The programme I have in mind has been most nearly formulated by David Bloor in Knowledge and social imagery (London, 1976). His ‘strong programme’ contains certain unnecessary prescriptions but is nevertheless the most forceful available statement.
2.
These terms are due to Karl Popper.
3.
Such as Farley and Geison's conclusion that it was only because Pouchet failed to take up Pasteur's challenge in the last instance, that Pasteur's (correct) view concerning spontaneous generation won the day so easily. It would seem that the planned experiments would have given support to the spontaneous generation view. FarleyJ. and GeisonG. L., “Science, politics and spontaneous generation in nineteenth century France: The Pasteur-Pouchet debate”, Bulletin of the history of medicine, xlviii (1974), 161–98.
4.
Of course there would still be a role for a notion of incorrect scientific method. One role would be as a description of cases where all parties were agreed that incorrect procedures had been used. And, of course, the notion of incorrect method would remain as a native members' device for describing experiments considered to be wrong, or incompetently performed — see below.
5.
The superstructure of Mertonian norms, such as universalism and organized scepticism, is built upon a base of assumptions of this sort. For a discussion of replication see CollinsH. M., “Upon the replication of scientific findings: A discussion illuminated by the experiences of researchers into parapsychology”, Proceedings of the 4S/ISA First International Conference on Social Studies of Science, Cornell University 1976 (available from author).
6.
Pickering in an unpublished paper, “Discovery claims, ad hominem arguments and research programmes: Case studies in elementary particle physics” (1978/9), draws a related distinction between what he calls the “local group” of competent experimentalists and the rest of the scientific community.
7.
The claims in this paragraph should more properly be cast as hypotheses since no systematic work has been done on frequency of replication and so forth. I put the claims forward with some confidence since they are confirmed by many of my scientist colleagues.
8.
Note that the definition involves two elements: (1) active involvement, (2) having an effect. Neither element on its own is a sufficient defining criterion. I believe that some have misread this paper, and tend to misread the case studies by taking (2) as a sufficient criterion. This is then read backwards and taken to imply that all those who have an effect on the outcome of a controversy are members of some small set of scientists close to the experimental work. On the contrary the body of work on scientific controversies, referred to here, shows that the activities of the ‘core-set’ alone cannot determine the outcome of controversies, so that the set of “those who have an effect on the outcome of a scientific controversy” may be very large indeed. It may include not only the wider community of scientists but also all those (often described in terms of the social structure) who contribute to our ‘mode of production’, ‘world view’, etc.
9.
Relevant papers will be found together in the tenth anniversary issue of the journal Social studies of science. This will be the first issue of the 1981 volume. It will comprise papers by Collins on gravity waves, Harvey on non-locality, Pickering on magnetic monopoles, Pinch on solar neutrinos and Travis on memory transfer. Vol. v of the Sociology of the sciences yearbook, due to appear in 1981, will contain papers by Harvey, Pickering, Pinch and Travis. Pickering has a paper on quarks, forthcoming in Isis. Relevant papers currently available include CollinsH. M., “The seven sexes: A study in the sociology of a phenomenon, or the replication of experiments in physics”, Sociology, ix (1975), 205–24, and CollinsH. M., “Upon the replication” (ref. 5).
10.
Collins, “Seven sexes” (ref. 9).
11.
Private correspondence.
12.
For the notions of constitutive and contingent forums see CollinsH. M. and PinchT. J., “The construction of the paranormal: Nothing unscientific is happening”, in WallisRoy (ed.), On the margins of science: The social construction of rejected knowledge. Sociological review monograph, xxviii (Keele, 1979), 237–70.
13.
Collins, “Upon replication” (ref. 5) for full account.
14.
MagyarG. in his “Pseudo-effects in experimental physics: Some notes for case studies”, Social studies of science, vii (1977), 241–67, has noted the progress of a number of controversies as recorded in the literature. He finds that such controversies die out very quickly. I am indebted to Magyar for evidence relating to the points in my next paragraph. Magyar, however, misses the fact that proponents of defeated views do not give up so easily, even though they may become invisible within the literature. Evidence on this point is available from the case studies, and from the editorial comments by R. N. Hall in his transcription of a talk by Irving Langmuir available from the General Electric Company: LangmuirI., “Pathological science”, transcribed and edited by HallR. N., General Electric Technical Information Series No. 68-C-035 (1968).
15.
Again, this claim has the status of a hypothesis since, so far as I know, no survey of the typical style of work of most scientists has ever been carried out. What I have very much in mind here, apart from conversations with colleagues, is Kuhn's characterization of most science as ‘normal science’. See his Structure of scientific revolutions (Chicago, 1962). Kuhn's characterization has been disputed of course and it may be that scientists experience controversy more frequently than this paper would imply. It is probably time to do some systematic research on the comparative experiences of scientists working in ‘normal’ and controversial areas.
16.
The suggestion that a written account should seem more compelling than the event it describes is not at all paradoxical within an ‘interpretivist’ approach. The interpretivist is aware of the infinite disorder—or meaninglessness of events until they are ordered by human activity. This infinite disorder is very conveniently reduced to apparent order in a written account. The written account hides the unformed nature of what it is supposed to represent. The process of ‘accomplishing objectivity’ is discussed, and illustrated with a case study of parapsychology, by CollinsH. M. and PinchT. J., Frames of meaning. The social construction of extraordinary science (London, 1981). It has been very well described, with reference to an empirical study of biological science, in LatourBruno and WoolgarS. W., Laboratory life: The social construction of scientific facts (Beverly Hills, 1979).
17.
I realise that no-one is surprised any more that scientists cannot give clear-cut answers to ‘foggy’ questions such as “How safe are nuclear power plants?”.
18.
One persistent criticism of this paper is that the uncompelling nature of experiment is known to a wider body of scientists than I suggest here. Latour has backed up this point by reference to his work on biochemical research (ref. 16). He claims that in this relatively uncontroversial area scientists were aware of the provisional nature of their findings and the need to give them social solidity before they became facts. Pickering has suggested, in correspondence, that the impression of solid reliability is widespread only outside the scientific community. These suggestions do not square with my experience, and again, I feel that more comparative research is required.
19.
Latour has tried to stress the point that socially constructed facts are as solid as anything can be in his paper entitled “The three little dinosaurs or a sociologist's nightmare”, Fundamenta scientiae, i (1980), 79–85.
20.
Ref. 3.
21.
Roll-HansenN., “Experimental method and spontaneous generation: The controversy between Pasteur and Pouchet, 1859–64”, Journal of the history of medicine and allied sciences, xxxiv(1979) 273–92.
