Automating Air Pumps: An Empirical and Conceptual Analysis

Abstract

Automating Air Pumps: An Empirical and Conceptual Analysis H. M . COLLINS AND M. KUSCH We want to introduce a theory of action in order to elucidate some old relationships. The way that technical skills, scientific closure, the building of technical devices, the automation of instruments, and the spread of scientific ideas fit together has been misunderstood. To clarify we need to understand how the skills needed to operate de­ vices change as the devices develop. We take as our example the automation of vacuum pumps. To see how automation is involved we need a new technical vocabulary. Operating an air pump today is different from operating an air pump in the time of Robert Boyle and Thomas Hobbes. We would expect that Boyle, or more particularly Boyle’s assistants, would need to be more noticeably inventive and skillful than those who operate air pumps nowadays; we would expect modern air-pump use to be less invested with doubts, less surrounded with the need to explore and establish what is going on, more typified by routines, and less obviously beset with the problem of tacit knowledge. We decided to find out what it was like to create a vacuum using a modern pump, and to record the experience. We would compare this with Boyle’s experience as described by Steven Shapin and Simon Schaffer.1 Dr. Collins is professor of sociology at the University of Southampton. His books include Changing Order: Replication and Induction in Scientific Practice (1985; Chicago, 1992), Artificial Experts: Social Knowledge and Intelligent Machines (Cambridge, Mass., 1990), and, with Trevor Pinch, The Golem: What Everyone Should Know about Science (Cambridge, 1993), which won the 1994 Robert K. Merton Award of the American Sociological Association. Dr. Kusch is lecturer in the Science Studies Unit of the Uni­ versity of Edinburgh. His books include Language as Calculus vs. Language as the Univer­ sal Medium: A Study in Husserl, Heidegger, and Gadamer (Norwell, Mass., 1989), Foucault’s Strata and Fields: An Investigation into Archaeological and Genealogical Science Studies (Nor­ well, Mass., 1991), and Psychologism: A Case Study in the Sociology ofPhilosophical Knowledge (London, 1995). The authors thank Bob Draper for his assistance and are also grateful to the Technology and Culture referees for useful and constructive criticism. 'Steven Shapin and Simon Schaffer, Leviathan and the Air Pump: Hobbes, Boyle and the Experimental Life (Princeton, N. J., 1987).© 1995 by the Society for the History of Technology. All rights reserved. 0040-165X/95/3604-0001$01.00 802 Automating Air Pumps: An Empirical and Conceptual Analysis 803 Both in describing what we did in using the pump and in compar­ ing it with the 17th-century experience, we find we need to define, explore, and develop a number of concepts. We look again at the concept of closure and its relationship to black-boxing and delega­ tion.2 We also need to use some concepts belonging to what has been called knowledge science.3 We define and show the application of the concepts polimorphic and mimeomorphic action, action plan, and action cascade.4 We elaborate the ideas of theoretical and behavioral indifference, which tie in with polimorphic and mimeomorphic ac­ tion, and we develop the concepts of active closure and passive closure and of mechanization. We conclude that understanding the historical development of a device such as the air pump requires us to separate the process of closure of debate from the process of mechanization, and separate them both from black-boxing and delegation. Mechani­ zation is not the direct consequence of closure even though it often follows chronologically; consensual science prior to mechanization is in a state of active closure; post mechanization we find passive closure. In the course of the article, we make use of these concepts to sketch out a “just so” story about air pumps. We do not claim this is history proper; rather, it is a model of what a history of a mechanical device might be like if it were informed by the theoretical structure we are trying to develop. We used two methods to find out what it was like to operate a modern vacuum pump. We looked at pumps and operated one our­ selves...