If there are no guides to the construction of the history of science: if the task is more than identifying precedents to the present, if narratives of class interest or other overriding social determinants of scientific knowledge are rejected, and if (as the Great Escape has it) philosophy is no guide to how knowledge develops and spreads, then the danger arises that history drops into a deeply contingent state that can only be successfully analyzed at the most local level.
Arriving in this position, one must resolve the absurdity by asking: what sorts of things can be the subject of historiography? I would reply that science studies has successfully argued that traditions of practice constitute the possible objects of historical inquiry. If definable physical conditions can persist through history (mountains are a nicely tangible example), then certain definable practical reactions to those conditions can also persist (climbing the mountain, digging mines, etc…). Such practices can be broken down into analytically useful categories: technology, technique, tactic, policy, arguments and knowledge claims, rhetoric, imagery, etc. Properly characterized varieties of these practices can be given useful labels (e.g. “empire-building” as a national policy).
Specific choices concerning how to deploy these practices in varying situations are informed by the history of ideas. Ideas may be decoded through a careful analysis of how practices are selected based upon historical appreciations of the character of situations faced (“imperialism” suggests territorial acquisition as a response to international economic and military competition). The outcome of choices is indeed contingent and local, but traditions in practice and the ideas that inform choices of practice may be traced, often for very long periods.
An understanding of practices, therefore, precedes a history of ideas, even if these practices are as cerebral as philosophical argumentation. Practices, unlike ideas, but like physical conditions, can be identified by their material characteristics. These characteristics can be obviously material (the design of automobiles through history), or materiality can be argued to inhabit even something like philosophy, through similarities in stock arguments, stock examples, lexicon, and grammar.
As I have suggested, in the historiography of science Peter Galison has offered both one of the best articulations of these methodological points and one of their best deployments. If he left How Experiments End (1987) with a vision of a local and contingent world where explanations of sequences of events could only be parsed by the historian—not the philosopher or the sociologist—by Image and Logic: A Material Culture of Microphysics (1997), he had developed a set of ideas about how constructive historiography could be assembled. I want to highlight three main ideas.
1) Mesoscopy (p. 61), the persistence of traditions that nevertheless do not philosophically transcend history: “This book is a brief for mesoscopic history, history claiming a scope intermediate between the macroscopic (universalizing) history that would make the cloud chamber illustrative of all instruments in all times and places and the microscopic (nominalistic) history that would make Wilson’s cloud chamber no more than one instrument among the barnloads of objects that populated the Cavendish Laboratory during this century.”
2) Materiality (p. 435), a quality that identifies mesoscopic traditions and allows for their persistence and replication: “From the Latin for ‘handing down’ or ‘delivery’, the term tradition had, by the Renaissance, an interlinked set of connotations. First was the handing over of material, as in the delivery of goods. But there is also an early notion of tradition as a set of practices, customary behaviors that may or may not have the force of law…. ‘Handing down’ in the material culture of the laboratory has, I have argued, taken on many forms… It can be as literal as the passing on of laboratory equipment… It can be partial, as in the ‘cannibalization’ of instrument parts from one experiment to another… But the transfer can also be of a more subtle kind… [Here Galison talks about material ‘bricolage’, a term tracing its tradition to Lévi-Strauss].”
The concept of materiality has been imported into studies of the history of cerebral things like scientific theory and calculation. Of course, persisting with the physical connotation of materiality, the term “paper tools” has had some appeal. See especially David Kaiser on Feynman Diagrams and Andy Warwick on the written Cambridge mathematical tripos exam. Importantly, both scholars have been centrally concerned with pedagogy, a key issue in explaining the intergenerational persistence and spread of materially related practices.
3) Intercalation (pp. 781-803), the weakness of interaction between materially distinct traditions. “…I want to reflect … on a description of physics that would neither be unified nor splintered into isolated fragments. I will call this polycultural history of the development of physics intercalated because the many traditions coordinate with one another without homogenization. Different traditions of theorizing, experimenting, instrument making, and engineering meet—even transform one another—but for all that, they do not lose their separate identities and practices” (p. 782).
Distinct traditions may define sub-communities within a larger community, and individuals can bear multiple traditions for the same or separate communities. A tradition of practice can be singled out from other traditions by the material continuity it exhibits across discontinuities experienced in other traditions. Wholesale transformations in one tradition may have an impact in other traditions, but they do not presage the wholesale transformation of other traditions. I would argue that, in addition to the actual physical conditions of the world, distinct traditions practiced by others define part of the “physical conditions” toward or within which one must use one’s ideas to decide how to deploy the traditions of practice at one’s disposal (“resources” in Simon Schaffer’s lexicon)—they represent likely counterarguments, accusations, or other forms of socio-intellecutal resistance.
In Image and Logic Galison articulated a historiographical position that I would argue had been possible at least since the 1980s when narrating the history of scientific practices began to challenge a preoccupation with the establishment of a stock of transcendent theoretical and metaphysical ideas to which the history of experiment and instrumentation were always subservient. Historians could now attempt a difficult and laborious task: to characterize the features of pertinent traditions and identify periods and constituencies within which they were carried, and thus identify the ideas governing their historical deployment. Historians could also characterize and narrate important changes in both practice and ideas, as well as narrate the results of specific deployments of practices and ideas.
The chronological synthesis of the resultant narratives would constitute the primary historiographical problematic that consolidated historiographical knowledge and generated historiographical debate concerning proper characterization of practices, ideas, constituencies, and periodizations, and the manner of their effects and the ways in which they can change. For reasons I will attempt to uncover in subsequent posts, this task never became an imperative within the professional history of science.