Robert Cox began his canonical 1981 essay “Social Forces, States and World Orders: Beyond International Relations Theory” with the observation that it is “necessary and practical” for academic disciplines to “divide up the seamless web of the real social world”. We make these divisions, Cox wrote, in order to analyse the world and thus to produce practical knowledge of that world. It is not a stretch to suggest that the real social world of International Relations scholarship might also be approached as worthy of analysis and theory. Indeed, reflection on International Relations as theory appears in the field as part of the necessary and practical division of the complexity of the social and political world. Rare is the introduction to IR textbook that does not emphasize, and usually begin with, the “great (theoretical) debates” that have structured the field since it emerged as an academic discipline.

Thus theory itself has long and often been treated as an object for theoretical reflection in International Relations. Recently, we could point to both the founding of a specific section of the US-based International Studies Association dedicated to Theory – indeed, this section honoured Professor Cox at the ISA convention in Toronto in 2014 – as well as the special issue in September 2013 of the European Journal of International Relations on “The End of IR Theory?”, accompanied by a wide-ranging discussion of the papers collected there in the Duck of Minerva blog.

For some of the contributors to the special issue and to the debates on the symposium – and I hope I can be forgiven for making an impressionistic observation rather than an analytical one – it seems that the question mark at the end of “the end of IR theory?” was a sign of fatigue rather than a sense that the debates over theory needed to be renewed. Fatigue, in the sense of “are we still having this conversation?” or “haven’t we moved on yet?” Emblematic of this fatigue were the reflections of Professor Chris Brown in both his article in the EJIR and his contribution to the symposium. It’s not that Brown is hostile to theory; on the contrary, his contribution was a complaint that the critical theories that emerged in the 1980s had not fulfilled their potential and that problem-solving theory had contributed much more.

I disagree with his assessment of the status of critical theory, as do many of the contributors to the special issue and to the symposium, in particular with his claim that it has failed to live up to the promise it showed in the 1980s – but that’s a conversation for another time. What is most interesting here is how Brown takes up Cox’s analytical division of “critical theory” from “problem-solving theory”. Indeed, the trope of problem-solving versus critical theory is asserted quite often in discussions of the status of theory in IR: for example, in A. C. McKeil; in Robert W. Murray; or in Ali Diskaya, just to take a few examples appearing here in e-IR. It is this trope, along with Cox’s other oft-cited claim in the 1981 Millennium article that “theory is always for someone and for some purpose” (Cox 1981: 129) that have made his article canonical. Indeed, Cox’s categories of “critical” and “problem-solving” are now part of the very common-sense ordering of theory in IR.

It is therefore important to consider what Cox actually said about these categories. Problem-solving theory, first, “takes the world as it finds it, with the prevailing social and power relationships and the institutions into which they are organised, as the given framework for action. The general aim of problem-solving is to make these relationships and institutions work smoothly by dealing effectively with particular sources of trouble” (Cox 1981, 128-129). This definition, remember, follows Cox’s opening statements in the article about the importance of theory for the production of practical knowledge. Cox is often interpreted as elevating critical theory over problem solving theory – Brown takes him to do so, for example, in the symposium when he says Cox “compared ‘problem-solving’ theory unfavourably with ‘critical theory’” but I am not convinced that a careful reading of Cox’s article supports this (and Cox argues something similar). In addition to signalling the importance of theory for practical knowledge, Cox explicitly notes, for example, how the analytical procedures he sees as defining problem-solving theory are the source of its strength. He takes issue with the idea that problem-solving theory is value-free and asserts that it is conservative (Cox 1981: 129-130) but this is as close to a normative assessment of problem-solving theory as Cox gets.

Critical theory, in contrast, is holistic where problem-solving theory is analytic. It “does not take institutions and social and power relations for granted but calls them into question by concerning itself with their origins and how and whether they might be in the process of changing. … Critical theory is directed to the social and political complex as a whole rather than to the separate parts” (Cox 1981: 129). Cox also allows that where problem-solving theory might be seen as conservative, critical theory might be seen as utopian: “Its aims are just as practical as those of problem-solving theory, but it approaches practice from a perspective with transcends that of the existing order. … Critical theory allows for a normative choice in favour of a social and political order different from the prevailing order, but it limits the range of choice to alternative orders which are feasible transformations of the existing world” (Cox 1981: 130).

I suspect that it is this holistic and utopian range for critical theory as asserted by Cox that leads Brown to identify critical theory with Quentin Skinner’s notion of Grand Theory and leads so many of the rest of us to assume that critical theory is posited as a superior theoretical approach to problem-solving theory. I’ve tried to show that Cox, at least, makes no explicit claims to that effect. Nevertheless, the categories Cox bequeathed to us seem to encourage us to turn this binary, critical versus problem-solving theory, into a hierarchy. Readers of International Relations theory instinctively want to read it as political theory – though as Rob Walker might remind us, this does not make us careful readers of political theory either. Perhaps we need to acknowledge that the theoretical choice presented by the binary critical theory-problem-solving theory is, at least, theory as engaged in a political contest.

Should Cox’s categories be preserved, critiqued, or abandoned? Would doing so lend a clearer (and practical) view of theory in the field? One way to extend the engagement with the practice of theory would be to read Cox’s categories alongside the other famous definition of critical theory, in Max Horkheimer’s essay “Traditional and Critical Theory” (Horkheimer 2014 [1937]). Though Cox did not explicitly cite Horkheimer, the latter’s efforts to distinguish critical theory from what he called “traditional theory” make him an obvious interlocutor. (There are, of course, many other serious contributions to the effort to situate theory socially and politically, from Bourdieu to Foucault, from Bruno Latour to Walter Mignolo. Many of these efforts are considered in the EJIR special issue and the Duck of Minerva symposium.) I focus on Horkheimer here in part due to the clear affinities between his approach and Cox’s and in part because Horkheimer, like Cox and as I would like to do, provides an explicit defence of the enterprise of critical theory.

While the scale of Horkheimer’s critique, engaged as he was with theory per se, is much grander than the stage Cox builds when he focuses on IR theory in the 1981 article, their accounts of “traditional” and “problem-solving” theory are remarkably similar. Cox notes how the power of problem-solving theory stems from its methodological “fixing” of the social and institutional parameters surrounding the variables it examines. For Horkheimer, this method, rooted in Descartes and predominant not only in social sciences but in science generally in his time, stems from the ability of the scholar to abstract him or herself from these social and institutional parameters in the production of theories and analyses. In other words, just as in any production process where a division of labour separates the subjective functions of planning, designing, interpreting, and analysing from the executive functions, traditional theory renders the world under study as objective and passive and the scientist an active, analysing subject.

In their conceptualisations of critical theory, however, Cox and Horkheimer differ slightly but in an important way. While both are concerned to defend theory as an approach to a dynamic and interconnected totality, Cox does not foreground the status of the theorist, while for Horkheimer the critical theorist must engage with theory as a productive process. Cox does take neorealism to task for neglecting the production process in the constitution of national interest (Cox 1981: 134-135) but Horkheimer goes further: it is not a matter of adding another parameter or variable to the theoretical enterprise; it is a matter of understanding the theoretical enterprise itself in relation to and as a part of a general production process and division of labour. When Cox wrote in 1981, the prevailing epistemology in IR and the epistemological commitment of problem-solving or traditional theory was realist: the world exists independently of our thoughts about it and the task of theory is to make thought adequate to reality. What Horkheimer shows is that there is no neat division between thought and reality that can justify the privileged position of the theorist in the social division of labour: our thoughts are part of reality, as real as the city you live in or the job you work at and they must be analysed as part of the general social division of labour and of social reproduction.

Thus the problem-solving theorist becomes a functionary in the maintenance of social order. The critical theorist must understand the role of theory in social reproduction in order to break down the divisions between theoretical reflection and the making of the world.

I am not suggesting that Horkheimer was right where Cox was wrong. Cox was certainly aware of the – explicit or implicit – political commitments of the theorist when he said, “theory is always for someone and for some purpose”. The question for me here is whether our common-sense taxonomy of International Relations theory as “problem-solving” or “critical” remains appropriate. Given how embedded it is in our ways of seeing ourselves and our field, and given that it can be quite useful for teaching theory, I don’t think it should necessarily be abandoned. Given the implicit, instinctual way IR theorists tend to treat the division as a contest, as in Chris Brown for whom problem-solving has produced better results than critical theory, I wonder if it would not be better to try to make the political stakes of that contest more evident.

For me, reading Cox with Horkheimer provides an interesting start on this task. The description of the first category as “problem-solving” or “traditional” points to a theoretical practice where expertise rules, where specialists take up their specific tasks and succeed or fail on the basis of how powerful their explanations are and the impact of their work. Cox and Horkheimer both acknowledge the importance of this approach in terms of method and results. And they both hint at the cost: politically, we might better describe this approach as technocratic theory. The enterprise is to uncover the timeless essences of things and relations and to keep things working by keeping them in their place.

The contrasting approach to theory, which Cox and Horkheimer both dub “critical theory”, seeks instead to enable the transformation of things. But critical theory is more than this, too; after all, as social constructivism suggests, the transformation of things is just the normal state of affairs as people make themselves and make their history. Critical theory does more: it disagrees. As Rancière suggests, it disrupts the order of the “distribution of the sensible.” Critical theory works by making visible the relationships and the things that International Relations refuses to recognise and qualify as relationships or things. It makes audible the voices of people not qualified to speak in International Relations. Against the technocratic barriers to international living and understandings, critical theory identifies the arbitrariness and artificiality of barriers and explains them in relation to their roles in the division of labour, social reproduction, or system maintenance. Politically, critical theory must be democratic theory – not a theory of democracy, posed externally to its object, but a theory that is democratic in its everyday practices. These are the political stakes in our theoretical choices.

Further Reading on E-International Relations

• Alienation and Marxism: An Alternative Starting Point for Critical IR Theory 
• The ‘Failure of Critical Theory’ as an Ideological Discourse
• Reflections on Critical Theory and Process Sociology
• Why Is There No Minor International Theory?
• Rethinking Critical IR: Towards a Plurilogue of Cosmologies
• Against Mystification, or What Went Wrong with Critical IR