Reconsidering the policy coordination process

As noted above, the conventional definition of cooperation focuses on the emergence of coordinated policies, that is, mutually beneficial policy adjustments. In consequence, the account – like many scholars employing it in their works – is explicitly and deliberately agnostic about the kind of state interaction involved in the production of cooperation.

Over the years, however, a significant number of IR scholars have come to adopt a much more restrictive approach to defining interstate cooperation than originally suggested by the conventional account. Rather than being agnostic about how mutually beneficial policy adjustments emerge, they have come to understand interstate cooperation as implying that a particular process of policy coordination takes place between the actors involved.

A prime example can be found in Eric Grynaviski's work on the origins of international cooperation. Generally basing his conceptualization of international cooperation on the conventional account, he deviates from its original formulation by positing that cooperation entails ‘the explicit coordination of policies’ and ‘usually tak[es] the form of an international agreement’. He argues that ‘cooperation’ requires ‘the explicit formulation of policies that intend to generate outcomes that improve the condition of both partners’. As such, Grynaviski moves away from an understanding of cooperation that merely implies the implementation of a particular set of policies to one that conceives of cooperation as necessarily involving explicit negotiation and bargaining in the process of reaching mutually beneficial policy adjustments.Footnote ²⁹

Many other theorists employ a similarly restrictive definition of cooperation in their works, although some of them are much less explicit about it than Grynaviski. For instance, Michael Webb suggests in his discussion of cooperation in the international political economy that it entails ‘negotiated mutual adjustment that causes states to pursue different policies than they would have chosen had policy-making been unilateral’. He continues: ‘[i]t is important to note that according to this definition, coordination involves both negotiation and policy adjustment’.Footnote ³⁰ Along similar lines, Andrew Moravcsik conceives of ‘international cooperation as an effort to arrange mutually beneficial policy coordination among countries whose domestic policies have an impact on another’. For him, too, cooperation implies that states coordinate their policies through ‘interstate bargaining’ which ultimately produces ‘substantive agreements’.Footnote ³¹

To reiterate, what these scholars have in common is that, to them, interstate cooperation entails a claim about a particular process of explicit coordination taking place. In doing so, these theorists advance a markedly different conceptualization of interstate cooperation than those who, as previously discussed, focus on the emergence of a certain set of policy decisions and largely disregard what kind of process led to their realization. At first sight, especially given that these scholars frequently refer to the conventional account approvingly, this might appear to be nothing more than a semantic issue.Footnote ³² Yet, to the contrary, the turn toward an understanding of cooperating as necessitating an explicit coordination process has far-reaching conceptual implications.

To see what exactly those are, recall that the conventional account originated out of an effort to understand how international regimes might influence interactions of states under anarchy. Scholars working in this tradition paid close attention to situations involving ‘different configurations of state interests, i.e., … different types of “cooperation problems”’.Footnote ³³ Conceiving of a plethora of scenarios as game-theoretic problems of strategic interaction, they sought to understand under what conditions states implement policies that benefit all parties involved and, conversely, eschew strategies that would leave everybody worse off. Against this backdrop, it is evident why many IR theorists working on international cooperation avoided placing significant limitations on the process that precedes policy adjustments. After all, these scholars thought that many state interactions could be modeled as, for instance, iterated Prisoner's Dilemmas in which the actors' ability to communicate with each other was drastically limited. Whenever cooperation – again, understood as the implementation of mutually beneficial policies – emerged in these models, it was the result of a tacit adjustment process, that is, of unilateral, ‘independent policy-making that takes international factors into account’.Footnote ³⁴ The conventional account's permissive conceptualization of cooperation thus reflects a desire to capture such tacit forms of policy coordination in addition to those instances in which policy coordination involves an explicit negotiation process.

Yet, those scholars that have turned away from this permissive conceptualization have done so precisely because they do not want to diagnose cooperation in instances where coordinated policies emerge without direct communication and bargaining between the actors. For instance, Webb directly opposes defining cooperation in a way that ‘does not distinguish independent policy-making that takes international factors into account from internationally negotiated adjustments to national policy’.Footnote ³⁵ Similarly, Grynaviski highlights that, according to his understanding of cooperation, ‘cooperating in the Prisoner's Dilemma is not cooperating; the prisoners in separate cells do not have the ability to explicitly coordinate their actions’.Footnote ³⁶

Structuring the debate: ‘minimal’ vs. ‘thin’ cooperation

We argue that this substantial – albeit rarely recognized and thus not yet formalized – disagreement within the IR literature is best understood as suggesting a distinction between two types of interstate cooperation: ‘minimal’ and ‘thin’ cooperation. Both types of cooperation focus on mutual policy adjustments that create benefits for all sides. The crucial difference between the two lies in how these adjustments come about. In cases of ‘thin’ cooperation, the actors involved explicitly coordinate their policies, seeking to adjust them in such a way that mutual benefits are realized. What makes the coordination process explicit is that it involves direct communication and negotiation that aims at producing beneficial policy adjustments. In cases of ‘minimal’ cooperation, on the other hand, such policy adjustments emerge out of a tacit coordination process, that is, they are the result of each side's unilateral policy adjustments that are dictated by their individual calculus in a context of strategic interdependence. Instead of explicitly debating possible changes in policy, each of the actors involved is independently ‘acting in pursuit of those things she wants or values in part in light of what she believes the other is doing, and where the other's actions depend in part on what the other thinks she will do’.Footnote ³⁷ In contrast to instances of ‘thin’ cooperation, when states cooperate in the ‘minimal’ sense, there is thus no joint ‘design of cooperative arrangements and no direct communication’ about the possible realization of mutual benefits through policy adjustments.Footnote ³⁸ Rather, mutually beneficial policies emerge out of a tacit process of mutual adjustment in which the actors strategize about the possible effects of their own actions on other actors and how the latter might respond to them.

Formally, we can thus define two distinct types of interstate cooperation implied by the IR literature:

Importantly, the processes involved in ‘minimal’ and ‘thin’ cooperation – tacit and explicit policy coordination, respectively – should be understood as entailing satisfaction of the intention constraint.Footnote ³⁹ This is to say that, for instance, when states cooperate in the ‘minimal’ sense, they need to hold an intention to at least not obstruct the other actors' goal pursuit during the process of tacit policy coordination. This is crucial because, as we have seen in our discussion of ‘accidental cooperation’ above, in the absence of such basic cooperative intentions our conceptualization of interstate cooperation would run the danger of producing conceptual implications that contradict our intuitions about what cooperation is.

Furthermore, note that both processes, tacit and explicit policy coordination, will in many cases take place over an extended period of time. The kind of communication and bargaining characteristic of explicit policy coordination will usually unfold over time, as state representatives discuss, negotiate, and ultimately agree on mutual policy adjustments. Similarly, tacit policy coordination will often occur in a gradual, temporally extended process, as states observe each other's behavior, adjust their policies in response, observe these policy changes, adjust to them, and so on and so forth. It is in this sense that our conceptualization of ‘minimal’ and ‘thin’ cooperation reflects one of the key insights in the literature on international cooperation, namely that many, if not most, instances of cooperation result from an iterative process in which states engage each other repeatedly over time.Footnote ⁴⁰

‘Minimal’ and ‘thin’ cooperation are mutually exclusive types of cooperation: either the actors engaged in cooperation coordinate their policies explicitly or mutual policy adjustments occur through a tacit process of policy coordination. To be sure, in practice, any group of two or more states will often engage in both ‘minimal’ and ‘thin’ instances of cooperation: over time, and across different issue areas, states frequently engage in various kinds of cooperative activities. However, as we will now illustrate by discussing real-world examples of both ‘minimal’ and ‘thin’ cooperation, one can usually classify particular sets of interactions on a given issue as closely approximating either the ‘minimal’ or ‘thin’ type of interstate cooperation.

Real-world examples of ‘minimal’ and ‘thin’ cooperation

A turn toward real-world examples helps illustrate what this theoretical distinction looks like in practice. Examples of ‘thin’ cooperation are easily discerned. Consider, for instance, the Anti-Ballistic Missile Treaty concluded between the United States and the Soviet Union in 1972, which marked the end of negotiations designed to bring about a coordination in policies that would leave both sides better off.Footnote ⁴¹ Similarly, international coordination on trade policies that was formally codified in the General Agreement on Tariffs and Trade was the result of explicitly negotiated policy adjustments.Footnote ⁴² It is equally easy to identify attempts at ‘thin’ cooperation that ultimately failed. For instance, Axel Marschik points at the early 2003 UN Security Council negotiations regarding Iraq as an important example of great powers deliberately seeking to reach an agreement and coordinate their policy positions on a contentious issue that ultimately ended in ‘a breakdown of negotiations, an acrimonious blame game and recourse to force outside the system of the UN Charter’.Footnote ⁴³

Cases of successful ‘minimal’ cooperation, on the other hand, look markedly different. As an illustrative example, consider the informal restraint the two Cold War superpowers came to exercise after first experimenting with anti-satellite weapons (ASATs) during the 1960s and 1970s. As Nancy Gallagher explains, neither the USA nor the USSR ‘made a sustained effort to develop and deploy ASATs’ although they ‘were technically feasible and legally permissible’. No treaty specifically addressing the issue of ASATs was negotiated and, in fact, ‘it does not appear that the superpowers exchanged views about military space activities writ large’. However, over time, both sides ultimately came to demonstrate ‘reciprocated restraint’ with regard to ASATs: both anticipated that if one of them were to deploy them, the other would quickly follow suit and that, conversely, if one exercised restraint the other might reciprocate and thus enable the two sides to avoid a costly arms race that was not in the interest of either party.Footnote ⁴⁴

Along similar lines, Benjamin Miller points at US and Soviet policies in the Middle East after the Suez Crisis as yet another example of cooperation being the ‘outcome of unilateral moves’, that is, a tacit coordination process in which communication was limited to unilateral ‘show of force signals’. Despite the fact that both sides ‘fundamentally rejected the idea of spheres of influence’, in practice they came to tacitly coordinate their policies such that each power's sphere would be respected by the other. Such policy coordination, which helped the two superpowers avoid a direct confrontation, occurred ‘from necessity [rather] than by choice’, in that it resulted from the actors' unilateral decisions taken in a situation of strategic interdependence. Miller summarizes aptly: the two great powers ‘tacitly … recognized the stakes of the rival superpower, its sphere of influence, and its equal status’ as, over time, ‘it became easier for each superpower … both to identify and respect the interests and spheres of influence of their rivals in crisis situations’.Footnote ⁴⁵

Crucially, although no explicit policy coordination took place in these cases of ‘minimal’ cooperation, the emergence of coordinated policies was far from accidental: considerations of the effects their actions would have on the other side featured prominently in both superpowers' decision-making processes, and in adjusting their policies both states came to intend not to obstruct the other on the given issue, albeit for solely egoistic reasons. It is in this sense that even though policy coordination took place only tacitly, it still involved some form of basic intentionality as demanded by the intention constraint.

Philosophical research on sophisticated forms of cooperation

Over the past 30 years, philosophers have increasingly called attention to and examined phenomena that fall under the heading ‘joint action’. In these sorts of activities that are presumed to represent a vital aspect of human sociality, a set of two or more actors jointly pursues a common – or shared – goal.Footnote ⁴⁷ Philosophers highlight that paradigmatic instances of joint action occur in relatively mundane situations, for instance when actors paint a house together, lift a table, or sing a song together.Footnote ⁴⁸ Crucially, in these kinds of cases it is not ‘enough to say that [the actors] each act independently, where the sum of those individual actions involves the realization of a certain effect. In each case there is something that [the actors] together do; apart from [their] individual actions, there is an action that [they] together perform’.Footnote ⁴⁹

An important class of joint action refers to a distinctive mode of cooperation that philosophers occasionally refer to as ‘we-mode cooperation’ or ‘shared cooperative activity’.Footnote ⁵⁰ In such cooperative endeavors, the actors' goal pursuit is driven by a commitment to attain their shared goal together, that is, as members of a group with a joint responsibility for achieving that end. The joint pursuit of a shared goal so understood requires an intimate kind of collaboration. For instance, the actors involved need to engage in an ongoing process of determining and coordinating their respective part-actions necessary for achieving the envisaged, shared goal. This process can be labeled ‘we-reasoning’ and, inter alia, includes answering questions such as ‘What should we do in this situation?’ and ‘What should I do as a group member as my part of our group's action?’ instead of merely ‘What should I do in this situation?’Footnote ⁵¹ As such, cooperation on the basis of joint action clearly involves more than a set of actors ‘taking each other as features of the environment in acting to achieve [their individual aims]’.Footnote ⁵²

In order to clearly distinguish this kind of cooperative interaction from less elaborate forms of cooperation, philosophers have sought to identify what features capture its unique phenomenology. For example, Bratman characterizes what he calls ‘shared cooperative activity’ (SCA) through reference to three distinct features. First, the actors engaged in an SCA need to exhibit a ‘mutual responsiveness’ toward each other. This means that ‘each participating actor attempts to be responsive to the intentions and actions of the other’. Of course, this feature also seems to figure in ‘minimal’ and ‘thin’ cooperation and is therefore insufficient to distinguish SCAs from these less-involved types of cooperation.Footnote ⁵³ Yet, Bratman notes that SCAs exhibit two additional features. The actors in an SCA, second, must also each exhibit a ‘commitment to the joint activity’, and ‘their mutual responsiveness’ needs to be ‘in pursuit of this commitment’. Third, and finally, the actors engaged in an SCA each need to have a ‘commitment to mutual support’, meaning that they are ‘committed to supporting the efforts of the other to play [their] role in the joint activity’.Footnote ⁵⁴

Note that a major philosophical debate surrounds the question of how to best analyze the ontology underlying these cooperative interactions.Footnote ⁵⁵ Specifically, philosophers controversially discuss the question of whether their analysis requires a reference to irreducible collective intentional states, or even group-level agency, or whether these phenomena can be understood merely by reference to individual actors and their intentional states.Footnote ⁵⁶ But, although philosophers ultimately continue to disagree on what is the most convincing ontological framework for joint action and the corresponding form of cooperation, it is almost orthodoxy among them that these phenomena are genuine and should be taken seriously.

Before demonstrating the relevance of cooperation based on joint action for contemporary IR theory, it is worth stressing that the phenomenon has already been fruitfully integrated into social scientific debates. As Raul Hakli and his co-authors note, this sense of cooperation is already a salient issue in disciplines such as economics and decision theory, as is evidenced by the highly regarded works of Robert Sugden and Michael Bacharach.Footnote ⁵⁷ Hakli et al. themselves illustrate how an appeal to cooperation as involving ‘we-reasoning’ leads to more faithful decision-theoretic models of actors' behaviors. In particular, they argue that appeals to cooperation in the we-mode have the potential to remedy well-known problems for standard non-cooperative game theory such as its predictions in games like the Prisoner's Dilemma or Hi-Lo that ‘seem to clash with either empirical results or intuitive rationality judgements’.Footnote ⁵⁸ Drawing on both economic and philosophical work on joint action, they argue that these decision-making problems can be solved if we suppose that the actors frame the particular problem as one to be solved by ‘we-reasoning’ and not solely by individualistic strategic reasoning. As Hakli et al. summarize, each ‘agent conceives the situation – not as a decision-making problem for individual agents – but as a decision-making problem for the group conceived as an agent’.Footnote ⁵⁹ Adopting such a collective stance toward decision-making problems promises to more adequately model actors' behavior as well as provide guidance for decision-making procedures.Footnote ⁶⁰

Sophisticated forms of cooperation in IR theory

The relevant literatures in philosophy, economics, and decision theory primarily focus on instances of joint action as they occur in interpersonal relationships. However, given that large sections of IR scholarship treat states as unitary actors – or even persons – and find this assumption to be compatible with their commitment to scientific realism, it is unsurprising that many scholars contend that cooperation based on joint action also occurs between states in the international arena and thus should be recognized as a distinct phenomenon in need of theorization.Footnote ⁶¹ In fact, Stacie Goddard and Daniel Nexon have gone as far as noting that ‘beneath their disparate analytical frameworks, social ontologies, and even conceptions of power, many of the different “paradigms”, “turns”, and “isms” found in contemporary … international relations scholarship … share a focus on … mechanisms and processes by which actors engage in, seek to achieve, or orient joint action’.Footnote ⁶²

It is especially work in the social constructivist tradition, however, that first comes to mind as invoking cooperation based on joint action as an essential phenomenon in IR. Most notably, in her work on global governance, Jennifer Mitzen makes a strong case for viewing ‘concerted action’ or ‘doing something together’ as an important, often overlooked mode of state interaction that ‘is not merely cooperation as IR scholarship has [traditionally] understood it’. When engaging in global governance, she posits, states intentionally design order at the macro-level, ‘concerting their power for common public interest’. Rather than tacitly or explicitly coordinating their individual policies as is the case in what we introduced as ‘minimal’ and ‘thin’ cooperation, Mitzen observes states as at times ‘commit[ting] to address particular problems together’ and thus ‘constitut[ing] a public power beyond the state with the capacity to steer international political outcomes’.Footnote ⁶³

Along these lines, John Ruggie has famously called for greater attention to ‘collective intentionality’ as an important phenomenon in IR that he considers unduly overlooked by large parts of traditional IR theory. Ruggie specifically notes that states can hold ‘intersubjective beliefs’ of the form ‘we intend’ and ‘I intend only as part of our intending’. He, thereby, suggests that – at least at times, if not regularly – a group of states can closely resemble a group of individuals that jointly pursue a shared goal and work toward its achievement as a collective.Footnote ⁶⁴

Similarly, Wendt criticizes work ‘which takes self-interested actors as constant and exogenously given and focuses on the selective incentives that might induce them to cooperate’ for ignoring the possibility that ‘states might form collective … interests’. He notes that states often exhibit ‘an empathetic rather than instrumental or situational interdependence between self and other’. The formation of a ‘collective identity’ blurs the distinction between self and other and, in turn, ‘is a basis for feelings of solidarity, community, and loyalty and thus collective definitions of interest. Having such interests does not mean that actors … no longer calculate costs and benefits but, rather, that they do so on a higher level of social aggregation’. For Wendt, it is precisely the emergence of such collective identities and interests that ‘provides an important foundation for’ certain forms of joint action in which states commit to managing a particular policy issue as a group.Footnote ⁶⁵

Finally, the literature on security communities has pushed for an introduction of the ‘language of community to understand international politics’ properly.Footnote ⁶⁶ They build on the seminal work of Karl Deutsch who famously defined security communities as ‘a group that has become integrated, where integration is defined as the attainment of a sense of community, accompanied by formal or informal institutions or practices, sufficiently strong and widespread to assure peaceful change among members of a group’.Footnote ⁶⁷ Communication between members of such groups, Deutsch and his co-authors note, allows them ‘to think together, to see together, and to act together’.Footnote ⁶⁸ Along these lines, constructivist scholars integrating this line of scholarship into modern IR theory have noted that groups of states sometimes exhibit ‘a compatibility of core values derived from common institutions and mutual responsiveness – a matter of mutual identity and loyalty, a sense of “we-ness”’.Footnote ⁶⁹

Social constructivist work alone, thus, suggests the need to expand our conceptual toolkit in such a way that it is able to accommodate cooperation based on joint action as a distinct type of interstate cooperation. Indeed, although Goddard and Nexon's aforementioned reference to the centrality of joint action to all traditions of IR theory rightly implies its relevance for the discipline at large, explicit discussions of the phenomenon have been mostly limited to work within this theoretical tradition: it is social constructivists that have put debates about collective intentionality, community, and joint action in IR center stage, and it is them who have been most successful at integrating these discussions into their overarching theoretical frameworks.

However, it is worth noting that work within other branches of IR scholarship, too, hints at the possibility of conceiving of certain cooperative state interactions as involving joint action rather than ‘minimal’ or ‘thin’ cooperation as defined here. For instance, realist work on economic and political integration has more or less explicitly recognized that interstate cooperation at times goes beyond mere ‘minimal’ or ‘thin’ cooperation. As an example, Sebastian Rosato notes that, under certain circumstances, ‘states agree to joint control over their militaries or economies’ and decide to ‘make policy jointly’. In these cases, ‘rather than resting in any one state's hands [policymaking] is determined by the member states acting together’.Footnote ⁷⁰ When engaging in this specific ‘form of centralized cooperation’, states interact with each other in the kind of we-mode characteristic of joint action: ‘if they are to achieve their common goal’, Rosato posits, ‘they need to know what issues are covered, what is expected of them, what they can expect from their partners, how decisions are going to be made, and so on’.Footnote ⁷¹

Work in normative IR theory has also analyzed the forms of cooperation that go distinctly beyond the two types of cooperation – ‘minimal’ and ‘thin’ – that have received most attention in mainstream IR theory. As an example, Toni Erskine highlights that states often concert their actions ‘in a way that cannot be simply reduced to the actions and intentions of individuals acting in isolation’. Specifically, she notes that when they ‘share a common purpose and have developed’ at least ‘an informal organizational structure and deliberative capacity in order to pursue it’, states can sometimes engage in what she terms ‘joint purposive action’, that is, ‘sophisticated forms of cooperation’ that emerge when actors ‘come together to participate in a common project’, are ‘united in pursuit of a common purpose’, and engage in ‘deliberation and special coordination’.Footnote ⁷²

Overall, important branches of IR scholarship, thus, suggest that states at times engage in more sophisticated forms of cooperation than those involved in ‘minimal’ or ‘thin’ cooperation. To be clear, not all of them explicitly discuss these interactions as involving collective intentionality and joint action; in fact, few outside of the social constructivist and normative IR literature do. Yet, what the above examples illustrate is that other scholarly traditions at times portray state interactions along similar lines and at least latently acknowledge the existence of such more sophisticated forms of cooperation. Hence, any typological framework of interstate cooperation seeking to supply IR scholars with the conceptual tools to guide their substantive works should be able to accommodate such cooperation based on joint action.

Defining ‘thick’ interstate cooperation

Scholars within philosophy, economics, and IR, thus, all identify a distinct sense of cooperation that does not merely refer to actors (tacitly or explicitly) coordinating their policies while pursuing their individual goals, but instead involves actors who jointly work toward the attainment of a shared goal. In Margaret Gilbert's words, the kind of cooperation at stake is coupled with a unique sense of ‘acting together’ in which ‘[t]wo or more people … collectively espouse a certain goal, and each one is acting in a way appropriate to the achievement of that goal, where each one is doing this in light of the fact that the goal is their collective goal’.Footnote ⁷³ Based on the foregoing analysis, we propose the following definition of a distinct type of cooperation that we term ‘thick’ interstate cooperation:

A couple of clarifications about the reasoning behind and implications of this conceptualization are in order. First, note that ‘thick’ cooperation – like ‘thin’ cooperation – entails an explicit process of coordination. To see why this is the case, note that for ‘thick’ cooperation to take place the actors need to intentionally create a shared goal that they subsequently pursue jointly. It is hardly conceivable that such shared goal creation and pursuit would ever occur without any direct communication and negotiation between states: at some point, they will need to explicitly discuss – whether in a formal or informal setting – how to begin and organize the joint endeavor. Against this backdrop, it is unsurprising that Erskine decidedly excludes groups that lack the capacity to directly communicate with one another from what she calls ‘joint purposive action’. As she explains, ‘members of a collective must have … the capacity to deliberate (however informally) in order to coordinate their actions (even imperfectly) in circumstances in which the required collective action is not obvious but, rather, open to disagreement’.Footnote ⁷⁴ This means that although, strictly speaking, it is logically conceivable that actors could engage in ‘tacit we-mode cooperation’ and jointly pursue a shared goal without any direct communication between them, it is unlikely that this kind of cooperative interaction would ever be observed in the real world.Footnote ⁷⁵ Accordingly, we do not further distinguish between two types of ‘we-mode’ cooperation – one involving explicit and one involving tacit coordination – as we did in the case of ‘I-mode’ cooperation. Rather, we recognize only one distinct type of ‘we-mode’ interstate cooperation, ‘thick’ cooperation, which assumes an explicit process of policy coordination to take place.

Second, ‘thick’ interstate cooperation paradigmatically satisfies the intention constraint. Applied to ‘thick’ cooperation, the intention constraint requires that the actors hold an intention to not obstruct the other actors' efforts in achieving the respective shared goal.Footnote ⁷⁶ Since both actors share the respective goal, they each have an interest in its attainment and will therefore not intend to obstruct the other actors' pursuit of it.Footnote ⁷⁷

Finally, ‘thick’ cooperation is conceptually distinct from, and cannot be reduced to, the types of interstate cooperation discussed previously, that is, ‘minimal’ and ‘thin’ cooperation. Recall that both these types of cooperation characterize cooperation as essentially being a matter of mutual policy adjustments conducive to the actors' individual goals. They are, therefore, based on what can be called merely individualistic conceptualizations of cooperation. The main reason why any individualistic conceptualization will fail to elucidate ‘thick’ cooperation is that its defining feature – the joint pursuit of a shared goal – does not necessitate that any of the actors' individual goals are facilitated during their ‘thick’ cooperative endeavors. This independence allows for two conceivable cases. In the first, the joint pursuit of a shared goal does not entail any benefits for the actors' individual goals.Footnote ⁷⁸ Here, merely individualistic conceptualizations would plainly disregard relevant instances of cooperation. In the second, pursuing a shared goal coincidentally does lead to mutual benefits for the actors' individual goals. Although individualistic conceptualizations might, in this case, formally attest cooperation, they would do so for the wrong reason. After all, the reason for which cooperation would emerge would be the coincidental promotion of the actors' individual goals, but not their joint pursuit of a shared goal. Therefore, the possibility of ‘thick’ cooperation ultimately presents merely individualistic conceptualizations, such as those underlying ‘minimal’ and ‘thin’ cooperation, with a dilemma: either they cannot identify instances of ‘thick’ cooperation to begin with, or – in cases where they may identify those – these conceptualizations would provide the wrong kind of explanation.

Real-world examples of ‘thick’ cooperation

Let us now turn to several real-world examples of ‘thick’ interstate cooperation, that is, instances in which states jointly worked toward the attainment of a shared goal. A first set of illustrative examples of ‘thick’ cooperation can be found in recent instances of so-called ‘coalitions of the willing’ jointly working toward a shared goal. The term, Erskine explains, ‘connote[s] a temporary, purpose-driven, self-selected collection of states’ that ‘come together to respond to a specific crisis’. Coalitions of the willing are an increasingly ‘common phenomenon in world politics’, and are most readily recognized when they ‘are summoned and established in cases of military intervention.’ The term first gained prominence when a coalition of the willing under US leadership responded to Iraq's 1990 invasion of Kuwait and fought together in the Gulf War. Other examples involving such a group of states working together for the achievement of a shared goal include the 1999 humanitarian intervention in Kosovo, or the 1999 intervention in East Timor. In each of these cases, the respective states working together formed a group that ‘seem[ed] to possess something that at least resembles a capacity for purposive action’. The mode of collaboration in the coalitions went beyond mere ‘minimal’ or ‘thin’ cooperation. In Erskine's words, any plausible ‘description[…] of actions and intentions’ of the states taking part in these joint endeavors differs ‘in important ways from accounts of individuals acting independently or in isolation’.Footnote ⁷⁹

In instances in which joint efforts take place in a more formalized and institutionalized setting than that present in the context of coalitions of the willing, ‘thick’ cooperation is even more easily discerned to take place. Consider, for instance, joint Anglo-American efforts to defeat Nazi Germany in World War II. Instead of merely coordinating and pursuing their individual policies in ‘I-mode’, the US and UK leaderships created the Combined Chiefs of Staff which integrated the Western Allies' decision-making procedures. Working within this formal framework, the two states were engaged in far-reaching cooperation, as they collectively worked toward the achievement of their shared goal.Footnote ⁸⁰ Similarly, in the wake of World War II, the minor Western and Central European powers' joint pursuit of protection from the Soviet threat led to the creation of an institutional framework that ‘provide[d] members with enhanced strategic, operational, and tactical coordination’ and thus paved the way for engagement in ‘we-mode’ cooperation.Footnote ⁸¹

The arguably most emblematic example of ‘thick’ cooperation at work, however, is the Concert of Europe. Following Napoleon's defeat, at the Congress of Vienna (1814–15) the five European great powers – Austria, Britain, Prussia, Russia, and France – reached the Vienna Settlement which ‘established the great powers’ intention to govern together’. Over the subsequent years, they managed European security affairs together, closely collaborating to avoid another devastating great power war. Importantly, none of the states involved abdicated ‘their independent capacity and authority to act for their own population’. However, ‘the great powers interacted in a markedly different manner from how they had acted in the eighteenth century or before’. Mitzen aptly summarizes what made this ‘experiment in international cooperation’ so unique: ‘the concerting of nineteenth-century Europe can be interpreted as not “just” cooperation based on overlapping individual preferences and in the pursuit of self-interested goals’. In other words, the Concert was not just an instance of ‘thin’ cooperation, that is, of states explicitly coordinating their individual policies for mutual benefits. Rather, it represented ‘a purposive political project articulated and pursued together’. The great powers, forming a collective intention ‘to manage Europe together’, acted together in the joint pursuit of a shared goal.Footnote ⁸²

Three types of interstate cooperation: a schematic overview

The conceptual framework developed in the preceding sections can be summarized in a matrix that distinguishes between different types of interstate cooperation through reference to the following two dimensions (see Figure 1).

Fig. 1.

Types of interstate cooperation.

The first dimension, represented by the rows, distinguishes the different kinds of goals that actors pursue in their cooperative endeavors. Both ‘minimal’ and ‘thin’ cooperation are centrally defined in terms of the actors' individual goal pursuit and therefore located in the upper row. In contrast, ‘thick’ cooperation is essentially a matter of the actors' shared goal pursuit, thus located in the bottom row, and in this sense conceptually distinct from both ‘minimal’ and ‘thin’ cooperation. Put differently, ‘minimal’ and ‘thin’ cooperation take place in I-mode, which is to say that the actors seek to realize their own individual goals, whereas ‘thick’ cooperation entails a we-mode of interaction.

The second dimension, represented by the columns, distinguishes the kind of coordination process involved in the actors' goal pursuit. ‘Minimal’ and ‘thin’ cooperation are primarily distinct from one another in that ‘minimal’ cooperation involves a tacit process of policy coordination, whereas ‘thin’ cooperation entails an explicit process that includes procedures such as explicit negotiation or bargaining. As discussed above, ‘thick’ interstate cooperation also involves an explicit process of policy coordination since it is hardly conceivable that the kind of ‘we-mode’ cooperation it entails would ever take place through a tacit coordination process alone. Accordingly, the bottom-left cell in our two-by-two matrix is left empty.Footnote ⁸³

Note that our integrative schema highlights that there is a clear sense in which all of the included types can be considered united, that is, instantiating the same general phenomenon of cooperation. At bottom, they all characterize cooperation as deliberative attempts by states to facilitate their (individual or shared) goals through policy coordination. Although this clearly seems reminiscent of the conventional account as originally put forth, our typology goes beyond it by vividly illustrating that there exists a variety of distinct senses of cooperation that differ in important, theoretically relevant ways from one another. As we will now show, it thus provides IR scholars with considerable theoretical leverage.

The typology's contribution to IR theory

Political methodologists have long recognized that descriptive typologies can crucially contribute to theorization in the social sciences.Footnote ⁸⁴ In fact, within the discipline of IR, similar conceptual frameworks on other foundational phenomena have led to important theoretical advancements and have done much to improve our understanding of international politics. Notable examples include John Vasquez’ disaggregation of the concept of war into several different subtypes as well as Thomas Wilkins' distinction between different types of interstate alignment.Footnote ⁸⁵ Similar to these studies, our three-way typology of interstate cooperation aids theorizing about international politics in two main ways: it facilitates conversation across traditionally rigid theoretical boundaries, and it allows for better theorization of the causes of cooperation through disaggregation of the overarching concept.

First, the conceptual tools developed here facilitate conversation across different branches of IR literature and allow scholars to explore fruitful ‘links [between] previously unrelated theories and strands of research’.Footnote ⁸⁶ The study of international politics today often takes place within strictly delineated theoretical camps and without much interaction across them. Indeed, at times it appears as if scholars treat different theoretical approaches to the study of IR as incommensurable. Consequently, in Thomas Kuhn's words, ‘the proponents of competing paradigms’ have ‘fail[ed] to make complete contact with each other's viewpoints’.Footnote ⁸⁷

Against this backdrop, our typology of interstate cooperation facilitates productive conversation between different bodies of IR theory by providing them with a common, more specific conceptual language vis-á-vis interstate cooperation. Specifically, our typology allows scholars to contrast their differing viewpoints on important instances of interstate cooperation more precisely and effectively. Take, for example, the literature on the Concert of Europe. Much of the theoretical disagreement there finds its roots in competing judgments of what kind of interaction took place between the European great powers in this period. For many balance-of-power theorists, the Concert was merely an instance of ‘thin’ cooperation in which the great powers, for a brief period of time, explicitly coordinated their policies in such a way as to facilitate the attainment of their overlapping individual goals.Footnote ⁸⁸ Social constructivist theorists, on the other hand, employ a vastly different outlook on the Concert that aligns with the interpretation we presented above, noting that ‘phenomenologically, it [the balance of power story] overlooks the fact that the leaders intended to manage Europe together’.Footnote ⁸⁹ In other words, like us, they observe an instance of ‘thick’ cooperation. The fine-tuned conceptual toolkit provided here, thus, allows theorists to locate important sources of disagreement and specifies along which dimensions their perspectives on the world differ from one another. A more generalist conceptualization of interstate cooperation would struggle to lay out the fault lines of disagreement in a similarly specific manner.

Conversely, our typology also helps scholars uncover previously unrecognized theoretical common ground. As discussed above, both social constructivists working on global governance and rationalist scholars working on economic and political integration at times allude to cooperation in the ‘we-mode’ as an important type of state interaction that differs in kind from other, less involved forms of cooperation.Footnote ⁹⁰ Yet, since the existing conceptual literature on interstate cooperation lacked the tools to delineate ‘thick’ cooperation as a distinct phenomenon in international politics, this common theoretical interest has not been recognized so far, and consequently these bodies of IR theory have continued to develop largely independently of each other.

Second, by disaggregating the concept of ‘cooperation’ into three subtypes, our framework allows for better theorization about the causes of interstate cooperation. As Derek Beach and Rasmus Pedersen explain, the ‘population[s] of cases’ social scientists are interested in can often be ‘very causally heterogeneous’. This is to say that ‘a given cause might have many different effects across different cases, in terms of either the same cause producing a different outcome or the same cause being linked to the same outcome through different causal mechanisms’. Scholars run the danger of focusing on such causally heterogeneous populations of cases, Beach and Pedersen warn, when ‘ignoring causally important differences between cases by lumping them together using excessively broad definitions of concepts’. This issue can be addressed by disaggregating the overarching concept into several sub-types and engaging in ‘typological theorization’.Footnote ⁹¹ Notably, Vasquez's well-known study of the causes of war rests on this very idea and utilizes it with great success. As he notes, a review of the historical and scholarly record suggests that ‘there are different kinds of wars and … a theoretical typology of wars … would be conceptually useful in explaining findings’.Footnote ⁹²

Given that the different forms of cooperation highlighted here significantly differ in character, it appears likely that conceptual disaggregation will provide similarly important theoretical leverage in the debate about the causes of interstate cooperation. After all, considering that ‘minimal’, ‘thin’, and ‘thick’ cooperation each entail a unique kind of interaction between states, it is very well possible that causal theories which, for example, are apt at explaining the emergence of ‘thin’ cooperation might struggle to explain instances of ‘thick’ cooperation, or vice versa. Our framework enables the kind of nuanced comparison of ‘concrete cases, both cross-sectionally and over time’, that is necessary for engaging in such type-specific causal theorization.Footnote ⁹³ As an example, it allows scholars to discern that interstate cooperation in Western and Central Europe has vastly varied in kind over time: at times only sporadically coordinating their actions in tacit ways, the European states at other times collaborated in a much deeper sense and even managed their security environment together.