Towards International Relations beyond the mind

Introduction

The article takes issue with the centrality of the mind in International Relations (IR) theorizing. It queries both the dualism between a knowing subject and an external object, and the monism of reducing the real to thought. While dualist positions are often questioned by pointing to the transitive quality of allegedly intransitive natural objects, monist positions seem harder to grapple with at least at first sight, for IR specialists as well as ordinary people talk about human minds as containing ideas, beliefs, apprehensions, desires, ideas and knowledge, all of them capable of inducing some kind of social behaviour. However, a direct causality between the ‘world’ and the mental states that represent this world has been discounted by numerous philosophers of mind after German idealism, for it implies two serious problems: the risk of error and the possibility of non-existence. Mental capacities, given that they are autonomously functioning, might simply be wrong in their representation of the world, and they might also be about things which do not exist. More crucially, however, a metaphysical humanism relying on a hierarchically reductionist depiction of mental explanations that start with human beings, focus on their minds, which are composed of neurons, and which can then be explained by physics, is nowadays often replaced in the philosophy of mind by a focus on the social as a starting point. The sequence must then be ‘social-mental-biological-physical’ (Brassier, 2017; Clapin, 2002: 18; Morton, 1997: 117–133) instead of ‘physical-biological-mental-social’. In this case, the question remains why one should not start with the functioning of the social from the outset if one wants to understand global politics as a social process.

In this article, I will employ the notion of discourse to develop this argument. While many forms of discourse theory are linked with anti-materialist idealism, the analysis puts forward exactly the opposite argument, that is, that discourse theory should primarily be situated beyond the mind. To develop more thoroughly the argument that ideas, fears or desires – as mental products – do not independently constitute the experienced (whether observable or not), but are themselves an upshot of discursive variation, the analysis starts with a discussion of prominent IR work on ideas and discourse in the next section. While not calling into question that minds exist, the crucial question that surfaces is what makes them work (Clapin, 2002: 2; Dupuy, 2009; Fodor, 1989). A hint may be found in the claim that ‘[t]hought is nothing without something that forces and does violence to it. More important than thought, there is “what leads to thought”’ (Deleuze, 2000: 61). This abstract thing that leads to thought, is precisely what I call discourse. I will show that a large segment of IR work is insufficiently clear on this crucial notion. Ideas and discourse are all too often employed coterminously, which makes the use of discourse redundant and superfluous. I therefore contend subsequently that this strange state of the art is reflected in how the philosophy of science and the philosophy of the mind have been treated in prominent IR work, epitomized in the work of Patrick Thaddeus Jackson and Alexander Wendt, by following a particular version of Cartesian rationalism. Unsurprisingly, the focus on the mind and on ideas as products of the mental forecloses an engagement with discourse as an ontological level beyond the mind. It is on this basis that the article proposes to transcend the dichotomies between mind and world as well as ideas and materiality by advancing a political ontology that stresses a poststructuralist concept of discourse in the final section. Although the disjointedness of the various theoretical strands dubbed ‘poststructuralist’ leads many of its alleged advocates to reject the label, the following analysis will develop a definition that works neither at the level of the ‘real’, nor at the level of the ‘imaginary’, but requires a third order that produces the unity of a theoretical object. This part of the argument is greatly indebted to the work of the late political theorist Ernesto Laclau and his co-authored work with Chantal Mouffe. In thoroughly engaging their argument, it will become possible in the conclusion to summarize a path towards an IR beyond the mind that engages in the study of the political more seriously.

Ideas, discourse and the mind

One important issue to be problematized in the following lies in the separation of ‘mind’ and ‘world’ that has been taken for granted in large parts of IR. First of all, what do these two concepts – ‘mind’ and ‘world’ – signify? Does it make sense to draw a clear boundary between mind and world and treat the two concepts independently? Is it in fact possible to speak of reality as existing independently of the mind? Or can we reduce ‘world’ to concepts, as has been popular among idealist philosophers in the tradition of Georg Wilhelm Friedrich Hegel? Although these are mainly philosophical questions, their discussion has stimulated the field of IR from the outset. The debate can be traced back at least to the early work of Hans Morgenthau and Kenneth Waltz. The former, perhaps the most important figure in the years after World War II, demonstrated a distinct philosophical proclivity that seems to combine a focus on the mind with an emphasis on constitution and contingency. Morgenthau’s (1947) theoretical stance is clearly outlined in Scientific Man vs. Power Politics:

Thus, the physical world, as we are able to know it, bears in a dual sense the imprint of the human mind; it is in a dual sense its product. We are able to know it only within the limits of our cognitive faculties […]. (1947: 123)

Morgenthau’s doctoral dissertation had explored the psychological concept of the will (Jütersonke, 2010: ch. 2), and he time and again pointed out that the laws of international politics were rooted in human nature, that is, in ideas of good and evil and a lust for power deeply engrained in the human mind (Morgenthau, 1948: 4 and 29; Petersen, 1999). It is thus perhaps against the background of Morgenthau’s attempts at setting up an IR theory that the preoccupation with the mind and with ideas as a progeny of the mind in later decades can best be grasped. This finding applies even to those structuralist theories that depart from Morgenthau’s focus on human nature. In that vein, Kenneth Waltz (2001 [1959]) summarizes his idea of theory, emphasizing the concept of ‘images’ that his work builds upon:

The word ‘image’ suggests that one forms a picture in the mind; it suggests that one views the world in a certain way. ‘Image’ is an apt term both because one cannot ‘see’ international politics directly, no matter how hard one looks […]. (2001 [1959]: ix (my emphasis))

By emphasizing the mental capacities of humans, Waltz situates his theory beyond the positivist mainstream, blaming the bulk of IR researchers for neglecting philosophical questions. He goes on to specify his mind-dependent definition of theory in his opus magnum:

If a theory is not an edifice of truth and not a reproduction of reality, then what is it? A theory is a picture, mentally formed, of a bounded realm or domain of activity. […] theories construct a reality, but no one can ever say that it is the reality. (Waltz, 1979: 8–9)

In this quote, the ‘mental’ assumes central significance; ‘reality’ is exclusively formed within the mind. In his work, Waltz has continuously made philosophically grounded statements related to the distinction between laws, models and theories, the latter at all times falling into the realm of the created and constructed by the mind (for a critique Waever, 2009). Later, even strictly materialist researchers like Robert Cox – as a historical materialist and Gramscian IR theorist – took up the differentiation between material factors and ‘ideas’ (Cox and Sinclair, 1996), and Marxist sociologist Martin Shaw (2000), in a similar vein, treats ‘ideas’ as mental products.

The development culminated in the 1990s in Alexander Wendt’s influential, realist-inspired ‘idealist ontology’ (Wendt, 1999), in which ideas and discourses appear to be treated as coterminous. ¹ Although Wendt acknowledges that ‘ideas exist and have effects because of the discursive forms (norms, institutions, ideologies) in which they are embedded […]’ (Wendt, 1999: 495), he does not explore the notion of discourse any further (for a critique Zehfuss, 2002). In contrast, I claim in the following that if discourse is in any meaningful way distinct from the distribution of ideas, it cannot derive its definitional content from any sort of idealism, for no more than a tautological statement could be established. If, however, ideas and discourse are seen as idem, serious metatheoretical as well as substantial and methodical objections must be raised.

These objections certainly do not remain restricted to Wendt. Crucially, critical realism underlines the importance of the mind for the social sciences in general: ‘Realism normally implies that objects have a mind-independent existence. Social objects clearly violate this principle: no people, no social objects. Social objects depend upon minds’ (Wight, 2006: 26). In this quote, agents are described as possessing autonomously functioning mental capacities, which operate independently of any further ontological level that creates these ideas in the first place. Writing about ‘conceptions’, Colin Wight (2006) maintains that: ‘It is important when discussing this issue that the centrality of agents’ ideas and concepts is not lost’ (2006: 56), and he adds that: ‘Unlike the natural sciences, however, the object studied by social science includes ideas; people act in accordance with ideas’ (2006: 57).

From this perspective, ideas are conceptualized as progenies of seemingly self-governing mental capacities, which also leaves Wendt (1999), in his search for a via media, to ‘turn around and defend the individualist intuition that mental states have an independent explanatory status’ (1999: 178). In consequence, it becomes a tricky endeavour to disentangle the relationship between ideas, cognitions and mental capacities on one hand and discourse and materiality on the other. If we refer to discourse as ‘a cohesive ensemble of ideas, concepts, categorizations about a specific object that frame that object in a certain way and therefore delimit the possibilities for action in relation to it’ (Epstein, 2008: 2), and accept the significance of ‘meanings that are constituted prior to a discursive intervention’ (Wiener, 2008: 74), then what does this leave for the status of discourse independent of any mental instantiation?

In consequence, some discourse analyses put their focus on ‘interventions made by individuals’ (Wiener, 2008: 71, and the critique of ‘intersubjectivity’ in Kessler, 2007). Individualism is then coupled with a tentative realism, for it is not only the reference to ideas which reminds us of idealism, but the relation to a ‘specific object’ that ideas are related to. When Antje Wiener (2008) refers to discourse as the ‘space in which collective perceptions are present’ (2008: 75), this brings the connection of idealism and externality to the fore, as perceptions are usually directed towards external objects which are then translated into ‘ideas’ by mental capacities. The reference to the mental has become hegemonic in IR to a degree that not even the most intriguing discourse theoretical work can entirely free itself from veiled forms of idealism. For instance, David Campbell (1998 [1992]), perhaps one of the most widely acknowledged discourse theorists in the field, maintains that ‘the dramatic changes in the international order currently underway are to be understood as a victory of one idea over others’ (1998 [1992]: 17), yet at the same time he takes a strict stance against all forms of ‘philosophical idealism’ and develops a stringent way of theorizing discourse. In a similar vein, Iver Neumann (1996), in his ground-breaking work on Russia and the Ideas of Europe, advances notions such as ‘the Russian discourse on Europe’, ‘the domestic political discourse’ and ‘the public discourse’, but equally emphasizes that ‘ideas about Europe emerge as a key background determinant for both domestic and foreign policy’ (1996: 1). In Uses of the Other, Neumann (1999) traces ideas connected to identity formation, claiming that ‘[the] collective self is predicated on certain key political ideas’ (1999: 31), leaving open whether ideas themselves are to be seen as a progeny of discourse. While both Campbell and Neumann have offered pioneering work on discourse, they also imply the significance of ideas and their causative, individualist as well as sometimes essentializing propositions.

We will show in the following that conceptualizations of discourse which lean to either side – idealism or materialism – must surrender at this juncture. This is also evident in the otherwise ostentatiously non-idealist IR discussion of ‘new materialism’. For instance, Tom Lundborg and Nick Vaughan-Williams (2014) convincingly show that new materialist theories suffer from the continuing split between ideas/language and brute materiality. They go on to maintain that the political power of materiality that new materialism focuses on cannot be understood without an understanding of the materiality of discourse. Yet, it is this kind of materiality that the authors might not be able to capture with their reading of Foucault and Derrida. When investigating the ‘interplay of language and materiality’ (Lundborg and Vaughan-Williams 2014: 17, emphasis added), I would insist that the difference between the two is resolved on a different ontological level. In fact, one might eventually repeat Wendt’s idealist-interactionist argument by reiterating the claim that ‘language and ideational factors can be said to interact with material practices of rendering things visible’. By emphasizing that ‘[b]oth forms of practices must be taken into account and seen as important’, Lundborg and Vaughan-Williams (2014) perhaps draw an all too well-defined line between the two spheres that they attempt to synthesize (2014: 18). The clear separation of ‘material’ and ‘ideational’ factors that they espouse is only consequential: ‘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’ (Lundborg and Vaughan-Williams 2014: 22). It is exactly this prevailing dichotomy, constructed in German idealism in the nineteenth century as a first alternative to a naturalist conception of reality, that will be deconstructed in the remainder of this article. To understand the prevalent focus on ideas in IR theory, one needs to leave the familiar terrain of IR theorizing and inquire into how philosophy-of-science issues have been received by the discipline.

IR philosophies of the mind

The separation between ‘mind-world dualism’ and ‘mind-world monism’ – which in IR has been quite prominently and rigorously discussed in Patrick Thaddeus Jackson’s widely discussed The Conduct of Inquiry in International Relations ² – can be questioned if, first, one takes discourse to be the constitutive sphere of all meaningful reality; second, if one assumes that subjects are not pre-existing, self-contained entities, but are to a large extent generated by society and third, if one reformulates the notion of reflexivity (the historical situatedness of the researcher within the ‘object of inquiry’) in a non-humanist fashion. Following Martin Heidegger’s critique of subjectivist voluntarism, many philosophers from different origins have argued that the very distinction between something that is ‘known’ and the various ways of knowing it ignore the discursive conditioning that takes place previously (Dupuy, 2009; Heidegger, 2010 [1953]). The kind of ‘conditioning’ referred to in this context is certainly not an individual but a social endeavour. To put it simply: is the experience of concrete phenomena dualist or monist? Does not the position that all reference to an external reality must be relinquished and knowledge must be restricted to sense-impressions require some sort of dualism? As Jackson himself concedes with reference to Kant, an unmediated perception that eludes language and social convention is nonsensical (Agar, 2006: 12–15; Hansen, 2006: 17–18; Jackson, 2011: 60; Kurki, 2008: 44; Zehfuss, 2002: 198).

In fact, however, Jackson’s account cannot liberate itself from Karl Popper’s categorization of science on the basis of ‘three worlds’ – World 1 (the physical world), World 2 (the subjective world) and World 3 (an objective body of knowledge that cannot be reduced to World 1). This categorization refers back to René Descartes’ distinction between res cogitans and res extensa (the latter denoting mere material substance), which subsequently paved the way for the focus on ideas and the mind as described in the previous section of this article (Bernstein, 1983; George, 1994: 46–49; Popper, 1972: 69–70, 115–118). Ideas take stable frontiers between body and mind for granted, and rely on the problematic notion of a pre-existing, autonomous subject capable of ‘perception’. If, then, the empirical applies to those domains that are experienceable or in some form intelligible, then the transcendental is equally excluded as the unthinkable, unstable, unidentified and unfinished.

This principal claim of the present analysis is therefore built on the argument that both ‘mind-world dualism’ and ‘mind-world monism’ take their cues from the centrality of the mind – from an emphasis on humanism, which conceptualizes the social and political as exclusively the results of the mental capacities of human beings. In this context, monism itself can be understood as an essentializing and totalizing concept, since it reduces existence to a single principle, that is, the mind. While appearing maiden-like and neutral, for it seems as if the mind represented a sphere that is not tainted by ideology, the ‘reality’ of the mind is in fact deeply ideological and embedded in dominant ‘realities’. In consequence, intersubjectivity is neither politically neutral nor a pre-discursive category (Edkins, 1999: 8–9, 26; Mackenzie, 2001; Zehfuss, 2002: 229). Likewise, ideology always restricts a universe of meanings to a particular one, and it is the mind which produces this fixture. For Jackson, autonomously functioning minds ‘generate’ theories; the structure of the social, which might provide an independent horizon for what is possibly generated by the mind, is neglected. ³ Monism has got nothing in common with post-foundationalism, as Christine Sylvester elucidates in her review of Jackson’s approach. ⁴ In contrast, it must be seen as seriously foundational, if not fundamentalist, if it involves the reduction of reality to some subjective or intersubjective truths which are not politically, that is, discursively mediated.

This level of mediation is lost in many treatments of philosophy of science questions within or at the edges of IR. It is without doubt that to propose a rigidly anti-mentalist and anti-idealist standpoint, one must also go beyond what Wendt recently proposed in Quantum Mind and Social Science, which sticks to a focus on the mind and literally attempts no less than an explanation of consciousness (or experience, which he treats synonymous) from a scientific perspective (Wendt, 2006, 2015). Wendt’s work in this context shifts the problem into another direction, but does not seem to solve it, which led Oliver Kessler (2007) to propose early on that ‘abandoning the distinction between ideas and matter’ (2007: 245) would be a way forward in IR theorizing. This would eventually make it possible to overcome the combination of a rump materialism with mental causality. In quantum theory, Kessler maintains, the distinction between consciousness and matter is upheld before it is finally fused conceptually. The fundamental problem with Wendt’s Quantum Mind thus lies in its continuing reliance on the combination between ideas as mental products and a rump materialism that had already been prominent in his earlier work.

In his bold endeavour, which is ‘to find the Undiscovered Country of philosophical clarity’ (sic!) (Wendt, 2015: 2), Wendt (2006) still sees only two metatheoretical options to social theorizing, one positivist, the other interpretivist, ‘and these are the only two worldviews we have’ (2006: 183). His starting point is an allegedly more refined dualist solution to the mind-body problem, enabling us to finally coalesce ideas and materiality within the mind. This is achieved by repealing the long-standing materialist assumption that the fundamental constituents of all objects must be material (Wendt, 2015: 14–15). Although he criticizes the mainstream view of linguistics as a cognitive science, that is, language as a mentally processed phenomenon, and briefly touches on Ferdinand de Saussure’s work (Wendt, 2015: 210–211), he does not engage with the notion of discourse as a transitive and public fact. Instead, he construes language as a ‘langue in the head’ (Wendt, 2015: 211, emphasis in original), because he cannot seem to escape the focus on the mind, and he sticks to the centrality of a ‘physical basis’ (which can, of course, be both classically and quantum physically conceived).

What I would dispute here is his claim that ‘physics should have the last word on reality’, and ‘that consciousness is a macroscopic quantum mechanical phenomenon’ (Wendt, 2006: 180). It remains one of the great mysteries of IR research that scholars are so obsessed with the way natural scientists conduct their inquiries, mostly at the expense of developing a proper theory of the social, political and political constitution of the social. By stressing that ‘consciousness is the basis of social life’ (Wendt, 2006: 181), the mind as an autonomous functioning entity keeps its central focus, and the question of what social factors constitute mental processes is once more ignored. In fact, poststructuralist theories are provided with a natural science basis in Quantum Mind, which is unnecessary from a genuine poststructuralist perspective. While Wendt contemplates why it is so difficult to explain consciousness, the question here is why – as social scientists – we should be interested in how the mind works in the first place? Undoubtedly, some of the tenets of quantum (meta-)physics correspond to dominant poststructuralist precepts, like Werner Heisenberg’s Uncertainty principle, which entails a fundamental insecurity and contingency with regard to both the location and the motion of a subatomic particle, calling into question basic principles of materialism/naturalism, determinism and objectivity. Furthermore, the collapse of the subject–object distinction, the notion of non-locality, the contradictory nature of subjective properties, the correlation between observation and scientific measurement, the argument ‘that individuals’ desires and beliefs are highly sensitive to context and framing effects’ (Wendt, 2006: 195) as well as the accompanying critique of individualism/atomism have been common in poststructuralism since the 1950s, with the qualification that ‘framing’ puts focus on individual agency and must hence be regarded as foreign to the poststructuralist vocabulary. While an electron is conceptualized as a particle and a field, and the field provides the data about the whole environment of the particle, Pierre Bourdieu’s (1982) habitus, for instance, can be seen as an incorporation of the social field into the practices of subjects, thereby creating what can speculatively be called consciousness. If Wendt (2015) takes as his central concern ‘to give meaning to events’ (2015: 28), one might claim that ‘events’ can only be grasped if inscribed in sedimented discourses. ‘Events’ are never situated outside of a discourse. They neither have a temporal nor a spatial location outside of the discursive articulations that translate it from a ‘pure event’ – with all its ambiguities, displacements and disruptions – into a ‘historical event’, which turns subjects into well-defined and sovereign beings, and establishes a homogeneous timeline which does not exist in the ‘pure event’ (Laclau, 2014; Lundborg, 2012: 92, 96).

Quantum science apparently has no answers that go beyond what has already been presented in continental philosophy and political theory in the past 100 years. Instead, it sticks to the naturalist conjecture that social science must be informed by quantum physics, without delving too deeply into what should concern a social or political theorist first and foremost, namely a concept of the social and the political. Moreover, hypotheses such as the ones derived from panpsychism cannot yet be substantiated by contemporary neuroscience. In fact, both parts of the human consciousness hypothesis that Wendt presents – quantum brain theory and panpsychist metaphysics – function well without reference to the political. At least three comments are therefore in order in this context: first, although panpsychism – the hypothesis ‘that consciousness goes “all the way down” to the sub-atomic level’ (Wendt, 2015: 5) – supposedly distances itself from idealism, it represents no more than the critical realist claim to combine a dualist ontology with what Jackson calls phenomenalism, and hence subscribes to a continued focus on the mind. Second, it is also monist, for matter must eventually be seen through the mind, as a ‘minded’ phenomenon. Third, it corresponds to some tenets of ‘new materialism’ in contending that brute materiality entails ‘some level of subjective experience or interiority’ (Wendt, 2006: 192), which leads Wendt to refer to some of the same philosophers that, for instance, Jane Bennett bases her concept of ‘thing power’ on. ⁵ For a dissolution of the distinction between mind and matter to become possible, it is argued here, we do not need a quantum social science which sees matter through the mind, but an ontology which transcends both matter and mind.

This means, from our perspective, that a quantum social science collapses with its premises, and eventually sticks to a focus on the mind, while this focus must be seen as problematic from a more sociological or specific discourse theoretical perspective. When Wendt (2015) laments in his latest work that classical physics ‘doesn’t tell us anything about meaning, discourse, and other intentional phenomena’ (2015: 14), the question may be permitted at this point whether he tells us anything about those important aspects of the social. When claiming that, ontologically speaking, language is a part of nature, he seems to omit important aspects connected to the notion of difference, which (as we will see subsequently) must be understood as that empty space within societies which makes their material structuring possible in the first place. When, on the other hand, he claims that ‘one needs particular forms of language, such as discourses about “capitalism’ or ‘marriage”’ to understand how society works (Wendt, 2015: 243), the paradoxical question remains if he can present a theory of how society works without engaging further with the very notions (i.e. discourse and politics) that makes this understanding possible.

In that sense, the contribution of this article starts where Jackson and Wendt stop. In his discussion of Clifford Geertz’s work, the former takes a crucial step in acknowledging that it might not be of principal importance to analyse ‘what people perceive, but on (so to speak) what they perceive with’ (Jackson, 2011: 134, referring to Geertz, 2000: 56–58). However, Jackson (2011) does not proceed further along this path, as he continues by arguing that ‘[s]omething like this is the basic conjecture of a certain kind of IR constructivism, which privileges the ideational and in this sense subjective aspects of social life’ (2011: 134). Such a conclusion is only consequential within Jackson’s typology, as constructivism epitomizes ‘the highest of highs’ in the matrix of monism and dualism. Political discourse theory, to which I will now turn in order to elicit the potentials of an IR beyond the mind, is clearly situated outside this matrix.

Towards an IR beyond the mind

As argued in the previous two sections, both dualists and monists are confronted with the same kind of problem: how are the world-independent ‘mind’ and the mind-independent ‘world’ mutually imbricated and constituted? To pick up from where we started this discussion: if discourse mediates any contact with reality and at the same time is not a mental product, then the very distinction between monism and dualism can be replaced by a third alternative. If, in other words, both materialism (the reduction of reality to the physical) and idealism (the reduction of reality to thought) are ruled out, the question needs to be raised why we need the mind-world distinction at all to build a better theory of the social in its global dimension. In other words: ff Wendt (2015), in his work on a quantum social science, claims that ‘the content of mental states is individuated by what “conceptual grid” is used’ (2015: 253), then does it not make more sense to develop thoroughly this ‘conceptual grid’ which is deemed so important?

To sketch such a path towards a poststructuralist IR theory beyond any reference to the mind, Martin Heidegger’s initial argument that the being of an object must be distinguished from its existence, while existence can only be grasped as constituted within a certain structure in which it is embedded, can be employed as a foil. Structure, according to Heidegger (2010 [1953]), ‘aims at the analysis of what constitutes existence’. He calls ‘the coherence of these structures existentiality’, and continues to explain that, ‘since existence defines Dasein, ⁶ the ontological analysis of this being always requires in advance a consideration of existentiality’ (2010 [1953]: 11, 12, italics in original). Heidegger’s notion of ‘ontological difference’, denoting the very difference between being (Sein) and ontic things-which-are (Seiendes) can hence be employed to overcome the monism-dualism dichotomy. In a nutshell, the difference between discourse theory – as proposed in the following – and new materialism as well as critical realism would be, in Heidegger’s terms, that being never stands in a direct relationship to existence and vice versa. No concept of the social or the political follows from being. Instead, as Laclau and Mouffe (1987) contend in Heideggerian fashion, being ‘depends on the reinsertion of that being in the ensemble of relational conditions which constitute the life of a society as a whole’ (1987: 91).

Crucially, affirming an understanding of society in this way implies that discourse is anti-mentalist and unconscious. While idealism implies a reduction of the real to thought, realism would establish an identity between thought and thing, and materialism would focus on the causal properties of physical materiality, discourse theory as presented here would call into question the unity of the mind and the thing. Let us draw on a famous example that Laclau and Mouffe (1987) present in their reply to Norman Geras’s critique, which is inspired by Wittgenstein and is worth quoting at length, for it clarifies the much conflated dichotomy between ideas and materiality:

Let us suppose that I am building a wall with another bricklayer. At a certain moment I ask my workmate to pass me a brick and then I add it to the wall. The first act – asking for the brick – is linguistic; the second, adding the brick to the wall, is extralinguistic. Do I exhaust the reality of both acts by drawing the distinction between them in terms of the linguistic/extralinguistic opposition? Evidently not, because despite their differentiation in those terms, the two actions share something that allows them to be compared, namely the fact that they are both part of a total operation which is the building of the wall. […] Obviously, if this totality includes both linguistic and non-linguistic elements, it cannot itself be either linguistic or extralinguistic; it has to be prior to this distinction. This totality which includes within itself the linguistic and the non-linguistic, is what we call discourse. (1987: 82)

The dichotomy that Laclau and Mouffe are trying to break in this quote is exactly what concerns me in this article. The ‘building of the wall’ can neither be seen as a material operation, nor can it be analysed as a linguistic phenomenon built on pre-existing mental properties that make speaking possible in the first place. This apparent dilemma leads to the necessity of introducing a further ontological level which combines the material and linguistic levels, while at the same time being independent of it. While a first wave of poststructuralist IR discourse theory shifted our attention away from idealism towards intertextuality and still seemed to draw a widely accepted line between ‘discourse and text’ (in particular Der Derian and Shapiro, 1989: ix), it is argued here that there must be a further level between text and brute materiality which mediates between the two, which cannot be the quantum solution of the brain as a physical infrastructure. Yet, this level is hard to detect, since it has for a long time not played an appropriate role in Western philosophy and the humanities, which have instead been occupied with what is visible, perceptible and/or experienceable. In order to close this gap and introduce the missing level, we have to engage with the notion of difference as introduced by Heidegger (1969), for it is the German philosopher who prominently claimed that beings, be they conceptual or material in the common sense of the terms, can only come into being ‘by virtue of the difference’ (1969: 64). Insofar as beings can only be thought in its difference to the general Being, the single element can only be grasped in its relationship to the whole. In this perspective, Being takes precedence over being; Dasein only follows on Mitsein. In contrast to Wendt, who can only grasp difference as ‘physical difference’ in either classical or quantum theoretical form, Being is the unthematizable horizon, against which beings gain meaning. Following this view, Deleuze claims that the subordination of difference by identity rests in a particular concept of the subject, which precludes a perspective on the difference between subjects (Connolly, 1991; Deleuze, 1994: xv–xvi; in IR, see Behr, 2014: 11). On this basis, Deleuze has maintained that any society is grounded in the repetition of difference and therefore remains groundless. This claim comes down to a prioritization of difference over identity.

Difference must thus be comprehended as an ontological level that neither has a physical basis nor is it rooted in thought. Discourse is a system of differences without any positive content. This implies a shift in the past century’s philosophy from things in themselves, over the emphasis on ideas, to a focus on relations between what is described as factual in language and discourse (Agar, 2006: 12–14; Currie, 2004; Derrida, 1980; George, 1994: 55–56; Hurst, 1998: 8–9; Walker, 2010: 150–183). Martin Heidegger’s post-Kantian existential philosophy and Ludwig Wittgenstein’s philosophy of language set the stage for poststructuralist critiques of structural linguistics. On the account outlined here, ‘knowledge’ of the world is thus impossible beyond the discursive schemes that create this world in the first place. Moreover, identity will no longer be prioritized over difference. In fact, identity loses its static, antecedent quality and cannot be isolated as an independent variable any longer. What the Cartesians aspired to attain by the search for foundations is replaced through the trust in impressions employed by empiricists. Both an idealist tradition, which tends to see reality as ‘what we think’, and the empiricist convention, which is inclined to see reality as ‘what is perceived’, seem to be trapped in the Cartesian dilemma. In contrast, one can reasonably argue that there is no need for an ultimate foundation or universal truth. While some contingent foundations need to be referred to in order to avoid pure relativism, these foundations are never ultimate.

To this we can add a notion of non-linear temporality. Crucially, Derrida’s notion of différance constitutes a critique of the non-temporal nature of structuralist analyses. Writing about conceptual oppositions, Derrida explains that: ‘The one is only the other deferred, the one differing from the other’ (Derrida, quoted in Spivak, 1976: xliii; also Behr, 2014; Dillon, 2013; Doty, 1996). Derrida maintained that signs are always shifting, and that any sign bears the traces of infinite other signs that surround, precede and follow it. Due to these traces of the Other in the present, the present itself remains an illusion; différance signifies ‘the non-existence of presence’ (Currie, 2004: 55, 48–60; also Edkins, 1999: 12–13; Behr, 2014: 11; Bennington, 2006: 194–199); hence any emphasis on either physical or mental bases must be considered pointless. Instead, Derrida’s concept of différance reveals the potential that comes with a deeper analysis of differential relations, their contingency and hierarchical quality. As Derrida (1976) concludes:

At the point where the concept of differance intervenes […] all the conceptual oppositions of metaphysics, to the extent that they have for ultimate reference the presence of a present, […] (signifier/signified; sensible/intelligible; writing/speech; speech [parole]/language [langue]; diachrony/synchrony; space/time; passivity/activity etc.) become non-pertinent. (1976: lix)

That said, it is only consequential to repudiate the distinction between the transitive and the intransitive, as widely employed in realist or materialist ontology. The intransitive, understood as real objects which exist independently of mental activities, creates an independent level of causality, and is seen as dogmatic by discourse theorists, ‘because why exclude from the transitive variation this object which is beyond the realm of transitivity?’ ⁷ The distinction between the transitive and the intransitive dissolves insofar as the intransitivity of objects is nothing else but another discourse constituting the object in the first place. In other words, materialism would be no more than a form of the discursive constitution of materiality.

Thus, discourse must be seen as material in a post-Gramscian sense, as it is ‘embodied in institutions and apparatuses, which welds together a historical bloc around a number of basic articulatory principles’ (Laclau and Mouffe, 2001 [1985]: 67). Linguistic and non-linguistic properties of a discourse are not played out against each other, but jointly constitute a structure of differential articulations which are only graspable within this structure, not from a position external to it. So-called ‘sedimented practices’ organize themselves within the framework of a particular symbolic order, and are eventually able to generate institutionalized structures which can be rather durable. In the disruption of sedimented practices, however, the contingent character of discourse becomes visible, too, as institutions are no longer able to represent the demands of the political sphere. Importantly, materiality must not be misunderstood here, for it is part of a discourse: any object can only constitute itself within the framework of articulatory differences, while the primacy of the political is always threatened by the quasi-naturalness of established social institutions, which in many cases involve bureaucratization and technologization. The prime example of this danger is the undoubted nature of the modern nation state, despite its only rather recent evolution, that is, after the Westphalian peace in 1648. Yet, the most brutal forms of war have since been fought in its name, and its almost natural legitimacy is institutionalized in international law. Discourses are generating, anchoring and dissolving institutions, and in this sense they are highly material in themselves. In this view of the materiality of discourse, the sedimentation of articulatory practices becomes existential of the social.

What is presented here is but one, notably poststructuralist form of discourse theory. The combining element of all the different theories labelled poststructuralist is a reliance on structural theorizing and the relationship between social structure and subjectivity, time, power and truth, summarized here under the label ‘discourse’. The notion of difference is fundamental in this context. Some works in IR and beyond already bear the concept in their title (e.g. Behr, 2014; Connolly, 1991; Inayatullah and Blaney, 2004; Young, 1990). While these are promising signals, a political discourse theory that explores the full potentials of an IR beyond the mind is yet to be developed.

Conclusion

The article set out to propose an IR beyond any reference to the mind. It was argued that work in IR all too often refers to ‘ideas’ when discourse was actually implied. If we accept the more than 200-year old conjecture that the real has no intrinsic quality, the idealist conclusion that it must be the mind that produces this quality may still be qualified on discourse theoretical grounds. What is more, prominent philosophy-of-science work within the discipline was of no great help in clarifying the fundamental difference between ideas as mental products and discourse as the fundamental logic of the social, in which any form of subjectivity is generated. The article thus drew on sources from continental philosophy and political theory to elucidate the potentials of discourse theory for IR theory. However, a fully developed theoretical approach to global politics in this strand of theorizing would require a deeper engagement with at least four arguments:

This kind of global politics can certainly not be reduced to individuals, their mental capacities or the ‘true’ nature of things. In conclusion, for our field of study, an IR beyond the mind involves a renunciation of a long tradition in Western metaphysics that takes a ‘true’ conception of reality or ‘true’ knowledge as its point of origin and adopts a stable, unified and integrated conception of subjectivity. Instead, it brings the political constitution of society, its contingent and ethical nature to the fore.