Loading field map
The sociology of science is the study of science as a social activity. It examines how scientists actually work, how scientific knowledge is produced, validated, and changed, and how the institutions, norms, and cultures of science shape what comes to be accepted as true. Rather than asking what makes a claim scientifically correct in the abstract, the field asks how a claim becomes credible, who gets to make it, and under what social conditions it is accepted or rejected. Its central subject matter is not nature itself, but the human practices through which claims about nature are made, contested, and stabilized.
The field’s founding puzzle is often called the problem of scientific knowledge: if science is a social activity, how does it produce knowledge that appears to be objective, universal, and independent of the people who make it? Early sociologists of science did not deny that science works; they wanted to explain how it works as a social system. The key move was to treat scientific facts not as discoveries that force themselves upon passive observers, but as achievements that require coordinated human effort. This raised a question that has divided the field ever since: does the social character of science merely explain its errors, biases, and institutional quirks, or does it also explain the content of successful scientific knowledge itself?
The first sustained answer came from the functionalist school, associated with Robert K. Merton in the mid-twentieth century. Merton asked what made science such an effective institution for producing reliable knowledge. He identified a set of norms that he argued governed scientific communities: universalism (evaluating claims by impersonal criteria), communism (sharing findings openly), disinterestedness (seeking truth rather than personal gain), and organized skepticism (scrutinizing claims before accepting them). These norms, Merton argued, were not just ethical ideals but functional requirements for a community whose goal was cumulative knowledge. He also studied the reward system of science—priority disputes, citation patterns, and the prestige attached to being first—showing that scientists’ pursuit of recognition could align with the collective goal of advancing knowledge.
The Mertonian approach had a clear limit: it treated the content of scientific theories as largely outside its scope. It could explain why scientists followed certain rules and how the institution rewarded certain behaviors, but it took for granted that the methods of science, once properly followed, would lead to truth. The social dimension was seen as a framework around knowledge, not as something that entered into the knowledge itself. This left a gap that a later generation would exploit.
In the 1970s, a group of scholars at the University of Edinburgh, including David Bloor and Barry Barnes, challenged this division. They proposed what they called the strong programme in the sociology of scientific knowledge (SSK). Its central demand was symmetry: sociologists should explain true beliefs and false beliefs using the same kinds of causes. One should not say that true beliefs are caused by evidence and rationality while false beliefs are caused by social pressures. Instead, all beliefs—including those that later became accepted as scientific facts—should be explained by the same social and cognitive processes. The strong programme also insisted on impartiality (not taking sides on whether a belief is true), reflexivity (applying the same explanations to one’s own sociological claims), and causality (seeking the causes of belief).
This was a radical break. It meant that the content of scientific knowledge—the actual theories, models, and classifications—was now a legitimate object of sociological explanation. Bloor’s influential example was mathematics: even the most abstract mathematical truths, he argued, could be studied as social conventions that had been negotiated and institutionalized. The strong programme did not claim that scientific knowledge was false or arbitrary; it claimed that the distinction between true and false was not itself a sufficient explanation for why people came to believe one thing rather than another. Evidence alone, in this view, underdetermines belief: the same evidence can support different theories, so social factors must be invoked to explain which theory wins.
The strong programme’s most famous empirical studies examined historical controversies in science. For example, studies of the debate between Newton and his continental rivals over the nature of gravity, or of the early twentieth-century dispute between biostatisticians and Mendelians over the mechanisms of heredity, showed that the resolution of these debates depended not only on experimental results but on the social positions, institutional alliances, and rhetorical strategies of the participants. These case studies demonstrated that scientific controversies were not settled by a simple appeal to facts, because the facts themselves were often ambiguous or in dispute.
The strong programme was heavily criticized, often by scientists and philosophers who saw it as relativist or as denying the reality of scientific progress. Its defenders responded that they were not denying reality but insisting that reality alone cannot explain why people believe what they do. The debate was sharp, but the strong programme’s influence was lasting: it established that the content of science could be studied sociologically, and it opened the door to a wide range of empirical studies of scientific practice.
A second major approach emerged in the late 1970s and 1980s, partly in response to the strong programme’s focus on controversies and beliefs. Rather than studying disputes after the fact, a group of sociologists and anthropologists began to enter laboratories and observe scientists at work. This approach, often called laboratory studies, treated science as a form of craft work. Bruno Latour and Steve Woolgar’s study of a neuroendocrinology laboratory, published as Laboratory Life, showed how scientific facts were constructed through a series of material and rhetorical operations: instruments produced inscriptions, inscriptions were compared and interpreted, and claims were gradually hardened from tentative statements into established facts. The laboratory was not a place where nature was simply observed; it was a place where nature was transformed into texts, graphs, and numbers that could be circulated and debated.
Latour, along with Michel Callon and John Law, developed this into actor-network theory (ANT). ANT’s central claim was that scientific knowledge is produced through networks that include both human and non-human elements. A scientific fact is not created by a lone genius or even by a community of scientists; it is created by assembling a network of instruments, funding agencies, journals, animals, chemicals, and other scientists. Each element in the network—a microscope, a grant, a laboratory mouse—plays a role in making the fact credible. The term “actor” was deliberately extended to non-humans: a piece of equipment or a chemical compound could be an actor in the sense that it made a difference in the network. The task of the sociologist was to trace how these networks were built, how they were stabilized, and how they sometimes fell apart.
ANT differed from the strong programme in a subtle but important way. The strong programme still assumed that there was a social realm distinct from a natural realm, and that the sociologist’s job was to explain beliefs about nature using social causes. ANT refused this distinction. It argued that the division between nature and society was itself an outcome of scientific work, not a precondition for it. When a scientific claim becomes accepted, it is because a network has been built that makes the claim robust; the claim is not accepted because it corresponds to a pre-existing nature. This made ANT more radical than SSK, but also more difficult to use for explaining why some claims succeed and others fail, since it seemed to rule out any stable ground from which to make such judgments.
Laboratory studies and ANT were criticized for their focus on small-scale settings and for their apparent indifference to larger questions of power and inequality. A study of one laboratory could show how a fact was constructed, but it could not easily explain why some laboratories had more resources than others, or why some research areas were funded while others were not. These criticisms led to a partial convergence with other traditions, discussed below.
Alongside these studies of knowledge content, a parallel tradition continued to examine the institutional structures of science. This tradition, rooted in Merton’s work but extending far beyond it, studies the organization of scientific careers, the allocation of resources, the functioning of peer review, the structure of scientific disciplines, and the relationship between science and the state or the market. It asks questions like: How do scientific fields form and split? What determines who gets funded? How do norms of openness and secrecy vary across fields and countries? How has the commercialization of research changed the production of knowledge?
This institutional tradition has remained important because it addresses questions that the knowledge-focused approaches tend to ignore. The strong programme and ANT can explain how a particular claim becomes accepted, but they have less to say about why some research programs receive massive funding while others are starved, or why certain disciplines are dominated by particular countries or social groups. Institutional sociology of science fills this gap by studying the macro-level structures within which scientific work takes place. It has also been the site of important work on gender and race in science, showing how exclusionary practices have shaped who becomes a scientist and what questions are asked.
The relationship between the institutional tradition and the knowledge-focused traditions is not one of simple opposition. Many contemporary scholars draw on both. A study of a scientific controversy might use the strong programme’s symmetry principle to analyze the arguments on both sides, while also using institutional analysis to explain why one side had more resources or better access to journals. The two approaches answer different questions, and the field has largely settled into a division of labor rather than a continuing war.
A third major development, beginning in the 1980s and continuing today, is the turn to practice. This approach, associated with scholars such as Karin Knorr Cetina, Andrew Pickering, and Hans-Jörg Rheinberger, focuses on the material and embodied dimensions of scientific work. It argues that science is not primarily a matter of beliefs or arguments but of practices: manipulating instruments, preparing samples, writing papers, building machines, and developing skills. Scientific knowledge, in this view, is embedded in these practices and cannot be separated from them.
Knorr Cetina’s concept of epistemic cultures captures this idea: different scientific fields have different ways of producing knowledge, and these differences are not just superficial. A high-energy physics laboratory, a molecular biology lab, and a field ecology station each have their own characteristic practices, instruments, and forms of collaboration. The sociology of science, in this view, should not look for universal features of science but should describe the specific cultures of specific fields. Pickering’s work on the history of physics emphasized the role of machines and instruments in shaping what could be known: the development of particle accelerators, for example, did not just allow physicists to test existing theories; it created new objects of study and new possibilities for knowledge.
The practice turn has also been influenced by work in the history of science and by feminist scholarship. Feminist scholars such as Donna Haraway and Sandra Harding argued that the traditional image of science as a detached, objective pursuit concealed the ways in which scientific knowledge was shaped by gendered assumptions and by the social positions of scientists. Haraway’s concept of situated knowledge proposed that all knowledge is produced from a particular standpoint, and that acknowledging this situatedness is a better route to objectivity than pretending to be nowhere. This line of work connected the sociology of science to broader debates about knowledge and power, and it remains influential in contemporary discussions of science and society.
The sociology of science today is a diverse field with no single dominant paradigm. The strong programme’s insistence on symmetry is widely accepted as a methodological principle, even by scholars who do not identify with the Edinburgh school. Laboratory studies and the practice turn have made the detailed observation of scientific work a standard method. Institutional analysis continues to provide the macro-level picture. The field has also expanded in several directions.
One important expansion is the study of science and technology studies (STS) more broadly, which includes the sociology of technology and the study of science in public life. The sociology of science proper remains focused on the production of scientific knowledge, but it increasingly overlaps with work on scientific controversies in the public sphere, on the role of expertise in democratic decision-making, and on the relationship between science and social movements. The COVID-19 pandemic, for example, generated a large body of sociological work on how scientific advice was produced, communicated, and contested in real time.
Another expansion is the growing attention to the global dimensions of science. Much of the classic work in the field was based on studies of science in Europe and North America. Contemporary scholars have examined how scientific knowledge travels across borders, how colonial histories have shaped the global distribution of scientific institutions, and how local knowledge traditions interact with international science. This work has challenged the assumption that science is a single, universal enterprise, showing instead that it is a set of practices that are always locally situated even when they claim universal validity.
A third area of growth is the study of data and computation. The rise of big data, machine learning, and algorithmic decision-making has raised new questions for the sociology of science. How are data sets constructed, and what assumptions are built into them? How do algorithms shape what counts as evidence? How do new forms of automated knowledge production change the social organization of science? These questions are being addressed by a younger generation of scholars who draw on the field’s established tools while adapting them to new objects of study.
The field’s central tension remains what it has been since the 1970s: how to reconcile the social character of scientific knowledge with its apparent success in predicting and controlling the natural world. Some scholars continue to argue that the social construction of scientific facts is compatible with their truth; others maintain that the very idea of truth needs to be rethought. This debate is unlikely to be resolved, and its persistence is itself a sign of the field’s vitality. The sociology of science does not offer a single answer to the question of what science is; it offers a set of tools for investigating how science works, and it insists that this investigation is itself a scientific enterprise, subject to the same scrutiny it applies to its objects of study.