In an important sense we are all constructivists now. The vast majority of International Political Economy (IPE) scholars would readily agree that interests are not natural or pregiven but constructed and bound up with identities; that ideas have a certain degree of independent causal efficacy; that values are not elements in a transcendent normative order but contingent social principles; that instrumental rationality is a historically specific institution; that the ways in which humans reflect on their own practices has a constitutive effect on those very practices; and that a social science worth its name should not approach its object as a collection of brute data but require a minimal degree of hermeneutic sensibility. Disagreement with such propositions is increasingly considered reflective of an oddly doctrinaire mindset, be it of a structuralist, rationalist or positivist persuasion. The major contribution of constructivism is to have brought this philosophical theme of the constructed nature of institutional facts (their observer-dependent character) into the mainstream of IPE.
And yet, it is not always clear what substantive difference constructivism has made: all too often, it is hard to avoid the impression that much has remained materially the same in the wake of the constructivist intervention. If we take Robert Gilpin’s categorisation of IPE theories in The Political Economy of International Relations (Realist, Liberal Institutionalist, Marxist) as a key point of reference we might say that the differences between each of these approaches can be expressed in a limited number of specific propositions about the relationship between states and the global economic system. Each of them suggests a particular angle from which we should look at reality. But it is hard to say the same thing about constructivism: somehow, the notion that human institutions are “made” does not automatically generate a distinctive lens through which to view the world: it does not give us a set of guidelines that tell us what kinds of actors or institutions we should foreground when we try to make sense of historical processes. That is, of course, not to preclude the possibility that such an angle might still emerge, but for the time being it is striking that those who adopt a broadly constructivist methodology have tended to generate substantive analyses that are remarkably in tune with the core tenets of liberal institutionalism (illustrated by the intense concern among leading constructivists with the role of international organisations).
This problem is illustrated by the responses that the constructivist challenge has tended to generate: those who are committed to the substantive claims of realism or marxism often find it hard to see what all the fuss is about. Sure, social facts are constructed – so what? One response to such concerns about the lack of distinctiveness of the constructivist approach would be to say that it provides a welcome opportunity to get away from seemingly unresolvable debates and that the value of constructivist methods consists in their ability to provide a firm methodological grounding for socially relevant empirical research. This is certainly the direction that the constructivist project seems to have taken in recent years: constructivist IPE is increasingly a combination of somewhat inconsequential meta-theory, on one hand, and empirical research conducted in a positivist frame, on the other hand, treating as brute facts the very phenomena that it claims are discursively and socially constructed, performative and observer-dependent.
This is an awkward direction for a project that takes itself to be deeply concerned with the constitutive powers of human reflexivity. It perhaps explains the somewhat defensive attitude that increasingly accompanies the constructivist project in IPE. Many constructivist articles and books still start with a list of all the naïve assumptions that they reject. In a review of the edited volume Constructing the International Economy, John Boli expresses concern in the journal Contemporary Sociology with the fact that most of the contributions:
are overly concerned with justifying the ‘sense and sensibility’ of the constructivist approach in the abstract. Norms matter; discourse has effects; material interests are socially constructed, not given in the nature of things; interests are less stable than rationalists assume. This form of argumentation, particularly when repeated (albeit in different terms and from different angles), becomes rather tiresome.
It would be hard to argue that this defensiveness stems from a general hostility towards the constructivist project: the more common attitude from those who are not on board is indifference or puzzlement, and in the meantime constructivists have been remarkably successful in terms of institution-building and paradigm-making. This suggests that the defensiveness has intellectual rather than sociological origins.
Perhaps we can shed some light on the problem by noting that there exists a certain odd self-limitation at the heart of the kind of IPE constructivism: mainstream constructivist scholars have remained highly reluctant to embrace the idea that everything is socially and discursively constructed. We may take some guidance here from scholars who are generally sympathetic to the constructivist approach but have suggested ways in which it can be enriched and moved forward. Wesley Widmaier’s article in Millennium: Journal of International Studies on the social construction of crises suggests that constructivism has so far focused too strongly on the role of ideas and cognitive frameworks and needs to do more to concern itself with emotions (such as the anxieties and resentments generated by economic problems and the way these are exacerbated by crises). Charlotte Epstein in the European Journal of International Relations has argued that constructivism has worked with problematic notions of identity and that a shift is required to the reflexive and interactive dynamics of identification and the way this revolves around an element that always eludes the constructive effects of norms, values and discourse (the Lacanian real). On one hand, these scholars hint at a constructivism that goes deeper (if perhaps not all the way down). On the other hand, they suggest that this would have to operate with a more complex understanding of the dynamics of construction. In other words, they express a concern with the artificial “neatness” of the constructivist perspective and its eagerness to re-bifurcate the world into objects and subjects just after we discovered intersubjectivity and constructedness.
To my mind, these are convincing arguments. But of course they chip away at the promise of formal and quantitative rigour, and in this respect they very much go against the general direction of the constructivist project in IPE. Key here is of course the concern that allowing for the possibility that construction is involved at all levels of reality-making would land us in the muddy methodological waters and pointless sophistry of postmodernism. And this in turn reflects the conviction that in the end constructed entities are less real or objective than natural or material facts. But this raises an important question: if “construction” only results in weak coherence and low levels of facticity, why should it be central to our theorising in the first place?
I would like to focus therefore on what exactly the constructivist claim is. The constructivist communicates to others (both scholars and social actors) that they are in the grip of a certain essentialism, that the phenomena that they think are natural, pregiven, objective are in fact constructed, made by human hands or minds. The constructivist points to a phenomenon in the world and says to others, “You think this is a simple, self-sufficient, objective fact, but you’re wrong: it’s socially constructed, composed of different heterogeneous elements; it’s not unitary but consists of ideas, interests and a whole range of other elements that you have lost track of”. To employ the language of actor-network theory, the constructivist says, “This is a complex assemblage that you have come to treat as a black box”. In other words, the central claim of constructivism is that the way in which other scholars and human actors relate to the world involves some kind of idolatry: we are seen to have forgotten our own role in the making of the object and now attribute to it intrinsic powers and capacities. The constructivist reminds us that we are the ones who bestowed those powers on the object in the first place through the way we were involved in its assembly.
So the force of the constructivist argument is dependent on us having forgotten that the facts of social life are conventions, contingent, cobbled together from heterogeneous materials. But it not clear that this is really how we relate to institutional facts. In fact, we generally have very little difficulty acknowledging that the kinds of phenomena that IPE studies—regulatory institutions, states, banks, corporations, contracts, markets—are far from natural, pre-given or monolithic but precisely constructed, internally complex and dependent on our ongoing ability and willingness to follow institutional rules and perform functional roles. This is why Realists and Marxists have so far failed to be suitably impressed by the argument that states and markets are social constructions: they never really denied that they were. In that sense, it’s not just that we’re all constructivists, but that there never was a time when we weren’t constructivists.
But if we are in this sense all constructivists, the point is that, like the Realist and the Marxist, we do not really view this as diminishing the facticity of these phenomena. This is key to our relationship to the modern fact: we are capable of seeing it as both a self-contained, autonomous phenomenon with a coherent and self-evident identity, and as something that has been built over time, is complexly layered and would crumble tomorrow if we collectively decided to no longer believe in it; both as a thing in itself and a contingent configuration of connections. We can view something as either a complex network or as a coherent actor. We ‘bracket’, in Anthony Giddens’ terms in Central Problems in Social Theory, or treat a complex historical process as a unitary identity. We treat phenomena as black boxes even though we are perfectly aware that that is what we are doing: it is not that we do not know there are things inside the black box that have historically shaped its characteristics; it’s just that we feel that knowing what exactly those things are would not make any difference to how we relate to the black box. A Realist does not deny that national states have complex histories, merely that uncovering these would add nothing consequentially new to what we presently understand to be the national interest.
This is not naïveté, but a substantive hypothesis about what the relevant units of social life are, the appropriate starting-points. It involves a productive move and is the stuff of conceptual progress: if we were always forced to return things to their most elementary, atomic level, we would never be able to learn anything about anything. As Bruno Latour has argued in We Have Never Been Modern, moderns are forever engaged in the twin projects of, on one hand, creating complex new networks and assemblages and, on the other hand, purifying them; constructing new configurations and finding ways to treat them as entities in themselves. While this process involves elements of disavowal and “forgetting”, it is important to understand it in the first instance as a productive moment that builds new capacities. Purification plays an important role in facilitating the construction of complex heterogeneous entities: things would quickly become unwieldy if our networking practices did not somehow ‘cluster’, produce new coherent facts with specific identities. This involves some kind of forgetting, but what we forget is not really the constructed nature of our own creations. What we lose sight of is not so much the fact of constructedness but its precise modalities and configurations, i.e. the how of constructedness. Certain aspects of public authority come to be consolidated in such a stable social construction that we can refer to it as “the government”, but it doesn’t take all that much for us to realise that it is shorthand for a complex constellation of institutions, norms and practices. It’s just that this realisation by itself does little to improve our understanding of how the government works.
Constructivism’s theoretical contributions have focused heavily on emphasising the fact of constructedness. Clearly there is a worthwhile political impulse here: a key concern of constructivist IPE has been to assert the contingency of what in the wake of neoliberalism and the collapse of communism had come to appear as the hard facts of capitalist life, and so to identify opportunities for politics and agency in a world increasingly governed by the dismal logic of markets. Constructivists tend to stress the constructed nature of things precisely when they want to emphasise their contingent and changeable nature. The constructedness of something is seen to reduce its degree of reality, to make it less than a natural or material fact. The constructed nature of a phenomenon is taken to mean that it could have been otherwise. The problem is that this has given constructivism a strongly “counterfactual” bent: it has tended to be concerned more with the difference that agency might make than the difference that it has made. And as long as the claim that reality is socially constructed is taken as emphasising its changeability, we will have little choice other than to impose strict limits on the phenomena that we can consider to be “socially constructed” and return to positing the existence of hard, pre-discursive facts.
This is not to deny that constructivist authors have investigated the actual processes whereby social facts are constructed. But it is striking how abortive this project has been, which explains the sense that there is nothing new on offer. Constructivist research tends to take one step back and disaggregate a particular institutional actor in terms of the ideas, interests and identities that have gone into its making; but these factors have quickly ended up being treated as givens (data) without a history of construction. In this way, the constructivist project has taken on a distinctly positivist flavour, except that there is now a sense that somehow agency, politics and morality still matter. Thus, on one hand, we have things that could have been otherwise, and on the other hand things that are just the way they are. The facts of social life are polarised into idols (things that we believe in and so acquire an ultimately contingent and changeable existence) and brute facts (things that exist and that is why we believe in them).
For Latour, in On the Modern Cult of the Factish Gods, the idea that idols / fetishes and facts are essentially different things is the ultimate modern conceit, an inability to recognise that networking and purification have always gone hand in hand. All we have are “factishes,” phenomena that have been constructed and in the process have attained a certain degree of reality or coherence that is not evident from their constituent parts. These constructions go beyond the purposes and intentions of their makers and so are generative of real effects and novelty. The idea that some things are just “made” through sheer intentionality is as much a fantasy of the modern mind as the idea that other things are just out there, waiting to be discovered. The signal discovery of constructivism should have been the discovery of factishes, not a return to a world that is cleanly divided between objects and subjects, between objective facts and subjective interpretation. In this sense, constructivism in IPE offers an unreflexive manifestation of the duality of the modern fact rather than an incisive analysis of it.
This, to my mind, is what the arguments of Widmaier and Epstein hint at, and it is something that a constructivism that goes “all the way down” could potentially come to terms with. But of course the actual direction of constructivist theory in IPE is in the diametrically opposite direction, namely to uphold the commitment to a distinction between material facts and idols. Although I have relied on the ideas of Latour here (also see the interview with him on ‘Modernity is Politically a Dangerous Goal’), I certainly don’t mean to imply that constructivism should turn itself into a branch of actor-network theory. Perhaps the real irony in the constructivist IPE project is that it has so little sense of its own intellectual lineages that it systematically neglects highly pertinent contributions—famous works that come to mind here include Peter Berger and Thomas Luckmann’s The Social Construction of Reality and Richard Bernstein’s Between Objectivism and Relativism—that would make the tendency to quickly relapse into subject-object dichotomies seem deeply problematic. In that sense, constructivism in IPE is rapidly becoming notable above all for its demonstration of the effectiveness with which neoliberal academia can take a critical impulse and turn it into the next paradigmatic fad.