Collins, H. M., (1975) `The Seven Sexes: A Study in the Sociology of a Phenomenon, or the Replication of Experiments in Physics', Sociology, 9, 2, 205-224.
It is very difficult to resolve a scientific controversy through replication of experiments alone. This is because a scientist whose paper has been negatively replicated will argue that the second experiment was not properly carried out. This view can be supported almost indefinitely because experimentation is a skillful practice; there are no direct measures of the proper execution of a skill except getting the right result. But in a scientific controversy the notion of `right result' is itself the subject of dispute. In my 1985/92 book and subsequently I refer to this problem as `The Experimenter's Regress.' This early paper was trying to establish a tradition which would deal with the problem, which was set out as follows on its first page:
`... when we consider the grounds of knowledge we do it within an environment filled with objects of knowledge which are already established. To speak figuratively, it is as though epistemologists are concerned with the characteristics of ships (knowledge) in bottles (validity) while living in a world where all ships are already in bottles with the glue dried and the strings cut.'
The idea of this paper, and most of my work on science, is to get the ship out of the bottle!