblages require work in the sense of physics and political economy. Mapping the flows of wage-labour goes a long way to appreciating an assemblage’s contingencies and disparities.

It is a pity therefore that a certain anti-Marxism characterizes most of the work cited in Anderson et al.’s paper. DeLanda, Latour, Stengers, John Law, Jane Bennett, William Connolly, Graham Harman and in geography Nigel Thrift and Kevin Hetherington all have made their name partially by dispelling the critique of class and empire from the intellectual agenda, even when these were always implicit concerns for their philosophical predecessor Deleuze and more explicitly for his friends Guattari and Negri. Are the assemblages we see on the news – the housing market, the global food system, refugee camps, Norwegian national identity – only interesting when they collapse? Collapse, as Marxism has always held, is endemic. It is often noted the materiality of capital is the strangest of all which makes it essential to any materialism. Certainly capital ‘relates’ places and populations to each other, but its uncontrollable ephemerality far exceeds actual relationships and has to be thought of more like a vast self-propelling reserve of possible accumulations and crises. Over a few centuries almost all assemblages on the planet have become crisscrossed and gambled by capital, though, as many geographers are quick and correct to point out, capital never could fully determine any of them.

This brings me to a thorny question elided in the paper: the relative weight of assemblage components. The so-called flat ontology of assemblage theory does away with cherished concepts such as scale and society as well as the traditional verticalities between base and superstructure, human and non-human, engineering and art, matter and language, power and meaning. Yet such metaphysical egalitarianism begs the question how reality could function let alone be explained if everything has equal importance in it. When in a wooden hut termites become ‘as forceful’ as its human inhabitants, would it still hold together? Is language not more important in a courtroom than in a diving contest? The age-old debate between nominalism and realism pops up here, and I think a materialist realist has to concede that the absolute incommensurability of the contribution of components to the holding-together of an assemblage requires accepting qualified reduction as unavoidable. Pure flatness is an epistemological illusion, ironically borne from trying to escape anthropocentrism. For it can only be the human perspective that calls everything ontologically flat. Flatness leads to a relativism at odds not only with the professed realism but also our ethico-political commitments to intervention. Ultimately assemblages are mapped and altered by humans who have to decide whether the primary component is capital, language, gender, technology or termites, while understanding that once it is decided they are less relevant, the secondary components will inevitably raise their voices too, because that lies in the nature of the assemblage and academic discourse.

Without some reductionism it is then impossible to describe, map or diagram an assemblage. Likewise formalism is not the enemy of assemblage theory most of its advocates claim it to be. What from a Deleuzian point of view is wrong with the formalisms of Lévi-Strauss, Lacan and Badiou is that they focus on components and relationships privileged in hegemonic European thought (minds, binaries, signifiers) at the expense of their continually changing obscure and remote backgrounds. Deleuze wants to think the obscure and remote, which can be even be formalized albeit in unconventional ways, as the diagrams in A Thousand Plateaus and the tables and axioms of Guattari’s The Machinic Unconscious (2011) illustrate. Furthermore, a Deleuzian formal system has to continuously reinvent its objects and internal cohesion to remain immanent to the variation it captures. But some fidelity to what previous research and generations have conventionalized is necessary or there could be no formalization. The task of assemblage theory is to conjure a small number of minimal general principles which any assemblage whatsoever has to comply with. Assemblage aspects such as its mobility (a ship versus a factory), demographics, involvement of capital, expenditure of electricity and other resources, level of inequality and segregation, carbon footprint and other quantifiables are easily drawn into a loose formal system. But the affects and signs circulating within assemblages can also be ‘measured’ or formalized in their efficacy, for example through the precision of a poem or a soundscape recording. It was formalism that buttressed the revolutions of spatial science and GIS. Suppose now a renewed concept of how to approach the formalization of spatiotemporal assemblages would allow for a similar integration of research in left-leaning human geography?

The surprising inclusion of the problematic of race may not only reflect the research interests of one of the four authors (Swanton) but point to a deeper shift in thinking the social in an age marked by biotechnology, new questions of human origins and futures, and widespread anxieties in the more privileged states from Poland to South Africa and the United States about which bodies to assemble under the flag (Amin, 2010). Given the silence on race in DeLanda and Latour (but not Deleuze: see Saldanha and Adams, 2012), I much welcome this gesture but nonetheless would like to point out one omission and one worry. The omission is sexual difference: the eons-old, creative, intractable division of the sexes as well as the couplings, violence, symbioses and combinatory newness it gives rise to. The worry is political. When Anderson et al. observe race like any assemblage is fundamentally provisional by virtue of its heterogeneous composition, this is a first step, but staying with provisionality and fluidity is politically no different than what commercial multiculturalism promotes. The next step is to show how this is exactly what makes racial differences persist.

 If institutional racism is an indirect effect of neocolonial flows of capital, as we know since Foucault, there is also no biopolitics without sexuality. Tending to the materiality of sex takes us further afield. When Anderson et al. cite Deleuze stating an assemblage consists of ‘relations across ages, sexes, and reigns’, and its unity is one of symbiosis, his proximity in all matters biophilosophical to the work of Lynn Margulis is evident (Margulis and Sagan, 1990). In a different vein, Guattari too never lost sight of sexual difference, trained as he was in psychoanalysis, though he radicalized the uncertainties at the heart of it. Deleuzian feminists like Elizabeth Grosz (2009) have deepened conceptualizations of the interplay of sex, earth, time and race, opening space for collectivities to organize themselves along vectors other than the biopolitical, the territorial and the monetary.

On a general level, assemblage theory, though indebted to Darwin’s anti-essentialism, does not consider the millions and billions of years Nigel Clark (2011) reminds us go into constituting human worlds. Human phenotype is mostly a product of evolution as is the oil industry and the distribution of influenza. Perhaps assemblage theory is a more synchronic endeavour (a contribution to modern sociospatial theory, as the paper says, more than evolutionary metaphysics) and does not seek a deep sense of time. It does seem opportune to me ‘assemblage’ be constricted to systems and processes involving humans. This will require a more exacting differentiation between assemblage and cognate notions like ecosystem, bio- or nanotechnological artifice, weather system, Umwelt, niche, cellular system, astrobiological environment and simulated interaction space.

Assemblage theory should therefore not forsake the radical ruptures and frustrations that sexual difference and instincts bring to bear to any human assemblage. As with the elision of class, avoiding desire and violence in their many forms results in a somewhat sterile and safe mapping of local linkages, without disturbing the ideological preconceptions they are based on and the distant corporeal effects they produce. One could show how masculinist desires driving the Wall Street assemblage lead to food shortages in Africa and Asia. Anyone claiming to follow Deleuze and Guattari is to take care not to parrot public opinion’s awe for new technologies, markets and fashions (see Dewsbury, 2011). What makes the concept of assemblage compelling is precisely the unexpected, jarring, non-local elements it uncovers. The ‘so what?’ question theoretically ambitious geographers often get should by assemblage geography be straightforwardly answered ‘to disturb hegemonic desire! to better intervene!’. Materialists from Democritus to Freud have always understood they were inevitably part of a world they had the power to change. When Guattari first started talking about ‘machinic assemblages’ it was explicitly to improve on his psychiatric and activist practices. Rather than ending with questions ethical and political, perhaps we should start with them.