Understanding Social Science

A common objection goes something like this: philosophy is fine for armchair speculation, but real science is about data, experiments, and results. This objection misunderstands what philosophy actually does in relation to science. Philosophy does not compete with empirical research. It examines the assumptions that make empirical research possible in the first place. Before you can collect data, you have to decide what counts as data. Before you can run a regression, you have to decide what you are measuring and whether that measurement captures the concept you care about. Those prior decisions are philosophical, and they shape everything that follows.

Consider a concrete example. Suppose a researcher wants to study whether income inequality causes social unrest. Before a single survey is sent out, the researcher must answer several philosophical questions. What does "income inequality" mean, and how should it be measured? Is the Gini coefficient the right tool, or does it obscure important distributional details? What counts as "social unrest"? Does it include peaceful protest, or only property destruction? And what would it mean for inequality to cause unrest, rather than merely correlate with it? Each of these questions has a philosophical dimension, and answering them poorly produces research that looks rigorous but rests on shaky ground.

The history of social science is full of cases where unexamined philosophical assumptions led researchers astray. Early twentieth-century eugenics programs drew on social science research that treated race as a natural biological category and intelligence as a fixed, measurable quantity. Both assumptions were philosophically and empirically wrong, but they were rarely questioned because they seemed obvious to the researchers who held them. The damage those programs caused was not just a failure of ethics. It was a failure of philosophical scrutiny.^[1]

More recent examples are less dramatic but equally instructive. Replication crises in psychology and economics have revealed that many celebrated findings do not hold up when studies are repeated under slightly different conditions.^[2] Part of the explanation is statistical, involving underpowered studies and publication bias. But part of it is philosophical. Researchers were often unclear about what exactly they were measuring, whether their operationalizations were valid, and what their statistical results actually licensed them to conclude. These are questions about Measurement Theory, Validity of Measurement, and Statistical Inference, all of which belong to the philosophy of social science.

Philosophy also matters because the social sciences study something that the natural sciences do not: beings who interpret their own situation and act on those interpretations. A rock does not change its behavior because it has read a geology textbook. But people do change their behavior when they learn what social scientists say about them. This reflexive quality of social life creates philosophical challenges that have no parallel in physics or chemistry. Understanding those challenges requires tools that go beyond standard scientific methodology.

None of this means that social science is impossible or that its findings are unreliable. It means that doing social science well requires philosophical awareness. Researchers who understand the assumptions behind their methods are better positioned to use those methods appropriately, to recognize their limits, and to communicate their findings honestly. That kind of awareness is what separates rigorous social inquiry from the production of impressive-looking numbers that mislead more than they inform.

For you as a reader, philosophical literacy about social science has a different but equally important payoff. You encounter social science claims constantly, in news articles, policy debates, advertising, and public health guidance. Being able to ask the right questions about those claims, about what was measured, how causation was established, whose perspective shaped the research, and what assumptions were built in from the start, gives you a form of critical intelligence that is genuinely useful in navigating a world saturated with data.

Every research study rests on a set of assumptions that are rarely stated explicitly. Some of these assumptions are methodological, concerning how data should be collected and analyzed. Others are epistemological, concerning what counts as knowledge and how it can be justified. Still others are ontological, concerning what kinds of things exist in the social world and how they relate to each other. Most researchers absorb these assumptions from their training and disciplinary culture without ever examining them directly. That is not a criticism. It is simply how disciplines work. But it does mean that the assumptions are there, operating quietly beneath the surface of every study.

Take the assumption of Methodological Individualism, which holds that social phenomena should ultimately be explained in terms of the beliefs, desires, and actions of individual people. This assumption is deeply embedded in mainstream economics and much of cognitive psychology. It shapes which questions researchers ask, which variables they measure, and which explanations they find satisfying. A researcher who assumes Methodological Individualism will tend to explain rising divorce rates by pointing to changes in individual preferences or incentives. A researcher who rejects that assumption in favor of Holism might instead point to shifts in labor markets, housing costs, or cultural norms as structural forces that shape individual behavior from the outside. Both researchers might be looking at the same data, but their hidden ontological assumptions lead them to very different places.

Another set of hidden assumptions concerns the relationship between facts and values. Many social scientists were trained in a tradition that treats objectivity as the absence of values, as if a truly scientific researcher would approach their subject with no prior commitments or perspectives. This view, sometimes called the fact-value distinction, has been challenged extensively in the philosophy of science. The choice of what to study, which populations to include, which outcomes to measure, and how to frame findings all involve value judgments. Pretending otherwise does not eliminate those judgments. It just makes them invisible, which is worse.

Consider how these hidden assumptions show up in practice. Here are three common types and what they look like in real research:

• Ontological assumptions: A study of poverty that measures only individual income and ignores neighborhood infrastructure, social networks, and institutional access is implicitly assuming that poverty is primarily an individual condition rather than a structural one.
• Epistemological assumptions: A study that treats survey responses as direct windows into people's beliefs is assuming that self-report is a reliable form of evidence, without asking whether people's stated beliefs match their actual motivations or behaviors.
• Causal assumptions: A study that uses a randomized controlled trial to test a social intervention is assuming that the causal mechanism works the same way across different contexts, populations, and time periods, an assumption that is often questionable in social settings.

The Demarcation Problem is another area where hidden assumptions do significant work. This is the question of what distinguishes genuine science from non-science or pseudoscience. Karl Popper's famous answer was Falsifiability: a claim is scientific if it can, in principle, be proven wrong by evidence.^[3] This criterion has been enormously influential, but it creates difficulties for social science. Many important social theories, including psychoanalytic theory, certain versions of Marxist historical analysis, and some rational choice models, are structured in ways that make them very hard to falsify. Does that mean they are not scientific? Or does it mean that Falsifiability is not the right criterion for demarcating science in the social domain? These are not trivial questions, and how you answer them affects what kinds of research you take seriously.

Hidden assumptions also operate at the level of language. The concepts social scientists use, things like "social capital," "rational actor," "culture," or "class," are not neutral descriptions of pre-existing reality. They are theoretical constructs that carve up the social world in particular ways, emphasizing some features and obscuring others. When economists model people as rational utility-maximizers, they are not simply describing how people behave. They are adopting a theoretical framework that makes certain kinds of analysis possible while making others difficult or invisible. Understanding that these concepts are constructs, not mirrors of nature, is a key step in philosophical literacy about social science.

One practical way to start uncovering hidden assumptions in any piece of research is to ask a set of diagnostic questions. These questions will recur throughout this book in various forms:

1. What is being measured, and does the measurement actually capture the concept the researcher claims to be studying?
2. What causal claims are being made, and what evidence supports them beyond correlation?
3. What is assumed to exist in the social world, and are those assumptions justified?
4. Whose perspective shaped the research design, and whose was excluded?
5. What would it take to show that the main finding is wrong?

These questions are not meant to be destructive. They are not a way of dismissing research or claiming that nothing can be known. They are tools for reading research more carefully and for distinguishing findings that rest on solid philosophical foundations from those that do not. Developing the habit of asking them is one of the most valuable things this book can help you do.

The chapters ahead will give you the vocabulary and conceptual frameworks to ask these questions with precision. Each chapter focuses on a specific philosophical domain, from Epistemology and Social Ontology to causal reasoning, methodology, objectivity, and explanation. Together, they build a comprehensive picture of the philosophical landscape that every social scientist navigates, whether consciously or not.

This book is organized to move from foundational questions toward more applied ones. The early chapters deal with the most basic philosophical issues: what knowledge is, what exists in the social world, and how causation works. The middle chapters examine how those foundational issues play out in specific methodological contexts, both quantitative and qualitative. The later chapters address the broader questions of objectivity, explanation, and the major metaphysical positions that researchers adopt when they think about the status of their own findings. Each chapter builds on the ones before it, but they are also designed to be read independently if you come to the book with a specific question in mind.

The chapter on Epistemology examines what counts as knowledge in social science and how the debate between Empiricism and Rationalism shapes what researchers accept as valid evidence. The chapter on Social Ontology explores what kinds of things actually exist in the social world, from individual actors to Social Structures and emergent phenomena, and how different answers to that question lead to different research designs. The chapter on causation unpacks the logic of Causal Mechanism and Counterfactual Causation, explaining why establishing causation in social science is so much harder than it looks and why Correlation vs. Causation remains one of the most misunderstood distinctions in public discourse about research.

The chapters on quantitative and qualitative methods are not how-to guides for running regressions or conducting interviews. Those skills are covered in methods textbooks. Instead, these chapters examine the philosophical foundations of each approach. The quantitative chapter addresses Measurement Theory, Operationalization, and the limits of Statistical Inference. The qualitative chapter explores Hermeneutics, the Hermeneutic Circle, Verstehen, and what Thick Description actually means as a methodological commitment. Both chapters are designed to help you understand not just what researchers do, but why they do it and what philosophical commitments those choices reflect.

The chapter on objectivity and values addresses one of the most contested issues in the philosophy of social science. The concept of Value-Ladenness, the idea that research is always shaped by the values of the researcher and the broader culture, is examined alongside Standpoint Epistemology and Positionality. The goal is not to conclude that objectivity is impossible, but to develop a more sophisticated understanding of what objectivity can realistically mean in social inquiry and how researchers can manage rather than pretend to eliminate the influence of values.

The chapter on explanation examines three major models: the Covering-Law Model, which treats explanation as subsumption under general laws; interpretive understanding in the tradition of Verstehen; and Narrative Explanation, which explains events by placing them in a meaningful sequence. The final substantive chapter examines the three major metaphysical positions that researchers adopt: Scientific Realism, Social Constructivism, and Critical Realism. Understanding these positions is important because they shape not just how researchers interpret their findings, but what they think their research is ultimately about.

A few practical notes on how to get the most from this book. First, the philosophical concepts introduced here are tools, not ends in themselves. The goal is not to memorize definitions but to develop the ability to apply these concepts when you encounter real research. As you read each chapter, try to connect the philosophical ideas to specific studies or claims you have encountered in your own reading or work. The concepts become much more useful when they are attached to concrete examples.

Second, the chapters are designed to be read actively. When a chapter presents a philosophical position, ask yourself whether you find it convincing and why. When it presents a counterargument, take it seriously rather than dismissing it. Philosophy of Social Science is a field where reasonable people disagree, and the disagreements are often productive. Engaging with them honestly is more valuable than arriving at a fixed set of conclusions.

Third, the canonical terms used throughout this book are precise and consistent. Terms like Falsifiability, Methodological Individualism, Hermeneutic Circle, Operationalization, and Generative Mechanisms have specific meanings that matter. When you encounter them, pay attention to those meanings rather than substituting everyday synonyms. Precision in philosophical vocabulary is not pedantry. It is what makes it possible to think clearly about difficult questions.

The goal of this book is not to make you skeptical of social science. It is to make you a more discerning consumer and, if you are a researcher, a more self-aware practitioner. Social science at its best is one of the most important intellectual enterprises humans have developed. Understanding its philosophical foundations is what allows it to live up to that potential.

Finally, the conclusion of this book does not simply summarize what came before. It examines how the philosophical insights from each chapter connect to each other and points toward open questions that the field is still working through. The philosophy of social science is not a settled body of doctrine. It is an ongoing conversation, and the most interesting parts of that conversation are still happening. By the time you reach the end of this book, you will have the conceptual equipment to follow that conversation and, if you choose, to contribute to it.

What follows is not a survey of social science findings. It is an examination of the ground those findings stand on. That ground is less stable than it sometimes appears, but understanding its contours is what makes it possible to build on it well. The questions ahead are genuinely hard, and they do not all have clean answers. That is not a reason to avoid them. It is precisely the reason to take them seriously.