Repair Work in the Sociology of Science

In The Social Construction of Housing Facts I kept close to Latour and Woolgar’s approach in Laboratory Life: the social construction of scientific facts (1979). I considered using some of Goffman’s concepts to enrich the analysis but decided that it would be wiser to stick to the text of Laboratory Life as I was concerned that even that might be hard to swallow, without introducing the further complication of Goffman.

The reason for this is that the Gouldner (1971) critique of symbolic interaction was unchallenged and widely accepted in urban political economy which was the dominant perspective in housing research until the mid-1980s, with the start of the sea-change marked by Esping-Andersen (1985) and Dickens et al (1985).

Constructionist Sociology of Science, and the work of Edge and Mulkay (1976), and Latour and Woolgar (1979) as well as the journal Social Studies of Science, was sufficiently distinct to enable me for the most part to keep the analysis within the new perspective.

Here I want to consider one issue – the repair work (Goffman, 1959) that quantitative housing researchers undertake to outflank criticisms that threaten to destabilise empirical categories of analysis, as a means of maintaining face in a potentially embarrassing situation, as well as the collusive co-operation of those making constructionist critiques, so as not to undermine potential applications for funding of research.

For the starting point of developing basic conceptual categories is a process that Latour and Woolgar term “the manufacture of data”. The equivalent in housing research are fundamental concepts such as household and dwelling which are in turn combined to generate second-order concepts such as overcrowding.

In the social sciences we manufacture data by the categories we choose and how we define them, not as in Latour and Woolgar (1979) in test-tubes:

“But the process of stabilisation is not just one which takes place as a once-and-for-all event. Rather, stabilisation must often be worked at to sustain the definition which has been agreed. Stabilisation is therefore also a micro process of interaction between individuals. A simple example can illustrate this. At a seminar held to discuss the purchase of data much of the discussion ranged around the issue of how to define a “household”. Now as already indicated, the concept of “household” has been widely agreed as being unproblematic, in spite of the severe methodological and empirical problems associated with its derivation.

However, the context of a discussion to purchase data revealed and indeed amplified the problematic nature of the concept. From becoming an established and stable fact, the concept of household began to “destabilise” to such an extent that the concept almost dissolved into incomprehensibility, and faith in its veracity as a real and vital fact or object began to fade. This destabilisation was only corrected when once again the complexities were glossed over and a popular and common definition of household was reaffirmed. The reaffirmation left unanswered the problematic nature of the concept for the purpose of defining the data to be purchased, but clearly if any progress was to be made at all then that was a small price to pay.” (Kemeny (1984 pp. 177-178).

Another example of repair work (that happened in the late 1990s) was in a constructionist sociology of economics seminar at the Institute for Housing and Urban Research (IBF), where the presenter (from another university) used Granovetter (1973) to argue that economics ultimately needs sociology to understand it. That produced the brilliant one-line retort by IBF Economics Professor Bengt Turner that – and here I loosely translate from Swedish – “if you peel away the layers of an onion you are left with empty hands and tears in your eyes”. It was a very effective way of harnessing humour to enable laughter to defuse a potential embarrassment.

The essential point of this is that housing researchers will not hesitate to collude with one another to maintain a definition of reality so as not to risk undermining a research project, grant application, or the standing of the discipline in the research institute or university. At the same time it can be used to attack and undermine a perspective that is potentially a rival perspective in a situation of power.

Embedded Links:

Gouldner’s critique of Symbolic Interaction:


Dickens, Peter, Simon Duncan, Mark Goodwin & Fred Gray (1985) State, Housing and Localities Methuen, London

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

Esping-Andersen, Gösta (1985) Politics against Markets: the social democratic road to power Princeton University Press

Goffman, Erving (1959) The Presentation of Self in Everyday Life Anchor, New York

Gouldner, Alvin (1971) The Coming Crisis of Western Sociology Heinemann, London

Granovetter, Mark S. (1973) “The Strength of Weak Ties” American Journal of Sociology (May) Vol. 78 No. 6 pp.1360-1380

Kemeny, Jim (1984) “The Social Construction of Housing Facts” Scandinavian Housing and Planning Research Vol. 1 No. 3 pp.149-164

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