Theory and history in International Relations

Theory and history in International Relations (IR) stand in a relationship of co-dependence, and like all such relationships, mutual attraction and affirmation are tempered by anxiety, ambivalence, and denial. ‘History isn’t theory’. ‘Theory is an abuse of history’. We have all heard it. Yet two things seem undeniable. Theory is history-dependent. If theory is always ‘from somewhere’, then the universal ‘somewhere’ is history. Not only because theorists write in historical contexts they cannot escape, but because theories invariably rest on historical propositions. The most fundamental of these concern whether international relations, as a distinctive realm of human interaction, is characterized by continuity, change, or both: all of which are historical propositions. Yet if this is true, so too is the reverse. History is theory-dependent. Carr (Reference Braumoeller1961, 11) famously observed that ‘a fact is like a sack—it won’t stand up until you put something in it’. It is the historian, he argued, ‘who decides to which facts to give the floor, and in what order or context’. If theories are, among other things, assumptions that enable us to make sense of complexity, then all acts of historical selection and interpretation are theoretical, even if not explicitly so. But in IR, a field with a particular penchant for theory, the theoretical dependence of history is especially pronounced. We read history through our concepts (some facts are ‘international’, for example), and our theoretical commitments affect our interpretive strategies. If theory is about causal explanation, history is read one way; if it is about conceptual exegesis it is read another.

The principal contribution of this symposium is to bring this co-dependent relationship between theory and history to the fore. At first glance Buzan and Lawson’s The Global Transformation is an unlikely candidate for a symposium in International Theory. It is, first and foremost, a work of macro-historical sociology. Initially, the authors admitted to offering no ‘novel theoretical argument’ (a statement they now regret), and their conceptual innovations, such as the ‘mode of power’, are highly productive promissory notes, inviting further elaboration (Buzan and Lawson Reference Bilgin2015, 10). Yet, in mounting their sweeping case for the long 19^th century (1776–1914) as the epochal transition in international relations, the period that ‘profoundly influenced the modern international order’ (Buzan and Lawson Reference Bilgin2015, 44), Buzan and Lawson have provoked among my fellow contributors a set of reflections that together betray the highly variegated ways in which today’s IR scholars build theory on history and, conversely, read history through theory.

That Buzan and Lawson’s book provokes such responses is hardly surprising. Defining origins is as much a penchant of IR as theorizing is: indeed, the two are deeply connected. Making the case for one origin over another is a theoretical act (we identify the origin of something we have already conceived), and defining an origin constitutes the object we seek to theorize (a system, structure, society, etc.). For Buzan and Lawson, the 19th century is the origin of global modernity, the structuring context of the contemporary international order. But for others, the origins are the 16th century (the modern world system), the 17th century (the system of sovereign states), the 18th century (the structure of great powers), and the 20th century (the American world order) (see Hinsley Reference Buzan and Lawson1967; Wallerstein Reference Reus-Smit and Snidal1974; Holsti Reference Carr, Edward1991; Ikenberry Reference Cox2011). Historical origins both reflect conceptual commitments and constitute the subject of theoretical inquiry, and bold acts of epochal declaration like Buzan and Lawson’s are lightening rods for debate over theory and history.

The following contributions engage The Global Transformation from multiple directions, applauding its many strengths and exposing the flaws and blind spots that invariably characterize a work of such ambition. What interests me, though, is the varied assumptions about theory and history that undergird these responses. Both faces of the co-dependence of theory and history are evident: theory’s dependence on history, and history’s dependence on theory. Examples of the former include the dependence of Buzan and Lawson’s concept of the mode of power on the historical proposition that international orders are the product of wholesale, epochal transformations, and Musgrave and Nexon’s (Reference Rosenau2016) call for ‘exotic international theory’, which depends on the historical proposition that today’s drivers of international change are not simply downstream products of a 19th century great transformation. The contributions reveal more, however, about the second face of the co-dependent relationship: history’s dependence on theory. In responding to Buzan and Lawson’s declaration of the long 19th century as the origin of modern international relations, they express very different conceptions of change and understandings of theory. These have two implications for IR theory more broadly. They demand a richer mapping of approaches to change and theory than our existing dichotomous distinctions allow. More importantly, they challenge the long-established distinction between systems and systemic change, which undergirds so much of our theory, and prompt debate about what constitutes proper theorizing in IR.

Conceptions of change

In the past two decades, IR scholars have become increasingly interested in the dimensions and dynamics of international change. The old notion of international relations as the realm of ‘recurrence and repetition’ has been rattled by a tightly compressed sequence of developments, all suggestive of change: the end of the Cold War, accelerating globalization, the flip from ‘unipolarity’ to ‘the rise of China’, the resurgence of transnational violence, the advent of virtual politics, etc. Yet, much of what we say about international change remains indebted to Gilpin’s (Reference Buzan and Lawson1981) oft-quoted distinction between ‘systems’ and ‘systemic’ change: the former, change from one international system to another (e.g. medieval heteronomy to a system of sovereign states); the latter, change within a system (e.g. shifts in polarity). For most of the field this distinction usefully fences off their primary interests – the rise and fall of great powers, change in issue-specific institutions and organizations, regional politics, etc. – from larger macro-historical transformations. And for those interested in such transformations, it renders systems change compatible with the prevailing structural mode of theorizing. If the behaviour of states is conditioned by the material or institutional structures of an international system or society, then the macro-historical transformations that matter involve the rise and decline of such systems or societies, understood as structural orders. Gilpin’s project was realist, but this way of conceptualizing change also informs Cox’s (Reference Bull and Watson1986) distinction between critical and problem-solving theories (the former concerned with where a system came from and how it might be transformed, the latter with a system’s management), Ruggie’s (Reference Philpott1993) account of the shift from heteronomy to sovereignty, and constructivist and English School accounts of the development of modern international society (Bull and Watson Reference Biersteker and Weber1984; Reus-Smit Reference Musgrave and Nexon1999; Philpott Reference Kuhn2001).

Yet, the adequacy and coherence of Gilpin’s dichotomy is tested by the contributions to this symposium. Let us assume, to begin with, that Buzan and Lawson, as well as my fellow contributors, are all concerned with systems change: the emergence of global modernity, and its associated international order, out of some prior system or systems. This leaves open the question, though, of how such change unfolds. It is useful here to distinguish between evolutionary, breakpoint, and processual modes of change. Rosenau distinguishes between a system’s parameters, its ‘basic rules and organizing principles’, and its variables, which ‘consist of both the nonhuman dynamics (such as technology and agricultural resources) and the human structures and processes (such as the composition and objectives of nation-states) through which collectivities interact’ (Reference Phillips1990, 79). Evolutionary change occurs when shifts in a system’s variables lead to adjustments in its parameters, usually in ‘some constant additive or subtractive pattern’ (Rosenau Reference Phillips1990, 82). Breakpoint change, by contrast, involves dramatic, system transforming instability, as ‘the parameters undergo a transformation that pushes their values beyond the breakpoint threshold’ (Rosenau Reference Phillips1990, 82). Processual change is emphasized by diverse process sociologists, from Norbert Elias to Charles Tilly, and finds expression in a range of recent IR scholarship (Nexon Reference Krasner2009; Linklater Reference Ikenberry2011, for example). Here, change is understood as multivariate and constant, the product of shifting networks of agential relations that coalesce in overlapping social ‘assemblages’ or ‘figurations’.

For my fellow contributors, The Global Transformation is about breakpoint change. The long 19th century produced a new social, economic, and political order – global modernity – which transformed the nature of international relations: an old system was displaced by a new one. In the book, Buzan and Lawson insist that this transformation unfolded over a century, and that it was dependent on a long chain of prior developments and transformations. And in their reply in this symposium they argue that they are actually concerned with all three modes of change: evolutionary, breakpoint, and processual. For most readers, however, it is the breakpoint features of their argument that will standout: ‘The nineteenth century is thus close kin to the twentieth and twenty-first century, and quite distinct from previous periods in world history’ (Buzan and Lawson Reference Bilgin2015, 5).

All of our contributors accept that far-reaching social, economic, and political changes unfolded across the nineteenth century: how could one argue otherwise? They are divided, however, over whether this amounted to a breakpoint change. Several accept Buzan and Lawson’s thesis, choosing to highlight its analytical utility or expose its neglected dimensions. Braumoeller does the former, arguing that by identifying the origins of the modern international system, Buzan and Lawson define ‘the scope conditions within which more specific theories of international interaction are operative’ (Reference Ruggie2016). Patricia Owens is more critical. While agreeing that the 19th century constituted a breakpoint change, she criticizes Buzan and Lawson for failing to grasp key dimensions of this change; most notably, the emergence of the idea of the ‘social’, a development she associates with new governance practices of regulation, pacification, and de-radicalization (2016). Buzan and Lawson’s (Reference Ruggie2015) text is peppered with phenomena they term ‘social’, but for Owens they fail to understand how the identification of a social realm, made up of social actors, social institutions, social structures, and social practices, was intimately connected to, and enabling of, distinctly modern forms of rule. Bilgin’s position is similar, accepting that the 19th century constituted breakpoint change, but critical of blind spots in Buzan and Lawson’s (Reference Ruggie2015) account, in this case their insufficient attention to contending conceptions of the international.

Our remaining contributors adopt a very different conception of change to Buzan and Lawson, subsuming the transformations of the long 19th century within longer processes of change. Musgrave and Nexon, Anievas, and Phillips all locate Buzan and Lawson’s triad of ‘interlinked processes’ – industrialization, the rational state, and ideologies of progress – in the antecedent dynamics of a prior global transformation. Anievas criticizes Buzan and Lawson for their ‘precarious engagement with the longer-term (often structural) sources of the dynamics they conceive as driving the global transformation’, a consequence, he argues, of their ‘contingent-based, conjunctural framework’ (Reference Skinner2016). Casting industrialization, the rational state, and ideologies of progress as ‘a contingent concatenation of social forces’ prevents Buzan and Lawson from grasping the pre-existing structural forces that generated this crucial conjuncture. Musgrave and Nexon are less concerned with the play of large-scale structural forces than Anievas, but they too challenge Buzan and Lawson’s neglect of the pre-19th century transformations. The European political order experienced a pre-industrial expansion that constituted ‘a genuinely global transformation only loosely connected to ideologies of progress or industrialization’, and by failing to untangle the ‘unique causal chains’ that link this earlier transformation to 19th century developments, Buzan and Lawson (Reference Ruggie2015) are left with a description, not an explanation, of the rise of global modernity (2016). Phillips enlists Buzan and Lawson’s conception of ‘configurational change’, but insists that 19th century changes were a product of an earlier configurative transformation, the effects of which still structure world politics. The ‘world that the global transformation re-shaped was already marked by considerable global connectivities’ (Phillips Reference Swedberg2016). It was also one that had already spawned regionally specific ‘multiple early modernities’ (Phillips Reference Swedberg2016).

In some of these accounts, change looks more evolutionary, in others more processual. Yet none are easily categorized. If evolutionary change occurs when shifts in a system’s variables lead to adjustments in its parameters, Anievas’ emphasis on structural processes inherent to capitalism may be more evolutionary, while Musgrave and Nexon’s argument is more clearly processual. What matters here, though, is that the further we move away from a breakpoint conception of change, what constitutes an international system, order, or society becomes more opaque to the point of disappearing in some accounts. To be sure, Buzan and Lawson are themselves unclear about what precisely they mean by terms such as ‘international system’ or ‘international order’. But while this may result from a simple lack of conceptual clarity, with Musgrave and Nexon, Anievas, and Phillips it is a function of their understandings of change. Anievas emphasizes the structural connection, and dynamic interplay, between capitalism and industrialism on the one hand, and the core-periphery world market and the rational state on the other, which arose through the uneven and combined development of global modernity. The ‘system’ here is always emergent, the ongoing product of evolving structural forces. Phillips embeds the global transformation of the 19th century in an earlier Eurasian transformation. The ‘industrious revolution’, absolutist state building, and large-scale normative pacification of this earlier transformation pre-structured later developments: Eurasian multiple modernities produced diversity in global modernity. Phillips is explicit that this entailed a shift in systems – a revolutionary change did occur. But he also argues that this change occurred through reconfiguration, that it was processual not breakpoint (Phillips Reference Swedberg2016). These positions sit together uncomfortably, though. The stronger the processual argument, the harder it is to talk about systems as discrete entities. This is most clearly apparent with Musgrave and Nexon. Closest to the canons of processual sociology, they resist Parsonian conceptions of system, focussing instead on variegated networks of agential relations, forged, sustained, and eroded by diverse, yet concrete causal processes (Mjusgrave and Nexon 2016).

Theorizing and theory

IR is a field that privileges theory, but it is also a field that spends little time reflecting on the nature of theory or the process of theorizing. Indeed, just as our thinking about change is shackled by Gilpin’s old systems/systemic distinction, our reflections on theory are caught in a worn set of distinctions between empirical (positive) vs. normative theory, critical vs. problem-solving theory, and explanatory vs. interpretive theory. The precarious nature of these distinctions is well understood. All empirical theories include normative assumptions, and all normative theories build on empirical propositions (Reus-Smit and Snidal Reference Nexon2008). Critical theories claim to aid human freedom, but problem-solving theories can have these effects as well. Explanatory theories are said to address ‘why’ questions, while interpretive theories answer ‘how’ questions. Yet, how and why questions are almost always connected with one another, with answers to the former contributing to answers to the latter, and vice versa (Price Reference Linklater1994). The problems extend beyond the blurring of routine dichotomies, however. These categories of theory do not exhaust the diverse forms of theory and theorizing currently undertaken by IR scholars, something clearly apparent in the responses of my fellow contributors to The Global Transformation.

Let me focus on different ideas of theorizing expressed in these responses, as these imply different understandings of what constitutes theory. To state the obvious, theorizing is a process, and theory is the end product of this process. ‘While theorizing, then, can be playful and full of movement’, Swedberg observes, ‘theory signifies the freezing of a set of thoughts. Theory is frozen into written language or a symbolic language such as mathematics. While theorizing never ends, theory is stuck forever in its formulation’ (Reference Reus-Smit2012, 15; also see Swedberg Reference Price2015). Swedberg criticizes his fellow sociologists for neglecting the art of theorizing. They have become very good at methods, and perhaps methodology, but theorizing has been reduced to choosing between pre-existing theories, as one chooses cereal in a supermarket. IR is vulnerable to precisely the same criticism. Seldom do we reflect on what theorizing might entail, and in our practices we treat it in much the same way as we treat methods, as preconstituted tools in a toolkit, and the task of theorizing is simply choosing the theoretical tool that best fits the problem, the one with the greatest analytical leverage or heuristic power. Yet, the most important differences between my fellow contributors lie not in their preferred theories, but in what they imply is proper theorizing.

Swedberg seeks to steer sociologists back to theorizing by breaking the process down into a series of stages, insisting throughout that theorizing is a creative, imaginative process, not a scientific one: ‘[C]reativity is primarily what matters when a theory is devised; and scientific logic and rigor is primarily what matters in the context of justification’ (2012, 16). Persuaded as I am by this, Swedberg’s specification of the process misunderstands a key component of theorizing, a component important to understanding our contributors’ responses. Swedberg identifies four stages in the process of theorizing: observation, concept formulation, the ‘building out’ of a theory, and the development of a final theoretical explanation (Reference Reus-Smit2012, 170). Lying behind this is the assumption that theorists observe empirical facts with clear minds, and that the process of conceptualization follows observation. Concepts here are theoretical responses to the observed: the observed is not determined by theorists’ concepts, and valid concepts refer only to the observable.

This clearly sets the cart before the horse. It is undoubtedly true that theorists often hone their concepts in response to their empirical findings. The recent shift from categorical conceptions of sovereignty to conceiving it as a practically constituted institution is a good example (Biersteker and Weber Reference Anievas1996; Krasner Reference Hinsley, Francis1999; Reus-Smit Reference Musgrave and Nexon1999; Philpott Reference Kuhn2001). Yet, this all occurs after a process of conceptualization has already determined how the empirical world is observed – observation is concept-dependent. This has been noted many times, by many scholars. Kuhn stressed the theory-dependence of observation, how the pre-existing conceptual and theoretical beliefs of scientists affect how they observe empirical phenomena: ‘No wonder, then, that in the early stages of the development of any science different men confronting the same range of phenomena … describe and interpret them in different ways’ (Reference Holsti1970, 17). The same is true of historical observation. As Skinner observes, an historian observing the art in a stately home will observe only things consistent with her conception of what constitutes art: another historian, with a different conception of art, will see different things. One might see the furniture, one might not (Skinner Reference Lukes2002, 20).

This argument about the primacy of conceptualization in the process of theorizing can be taken one step further. Not only does conceptualization condition observation, theorizing frequently involves conceptualizing phenomena we cannot observe. Lukes opened his oft-quoted mediation on power with precisely this point: ‘[W]e need to think about power broadly rather than narrowly—in three dimensions rather than one or two—and that we need to attend to those aspects of power that are least accessible to observation: that, indeed, power is at its most effective when least observable’ (Reference Keohane2005, 1). Conceptualizing the unobservable is in fact standard fare in IR, just as it is in other social sciences. Many of the most fundamental conceptual building blocks of our theories – interests, norms, structures, power, etc. – cannot be observed directly and require artful methodological strategies for their analysis.

A persistent theme in the following articles is Buzan and Lawson’s failure to theorize the global transformation. Yet, my fellow contributors betray very different views about the kind theorizing Buzan and Lawson should have undertaken, views hinging very much on the priority each attaches to conceptualization in the process of theorizing.

When Musgrave and Nexon, Braumoeller, Anievas, and Phillips criticize the limits of Buzan and Lawson’s theorizing, they work largely within Swedberg’s designated stages (observation, conceptualization, building out of a theory, and development of a final explanation), focussing in particular on the post-observation stages. For Musgrave and Nexon and Anievas, Buzan, and Lawson fail to provide an explanatory theory of the 19th century transformation, with theorizing understood as the identification of causal factors and the weighing of their relative causal effects. Anievas (Reference Skinner2016) laments Buzan and Lawson’s lack of theoretical ambition, their provision of a conceptual framework for nothing more than ‘analytical–heuristic’ purposes and their easy appeal to contingency and multi-causality. Similarly, Musgrave and Nexon (Reference Rosenau2016) claim that Buzan and Lawson provide a conceptual apparatus for approaching change, but no explicit ‘theory of change’. Yet, while they end up calling for more radical modes of theorizing, their underlying model of theorizing is utterly conventional: the specification of causal factors, and careful weighing of their relative causal impact (through counterfactual reasoning if needed). To be sure, their call for ‘exotic’ IR theory rests on a non-linear conception of global change, the process of theorizing remains the identification of causal variables and mechanisms. In their reply, Buzan and Lawson reject this model of causation, arguing at some length that their approach ‘relies on generating an “intelligible connection” between “closely related bundles” of historical events’ (Reference Ruggie2016).

Braumoeller and Phillips are less animated by Buzan and Lawson’s failure to build a conventional causal theory, but their understanding of theorizing is still confined within Swedberg’s schema. Phillips’ critique focusses on the adequacy of Buzan and Lawson’s sphere of observation and the temporal application of their conceptual framework. He calls for broadening of their historical gaze to see Mughal India and Manchu China, and the application of their configurational conceptual framework to an early modern ‘Eurasian transformation’ that prefigured Buzan and Lawson’s 19th century transformation (Phillips Reference Swedberg2016). Braumoeller, by contrast, is animated neither by the absence of clear explanatory theory nor by the limits of Buzan and Lawson’s historical gaze. In fact, of all our contributors, he is perfectly content with their conceptual framework and descriptive analytics. Yet, this is only because the domain of theory he is concerned with lies elsewhere. He understands the process of theorizing as the specification of causal relations, but his focus is on systemic IR theory, on ‘the impact of the structure of the system of the behavior of states’ (Braumoeller Reference Swedberg2016). Irrespective of whether Buzan and Lawson adequately theorizing the global transformation, their account can help refine the scope conditions for systemic theory.

This tendency to think of theorizing as a process confined within Swedberg’s four stages is challenged head on by Owens and Bilgin. For both, conceptualization occurs prior to, or entwined with, observation. Proper theorizing thus demands that theorists exercise a high degree of reflexivity in the concepts they employ, as concepts determine what is observed. From this perspective the problem with Buzan and Lawson’s theorizing is not that they fail to specify causal variables and mechanisms, it is that they use concepts without the requisite reflexivity. Owens and Bilgin approach this from different angles and with different foci. Owens challenges Buzan and Lawson’s pursuit of an historical sociological account of the great transformation, criticizing them for failing to historicize the ‘social’ concepts that structure their analysis. Not only do these concepts have histories that have shaped their meanings, most of them are products of the very transformation they are being enlisted to understand: ‘In adopting rather than fully historicising the rise of and expansion of social theory such works of “historical social science” obscure rather than illuminates the historical and political origins of distinctly social forms of governance and thought; underestimates its significance for the history of international thought; and, more importantly, is unable to identify the more fundamental governance form of which the rise of the modern social realm is a concrete historical expression’ (Owens Reference Ruggie2016).

If Owens’ focus is the historical emergence of key concepts, their implication in practices of social and political ordering, and the baggage they carried forward into contemporary political and historical analysis, Bilgin is concerned with the application of concepts to diverse periods and processes of global history as though they have a universal meaning (2016). She is concerned with a deep form of Eurocentrism, in which global histories that privilege Europe are allowed to stand despite the existence of contending, non-European accounts. The hallmark of this Eurocentrism is the application of concepts fashioned in Europe to understand the experiences of non-European peoples. Buzan and Lawson are explicit in seeking a non-Eurocentric account of the global transformation, but while Bilgin credits this, she insists that they have not freed themselves of the deeper form of Eurocentrism. The Global Transformation is centrally concerned with 19th century changes in the ‘international’. Yet, how Buzan and Lawson conceptualize international order, international society, or international politics pays little attention to non-European understandings of the international, something they acknowledge in their reply. Focussing on Ottoman conceptions of modernity and the international, she points not only to the distinctiveness of these conceptions, but claims that ‘Inquiring into the international as viewed from the perspective of ‘others’ would allow further insights into the ideational structure underpinning international society’ (Bilgin Reference Wallerstein2016).

Two challenges for IR theory

Reflecting on Krasner’s influential article ‘State Power and the Structure of International Trade’, Keohane extolled the virtues of arguments characterized by what he termed ‘problematic lucidity’. He argued that ‘the study of world politics is likely in the future to progress less through incremental additions to our knowledge than through leaps of imagination and the critical, dialectical discourses these flights of fancy generate’ (Keohane Reference Gilpin1997, 170). Krasner’s argument transformed debate in 1970s international political economy not because he was right – by his own admission, his theory did not explain half of his cases – but because he lucidly presented a bold new thesis that inspired debate, debate driven by gaps in the argument and empirical anomalies that ‘practically leap off the page’ (Keohane 1997, 153).

The Global Transformation is likely to have considerable impact on debate about the origins and future development of the modern international system. Yet, this impact will derive less from the robustness of Buzan and Lawson’s argument than its problematic lucidity. Their argument is bold and artfully articulated – modern international relations is the product of a 19th century global transformation, a transformation wrought by interrelated processes of industrialization, rational state building, and ideologies of progress. It is the very ambition of their thesis, however, and the sweeping nature of their historical synthesis, that simultaneously captures the imagination and provokes debate.

As this symposium demonstrates, however, such debate has quickly spread from the realm of history into the domain of theory. This is not just because my fellow contributors have written for a theory journal. Rather, as suggested earlier, the very act of specifying an historical origin is a theoretical act: it rests on a prior conception of the emerging unit (system, order, society, etc.), it constitutes that unit as a structural formation, and it assumes a model of historical causality. Theoretical questions were thus always going to dominate debate about this book.

What matters, though, is that the debate that unfolds in the following articles poses two significant challenges for IR theory – and theorists. First, most IR theory is systemic, in the sense that it assumes a given international systemic context, and then seeks to explain political behaviour within that context. Some theories are structural, some are agentic, some try to dodge the dichotomy by focussing on practices, but all assume a larger systemic context: a system of states, an international society or order. Yet as the previous discussion demonstrates, if we are sceptical about breakpoint conceptions of systems change, and find merit in evolutionary and processual conceptions, then what constitutes an international system is harder to grasp. This is not just an issue for those of us interested in theorizing the rise, development, and decline of international systems, societies, or orders. It challenges the fundamental theoretical boundary condition of systemic theorizing: the existence of a coherent systemic context with identifiable structural characteristics. Not surprisingly, it is Braumoeller – the self-confessed ‘quantitative, empirically-oriented, plausibly rationalist systemic theorist’ (2016) – who is happiest with Buzan and Lawson’s breakpoint account, as it leaves a distinct systemic realm intact. The more IR theorists entertain alternate conceptions of large-scale change, particularly the processual, the more they are compelled to ask what constitutes the systemic context assumed by their work?

Second, as noted above, IR scholars spend little time reflecting on the nature of theorizing, very often assuming that it amounts to little more than choosing a theory from the available menu, or melding bits from different theories to solve a particular problem. Like Swedberg’s sociologists, we seldom think about theorizing as a process, one as creative as ‘scientific’. Yet as the preceding discussion explains, when my fellow contributors challenge Buzan and Lawson for not presenting a theory of change, they do not themselves offer theories. Rather, they present different perspectives on what constitutes proper theorizing. And in their reply, Buzan and Lawson do not disclose their long hidden theory; they describe their process of theorizing. The most important thing in all of this is that the range of possible forms of theorizing extend well beyond what mainstream IR scholars assume, or accept. This is clearest with Owens and Bilgin, both of whom stress the primacy of conceptualization, as well as its pre-observational nature. It is here that the challenge to the field is most pronounced. If observation is concept dependent, then our concepts come from somewhere other than brute empirical facts. Concepts have histories: path dependencies, hidden allegiances, and often dark, hard to acknowledge pasts. What is more, these histories blinker as much as they enlighten: they give concepts meanings that enable us to see some things but not others, to construct the unobservable in one way and not another. If this is true, then we should recognize that proper theorizing in IR encompasses not only the creative art of conceptualization, but also critical reflection on the conditioning histories of our concepts.