Social Construction of Science

In its simplest form, the social construction of science means that there is no direct link between nature and our ideas about nature -the products of science are not themselves natural. This claim can be taken to mean different things and a distinction is often made between strong and weak interpretations of social constructivism. The stronger claim would not recognize an independent reality or materiality outside of our perceptions of it, or at least dismiss it as of no relevance as we cannot access it. This stance is, however, not a very common one. A weaker social constructivism tends to leave ontological queries to one side and instead focus on epistemological matters – how we gain knowledge about the world. What we count as knowledge is dependent on, and shaped by, the contexts in which it is created. Knowledge is thus made by people drawing on available cultural material, not preexisting facts in a world outside of human action, waiting to be uncovered.

The philosopher Ian Hacking has discussed and criticized different uses of the concept social construction. Hacking (1999) takes apart and analyzes the many and varying meanings of social construction. According to Hacking, the concept is routinely used in a way that makes it devoid of meaning. ”The phrase has become code. If you use it favorably, you deem yourself rather radical. If you trash the phrase, you declare that you are rational, reasonable, and respectable” (p. vii). Furthermore, the concept often comes with an inbuilt value judgment that implies that things should ideally be constructed differently.

When teasing out different meanings that different authors have given to social construction, Hacking found three main types: contingency, nominalism, and external reasons for stability (Sismondo 2004). The first kind of social constructivism essentially comes to mean that things could have been different – there was nothing inevitable about the current state of affairs and it was not determined by the nature of things. The second kind of social constructivism focuses on the politics of categories and points to how classifications are always human impositions rather than natural kinds. The third kind of social constructivism points to how stability and success in scientific theories are due to external, rather than evidential, reasons.

Whereas the idea of science and scientific knowledge as socially constructed can be traced to many a scholar, the very concept of social construction was introduced into mainstream social sciences by Peter L. Berger and Thomas Luckmann in their influential book The Social Construction of Reality. A Treatise in the Sociology of Knowledge (1966). In it, the authors combine ideas from Durkheim and Weber with perspectives from George Herbert Mead, to form a theory of social action. This theory would not only deal with plurality of knowledge and reality – for example what counts as knowledge in Borneo may make little sense in Bath and vice versa – but also study the ways in which realities are taken as known in human society. How is it that a concept such as gender is taken to be “natural” and “real” in every culture, while at the same time it is perceived and per formed very differently in different cultures? Knowledge about the society in which we live is ”a realization in the double sense of the word, in the sense of apprehending objectivated social reality, and in the sense of ongoingly producing this reality.” An objectivated social reality is a reality that is not ”private to the person who produced it, but accessed and shared by others. As humans we are continuously creating and recreating reality, and the role of the sociologist is to analyze the process of how reality is constructed, that is, how knowledge becomes institutionally established as real.

One way of understanding science as socially constructed is to point to obvious and ”external social factors, such as funding structures or political influences. These affect the way in which science develops; business interests can deter mine which projects are pursued, policy decisions can effectively close down entire avenues of research, and so on. The way in which research is institutionally organized is another much cited example of ”external social shaping of science – for example how heavy bureaucracy and strict disciplinary boundaries render the pursuit of transdisciplinary science difficult. Another variety of this brand of social constructivism is the argument that only scientific knowledge deemed to be ”relevant   or interesting will be pursued. Social theorists such as Helen Longino and Evelyn Fox Keller have pointed to how male dominance in society in general, and in the scientific profession in particular, has resulted in certain kinds of scientific knowledge. The definition of scientific problems and framing of hypotheses come with an inbuilt gender bias. Male contraception is an under researched area because reproductive responsibilities are firmly placed with women in our society and it is thus assumed that it is the female body that is to be manipulated. Such social values are also reflected in the very methods that scientists will use – most human trials of medicines are performed on young men between 18 and 20 years of age. The generic ”human   is thus a young man, whereas elderly women are the more likely consumers of the medicines that are being trialed.

Theorists such as Sandra Harding have argued that certain social positions – such as gender, race, or class – will render particular epistemological perspectives. A science conducted and shaped by black women would not contain the same knowledge as a science created by white men. What we call the collective body of knowledge in our society is really the knowledge of a dominant group – in this case men. This is not only due to ”female” questions falling outside the framework of what is perceived to be ”real science, but because our entire view of knowledge is a (male) ideological construct. A Cartesian dualism such as body/mind is a construction built on the male experience of nature and culture as separate entities, as men tend to be free to engage in intellectual activity without having to take responsibility for their own or others’ bodies. The precondition for this male focus on matters intellectual is that women take care of the shopping, cooking, child-rearing, laundry, cleaning, and other tasks that subsequently are not included in men’s abstract conceptualizations of reality.

Other prominent feminist thinkers, notably Donna Haraway, instead view (scientific) knowledge as fragmented and physically anchored (but not necessarily the epistemological perspective of a particular social group). A knowing subject’s perspective can always be located to a specific field – there is no objective ”view from nowhere.” The arguments above differ both in their conceptualizations of the subject and of society, but have in common that they see scientific knowledge as dependent on the social frameworks in which they are produced. Science is not a neutral activity, but instead reflects institutional values.

Scientists tend to insist that their way of arriving at knowledge makes their claims more true and more valuable than other groups’ knowledge claims (who arrived at their conclusions by different means and on different grounds). They argue that while it may be the case that certain types of knowledge – such as ideas about morality – are socially constructed, scientific knowledge should be exempt from such a mode of analysis. Scientific knowledge has a special authority and status because of the way in which we arrive at such knowledge. The ”scientific method” – rigorous and systematic examination, testing, and replication – thus guarantees the veracity of scientific claims. ”Truthfulness” is taken to mean that the claim in question is a direct representation of a reality that exists outside of, and independent from, our perceptions of it. A social constructivist view of science instead holds that scientific knowledge is as ”social” as other types of knowledge.

A social constructivist perspective of science common in the field of science studies emphasizes the social influence at the very core of technical judgments. Scientific theories, it is argued, are always underdetermined by empirical data – there are a potentially infinite number of hypotheses that could serve to explain the same set of data. Despite this, scientists manage to ”gel” around a limited number of possible explanations and eventually agree on which one they consider to be true. This process of ”truthmaking” is a social activity where the meaning of data is continuously being negotiated and renegotiated.

Sociologists of science have shown that scientific work in practice is rather more messy than in theory (Knorr-Cetina 1981; Collins 1985; Fujimura 1988; Pickering 1992) and that data always require interpretation, that machines are continuously calibrated to generate information that ”makes sense” (i.e., fits into a given frame of meaning), and that tests and models build on the assumption that the circumstances correspond precisely to ”real world” circumstances. Furthermore, experiments routinely go wrong and scientists spend a substantial amount of their time attempting to discipline wayward material and tweak variables until they work (Knorr-Cetina 1981; Latour & Woolgar 1986). The success of an experiment is determined by its outcome and thus measured against a host of prior assumptions about what ”nature” looks like and whether the result at hand corresponds with that nature. If the result is deemed to fit into that framework it will eventually become a fact and taken to be not only a good model of nature, but part of nature itself.

Social constructivists of all extractions also tend to argue that the success of science in claiming to be the highest form of knowing in part rests on its ability to appear as though it lacks both temporal and spatial location. Harry Collins uses the metaphor of a ship in a bottle -once in place, science, like the ship, appears to have a timeless quality, as though it has always been there and always will be. The processes by which science and scientific knowledge are produced tend to disappear from later narratives when scientific ”discoveries” or ”facts” are presented. A scientific fact is like a ship in a bottle; it is near impossible to conceive of how the ship was ever outside the bottle, because the bottle neck is far too narrow for the ship to have been pressed through it. At one time it was, however, a mere pile of sticks outside the bottle (Collins 1985: preface). The mistake we make is to assume that the ship has always been a ship, and, in the case of science, that the fact has always been a fact.

Another common source of scientific authority is the notion of ”objectivity.” The ”human factor” – that is, the scientist(s) who produced the knowledge – is made invisible, as are the circumstances under which the work was conducted. Porter (1992, 1995) has suggested that an inherently social relationship such as “trust” has taken the shape of objectivity by means of the apparent removal of individual, and therefore subjective, assessments. Porter (1995) argues that objectivity has nothing to do with truth and nature, but that it is instead the effort to exclude subjectivity – the ”struggle against judgment.” As scientific communities are growing increasingly larger and span several continents, trust has to be achieved at a distance, without personal contact (Luhman 1979; Giddens 1989; Porter 1995). Trust relations, previously negotiated in direct interaction or via a personal contact, have been institutionalized -the checkpoints are no longer embodied in scientists but rather located in seemingly impersonal sets of procedures. One of the most successful examples of this is quantification, an almost ubiquitous feature in today’s natural sciences.

The interesting question to pursue is then -to speak with Berger and Luckmann – how knowledge becomes reality. Other theorists prefer not to speak of reality in such relativist terms, but still seek to study the genesis and development of facts. The task of the social scientist is to unpack so called black boxes (i.e., unproblematic givens that we no longer question) and analyze the processes that went before this fact, or set of facts, became taken for granted knowledge. In order to study how knowledge claims come to be established as facts, when they were once merely one of many competing theories, analysts need to go ”upstream” and examine a time when these claims were more contentious (Latour 1987; Sismondo 2004). Latour and Woolgar (1986) show how an initially ”nonsensical” statement gradually becomes a reasonable claim, to then be labeled ”false,” only to retrieve its air of probability, to finally take on the status of a fact. This is done through a series of operations that aim to mobilize and ”hook up” with other facts, scientists, and artifacts (Latour 1987).

So why is science a social construction, rather than just a construction? Woolgar has criticized the inherent asymmetry in the so called interest model, where scientific knowledge is explained by reference to social interests held by individual stakeholders or groups, but the social interests themselves are taken as ”real” and stable entities. Latour (2005) purports not to be a social constructivist, but ”certainly a full blooded constructivist.” The first edition of Latour and Woolgar’s seminal work Laboratory Life had as its subtitle ”the social construction of scientific facts,” which in the second edition was changed to merely ”the construction of scientific facts.” According to Latour, an ideal subtitle would have read ”the practical construction of scientific facts.” He has sought to clarify this position by arguing for a ”constructivist realism.” The notion of construction must, according to Latour, be reconfigured altogether if science in action is to be understood. The trouble with the social constructivist view is that it builds on a false dual ism – objects are taken to reside in nature, whereas subjects dwell only in society. Latour’s alternative actor network theory (ANT) is a materialist theory which puts ”social” and ”natural” on an equal footing – studying scientists’ practices should reconfigure what we traditionally think of as ”social” just as much as it challenges our traditional views of ”science” or ”nature.” Latour does not want to talk about nature in the way it is commonly understood, but equally he does not want to talk of society. ”Society” has been ruined by sociologists and social constructivists, as they have made sure that it has been purged of what Latour calls not objects, but ”nonhumans.” If the social constructivist is to be believed, says Latour, only social relations exist in society. Furthermore, as nature is not awarded a reality status in its own right, but is simply a series of social inscriptions, the entire project becomes tautological. Latour thus disputes what he calls a dualist paradigm and seeks to avoid a subject-object distinction altogether. As ”society” has become tainted, he prefers instead the notion of ”collective.” This collective is extended to include nonhumans as well as humans. Latour’s society is constructed, but not socially constructed.


  1. Collins, H. (1985) Changing Order: Replication and Induction in Scientific Practice. Sage, Beverly Hills.
  2. Fox-Keller, E. & Longino, H. (Eds.) (1996) Feminism and Science. Oxford University Press, Oxford.
  3. Fujimura, J. H. (1988) The Molecular Biological Bandwagon in Cancer Research: Where Social Worlds Meet. Social Problems 35: 261-83.
  4. Giddens, A. (1989) The Consequences of Modernity. Stanford University Press, Stanford.
  5. Hacking, I. (1999) The Social Construction of What? Harvard University Press, Cambridge, MA.
  6. Haraway, D. (1991) Simians, Cyborgs, and Women: The Reinvention of Nature. Routledge, New York.
  7. Harding, S. (1991) Whose Science? Whose Knowledge? Cornell University Press, Ithaca, NY.
  8. Knorr-Cetina, K. (1981) The Manufacture of Knowledge. Pergamon Press, Oxford.
  9. Latour, B. (1987) Science in Action. Harvard University Press, Cambridge, MA.
  10. Latour, B. (1999) Pandora’s Hope: Essays on the Reality of Science Studies. Harvard University Press, Cambridge, MA.
  11. Latour, B. (2015) Bruno Latour. Online.
  12. Latour, B. & Woolgar, S. (1986) Laboratory Life. Princeton University Press, Princeton.
  13. Luhman, N. (1979) Trust and Power: Two Works. John Wiley, Chichester.
  14. Pickering, A. (Ed.) (1992) Science as Practice and Culture. University of Chicago Press, Chicago.
  15. Porter, T. M.  (1992) Quantification and the Accounting Ideal in Science. Social Studies of Science 22: 633-52.
  16. Porter, T. M. (1995) Trust in Numbers: The Pursuit of Objectivity in Science and Public Life. Princeton University Press, Princeton.
  17. Sismondo, S. (2004) An Introduction to Science and Technology Studies. Blackwell, Oxford.

Back to Top

Back to Sociology of Science