home
   ::: about
   ::: news
   ::: links
   ::: giving
   ::: contact

events
   ::: calendar
   ::: lunchtime
   ::: annual lecture series
   ::: conferences

people
   ::: visiting fellows
   ::: resident fellows
   ::: associates

joining
   ::: visiting fellowships
   ::: resident fellowships
   ::: associateships

being here
   ::: visiting
   ::: the last donut
   ::: photo album


::: center home >> events >> conferences >> other >> 2007-08>> &HPS

Beyond case studies: history as philosophy
Hasok Chang

What can we conclude from a mere handful of case studies? This has been a vexing question for anyone attempting to make philosophy of science more historically informed, and history of science more relevant to philosophy of science. The field of HPS has witnessed too many hasty philosophical generalizations based on a small number of conveniently chosen case studies. One might even speculate that dissatisfaction with such methodological shoddiness contributed decisively to a widespread disillusionment with the whole HPS enterprise in the last few decades.

Emblematic of this disillusionment is Kuhn's paper "The Relations between the History and the Philosophy of Science". Kuhn argued that history of science and philosophy of science have different goals, and therefore "no one can practice them both at the same time" (p. 5). Although he did agree that philosophy of science should eventually be informed by some knowledge of scientific practice (historical or contemporary), Kuhn never quite specified a clear mechanism by which this history-philosophy interaction can take place. Without such a mechanism, we would be condemned to either making unwarranted generalizations from history, or writing entirely "local" histories that have no bearing on an overall understanding of the scientific process.

I propose a move away from the habit of viewing historical cases as an inductive evidence-base for general philosophical theses. (To that end, I prefer to speak of historical "episodes" rather than "cases".) The relation between historical and philosophical studies should not be seen as one between the particular and the general, but as a relation between the concrete and the abstract. An abstract framework is necessary for telling any concrete story at all; that is the key to the philosophy-history relation, to the extent that we see history as concrete and philosophy as abstract. I agree with Lakatos that all historiography of science is philosophical. Abstract ideas are needed for the understanding of even just one concrete episode. We cannot understand scientists' actions, not to mention judge them, without considering them in abstract terms (such as "justified," "coherent," "observation," "measurement," "simple," "explanation," "novel," etc., etc.). Any concrete narrative requires abstract notions in the characterization of the events, characters, circumstances and decisions occurring in it. If we extract abstract insights from our account of a particular episode, that is not so much a process of generalisation, as articulation of what was already put into it. It may even be an act of self-analysis, in case the episode was initially narrated without a good awareness of the abstractions that guided its construction.

Philosophers will be pleased but not surprised to hear that abstract conceptions are necessary for historiography. The more novel idea I would like to propose concerns the opposite direction of dependency: how doing history can help our philosophizing. In case there are no ready-made philosophical concepts through which a given historical episode can be properly understood, the historian is compelled to craft new abstract philosophical concepts. Therefore, history-writing can be a very effective method of generating fresh philosophical insights. Instead of sitting back and complaining that existing philosophical frameworks are inadequate for understanding real historical episodes, historians can actively engage in the creation of new philosophical ideas through their concrete investigations.

Once an abstract idea has been generated, it needs to show its worth in two different ways. First, its cogency needs to be demonstrated through further abstract considerations and arguments. Second, the applicability of the idea has to be demonstrated by showing that it can effectively be employed in the telling of various other concrete episodes. An abstraction becomes general only when it has been applied widely.

I will illustrate these claims through a discussion of two investigations in HPS from my own recent and current work. (This will also raise, and solve, a problem of reflexivity: how can we use case studies to show how to go beyond case studies?!)

(1) Temperature measurement and epistemic iteration. In this recent work, I started out with very concrete questions about the justification of measurement methods, such as: how can we judge whether the mercury in the standard thermometer expands uniformly with temperature, in the absence of an already-established thermometer which can give us the correct temperature values? I then made a detailed examination of the history of thermometry in order to see how scientists actually resolved this problem. I was quite puzzled by the historical material I was finding, until I hit upon the idea of "epistemic iteration". I could make sense of this particular scientific episode only when I understood it as a process in which scientists started out by presuming the validity of an untested measurement method, starting a process of inquiry which eventually doubled back on itself to refine and correct its own starting assumptions. This abstract notion of epistemic iteration has proven to be applicable quite widely.

(2) Analytical chemistry and the simplicity of substances. This is an investigation currently in progress, which began with a question about how we can determine the simplicity of substances in chemistry. Historians often credit Lavoisier with an operational definition of a chemical element, as a simple body that cannot be broken down any further. This assumes that one knows how to tell which chemical bodies are simpler than which, so that one knows how to tell when a chemical reaction constitutes analysis rather than synthesis. It may seem obvious that water, for example, is a compound because it can be broken down into hydrogen and oxygen. However, Ritter and others maintained that water was an element, which formed oxygen or hydrogen through electrolysis by combination with positive or negative electricity. Analytical chemistry could not make progress without an ability to judge the simplicity of substances. In order to come to a full understanding of this development, I will have to craft the right set of abstract concepts concerning elements, compounds, and the process of combination. The abstract scheme generated in this enquiry could have far-reaching applicability in understanding the construction of the basic ontology of the material world.

 
Revised 3/10/08 - Copyright 2006