International relations (IR) is broadly classified as a social science, and in most universities worldwide it sits within a political science or social science department. That classification reflects the field’s core activity: studying how states, organizations, and other actors interact using systematic methods that aim to identify patterns, test theories, and explain outcomes. But the label “social science” understates the range of what IR scholars actually do, because the discipline also draws heavily on philosophy, history, ethics, and law in ways that push it toward the humanities. Understanding where IR fits means looking at both what it studies and how it studies it.
Why IR Counts as a Social Science
The social sciences are generally defined by their subject matter (human societies, institutions, and behavior) and by their commitment to studying those subjects through observation, evidence, and theory-building rather than purely through speculation or textual interpretation. By those criteria, international relations checks every box. Its subject matter is the behavior of states and non-state actors in the international system. Its theories attempt to explain recurring phenomena like war, trade, alliance formation, and cooperation. And a large share of its research output uses empirical methods that would be recognizable to economists, sociologists, or psychologists.
Institutionally, this classification is well established. IR programs typically live inside departments of political science or government, and the major journals publish alongside other political science research. The field’s professional associations overlap heavily with those of political science more broadly. A study examining IR’s disciplinary placement across different academic systems concluded that while IR is sometimes treated as an independent discipline, it is fundamentally part of social science.1ResearchGate. International Relations as Part of Social Science or Humanities?
The Empirical Side of IR
A significant portion of IR research is quantitative and looks a lot like what you would find in other social sciences. Researchers collect data on things like armed conflicts, trade flows, treaty ratifications, refugee movements, and voting patterns at the United Nations. They use statistical analysis to test whether certain conditions make war more or less likely, whether economic interdependence promotes peace, or whether democratic countries behave differently from authoritarian ones.
This quantitative tradition has deep roots. By the early 1980s, there was already a substantial body of empirical research on the causes of war, drawing on conflict datasets and statistical techniques to look for patterns across hundreds of interstate disputes.2Journal of Peace Research. The Quantitative Study of International Conflict: Quantity and Quality? An Assessment of Empirical Research That tradition has only grown. Projects like the Correlates of War dataset, which catalogs armed conflicts going back to the early nineteenth century, have become foundational tools that researchers use the way an epidemiologist might use a disease registry.
Formal modeling has been another pillar. Game theory and rational choice frameworks, originally developed in economics and mathematics, have been adapted extensively to study international conflict. A widely used approach treats war as the result of a bargaining breakdown: states prefer negotiated outcomes but sometimes fight because of private information, commitment problems, or indivisible stakes.3Annual Review of Political Science. Bargaining Theory and International Conflict These formal models generate precise, testable predictions about when conflicts escalate and when they get resolved, which is exactly the kind of work that defines a social science.
The Humanities-Leaning Side
If IR were just about data and models, its classification as a social science would be uncontroversial. But a large and influential strand of IR scholarship operates in ways that look much more like the humanities. Normative theory in IR, for instance, directly engages with moral philosophy. It asks questions like whether military intervention to stop a genocide is ethically justified, what obligations wealthy nations owe to refugees, and whether the current international order is fair. This work borrows concepts like cosmopolitanism and communitarianism from political philosophy and applies different schools of ethical reasoning to problems that arise between states.4Oxford Academic. Normative International Relations Theory
These are not empirical questions with testable answers. They are questions about values, justice, and what ought to be, and answering them requires the kind of interpretive and argumentative work that characterizes philosophy and the humanities. A scholar writing about the ethics of drone warfare is not testing a hypothesis. They are constructing an argument about right and wrong, drawing on moral traditions that predate the social sciences entirely.
Critical and post-structuralist approaches push even further from the social-science mainstream. These scholars interrogate the assumptions embedded in IR’s own theories, asking whose interests are served by concepts like “national security” or “the balance of power.” Some critical IR theorists treat truth itself as something constructed through social and political practices rather than something discovered through observation.5Millennium: Journal of International Studies. Truth, Values and the Value of Truth in Critical International Relations Theory That stance puts them closer to literary theory or continental philosophy than to the hypothesis-testing model that most people associate with the word “science.”
How Theorists Think About This Internally
IR scholars themselves have spent decades debating whether their field is, or should be, scientific. For much of the twentieth century, the discipline went through what are sometimes called “great debates.” One of the most consequential was the clash in the 1960s between traditionalists, who favored historical and philosophical approaches, and behavioralists, who argued that IR should adopt the methods of the natural sciences as much as possible. That tension never fully resolved. It just evolved.
Today, the field’s theoretical landscape is sometimes described as “post-paradigmatic,” meaning it has moved past the era when a few big schools of thought (realism, liberalism, constructivism) competed for dominance. Instead, there are multiple clusters of theoretical work that differ in how they study world politics and what they think counts as good explanation. One assessment identified three major clusters: choice-theoretic approaches that use formal models, experience-near approaches that emphasize thick description and context, and social-relational approaches that focus on networks and social structures. These clusters disagree on fundamental questions, like whether actors can be studied as autonomous decision-makers or whether they are shaped so deeply by their social context that separating them out is misleading.6European Journal of International Relations. International theory in a post-paradigmatic era: From substantive wagers to scientific ontologies
That kind of methodological pluralism is unusual among the social sciences. Economics has its internal debates, but the discipline is overwhelmingly committed to mathematical modeling and econometrics. Sociology is methodologically diverse but still broadly empirical. IR is distinctive in that a respected scholar can publish a book that reads like philosophy and another can publish a paper full of regression tables, and both are considered mainstream work within the same discipline. This is part of why the question “is IR a social science?” does not have a clean yes-or-no answer. The field’s center of gravity is social-scientific, but its edges extend well into the humanities.
Computational Tools and the Data Revolution
One trend pulling IR more firmly into the social-science camp is the growth of computational methods. Researchers are increasingly using techniques like machine learning, natural language processing, and automated text analysis to study international politics at a scale that was impossible a generation ago. You can now scrape news reports from dozens of countries, process them algorithmically, and build near-real-time datasets tracking protests, political violence, and diplomatic events worldwide.
This computational turn is still relatively young, and practitioners have noted a significant gap between what the tools can do and how widely they are being applied in the field. One review observed that “computational international relations” remains underdeveloped compared to computational approaches in other social sciences, despite the potential for these tools to test more advanced theories with larger and better-organized data.7arXiv. Computational International Relations: What Can Programming, Coding and Internet Research Do for the Discipline?
There are also real methodological pitfalls. Automated text analysis, for example, often relies on English-language sources, which introduces bias when studying events in non-English-speaking countries. Researchers have flagged this as a serious concern, noting that the data sources used to track contentious political events like protests and riots are being diversified, but slowly.8Global Studies Quarterly. Bias in Text Analysis for International Relations Research If your conflict dataset is built mostly on Reuters and the Associated Press, you may be measuring media coverage patterns as much as actual conflict patterns. That kind of methodological self-awareness is itself a hallmark of social science: the field is actively scrutinizing the validity of its own instruments.
Where IR Meets Other Disciplines
One of the things that makes IR hard to pin down is how aggressively it borrows from everywhere. International law scholars contribute legal analysis. Historians provide case studies and archival research. Economists build models of trade and sanctions. Psychologists study how individual leaders make decisions under uncertainty. Anthropologists examine how local cultures shape the meaning of global governance. Even natural scientists are becoming part of the conversation, as IR scholars grapple with issues like climate change, pandemics, and resource scarcity that cannot be understood without some grasp of the physical world.
Climate change is a good example of how this interdisciplinary borrowing works and where it creates tensions. IR scholars have traditionally analyzed climate through the lens of state security: will droughts cause migration, will migration cause conflict, will resource competition destabilize regions? But some recent scholarship argues that this framing is too narrow, because it treats security as something defined by existing power structures rather than asking who is actually threatened and who gets to define the problem. Work that takes this broader view deliberately draws on insights from outside the IR discipline, blending climate science, critical geography, and political ecology.9International Relations. Fit for purpose? Climate change, security and IR
This interdisciplinary sprawl is both a strength and a complication. It means IR can tackle almost any question that involves actors crossing borders or institutions operating above the level of a single state. But it also means the field lacks the methodological coherence that disciplines like economics or psychology have (for better or worse) achieved. Whether you see that as intellectual richness or disciplinary confusion depends on what you think a social science should look like.
What Students and Outsiders Often Get Wrong
People encountering IR for the first time frequently expect it to be more predictive than it is. If it is a science, surely it should be able to tell us whether two countries are about to go to war, or whether a sanctions regime will work, or whether a peace deal will hold. In practice, IR’s predictive track record on specific events is modest at best. The field is better at identifying general patterns and risk factors than at forecasting particular outcomes. This is not unique to IR; most social sciences struggle with prediction, because human behavior is influenced by too many variables, many of which are impossible to observe in advance.
Another common misconception is that the theoretical schools in IR are competing to be “right” in the way that competing hypotheses in physics might be. Realism, liberalism, and constructivism are better understood as lenses that highlight different features of international politics. Realism focuses on power and security competition. Liberalism emphasizes institutions, trade, and cooperation. Constructivism examines how ideas, norms, and identities shape state behavior. Each captures something real, and most practicing scholars draw on multiple traditions depending on the question they are studying. The introductory textbook version, where you pick a team and argue for your paradigm, does not reflect how the field actually works.
A third misunderstanding is that IR is just a branch of political science with a global flavor. While IR is often housed in political science departments, many of its most active research areas involve questions that political scientists studying domestic politics would not recognize as their own. The study of international organizations, transnational advocacy networks, global health governance, and the laws of war are all central to IR but peripheral or invisible in domestic political science. Some scholars argue that IR is better understood as its own interdisciplinary field that overlaps with political science rather than being contained within it.10ResearchGate. International Relations as Part of Social Science or Humanities?
The “Science” Question Behind the Question
When people ask whether IR is a social science, they sometimes have an underlying concern: can I trust what IR scholars say the way I trust findings from a lab? The honest answer is that it depends on what kind of IR scholarship you are reading. A large-n statistical study linking economic sanctions to regime behavior is built on data, has defined variables, and can be checked by other researchers. A normative argument about whether humanitarian intervention violates state sovereignty is built on reasoning and values, and two thoughtful people can disagree about it indefinitely. Both are legitimate forms of scholarship. Only the first fits neatly into what most people mean by “science.”
The field is increasingly aware of this distinction and has seen growing interest in research transparency and replicability, following broader trends in the social sciences. Qualitative IR researchers have debated how to make their case studies and process-tracing more rigorous and open to scrutiny, while quantitative researchers face the same replication challenges that have roiled psychology and economics. None of this disqualifies IR from being a social science. It just means the field is grappling with the same methodological growing pains as its neighbors, while also containing a substantial wing of scholarship that was never trying to be scientific in the first place and does not need to be.
How Institutional Placement Varies Around the World
Where you study IR affects how it gets framed. In the United States and much of the English-speaking world, IR is overwhelmingly treated as a subfield of political science, and its methods lean heavily empirical. In parts of continental Europe, IR has stronger ties to law, philosophy, and history, and normative and critical approaches carry more weight. In some countries, IR operates as a fully independent department rather than a subsection of political science, which gives it more freedom to define its own methodological identity.
These institutional differences matter because they shape what students are trained to do and what counts as good research. An IR graduate student in a U.S. program will likely be trained in statistics and research design. A student in a European program may spend more time reading critical theory and international law. Both are studying international relations, but the methodological emphasis can be strikingly different, which is another reason the “is it a social science?” question does not have a universal answer. The discipline’s character shifts depending on where you stand within it.