Realism and liberalism are the two theoretical traditions that have shaped the academic study of international relations more than any others. Each offers an account of why states behave as they do, what makes conflict and cooperation more or less likely and what can reasonably be expected of the international system. They are taught as the starting point of the discipline, and most later approaches were developed either as extensions of them or as objections to them.
Both names are, however, unhelpful. In everyday English, being realistic means being hard-headed and unsentimental, and being liberal means holding a particular set of political opinions, usually on the left of centre in the political cube. Neither everyday meaning corresponds to the technical one. This article sets out what the two traditions actually claim, where the ordinary meanings of the words lead readers astray and where the real disagreement between them lies.
Two Familiar Words, Two Technical Meanings
Realism in international relations is not a personality trait, a preference for tough measures or a synonym for cynicism. It is a body of theory built on specific claims about the structure of the international system and the constraints that structure places on states. A realist analysis is not necessarily an argument for the use of force. Indeed, realists have often been prominent opponents of particular military undertakings: a number of well-known realist scholars publicly opposed the invasion of Iraq in 2003, on the grounds that it did not serve the security interests of the state that undertook it. Realism is a framework for identifying interests and constraints, not a recommendation to be aggressive.
Liberalism in international relations is likewise not the party-political term familiar from domestic debate, nor is it the same thing as the economic doctrine sometimes called neoliberalism. It descends from the older tradition of liberal political philosophy, concerned with individuals, property, rights, commerce and representative government, and it asks what follows for relations between states when these things are taken seriously. Liberalism in this sense is not pacifism, and liberal theory has been used to justify the use of force as readily as to counsel against it.
A further confusion arises from the label “idealism”. Early critics of liberal thinking, above all Edward Hallett Carr in The Twenty Years’ Crisis, published in 1939, attacked what he called utopianism: the belief that the harmony of interests among nations could be assumed and that international law and public opinion would suffice to keep the peace. Contemporary liberal theory is not that position. It is a rigorous body of work that shares a good deal of its method, and some of its assumptions, with realism.
The Core Claims of Realism
Realist theory rests on a small number of connected propositions. The first is that the international system is anarchic. Anarchy here is a technical term and does not mean chaos or disorder; it means simply that there is no authority above states, no world government able to make and enforce rules upon them. Within a state, disputes can be taken to courts backed by police. Between states, there is no equivalent. Thomas Hobbes was one of the first to treat this matter in his book Leviathan (1651), which remains one of the core pieces of literature in political science and beyond.
From this follows the second proposition, that states must ultimately rely on themselves, a condition described as self-help. Because no higher authority guarantees their survival (the leviathan), states must attend to their own security, and the means of doing so is power, understood principally as material and military capability.
The third proposition is that states are the principal actors and can usefully be treated as unitary and rational: as entities that pursue their interests coherently, whatever their internal arrangements. For most realists, the internal character of a state, whether its government is elected or not, matters far less than its position in the distribution of power.
Two further concepts follow. One is the security dilemma, a term introduced by the political scientist John Herz in 1950. Measures that one state takes to make itself more secure, such as strengthening its armed forces, appear threatening to others, who respond in kind, with the result that all end up less secure than before, even though none intended harm. The dilemma explains how arms races and spirals of hostility can arise between states with no aggressive intent. The other is the balance of power, the tendency of states to align against any power that grows strong enough to threaten them.
Realists also emphasise relative gains. In deciding whether to cooperate, a state on this account asks not only whether it will benefit but whether its partner will benefit more, since a partner’s advantage today may become a capability used against it tomorrow. This concern places a persistent obstacle in the way of cooperation.
Varieties of Realism
Classical realism, associated above all with Hans Joachim Morgenthau and his work Politics Among Nations, published in 1948, located the source of conflict partly in human nature, in an enduring drive for power and domination that political life expresses at a larger scale. Writers in this tradition drew on earlier thinkers, among them Thucydides, Niccolò Machiavelli and Thomas Hobbes, although describing those figures as realists is a retrospective classification rather than one they would have recognised.
Structural realism, or neorealism, marked a decisive change. In Theory of International Politics, published in 1979, Kenneth Neal Waltz argued that recurring patterns in international politics could be explained without reference to human nature or to the internal features of states, by attending instead to the structure of the system and the distribution of capabilities within it. On this account, states behave similarly under similar structural pressures whatever their character, because those that ignore the pressures do not survive.
Within structural realism, a further division opened. Defensive realists hold that states seek an appropriate rather than a maximum amount of power, since excessive expansion provokes others to combine against the expanding state. Offensive realists, of whom John Joseph Mearsheimer is the best known, argue in The Tragedy of Great Power Politics, published in 2001, that uncertainty about the intentions of others gives states reason to maximise their power and to seek dominance in their region where they can.
Neoclassical realism, a term introduced by Gideon Rose in 1998, reintroduces domestic factors, treating the perceptions of leaders and the capacity of states to extract resources from their societies as intervening between structural pressures and actual policy.
The Core Claims of Liberalism
Liberal theory does not deny anarchy. Most contemporary liberals accept that there is no authority above states and that states pursue their interests. What they dispute is the conclusion that the effects of anarchy are fixed. Their claim is that those effects can be mitigated, and that the extent to which they are mitigated varies with conditions that can be identified and studied.
Liberalism differs from realism in three main respects. First, it treats actors other than states as significant: individuals, commercial firms, non-governmental organisations and international bodies all shape outcomes. Second, it holds that what states want cannot be read off from their position in the distribution of power. As Andrew Moravcsik has argued, state preferences are formed through the relationship between a state and its society, as the interests of groups within that society are aggregated and represented; the resulting preferences, not capabilities alone, drive behaviour. Third, liberals emphasise absolute rather than relative gains: a state that gains from an arrangement has reason to sustain it, even if others gain more.
The Strands of Liberal Theory
Liberal theory is conventionally divided into three strands, each identifying a different condition under which conflict becomes less likely.
Commercial liberalism holds that economic interdependence raises the cost of war and so reduces its attractiveness. The argument was put in a widely read form by Norman Angell in a work first published in 1909 and expanded as The Great Illusion, which contended that conquest had ceased to be profitable for advanced economies. It is often said that Angell predicted that war had become impossible, which he did not; his claim concerned profit, not possibility. The modern treatment is Robert Owen Keohane and Joseph Samuel Nye Junior’s Power and Interdependence, published in 1977, which described conditions of complex interdependence in which multiple channels connect societies, military force loses much of its utility and issues cease to be arranged in a clear hierarchy.
Republican liberalism concerns the internal character of states. Its origin is Immanuel Kant’s essay Perpetual Peace, published in 1795, which set out conditions under which a lasting peace among states might be established. The modern form is the democratic peace proposition, revived in work by Michael Doyle in 1983. Here, precision matters because the claim is regularly overstated. The well-supported finding is dyadic: democracies very rarely go to war with one another. It is not the claim that democracies are peaceful in general, and the evidence does not show that they fight fewer wars overall; potentially, the opposite can be the case.
Institutional liberalism, often called neoliberal institutionalism, argues that international institutions make cooperation more achievable. Robert Owen Keohane’s After Hegemony, published in 1984, is the central work. Its argument is notable for how much it concedes: it accepts anarchy, accepts that states are self-interested and rational and does not suppose that institutions make states benevolent. Institutions matter, on this account, because they reduce the costs of reaching agreements, supply information about what others are doing, make cheating easier to detect and raise the expectation of continued dealings, so that reputations acquire value.
Where Realism and Liberalism Disagree
Because institutional liberalism accepts so many realist premises, the two traditions converged considerably during the 1980s and 1990s, and the ensuing exchange is sometimes called the neo-neo debate. Both sides in it treat states as rational actors pursuing interests under anarchy, and both use similar methods. The remaining disagreements are narrower than the labels suggest, but they are real.
The first concerns gains. Joseph Grieco argued in 1988 that liberal institutionalists had understated the relative-gains problem: states worried about their standing relative to others will forgo mutually beneficial arrangements in a way that institutional theory does not predict.
The second concerns whether institutions have effects of their own. John Joseph Mearsheimer argued in 1994 that institutions largely reflect the distribution of power among their most capable members and exert little independent influence on state behaviour. Liberals reply that institutions persist after the conditions that produced them have changed, which is difficult to explain if they are only a reflection of present power.
The third, and most fundamental, concerns preferences. Realism generally treats what states want as given by their situation, so that survival and security can be assumed. Liberalism treats preferences as variable, arising from state and society relations, and therefore as something to be explained rather than assumed. Much of what separates the traditions in practice follows from this difference.
Conclusion: Beyond the Two Traditions
Realism and liberalism are technical traditions whose names invite misunderstanding. Realism is not a disposition towards toughness but a structural account of how the absence of authority above states shapes their conduct, and it comes in classical, structural and neoclassical forms that differ over where the causes of conflict lie. Liberalism is not political liberalism or naive optimism but a set of arguments about how commerce, representative government and institutions alter what anarchy produces, and it too accepts that states pursue their own interests. Understood accurately, the two are less opposed than their reputations imply: they share a good deal of ground, and their genuine disagreements concern the weight of relative gains, the independent effect of institutions and whether what states want can be assumed or must be explained.
Neither tradition holds the field alone. The most influential challenge to both came from constructivism, and in particular from Alexander Wendt’s argument, published in 1992, that anarchy has no fixed logic of its own, and that the meaning states attach to it, and to each other, is produced through their interaction. On this view, the self-help behaviour realists treat as a necessary consequence of anarchy is one possible outcome among several. Marxist, feminist and postcolonial approaches have raised further objections, questioning the state-centred focus of both traditions and the interests their categories take for granted.
We at Essydo use these concepts rather than position ourselves within them. For our analysis, it is important to assess political developments from different perspectives and often in different paradigms. The paradigms of realism and liberalism help us to do so. However, both of them, as well as other paradigms, such as those mentioned in the previous paragraph, have structural deficits. Maybe the most striking shortcoming of all is their focus on what is and what happened. What sets our devletist approach apart is that we find out what ought to be.
