Some Common Ground for Liberals and Realists
In both Andrew Moravcsik’s article, “Taking Preferences Seriously: A Liberal Theory of International Politics,” and in his chapter, “Liberal International Relations Theory: A Scientific Assessment,” in Elman and Elman’s Progress in International Relations Theory, he makes a sound, though impetuous, case for liberalism’s place in the hall of international relations paradigms. He challenges the alleged primacy of realism and institutionalism, whose practitioners, he asserts, have done the field a disservice by failing to test their theories against liberalism. Importantly, he outlines the three core assumptions of the liberal research program. For clarity’s sake, I will restate them here. First, liberals believe that the most important actors in the international system are “rational individuals and private groups who organize to promote their interests.” Second, those individuals and groups vie for the expression of their interests in the state, which is a representative structure-though by no means necessarily egalitarian. The “weighted preferences” of individuals and groups are then aggregated in some way within the state and output as foreign policy goals of the state, or the state’s preferences. Third, those preferences and the preferences of other states interact in the international system-which can be conceptualized as a web of “policy interdependence” – and in that way shape state behavior.
As a first cut theoretical framework, this set of assumptions seems sufficiently flexible and abstracted that I believe we can in fact fit realism into it. To my mind, Moravcsik could have strengthened his argument if he had elaborated on one of his minor points that began to explore this possibility of merging realism within the liberal realm rather than keeping it separate (as in the diagram on page 545 of his article). In the article, he points out that the commonsensical liberal assumption of states’ preferences shaping behavior is actually implicit in realism as well, because ” the assumption of rationality or purposive behavior central to realismâÂ?¦implies action on the basis of a prior, specific and consistent set of preferences.”
With Moravcsik’s observation in mind, further, more systematic exploration of realism’s potential fit within this framework is warranted. Michael Doyle identifies three commonly agreed upon but critical assumptions that unite realist scholars. First, realists assume that the international environment is anarchic. Second, they believe that states are the most important actors, and that they are rational and unitary entities. Third, they believe that the international system is made up of interactions between states and that the most important attribute shaping state behavior is the distribution of capabilities. This last point, however, has gotten rather sticky with the arrival of the neoclassical realists on the scene, who – unlike structural realists such as Waltz – consider the international system to be the primary but not sole determinant of state behavior. Scholars like Zakaria believe that in some cases the international system imposes very few constraints, and that in those cases we must look to domestic factors to understand state behavior. This development, as I shall argue, has allowed realism to interact with other research programs and greatly facilitated the potential merger between realists and liberals.
The realist assumption of anarchy does not seem to be incompatible with the Moravcsik framework. In the liberal view, the constraints on the behavior of states are domestic factors and the preferences of other states. There is no global governance structure to impose hierarchy. When we come to the second tenet of realism, however, we run into our first problem.
Liberalism, after all, considers individuals or groups acting on behalf of certain interests to be the most important actors, whereas for realists, states have primacy. Yet, liberals also conceive of individuals acting through the state internationally; it is the state that ranks, aggregates, or eliminates individual preferences to be expressed in international politics. So, although preferences originate in individuals, who are thus in principle the most important actors, how the state aggregates, ranks, or eliminates individual preferences to be expressed in international politics is arguably what produces the most variance in outcomes of state preferences. So, we can conceive of the liberal framework as capable of explaining more of a two-step process than most variants of realism, in that it allows for the state to act as a more or less unitary actor on the international level, but can also account for the internal struggle that can produce state preferences that at times seem counterintuitive according to realist logic.
Moreover, there is implicit recognition of this domestic power struggle and its potential role in international politics in neoclassical realists’ focus on the state’s variable extractive capacity. And because liberalism at this level of abstraction makes no assertions as to the attributes of the state or whose preferences are expressed in foreign policy, there is in theory no reason to exclude other realists like Machiavelli, for example, for whom the state was essentially embodied in one leader. Just because the liberal framework gives us the flexibility to recognize non-state actors further down the ladder of abstraction in relation to more specific policy realms does not mean that it conflicts with realism. What is important is that realism can be contained by this level of liberal theoretical framework-not vice versa.
The last realist assumption-that of the system as the primary determinant of state behavior-seems most at odds with our liberal framework. Realists, after all, take a “top-down” approach to world politics, while liberals prefer a more “bottom up” point of view. In this instance, amendments to both schools’ rather unidirectional conceptualizations of international politics might be in order. It does not seem inconceivable to imagine that states’ preferences might be influenced by systemic conditions, like transnational social movements, and although Waltz may criticize them as reductionist, some realist theories do explicitly recognize the effect of unit level variables on the character of the system, such as Rousseau’s constitutionalism. In fact, to the extent that realist assumptions about human nature have impacted their understanding of the characteristics of the international system, realist theory has already been informed by unit level variation-even Waltz’s structuralism.
If we accept this more relaxed version of realist tenets, it is really only once we begin to move down the ladder of abstraction to “thicker” variants of liberal theory that we start to see theories that explicitly butt heads with realism. This is because in fleshing out the liberal framework of international politics, we must begin to examine the actual content of states’ preferences – to which we had previously only referred in the abstract – and what those preferences mean for the dynamics of the international system. For realists, states’ overriding preference in the international arena is the pursuit of security or power-depending upon whom we read, of course. This dark, rather threatening conception of international politics and states’ necessary quest for survival therein stems from the philosophical roots of realism: the Hobbesian conception of man fleeing from the anarchic and dangerous State of Nature, which translates into a similar environment for states in the international system.
As Moravcsik presents briefly, and Doyle explores more fully, different variants of liberalism lead to more variation in state preferences, and thus a potentially very different kind of international system. For “commercial pacifists” economic incentives and actors are the most important determinants of state preferences; for “ideational” liberals, social identities, which determine how public goods should be provided, shape state preferences; for “republican” liberals, it is the structure of domestic institutions and the ways in which they weigh societal demands that determines state preferences in foreign policy-so preferences will reflect the segments of society that have the most influence over those institutions. The priorities reflected in these strands entail a far different international view than the realist reasoning. This is what has led to the development of the theory of liberal democratic peace, of course. For commercial liberals, war is too costly, and states should and will avoid it against states with which they have close economic ties. For ideational and republican liberals, democratic polities-in their capacity to fill the state’s power vacuum with their pacifying influence-prefer not to fight each other either because they are risk- and cost-averse or they share values and are unlikely to have conflicting preferences.
Clearly, then, despite the commonalities between realists and liberals at a general level, research into more specific issue areas has led them to have very different views of the international system. And this is as it should be-those differences then allow us to test theories against each other to evaluate their utility. I should be clear that I am not arguing for the obsolescence, or even the weakness of realist theory. Instead, I am trying to emphasize the connections between the liberal and realist research program in the hope that merging them under one broad, general framework of assumptions will help to systematize the field insofar as it facilitates better, more extensive, theoretical testing and less reliance upon the kind of monocausality that Moravcsik rightly denigrates. Moravcsik offers a thoughtful critique of the Lakatosian philosophy of science in the Elman and Elman volume, but to me the most important flaw in this approach is the isolation and polarization it seems to encourage amongst schools of thought that are really not so far apart. The supposedly insurmountable differences between what realists and liberals, respectively, assumed to be the “hard cores” of their research programs have kept both on the defensive and hindered the progressive accumulation of knowledge.
Under a common umbrella, however, realism as a school could still focus its research on states, capabilities, and systemic dynamics, and such studies could complement the more liberal agenda containing non-state actors, commerce, and national attributes like institutions. One might ask why, then, realism’s three core assumptions should not constitute the highest level of abstraction. The answer is simple: they are too rigid to comfortably encompass liberalism’s major variants. But the reverse is not true; Moravcsik’s “liberal” framework, as he presents it, is generalizable enough to stand above realism without sacrificing too much of its elegant reasoning.
The fact is, realism and liberalism together can explain much more, in a more accurate manner, than they can separately. As Deudney and Ikenberry succinctly point out, there are certain empirical realities in international politics that do not sit well with realism’s theoretical precepts and predictions. The realist paradigm has yielded rich insight into phenomena like arms races, but liberalism is much better at explaining trade policy behavior, to use just one example. The international system is multi-faceted. To expect just one theory to be able to explain all of its dynamics and outcomes is unrealistic. Thus for each aspiring paradigm in international relations theory to blindly go its own way is not progressive or even helpful-it is simply bad science.
