Introduction
Foregrounding material factors and reconfiguring our very understanding of matter are prerequisites for any plausible account of coexistence and its conditions in the twenty-first century.Footnote 1
Materiality is seemingly back: matter matters after all in the study of contemporary political life. This is the mantra of a range of scholarship variously referred to as ‘New Materialisms’, ‘New Vitalism’, and the ‘Materialist Turn’ produced by diverse social and political theorists in recent years.Footnote 2 What binds this diverse literature together is a common attempt to thematise the concept of materiality, its relationship with politics, and how an emphasis on material factors might lead to a refashioning of our understanding of the concept of ‘the political’. In this context, ‘materiality’ is typically cast in deliberately broad terms: ‘objects, bodies, machines’;Footnote 3 ‘edibles, commodities, storms, metals’;Footnote 4 ‘micro-organisms … cellular reactions … cosmic motions’.Footnote 5 By now the ‘New Materialisms’ literature comprises a rather heterogeneous and not always compatible set of theoretical positions – an issue to which the discussion will return. However, a general point of convergence is not only the fact that human beings are surrounded by, immersed in, and indeed composed of matter understood in these terms,Footnote 6 but also the claim that the relationship between people, materiality, and sociopolitical life is intensifying. Thus, for example, Diana Coole and Samantha Frost argue that developments in natural science have led to the increasing ‘saturation of our intimate and physical lives by digital, wireless, and virtual technologies’.Footnote 7 Similarly, Bruce Braun and Sarah Whatmore claim that it is ‘perhaps no longer possible to imagine either the human as a living being or the collectivities in which we live apart from the more-than-human company that is now so self-evidently integral to what it means to be human and from which collectivities are made’.Footnote 8 For this reason, Coole characterises New Materialisms not only as an ontological diagnosis of contemporary political life, but also as a ‘political-ethical intervention’ and a ‘reckoning of the material circuits, flows and experiences that mark the 21st century’.Footnote 9
In this article we begin to explore what some of the theoretical and methodological implications of the ‘New Materialisms’ turn in recent social and political thought might be for scholars of International Relations (IR). To do this we situate the discussion more specifically in debates around the status of ‘discourse’ and ‘discourse analysis’ as it is often in this context that the question of materiality – particularly as it relates to language – is most actively and explicitly engaged in the discipline.Footnote 10 As such, although our investigation has implications for a range of empirical contexts central to the self-image of IR, the discussion proceeds primarily along conceptual lines.Footnote 11 The main questions that animate our enquiry are as follows: What do the insights of the ‘New Materialisms’ turn mean for theorists of discourse in IR? Does the New Materialisms literature offer a satisfactory response to the limits of prior conceptions of discourse or is there a need to find other critical resources apposite to this task?
In order to address these questions we begin by examining how discourse analysis has been conceptualised and operationalised in IR via a reading of several prominent works by Jennifer Milliken, Richard Jackson, Lene Hansen, and David Campbell.Footnote 12 This is an important task, because in order to understand the challenges posed by the New Materialisms turn it is first necessary to clarify what precisely this turn can be seen as a response to. In the first instance, we identify in Milliken and Jackson a tendency to conceptualise ‘discourse’ in the narrower sense of words: in this mode ‘textuality’ is understood as exclusively linguistic. Prima facie Hansen and Campbell broaden and deepen this conceptualisation of discourse to include nonlinguistic elements such as images and other visual and affective phenomena as part of an extended ‘intertextual’ milieu. However, we suggest that what these authors ultimately have in common is a shared emphasis on meaning-making practices and the politics of representation. In the final analysis they end-up relegating what New Materialists refer to as the ‘material realm’ of social life to the status of an inert and apolitical backdrop, which can only acquire political significance via linguistic or visual representation. This insight is instructive precisely because it helps to explain why the discipline of IR – with its enthusiastic embrace in recent years of both ‘textual’ and ‘intertextual’ approaches to discourse analysis – has become potentially vulnerable to New Materialist critique.Footnote 13
Working in many ways against the limitations of a concern with the politics of representation, the New Materialisms turn encourages a more direct engagement with the political force of materiality. The turn provides intellectual resources for investigating the material realm independently of the means by which language and nonlinguistic signs such as images come to construct the ‘meaning’ of this realm. For this reason, the New Materialisms literature can also be seen as a counterpoint to more traditional engagements with materiality in IR, which have historically been delimited by the boundaries of positivist social science and Realist assumptions of an independent material realm that can be ‘accessed’, ‘known’, and ‘represented’.Footnote 14 Therefore, in order to analyse some of the key implications of the New Materialisms turn and how it can be said to challenge both narrower conceptions of discourse and more traditional approaches to materiality, the second section examines in closer detail landmark contributions associated with this turn. We note that while this literature acts as a critical alternative to both the linguistic bias in some discourse analysis literature and the view that materiality is the preserve of Realist scholarship in IR, there are also various limits implicit in this move. For instance, in some incarnations it may risk reversing the emphasis so that materiality becomes central and questions of language, meaning, and the politics of representation fall out of the equation. As we will show, this tendency is also evident in extant attempts to incorporate the insights of the New Materialisms turn into the study of IR.
Another potential problem with the ‘New’ Materialisms turn is that in claiming ‘novelty’ some work in this vein arguably overlooks already existent resources for theorising the complex interplay between ‘language’ and ‘matter’ to be found in earlier poststructural thought.Footnote 15 In the third and final section we therefore turn to key poststructural texts and thinkers, and argue that there are critical resources in this genre to think beyond the dichotomy between language on the one hand, and materiality on the other. Drawing on the diverse works of Michel Foucault and Jacques Derrida, we seek to demonstrate how, though heterogeneous, what we might refer to as a broadly poststructural perspective can act as a corrective to the potential excess associated with both ‘sides’ of the above debate.
The explicit targets of our ‘poststructural rejoinder’ are twofold. First, we seek to challenge the idea that poststructuralism is somehow locked in a linguistic realm and ‘can't do/won't do’ materiality. This is a claim made not only by some advocates of New Materialisms, but also by several ‘Critical Realists’. For example, Coole and Frost argue that poststructuralism is ‘inadequate for thinking about matter, materiality, and politics in ways that do justice to the contemporary context’.Footnote 16 Similarly, Colin Wight accuses poststructuralism, particularly the thought of Derrida, of treating ‘all social life … as a text’ and having an ‘underlying ontology’ that is ‘flat, one-dimensional, and reductionist’.Footnote 17 By contrast, we aim to show that language and materiality are inextricably inseparable in key poststructural thinkers’ oeuvres. Second, despite the centrality of the inseparability of language and materiality in the works under consideration, the implications for discourse analysis have gone largely unnoticed in the theoretical and methodological literature produced by IR. Some IR scholars who describe themselves as poststructuralist in orientation have drawn upon Ernesto Laclau and Chantal Mouffe's famous formulation that while there is a ‘material realm’ it is only through meaning-making practices and the politics of representation that this realm comes into being and has any political relevance.Footnote 18 In our view, however, this is only one possible rendering of what a ‘poststructural’ approach to discourse analysis might involve and one based primarily on Laclau and Mouffe rather than a wider pool of thinkers commonly associated with this label. Indeed, by drawing more extensively on the primary works of Foucault and Derrida, we advance an alternative reading of where ‘poststructuralism’ leads: one that emphasises how these thinkers negotiate, problematise, and ultimately deconstruct the language/materiality binary while maintaining both as part of a complex and radical intertextuality. We therefore seek to demonstrate how our reading of poststructuralism offers IR scholars a broadened and deepened understanding of what discourse is and how discourse analysis might operate beyond the limits of the emerging debate.
Discourse analysis in IR
In order to provide the historiographical context of how ‘discourse’ has been treated in IR – and therefore why the discipline has arguably been so susceptible to the New Materialisms turn – it is helpful to go back to where the ‘Third Great Debate’ in IR appears to have ended. At stake in this debate, according to Yosef Lapid, was the crucial question of whether the hegemony of positivist methodology and epistemology could continue to hold its grip over the discipline, or whether the emergence of another set of methodologies in the social sciences, which strongly rejected the premises and core assumptions of positivism, could and should be adopted.Footnote 19 In his Presidential address to the ISA Convention in 1988, Robert O. Keohane invented the term ‘Reflectivism’ as a short hand for the variety of perspectives that sought to go beyond positivism. One of Keohane's challenges to this heterogeneous body of work was to develop causal hypotheses that might be tested rigorously in order to establish a coherent research agenda, thus reaching the same standards of social science research as neo-Realism and neo-Liberalism that had emerged during the 1980s. Keohane's distinction can be considered paradigmatic because it has since been taken up and reproduced in many discussions of the trajectory of theorising in the discipline.Footnote 20
By now, many authors Keohane labelled as ‘Reflectivists’ have introduced and formalised notions of ‘discourse’ in order to operationalise a research methodology commonly referred to as ‘discourse analysis’. Indeed, either directly or indirectly in response to Keohane's challenge, discourse analysis has in many ways become the standard bearer of Reflectivist research in IR. In an influential guide to the use of discourse analysis in IR, Jennifer Milliken offers a robust defence of discourse-based approaches from those who portray it variously as ‘bad science’,Footnote 21 ‘prolix and self-indulgent’,Footnote 22 and ‘deviant and marginal’.Footnote 23 One of the moves Milliken makes is to emphasise that discourse analysis does not constitute a single approach. Indeed, she argues that there is very little agreement about what discourse is or how ‘it’ should be studied. Rather, the discussion goes on to outline an array of perspectives including predicate analysis, metaphorical analysis, deconstruction, and genealogy. As an introduction to these perspectives, the article is widely acknowledged as a key reference point in sketching out different ways in which discourse can be understood and analysed.Footnote 24 In addition to this exegesis Milliken also advances a particular view of what discourse is and what discourse analysis should imply.
According to Milliken, discourse refers to ‘structures of signification which construct social realities’ so that ‘things do not mean (the material world does not convey meaning); rather, people construct the meaning of things, using sign systems (predominantly, but not exclusively linguistic)’.Footnote 25 Second, Milliken argues that discourse is itself a productive system that produces subjects and their authority to speak and act, shapes different kinds of knowledge practices, and enables/disables multiple ways of thinking and doing politics.Footnote 26 Third, discourse analysis is characterised as efforts ‘made to stabilize and fix dominant meanings’ through the subjugation or exclusion of other forms of knowledge.Footnote 27 On this basis, Milliken argues: ‘A discourse analysis should be based upon a set of texts by different people presumed (according to the research focus) to be authorized speakers/writers of a dominant discourse or to think and act within alternative discourses.’Footnote 28
Importantly, ‘discourse’ is here taken to be synonymous with ‘text’, which, in turn, is understood in the narrow sense of ‘words’, ‘language’, and ‘written/spoken claims’. Despite her disclaimer that sign systems are ‘predominantly, but not exclusively linguistic’, all the examples offered by Milliken to illustrate various discourse approaches are based upon ‘documents’ in the written form: speeches, policy papers, and so on. On the one hand, Milliken's disclaimer clearly indicates her awareness of the problem of excluding ‘material’ phenomena from discourse analysis and it may well be that her choice of examples is not deliberate in perpetuating that exclusion. On the other hand, however, the tension between principle and practice here does little to further explain and illustrate how discourse analyses can (and should) include material phenomena as objects of study in political analysis. Indeed, whether intended or not, Milliken privileges an interpretation of discourse as language, thus giving the impression that discourse analysts are disinterested in and/or unable to grasp materiality.
Another prominent example of discourse as language can be found in Richard Jackson's Writing the War on Terrorism: Language, Politics and Counter-Terrorism (2005), which claims to employ a ‘critical discourse analysis’ approach to the language of the war unleashed by the US in the aftermath of the events of 11 September 2001. Jackson argues that what marks out his approach as ‘critical’ is his insistence that ‘the language of the war on terrorism is not simply an objective or neutral reflection of reality’.Footnote 29 Rather, he emphasises that it is a ‘carefully constructed discourse’, which creates the world it refers to thereby harnessing one of the key insights of the linguistic turn that language is constitutive of reality. On this view, the discourse of the war on terrorism is said to be ‘a deliberately and meticulously composed set of words, assumptions, metaphors, grammatical forms, myths and forms of knowledge’.Footnote 30 Discourse, for Jackson, is understood as ‘a particular way of talking about and understanding the world that involves a limited number of statements and words’.Footnote 31 Echoing Milliken, discourse is taken to be synonymous with text and the realm of the textual refers to ‘any act of written or spoken speech, from speeches to interviews to postings on websites to emails between officials’.Footnote 32 The focus of Jackson's book, therefore, is also on discourse understood as language. Words are taken to be ‘vestibules of meaning’, which in turn hold the key to understanding how US foreign policy influences global security relations.Footnote 33
We find overlap between Milliken's and Jackson's understanding of discourse as language and the seminal account offered by David Campbell. In Writing Security: United States Foreign Policy and the Politics of Identity (1992) and National Deconstruction: Violence, Identity and Justice in Bosnia (1998) Campbell offers an examination of how the identity of subjects and the meaning of threats, dangers, and enemies are constituted through practices of representation and interpretation. Drawing on Laclau and Mouffe's concept of discourse, as well as Foucault's notion of the discursive formation of objects, Campbell says that he does not deny the existence of an external material reality. However, he argues that such a reality has no meaning in and of itself, which, in turn, entails that it cannot constitute a political force independently of the ways in which we speak and try to make sense of ‘it’: ‘the world exists independently of language, … but we can never know that (beyond the fact of its assertion), because the existence of the world is literally inconceivable outside of language and our traditions of interpretation’.Footnote 34 As Campbell puts it in National Deconstruction: ‘There's no way of bringing into being and comprehending non-linguistic phenomena except through discursive practices.’Footnote 35 By analysing practices of representation, he does not seek an objective way of representing or interpreting an external reality as found in some Realist IR scholarship. Rather, Campbell is concerned with exploring how different practices of representation and interpretation constitute aspects of social life that they otherwise merely purport to describe. The significance of language therefore lies in the performative work it does in bringing into being an otherwise inert and meaningless material background.
A similar approach is detectable in Lene Hansen's Security as Practice: Discourse Analysis and the Bosnian War (2006), which aims to explore ‘poststructuralist discourse analysis’ and its application to the study of Western debates about the Bosnian conflict. Hansen claims to build her theoretical approach on the work of poststructural thinkers including Foucault, Derrida, Kristeva, and Laclau and Mouffe. She invokes the concept of ‘intertextuality’, attributed to Kristeva, in order to refer to the way that ‘the meaning of a text is … never fully given by the text itself but is always a product of other readings and interpretations’.Footnote 36 However, as Claudia Aradau and Jef Huysmans have argued, Hansen's approach to ‘intertextuality’ appears to diverge somewhat from that set out by Kristeva and other thinkers commonly associated with the term ‘poststructural’. Whereas the latter typically treat intertextuality to refer to the uncertainty of identity and meaning arising from the radical relationality between subjects and objects (as we will go on to explore), Hansen arguably uses it in her analysis ‘as a device fixing identities’.Footnote 37 Furthermore, we would add that for Hansen the ‘text’ of ‘intertextuality’ also remains rather narrowly conceived of in terms of language rather than inclusive of a wider set of phenomena: ‘an inter-textual understanding of foreign policy argues that texts build their arguments and authority through references to other texts: by making direct quotes or by adopting key concepts and catchphrases’.Footnote 38 References in this context to ‘direct quotes’, ‘key concepts’, and ‘catchphrases’ are all indicative of a language-based understanding of ‘discourse’ throughout Hansen's book, which, as we will later argue, parts company from more radical notions of intertextuality. Such a view follows from Hansen's assertion that ‘discourse analysis has … a discursive epistemology, and its methodology is, as a consequence, located at the level of explicit articulations’.Footnote 39 Therefore, despite Hansen's ostensible commitment to intertextuality as understood in the context of poststructuralist scholarship, it is in fact a much narrower version of discourse that arguably characterises Security as Practice. Indeed, Hansen follows Milliken, Jackson, and Campbell in establishing and maintaining a prior distinction between materiality on the one hand and the role of language and representation on the other: ‘It is only through the construction in language that “things” – objects, subjects, states, living beings, and material structures – are given meaning and endowed with a particular identity.’Footnote 40
It might be argued that a potentially expanded treatment of the concept of discourse may be found in some of Hansen's and Campbell's more recent works. Both authors have made various attempts to conceptualise discourse so that it is inclusive of nonlinguistic dimensions of contemporary political life: for example, the effects of photographs in reproducing the violence of disasters, famine, and war in the case of Campbell;Footnote 41 and the otherwise overlooked role that visual phenomena play in acts of securitisation – as exemplified by the 2005 Danish cartoon crisis – in the case of Hansen.Footnote 42 However, while the move to consider visual phenomena may indeed extend discourse analysis beyond an exclusive focus on language, Campbell's and Hansen's treatments of images nevertheless continue to work within the framework of the politics of representation. In other words, despite adding visual phenomena to the study of discourse they recycle rather than displace the distinction between meaning-making practices on the one hand and an external material world ‘in need’ of representation on the other. Thus, despite the move to the visual, they continue to downplay the possibility that the discourse analyst may be interested in the role materiality might play politically beyond meaning-making practices. In other words, the place of materiality is still secondary to the politics of representation through which it acquires political significance.
The works of Milliken, Jackson, Campbell, and Hansen have yielded many significant insights into, for example, the role of identity construction, the importance of ideational factors, and the social manufacture of ‘danger’, ‘threat’, and ‘fear’, and so on. However, the conceptualisation of discourse with which they operate ultimately reflects a rather limited set of assumptions about what discourse ‘is’, how it ‘should’ be studied, and what makes discourse analysis a distinctive methodological approach. Common to the authors above is an assumption of the prior separation between ‘language’ on the one hand and ‘materiality’ on the other, which then leads ultimately to a privileging of the former over the latter. This allows, consciously or otherwise, for materiality to effectively drop out of their ‘discourse analyses’, which proceed by way of a much more circumscribed focus on language as distinct from materiality.
Our overarching point here is not that these prominent authors deploying discourse analysis in this mould are somehow ‘wrong’ and nor do we deny the importance and value of their work in the field. On the contrary, this strand of research has stimulated a productive subfield of knowledge within the discipline and has proven exceptionally popular among recent generations of IR students. However, the role of material objects in this framework plays second fiddle to the allegedly more important, active, and properly ‘political’ practices of linguistic utterances, articulations through which dull, inert, and otherwise ‘apolitical’ matter – gas pipelines, bridges, the fabric of cities, and so on – acquires any sort of relevance to the IR scholar. Such an approach is arguably derivative of a deeper anthropocentric understanding of the concept of the ‘political’ as something that only pertains to linguistic and thus human relations. It is primarily for these reasons, we argue, that the New Materialisms literature has recently found considerable traction in IR as an antidote to the perceived excesses of a focus on discourse as language and the twinned emphasis on the politics of representation and meaning-making practices.Footnote 43
The New Materialisms turn
A central point around which the New Materialisms literature converges is the argument that political analysis broadly conceived has traditionally failed to take sufficient account of materiality. While natural scientists, engineers, and geographers, for example, habitually work with and are sensitive to matter and nonhuman forces, political theory, according to Braun and Whatmore, has tended to ‘purify’ its analysis of human society from the nonhuman/material world.Footnote 44 For Coole and Frost this is partly an effect of the history of philosophy, which has typically focused on issues such as ‘language, consciousness, subjectivity, agency, mind, and soul’, presented these as distinct from matter in the first place, and hence marginalised ‘stuff’.Footnote 45 Much of this literature argues that the focus on meaning and practices of representation as a yardstick according to which we can judge the political salience of phenomena is a highly problematic starting point. Such a focus already works within and reproduces an assumed separation between the ‘human’ on the one hand and the ‘nonhuman’ on the other where the realm of the political is confined exclusively to the former. Instead of reducing the study of the political to meaning-making practices and the politics of representation, the New Materialisms turn prompts a reconsideration of matter and its political status: not as something intrinsically inert, nothing ‘in and of itself’, and without any function or implication; and neither as something whose political significance can only come about through linguistic forms of identity construction; but rather as an active, affective, and politically significant set of forces in its own right. Such a view presents a potentially radical challenge not only to political theory, but also to IR in general and the discourse analysis literature produced by that field in particular: namely, that matter has significance beyond the politics of representation.
Throughout the New Materialisms turn is the notion that the ‘stuff of politics’, the plethora of objects, materials, and forces around us ‘help constitute the common worlds that we share and the dense fabric of relations with others in and through which we live’.Footnote 46 In other words, ‘things’ condition the possibility of human interactions, shape political communities, and influence behaviours and outcomes – indeed, matter cannot be divorced from the ‘we’ it in part constitutes. Moreover, as Coole puts it, there are presently ‘material changes and processes underway’ – for example the effects of human behaviour on the eco-system in the context of the anthropocene – to which the New Materialisms literature seeks to give ‘renewed attention’.Footnote 47 On this view – and contra the perspectives examined in the previous section – ‘matter’ is not merely a static backdrop simply waiting to acquire meaning via human practices of representation. It is not understood as ‘the dead, inert, passive matter of the mechanist’, but rather as a ‘materialisation that contains its own energies and forces of transformation’.Footnote 48 Central here is an attempt to reconceptualise agency in contemporary political life such that, reflecting Bruno Latour's ‘Actor-Network-Theory’ (ANT) approach, the human no longer comes to occupy the ontological frame: ‘new materialisms recognizes agency as being distributed across a far greater range of entities and processes than had formerly been imagined’.Footnote 49 Coole thus calls for a ‘decoupling’ of anthropocentrism and agency in political analysis in order to better capture ‘agentic capacities’ – understood in her terms as the ability to effect some sort of change – within any given ‘field of forces’.Footnote 50 One recent attempt at theorising the agentic capacity of inanimate objects – inspired largely by Latour and the vitalist materialism of Gilles Deleuze – is offered by Jane Bennett.
In Vibrant Matter (2010), Bennett argues that traditional approaches to politics rely on a problematic, anthropocentric, and therefore highly political distinction between the supposedly dull life of things on the one hand versus the vibrant life of humans on the other hand. Such an approach is problematic, according to Bennett, because it ignores the ‘vitality of matter and the lively powers of material formations’, where ‘vitality’ refers to ‘the capacity of things – edibles, commodities, storms, metals – not only to impede or block the will and designs of humans, but also to act as quasi agents of forces with trajectories, propensities, or tendencies of their own’.Footnote 51 Politics is recast as a ‘political ecology’ on Bennett's view, thus drawing attention to the active role of nonhuman materials in producing affects in public life – a power of things in themselves or what she calls ‘thing-power’.Footnote 52 This concept is understood as ‘the curious ability of inanimate things to animate, to act, to produce effects dramatic and subtle’.Footnote 53 One example she gives of an inanimate-animate assemblage is that of an electrical power grid: ‘a volatile mix of coal, sweat, electromagnetic fields, computer programs, electron streams, profit motives, heat, lifestyles, nuclear fuel, plastic, fantasies of mastery, static, legislation, water, economic theory, wire, and wood’.Footnote 54 From Bennett's perspective, ‘this is not a world, in the first instance, of subjects and objects, but of various materialities constantly engaged in a network of relations’.Footnote 55
Bennett's move is to contest the common assumption – and one that undergirds the approach to discourse analysis as the politics of representation – that ‘things are always already humanized objects’.Footnote 56 On the contrary, inanimate things, she argues, possess a power in themselves – a power to produce different kinds of affects. Materiality, contra Marx, is therefore not immediately linked to a social and economic context on this view. Nor is the value (or ‘meaning’) of the material realm fully determinable by its connection to human bodies. The inanimate, according to Bennett, should not automatically be reduced to the animate in this way. Rather, it is important to maintain a distinction between the two in order to explore ‘the world of nonhuman vitality’ and allow ‘nonhumanity to appear on the ethical radar screen’.Footnote 57 This form of ‘naive realism’, ‘onto-story’, or ‘ecology of matter’, as Bennett calls it, thus emphasises the importance of exploring the potentiality and vitality of the nonhuman, but also of articulating ‘ways in which human being and thinghood overlap’.Footnote 58 In so doing, Bennett challenges the widespread tendency ‘to conclude the biography of an object by showing how it, like everything, is socially constituted’.Footnote 59 Countering this tendency by highlighting the force of materiality, Bennett provocatively challenges and puts forward an alternative to the narrow understanding of the material realm that arguably pervades IR. Moreover, by pointing to what she considers to be the irreducible nature of materiality, Bennett moves beyond the representational model of language: things are ‘not entirely reducible to the contexts in which (human) subjects set them, never exhausted by their semiotics’.Footnote 60
However, it is precisely around the question of ‘agentic capacities’ and the limits of this concept that key differences have also begun to emerge within the New Materialisms literature. Coole, for example, seeks to differentiate her approach somewhat from that of Latour and Bennett. While the latter ascribe agentic capacity to inanimate objects – or nonhuman ‘actants’ (such as the power grid in Bennett's case) – the former is sceptical of this move because it ‘attributes agency to inorganic matter that is indifferent to the impact of its efficacy’.Footnote 61 By contrast, Coole wishes to retain ‘reflexivity’ as a criterion for ascribing agentic capacity. This is important in the context of an allied position that Coole takes, which is to argue that a third criterion for attributing agentic capacity should be a sense of responsibility: not in the ‘sense of moral agency’, but rather, for example, in terms of being able to attribute the role that ‘(some) humans’ have played in ‘imperilling the entire eco-system’.Footnote 62 What Coole ultimately argues for is a ‘capacious historical materialism’; one that expands the ontological remit of political analysis to include material circuits, flows of matter, and nonhuman assemblages without losing sight of the impacts of human agency.
Some of the implications of a recovery of the political force of materiality for IR have already been explored empirically. Martin Coward's work, for example, has offered a sustained critique of what he considers to be the dominant anthropocentrism of IR and illustrates how a renewed focus on material and nonhuman factors opens up alternative avenues of enquiry. In Urbicide: The Politics of Urban Destruction (2009), Coward investigates the destruction of built environments in warfare. Drawing on examples from the Bosnian war, the Russian invasion of Chechnya, and the ongoing Israel-Palestine conflict, Coward shows how the deliberate destruction of built environments is increasingly an end itself in contemporary warfare and one which takes the materiality of the physical environment – rather than individuals or groups of people – as its ultimate target. Buildings play an active role in constituting who we think we are, where we think we are, and how we think about our existence in the world. The materiality of the urban environment is not then simply an instrumental backdrop, but something that establishes community understood as ‘the experience of fundamental heterogeneity that is an existential condition of all being’ and therefore worthy of violent military targeting in its own right.Footnote 63
More recently Coward has developed his treatment of contemporary urban politics via an explicit engagement with Bennett's work on the vibrancy of matter.Footnote 64 He draws on the idea of the city as a complex ecology to discuss how global urbanisation poses a direct challenge to the way contemporary citizenship is usually imagined. Instead of viewing the citizen as an autonomous, atomised individual – as Coward claims is usually the case in political theory – subjects are rethought as part of ‘an assemblage composed of human and non-human materials’.Footnote 65 Thus, he argues, ‘the shopper is … part of a wider assemblage that includes the trucks/planes that transport goods, the electricity station that lights the supermarket, and the personal computer through which consumption is shaped (and perhaps performed via online ordering)’.Footnote 66 What emerges from this vitalist-materialist perspective is a decentring of the citizen-subject such that life in the city is connected by and forms part of a complex infrastructural assemblage. This alternative diagnosis has manifold implications for the way in which we think about citizenship, ethical obligations to human and nonhuman others, and the politics of who and/or what is included/excluded from vital infrastructures that knit political communities together.
Similarly, Nadine Voelkner has applied a vitalist-materialist approach drawing on Bennett's work in order to analyse human security and what she calls the ‘migrant health assemblage’ in Thailand.Footnote 67 While material dimensions such as ‘small arms, carbon dioxide emissions, viruses, computers, and airplanes’ all form an integral part of the global human security discourse, Voelkner argues that ‘they tend to appear only as raw, brute, or inert objects whose existence and circulation either benefits or risks (global) human security’.Footnote 68 By contrast, her approach attempts to show how human-nonhuman alliances have agential capacities to shape human security strategies, produce different forms of political subjectivity, and provide the context for different scientific knowledge and intervention. Thus, for Voelkner, the ‘migrant health assemblage’ is constituted by diverse elements:
travelling pathogens, mosquitoes, crowded an unhygienic spaces, weakened and neglected refugee bodies, human intentions and desires, Thai refugee and migrant policy, ‘states of non-belonging’, fear and anxiety, failed government initiatives, sex work and body fluids, the problem of circulation, foreign aid capital, transnational agencies, inadequate health funds and the rise of global human security.Footnote 69
Voelkner shows how these elements frame, authorise, and shape knowledge in state responses to Burmese migrants, whose health becomes ‘manageable’ as a result.
In drawing attention to the material context in which certain security practices become possible, both Coward and Voelkner make an implicit critique of extant approaches that remain blinkered by a linguistic bias. This critique is made more explicit by Claudia Aradau's treatment of critical infrastructure protection in which she argues that ‘securitization has been seen as largely part of the linguistic and social constructivist turn in IR’.Footnote 70 For Aradau this approach has ‘largely ignored the role of “things” in the articulation of insecurities’.Footnote 71 Drawing on Karen Barad's materialist-feminist perspective she calls for a reappraisal of securitization as ‘a process of materialization that enacts a reconfiguration of the world in ways in which differences come to matter’.Footnote 72 The urban infrastructures to which Coward refers are increasingly ‘securitized’ in contemporary cityscapes and for Aradau the labelling of infrastructure as ‘critical’ is after all a securitizing move. But in a more vitalist-materialist sense she demonstrates how critical infrastructures designated as such also perform politically significant roles by, for example, separating ‘good from bad circulations and the associated forms of life’.Footnote 73 In this way, critical infrastructures themselves acquire generative and agentic capacities to create and sustain borders between people, places, and things.Footnote 74
As we have already noted, Coole, Bennett, and others associated with the New Materialisms turn have stimulated much-needed critical reflection on the limited treatment of materiality in Politics and IR. The work examined above simultaneously reveals the vulnerability of extant approaches to discourse analysis in IR and offers a corrective by opening up new avenues of enquiry into the performativity of ‘things’ that would otherwise remain closed off. Another achievement of the New Materialisms literature is that it has begun to challenge the notion that the Realist tradition has a monopoly on the ability to incorporate material factors into the study of IR.Footnote 75
However, despite the promises of the New Materialisms turn, there are reasons to pause and consider some of the entailments and possible limitations of this literature. Any move to simply overturn the hierarchy between language and materiality runs the risk of reifying – rather than stepping outside – the limits of the existing debate. For example, critics may argue that Bennett's analysis of the power-grid – despite its fleeting reference to ‘fantasies of mastery’ – tends to downplay the role of ideational factors such as ideas of profit and loss, speech acts that securitize energy as an existential threat, and articulations of blame, and so on. Similarly, it might be argued that the broader linguistic context in which the meaning of ‘urbicide’ takes form – including the usage of concepts of ‘ethnic cleansing’, ‘nationalism’, and ‘war’ – takes a back seat in Coward's analysis of violence in urban contexts. Indeed, the status and role of language in human/nonhuman assemblages are somewhat hard to discern in Bennett's, Coward's, and Voelkner's respective analyses. For Coward this is because ‘assemblages comprise a distinctive ontological entity’Footnote 76 and yet by reprioritising the constitutive role of materiality there is a danger of reproducing the distinction between materiality and discourse understood narrowly in terms of language. This issue is clearly a concern for Aradau who, while arguing that matter matters in the securitisation of critical infrastructure, notably takes great care to articulate a nuanced position in her analysis.Footnote 77 However, Aradau appears to approach ‘material’ and ‘discursive’ practices as if they were separate to begin with in order that they might then be brought together in her Barad-inspired analysis. We now seek to push Aradau's position further by developing a reading of discourse that already encompasses – and refuses to draw any operating distinction between – language and matter as part of what we call a radical intertextual approach.
Discourse as radical intertextuality
For some New Materialists, poststructuralism is seen as something of a linguistic dead-end that is ill-suited to the task of grappling with the material conditions of contemporary political life. Indeed, prominent works contributing to the turn claim to be animated in contradistinction to certain strands of poststructural thought. For instance, Coole and Frost argue that in recent times the marginalisation of matter in political theory has been exacerbated by the ‘exhaustion’ of existential phenomenology and Marxism and the subsequent critique of these approaches by poststructuralism.Footnote 78 On their view, ‘the constructivist orientation to social analysis is inadequate for thinking about matter, materiality, and politics in ways that do justice to the contemporary context’, and a ‘New Materialisms’ approach is therefore required as a corrective.Footnote 79 In a subsequent intervention, Coole tempers this position by characterising the New Materialisms turn ‘neither as a complete revolution back to older forms of materialism nor a complete rejection of the more constructivist approaches associated with poststructuralism’.Footnote 80 However, in the same article she also develops a critique of ‘deconstructionist approaches’ for their alleged ‘tendency to become imprisoned within self-referential circles of language or culture that are unable to give matter its due’.Footnote 81 Coole is by no means the only commentator to mount this critique of poststructuralism. In his critical commentary on ‘poststructuralist discourse theory’ (PDT), for example, Benjamin R. Banta makes a similar argument from a critical realist viewpoint: ‘While PDT does not deny the reality of a world outside of discourse, there is an important block thrown up against integrating it meaningfully into analysis.’Footnote 82
To some extent we find ourselves in agreement with the general thrust of Coole's and Banta's critique inasmuch as it may be said to apply to certain secondary works in IR. Indeed, their characterisation strikes us as a fair and important challenge to much of the literature on discourse analysis surveyed in the first section of this article. However, this characterisation is arguably less fitting of those writers who first imported to IR the insights of the broader ‘linguistic turn’ in social and political thought of the 1970s and 1980s. Landmark texts in this respect include: William E. Connolly's The Terms of Political Discourse (1974); Michael J. Shapiro's Language and Political Understanding (1981); James Der Derian's On Diplomacy (1987); and Shapiro and Der Derian's coedited volume International/Intertextual Relations (1989). Although ‘poststructuralism’ is often seen as one of the main representatives of – if not the representative of – language-centred approaches in IR, these authors espoused a nuanced stance, which sought to combine language and materiality as part of a radical intertextuality. For example, in his introductory chapter to International/Intertextual Relations, Shapiro argues that an understanding of ‘textuality’ should not be reduced to ‘specific instances of linguistic expression’.Footnote 83 Rather, as per Frederic Barthes work on ‘material texts’ such as toys, photographs, and forms of advertising, Shapiro refers to ‘text’ in a much broader sense: a ‘figuration related to distribution, exchange, and control’.Footnote 84 For Shapiro, it is precisely attention to the context in which this figuration takes place – a context that reads language and materiality as fundamentally interrelated and inseparable – that a ‘poststructural “textualisation” of phenomena involves’.Footnote 85 Similarly, in The Terms of Political Discourse, Connolly's exegesis of writers associated with ‘poststructuralism’ at no point invokes or presumes the language/materiality distinction. Connolly portrays deconstruction not as a linguistic ‘method’ of reading, but as a close analysis of the sort of ‘figuration’ to which Shapiro refers: ‘Deconstructionists show how every social construction of the self, truth, reason, or morality …, is actually composed of an arbitrary constellation of elements held together by powers and metaphors which are not inherently rational.’Footnote 86 Equally, Connolly refers to genealogy not as a form of ‘discourse analysis’ that merely traces the historical usages of different words, but as ‘a mode designed to expose the motives, institutional pressures, and human anxieties which coalesce to give these unities the appearance of rationality or necessity’.Footnote 87
In these earlier works the concept of radical ‘intertextuality’ enabled a folding of the language/materiality dichotomy and an expansive notion of discourse as encompassing the context in which the two are fundamentally inseparable. However, as we have already seen, subsequent discourse analyses in IR have – intentionally or not – narrowed their understandings of ‘discourse’ and left the discipline vulnerable to the critique of New Materialists. Inspired by this ‘first wave’ of poststructural scholarship in IR, we now (re)turn to two of the chief philosophical backstops underpinning this work – namely Foucault and Derrida – as the basis of a response to both the discourse analysis and New Materialisms literature. Instead of proceeding by way of a general exegesis of these thinkers – which would be well beyond the remit of this article – we focus on concepts central to their oeuvres and the themes of our preceding discussion: the visible and the articulable in Foucault; and force and the generalised text in Derrida. In each case we show how the thinkers under consideration radically challenge and seek to move beyond the limits of the language/materiality divide thereby opening up an alternative perspective to the potential excesses of the two positions discussed in the previous sections of the article.
The visible and the articulable
Foucault is sometimes associated with a narrow conception of discourse – one mainly concerned with language, meaning-making practices, and the politics of representation. This dimension of his work is perhaps most evident in The Archaeology of Knowledge (1969), where Foucault dispenses with ‘things’ and focuses instead on the linguistic conditions that enable certain objects to emerge, acquire meaning, and gain significance.Footnote 88 However, an over-emphasis on this aspect underplays the significance that material factors play in an inseparable relationship with language in his oeuvre.Footnote 89 Already in Madness and Civilization (1964), for example, Foucault examines how the discourse on ‘madness’ emerges from a complex field of human and nonhuman, linguistic and nonlinguistic elements that consist of social institutions, as well as art and visual representations, scientific statements, and so on.Footnote 90 Here we find that ‘discourse’ does not refer so much to the ways in which language assigns meaning to the ‘object’ of ‘madness’, as in the framework of the politics of representation. Rather, discourse has to be understood in expanded terms to include the ‘complex group of relations’ within which the ‘truth’ of ‘madness’ is produced and eventually comes to be accepted as such.Footnote 91 This radical attentiveness to the equal and inseparable roles of language and materiality in various human and nonhuman assemblages is even more pronounced in his later investigations into relations of ‘disciplinary power’ and its relationship with the body. One example of how Foucault explores the radical interplay of language and materiality is his examination of practices that render things visible as well as articulable.
Emerging in the second half of the eighteenth century, disciplinary power deployed punishment mainly as a technique for the coercion of bodies, with the aim of making those bodies into useful parts of society. Instead of being the king's property, the body of the ‘condemned man’ became the ‘property of society, the object of a collective and useful appropriation’.Footnote 92 The forms of punishment involved in this process included: the correction of behaviour; training the habits of the body; controlling the body through surveillance; and targeting the ‘soul’ of the body as something that must be punished and ultimately set free. When explaining the emergence of disciplinary power Foucault points among other things to the various practices involved in rendering the crime as well as the criminal visible. The primary example of such practices can be found in the spatial arrangement of the ‘modern prison’, which made it possible to ‘establish presences and absences, to know where and how to locate individuals, to set up useful communications, to interrupt others, to be able at each moment to supervise the conduct of each individual, to assess it, to judge it, to calculate its qualities and merits’.Footnote 93 The paradigmatic illustration of this kind of spatial arrangement is Jeremy Bentham's prison plan: the architectural figure of the ‘Panopticon’.
As Deleuze noted, by rendering the crime and the criminal visible through the architectural figure of the Panopticon the prison does not only ‘display the crime and the criminal but in itself it constitutes a visibility, it is a system of light before being a figure in stone’.Footnote 94 There is in this sense a certain materiality of the prison, which is directly linked to the production of a mode of visibility, gaze, or way of seeing. The materiality of the prison belongs to its architectural form and spatial arrangement as a specific ‘system of light’. It is a system that, moreover, must be grasped as an ‘instrument and vector of power’.Footnote 95 The same point can be made in relation to other material arrangements such as hospitals, factories, and schools. According to Foucault, all of these institutions were based on a particular gaze or mode of seeing that emerged in the eighteenth century, with the purpose of controlling and correcting the behaviour of patients, workers and schoolchildren, in accordance with the mechanisms of disciplinary power.
Importantly, Foucault's study of disciplinary power also includes an examination of different ways of categorising and naming the crime and the criminal, for example through the sentencing by a judge. Just like the architectural figure of the prison plan, such categorisation or naming can be analysed as an ‘instrument’ or ‘vector’ of power. In this way, rather than ‘representing’ the content and meaning of the ‘visible’, language and ideational factors can be said to interact with material practices of rendering things visible. Both forms of practices must be taken into account and seen as important when analysing the way disciplinary power works and manifests itself. The latter happens both by making it possible to speak in a certain way: to name and categorise criminals, students, and workers; and by imposing a particular gaze or mode of seeing to render their bodies visible through surveillance. Both seeing and saying – the visible and the articulable – participate in the mechanisms of disciplinary power.
Given Foucault's insistence on the importance of the materiality of the body in contemporary political practice it is unsurprising that some of the New Materialisms literature has drawn in part upon his work. For example, Coole and Frost find Foucault's approach to disciplinary power useful precisely because it ‘describes the kind of micropractices that are at stake in pacifying and reproducing social regimes in order to demonstrate how thoroughly our ordinary, material existence is affected by, and saturated with, power and how protean yet banal many of its tactics remain’.Footnote 96 On this basis, Coole has called for a ‘rediscovery’ of ‘the earlier, materialist Foucault’ in order to diagnose the manifold ways in which bodies are targeted by ‘the new productive machinery of capitalism’.Footnote 97 However, it is nevertheless possible to detect a certain tension between our reading of Foucault presented here and the kind of ‘ontological project’ that New Materialisms stands for.
Arguably one of Foucault's greatest contributions to philosophy as well as to the social sciences is his insistence on the mutual imbrication of ontology and epistemology. While highlighting the significance of the ontology of the body in relations of power, Foucault is always very careful in pointing out that this ontological dimension is dependent on an epistemology, which enables the body to be produced on the basis of particular ways of seeing and speaking in the social field. Power is thus inseparable from the production of knowledge. Here we detect a possible limit to the New Materialisms turn: in stressing the ontology of matter there is a tendency to give less emphasis to Foucault's important insight concerning the ways in which ontology and epistemology are always intertwined. It seems to us that this insight is crucial for analysing the ways in which language and materiality interact as parts of the same strategies of power, the same forms of knowledge, and the same production of the materiality of the body. Moreover, in contrast to the normative direction in which Coole and Frost ultimately take Foucault – a move which they claim is necessary in order to counter ‘Foucault's insistence on his own nonnormative positivism’Footnote 98 – we see ethical-political value in genealogy because of the ways in which it problematises practices of imposing hierarchical binaries and because it emphasises the openness of a general field of forces in which new subjects and events may constantly emerge.Footnote 99 By tying this general field of forces to a normative framework, which Coole and Frost suggest we should do, there is a risk of closing down not only the contingency and productive elements of power, but also the radical inseparability of language and materiality that informs much of Foucault's work and the expanded conceptualisation of discourse that we derive from it.
Force and the generalised text
Of all thinkers associated with ‘poststructural’ thought, the work of Jacques Derrida is perhaps most commonly portrayed as trapped in a discursive realm consisting entirely of linguistic texts.Footnote 100 This has led to the critique, as we have already seen from New Materialists among others, that ‘deconstruction’ is a method of reading that ultimately fails to offer an adequate account of materiality.Footnote 101 In this context, one of Derrida's most infamous sayings ‘Il n' ya pas de hors-text’ (‘there is nothing outside the text’ or ‘there is no outside-text’) is often cited as evidence of his hyper-textualism.Footnote 102 In the reading put forward here, however, we want to argue that this is a partial reading of what a deconstructionist approach might entail.
Derrida is not concerned with ‘discourse’ in the limited linguistic sense, as discussed in the first section of this article. Responding to many of his critics, he argues that it is ‘monstrous’ to say that deconstruction ‘confines itself to language and language games’.Footnote 103 While Derrida is certainly interested in language, he is also concerned with what lies beyond it and the aporia of trying to address that question from within the general problematic of linguistic structures. On this reading, it is arguably ‘materiality’ – understood as the Other of language – that Derrida is motivated by and continually seeks to bring to the fore in many of his works, albeit always from the perspective of their mutual imbrication.Footnote 104 In Specters of Marx (1993), for example, Derrida refers to his ‘obstinate interest in materiality without substance’.Footnote 105 Here the caveat ‘without substance’ is imperative when trying to grasp the implications of a deconstructionist approach for rethinking the relationship between language and materiality. As Pheng Cheah has recently pointed out, from a Derridean perspective the concept of materiality in the New Materialisms literature runs a risk of falling back into the status of a transcendental signified and thus the logocentrism of idealism.Footnote 106 In other words, and contra some of the ‘New Materialisms’ literature, it is impossible to assume any sort of pure ‘material realm’ that is uncontaminated by language (and vice versa). Matter ‘is’ not anything, per se. Rather, the language/materiality dichotomy breaks down in Derridean thought because matter can only be understood as part of a complex and radical intertextuality. Hence, it does not make any sense to think about either ‘language’ or ‘materiality’ in isolation. Both ‘elements’ are sutured into each other such that we cannot separate them or even speak of them as being straightforwardly ‘interrelated’ as this would again imply some sort of anterior distinction.
The move Derrida makes in order to ‘think matter outside the oppositions that have imprisoned it’ is to develop the idea of what he calls the ‘generalised text’, which helps us to understand what the notion of ‘radical inter-textuality’ refers to.Footnote 107 Contrary to certain portrayals of his work referred to earlier, when Derrida refers to ‘text’, ‘(inter)textuality’ or ‘the generalised text’, he does not limit this to ‘the graphic, nor to the book … and even less to the semantic, representational, symbolic, ideal, or ideological sphere’.Footnote 108 To do so would be to continue to set the textual ‘against the social, the political, and the historical, as if it were still the book on the bookshelf of the library’.Footnote 109 Instead, the concept of the generalised text opens up his analysis of différance – the endlessly differing and deferring nature of meaning in chains of signification – across those artificial divides. As Derrida states categorically in the ‘Afterword’ of Limited Inc: ‘What I call “text” implies all the structures called “real”, “economic”, “historical”, “socio-institutional”, in short: all possible referents.’Footnote 110
The generalised text within which Derrida identifies the restless play of différance is not, therefore, understood as a purely linguistic structure, but in terms of a field of forces that slices through the language/materiality divide in such a way that renders it impossible to uphold that distinction. Some of the ‘forces’ to which Derrida refers in this context include ‘libidinal forces, political-institutional or historical-socioeconomic forces, or concurrent forces of desire and power’.Footnote 111 One notable example in Derrida's vast oeuvre would be the work that alterity does as a ‘force’ that conditions the possibility of identity, ethics, friendship, and so on. Another would be totalitarianism understood in the specific sense of a force that works against alterity to stifle difference and close off all that it enables. Such forces are often invisible or impalpable, but they are nevertheless central in the production of, for example, different forms of political subjectivity. Texts of a linguistic and nonlinguistic nature are always mutually interdependent, coconstituting, and open to the other: ‘the force of materiality is nothing other than the constitutive exposure of (the subject of) power to the other’.Footnote 112
Derrida's treatment of the generalised text as a field of forces takes on added significance for political analysis and IR. It points to the way in which meaning, identity, and subjectivity are always constituted through force. This is where the ethical-political dimension of deconstruction is perhaps most obvious: it calls for detailed attention to all moves that entail closures in the attempt to delimit a specific context. Of course, this might apply to linguistic phenomena and many of Derrida's analyses of writers such as Saussure are based on careful readings of ‘texts’ in the narrower sense. Equally, however, this form of deconstructionist critique can be mobilised to analyse other forces in the generalised text such as the violent auto-authorisation of juridical-political structures, the denial of hospitality to the global sans papiers, or the media-theatricalisation of particular ‘events’ such as ‘9/11’.Footnote 113
What Derrida offers, then, is a deconstruction of the language/materiality binary within which both matter and language are typically framed. Far from negating material factors, a Derridean approach provides for an alternative theorisation of matter and language as mutually constituted. Coole sets the ontology of New Materialisms in contradistinction with what she considers to be the epistemological focus of poststructuralism when she argues that ‘rather than deconstructing oppositions – and especially rather than undertaking linguistic or textual deconstructions of texts – New Materialisms focus on the actual entwining of phenomena that have been historically classified as distinct’.Footnote 114 But rather what we see in Derrida is neither a reduction of the concept of the text to a linguistic form nor a denial of the work that binary oppositions do in structuring actuality. A concrete example of the kind of ‘capacious historical materialism’ that Coole calls for is to some extent prefigured in Derrida's sustained analysis of the work that the human/animal distinction does in shaping structures of sovereignty, subjectivity, and political thought. Beyond merely an identification and deconstruction of this distinction in various philosophical writings, Derrida also seeks to interrogate the violent socioeconomic and political practices that it gives rise to. For Derrida, the signifier animal is an ‘appellation that men have instituted’ in order to imbue humanity with the authority to authorise its own privileged status. In turn the human/animal distinction – neither a purely linguistic nor material ‘discourse’ – permits violence globally: ‘genetic experimentation …, the industrialization of what can be called the production for consumption of animal meat, artificial insemination on a massive scale, … the reduction of the animal not only to production and overactive reproduction … all … in the service of a certain being and putative human well-being of man’. Far from being locked inside a world of linguistics, in this example Derrida takes as his focus the violent zoopolitical architectures of late modern capitalism in order to diagnose what is at stake in the (re)production of the ‘human’ today – central not only for those interested in animal welfare, but the international politics of human rights, human security, and humanitarianism.Footnote 115
Conclusion
The New Materialisms turn has issued a number of pressing theoretical and methodological challenges for social scientists in general and IR theorists in particular. IR's own genealogy as a discipline has made it especially vulnerable to aspects of the New Materialist critique. This vulnerability is most acute in the context of debates surrounding the status of discourse and what it means to do discourse analysis in IR. While dominant approaches to discourse analysis in IR have questioned the traditional Realist focus on material factors by drawing attention to the politics of language, representation, and meaning-making practices, they have tended to operate within a narrow understanding of what discourse ‘is’. Most commonly, IR discourse analysts have taken language or occasionally images as their objects of study and left the question of materiality – and its relationship to ideational factors – unproblematised. As such, matter is not considered to be the ‘proper’ domain of the discourse analyst and instead remains the preserve of mainstream positivism.
The New Materialisms literature poses a serious challenge to both ‘mainstream’ and ‘critical’ strands of IR: it provides a counter-point to positivistic notions of materiality and promises a corrective to the linguistic excesses of discourse analysis. Its core message is that all aspects of international politics are always inevitably saturated in matter; that the nature of the relationship between human and nonhuman forces is changing; and that, in some accounts, stuff, objects, and things have a certain agentic capacity of their own. As such, this literature constitutes a major ontological project, which not only questions the prevalent linguistic bias in certain quarters of IR, but also calls for a wholesale re-evaluation of the anthropocentrism of the discipline.
In this article we have expressed sympathy with the combined insights of the New Materialisms literature and its rallying call for more serious engagements with materiality as an active political force. However, we have also shown that in certain guises, particularly extant attempts at applying the insights of New Materialisms to IR, the language/materiality dichotomy is recycled such that the latter becomes privileged over the former. The problem with this move is that it overwhelms an appreciation of the role of language in human/nonhuman assemblages and ultimately works within rather than challenges the limits of the debate whereby ‘language’ and ‘materiality’ are treated as separate – and indeed separable – to begin with.
By contrast, we have argued for a position that takes the debate a step further by drawing on earlier poststructural works in order to refuse to make a clear distinction between language and materiality. Contrary to the claims of certain New Materialists and Critical Realists, we have shown that two key thinkers associated with poststructuralism (notwithstanding all the problems of using this term) – Michel Foucault and Jacques Derrida – are not dismissive or neglectful of matter or ‘ontology’. Indeed, one of the key points of this article is that it becomes very difficult – and in our view impossible – to sustain the argument about the linguistic bias of poststructuralism when we return to the primary works of these thinkers. As we have shown in respect of Foucault's treatment of disciplinary power and Derrida's notion of the generalised text, there are already critical resources in these authors' works for analysing the inseparability of language and matter. If adopted, such a perspective entails a more inclusive notion of discourse than that currently found in IR. Pushing the limits of the current debate, this extended perspective on discourse ultimately assigns equal weight to linguistic and material dimensions as part of a radical intertextuality.
