Social Construction of Scientific Facts

Social Construction of Scientific Facts, first published in jimsresearchnotes 14 Jan 2010, revised 13 April 2015.

Background: One of my enduring interests has been the constructionist sociology of science, from the late 1960s by reading Kuhn (1962) and Fell (1960). Fell (1960) was a short article, being her Presidential address to the International Society for Cell Biology given at its Paris meeting that year.

These two publications – one with a broad historical sweep, the other by a senior cell biologist illustrating shorter cycles of what the “in” trend was in a specialised branch of biology – epitomised for me the way constructionism could provide a critical understanding of science.

Others like Merton (1973) bringing together and developing his earlier pioneering work, Crane (1965, 1969, 1972) and Restivo (1974) reflected and encouraged a growing interest in the sociology of science. But it was the interactionist developments taking place in the sociology of science that rekindled the enthusiasm I had felt in the late 1960s.

One of the originators and leading proponents of the new interactionist school from the late-1960s onward was Michael Mulkay, who early on understood the need for researchers who were knowledgeable of the life-worlds of the scientists they researched (see Mulkay, 1969; Mulkay and Williams, 1971; Mulkay, 1972 and 1975; Edge and Mulkay, 1976). The first issue of Science Studies was published in 1971, though it was renamed Social Studies of Science in 1975, the year the Society for Social Studies of Science was founded.

For what I had hitherto lacked in the literature, was sustained insider deconstructionist analysis. It was all very well to critically analyse sciences other than ones’ own, but to be convincing we should be able to apply the same kinds of perceptions to our own area of research, much as Honor Fell did for cell biology. That this began now to be done in empirical research on the physical sciences such as radio astronomy involving collaboration between sociologists and specialists in the science under examination was largely due to Mulkay.(1) I decided I would write a paper applying these ideas to housing research, using Latour and Woolgar (1979). In a research note in preparation I will supplement Kemeny (1984) with further observations. Here, by way of introduction I provide a summary of Latour and Woolgar (1979).

Latour and Woolgar on endocrinologists: One of the early social studies of science books, Latour and Woolgar’s (1979) Laboratory Life: the social construction of scientific facts is a study of scientists at work in the endocrinology laboratory of the Salk Institute, California. It used a multi-disciplinary team of researchers so that the participant observation included researchers familiar with the processes in the specific science (or sciences) they were studying.

Laboratory Life analyses how laboratory staff used what Latour and Woolgar termed “technological inscription devices” to produce “evidence” to support a theory and to publish the results in a scientific paper. The idea behind this was to negotiate a degree of temporary stability of how “real” any scientific fact under investigation can be made, given widespread and deep disagreement between groups of researchers – and then defend and try to maintain that facticity in further publications when challenged.

The manufacture of data: inscription devices: Latour and Woolgar (1979 pp. 45-53) describe the taken-for-granted process of “data manufacture” as “Literary inscription” in a manner that treats data manufacture as “anthropologically strange” (Garfinkel, 1967). They define an inscription device as follows:

“…any item of apparatus or particular configurations of such items which can transform a material substance into a figure or diagram which is directly usable by one of the members of the office space.” (Latour and Woolgar, 1979 p.51)

In the case of this particular laboratory the material substance to be transformed was a tiny amount or rat brain. The manufacture process generated computer data sheets which were finally summarised in the form of graphs. Then followed a period of assessment, discussion and testing of the argument internally among the scientists.

“Once again the focus of attention shifted: the computer data sheets were filed away and it was the peaks and slopes of the curve which excited comment from participants in their offices: “how striking”, “a well-differentiated peak”, “it goes down quite fast”, “this spot is not very different from this one”. A few days later, the observer could see a neatly redrawn version of the same curve in a paper sent out for possible publication. If accepted, this same figure would be seen by others when they read the article and it was more than likely that the same figure would eventually sit on some other desk as part of a renewed process of literary juxtaposition and construction.” (Latour and Woolgar, 1979 pp.49-50)

The Malleability of Facticity: The purpose of this activity around data manufacture is to create, assemble and use an inscription device to produce reliable evidence that can be used in a scientific paper to increase the strength of the argument along a continuum of facticity. Scientific papers, citation-rates, and collecting and displaying a Greek chorus of research references to back up every claim made, taken together comprise an important part in the social construction of a successful increase in facticity for every claim. This has parallels with other professions based on argumentation and dispute, such as law.

Latour and Woolgar (1979 pp. 75-88) divide this continuum of facticity into five very approximate types:
Type 5: the strongest (taken for granted, self-evident or at worst barely needs stating), mostly used as starting points in arguing for an attempt to redefine another, weaker, statement into a stronger category of facticity.

Type 4: statements about relationships generally accepted as uncontroversial of the sort found in science textbooks.

Type 3: statements about relationships of a type 4 character though without being able to claim what the explanation or reason for the relationship might be. The relationship has become generally recognised as “a fact”, but its explanation remains speculative.

Type 2: claims concerning an explanation of a relationship of the sort that: “there is evidence to suggest that..” How strong that evidence is depends on how large and impressive a Greek chorus of agreeing authors (citations) can be assembled.

Type 1: conjectures or speculations about a relationship with weak, little or no evidence.

Some more general observationsLaboratory Life was for me the book that completed the phenomenological turn in Social Studies of Science, by deconstructing the production process of a scientific fact. This means that scientific facts cease to be never-changing absolute truths that are “out there”, but variable and unstable and therefore vulnerable to being challenged. In short, they become something to be argued over by scientists.

For science is no friction-free idyll of collegiality, but a competitive career in which reputation is pursued in often intense and sometimes very personal conflicts. Strong allies, high income, power in terms of control of resources, professorships and headships of departments and of research institutes, recognition in terms of nominations to international prizes all contribute to scientists being deeply enmeshed in power struggles at many different levels.

Bucher and Strauss (1961) coined the term “Professions in process”, which might equally have been called “sciences in process”. Controversies in science are never completed and settled, but like all else in society need to be continually “worked at”, because every “discovery” or “breakthrough” will be challenged, and new technological inscription devices will be invented or assembled by ambitious researchers struggling to climb to the top of their particular ant-heap. Some of these inscription devices will enable new data to be manufactured, which in turn may have a knock-on effect of changing the facticity status of a whole range of theories and hypotheses. Acquiring access to such resources often involves moving to research institutions which have the most advanced transcription devices and some of the best researchers.

Embedded Links

(1) Michael Mulkay was lecturer in sociology at Simon Fraser University 1966 to 1969 and then briefly a sociology colleague at Aberdeen University from 1969 to 1973, then York University. At Cambridge he held a research position and used it to collaborated with natural scientists some of whom, like Andrew Webster, David Edge, Nigel Gilbert and Steve Woolgar made the disciplinary switch to sociology and became prominent in social studies of science, a new and expansive constructionist problem area that Mulkay played a central role in developing.

Postscript: Mike Mulkay has since been critical of his own approach, and indeed the whole project of deconstructing the sociology of science. This will be discussed in a later post.


Bucher, Rue and Anselm Strauss (1961) “Professions in Process” American Journal of Sociology Vol. 66 No. 4 (Jan) pp. 325-334

Crane, Diana (1965) “Scientists at Major and Minor Universities: a study of productivity and recognition” American Sociological Review (Oct) Vol. 30 No. 5 pp. 699-714

Crane Diana (1969) “Social structure in a group of scientists: a test of the “invisible college” hypothesis” American Sociological Review (June) Vl. 34 No. 3 pp. 335-352

Crane, Diana (1972) Invisible Colleges: diffusion of Knowledge in Scientific Communities University  of Chicago Press

Edge, David O. and Michael Joseph Mulkay (1976) Astronomy Transformed: the emergence of radio astronomy in Britain Wiley

Fell, Honor B. (1960) “Fashion in Cell Biology ” Science (2 December) Vol. 132. no. 3440, pp. 1625-1627

Garfinkel, Harold (1967) Studies in Ethnomethodology Prentice Hall, Englewood Cliffs, NJ

Hiner, N. Ray (1973)
“Professions in Process: Changing relations among Social Scientists, Historians and Educators, 1880-1920” The History Teacher Vol. 6 No. 2 (Feb) pp.201-218

Kemeny, Jim (1984) “The Social Construction of Housing Facts” Scandinavian Housing and Planning Research Vol.1 No.3 pp.149-64

Kuhn, Thomas S. (1962) The Structure of Scientific Revolutions University of Chicago Press

Latour, Bruno and Steve Woolgar (1979) Laboratory Life: the social construction of scientific facts Sage Publications (Library of Social Research Vol. 80) Beverley Hills

Merton, Robert K. (1973) The Sociology of Science: theoretical and empirical investigations Chicago University Press

Mulkay M. J. (1969) “Some aspects of cultural growth in the natural sciences” Social Research Vol. 36 No. 1 pp. 22-52

Mulkay, M. J. and Anthony T. Williams (1971) “A Sociological Study of a Physics Department” British Journal of Sociology (Mar) Vol. 22 No. 1 pp.

Mulkay, M. J. (1972) The Social Process of Innovation Macmillan, London

Mulkay M.J, Gilbert G. N., Woolgar S. (1975) “Problem areas and research networks in science” Sociology Vol. 9 pp. 187-203

Restivo, Sal P. and Christopher K. Vanderpool (eds) (1974) Comparative Studies in Science and Society Merrill Publishing Company, Columbus

This entry was posted in Sociology of Science and tagged , , , , , , . Bookmark the permalink.