The aim and structure of physical theory.

 

 

The aim and structure of physical theory.

Writer: Pierre Duhem. Field: Physical theory. Composition: two parts. Eleven chapters. Nine appendices.

Main theme : Structure of physical theory.

Secondary themes: mathematisation of physics. Sub-determination of theories. Primary qualities, quantity, measurement. Aristotle, Newton. Crucial experiment. Hypotheses.

Historical context: A practising Catholic and believer, anti-Republican and anti-Dreyfusard, Pierre Duhem was nonetheless a major scientist and a thinker on the nature of physical theory. His approach is still of interest to historians today, but also to epistemologists and some scientists. It is interesting to see that the notion of structure is put forward by Duhem, in the same way as it was in another scientific field by Ferdinand de Saussure. Since this paradigm of structure was to become dominant among many French intellectuals in the course of the twentieth century, it is worth noting that this theme was to have a major influence, even if the authors in question were more willing to acknowledge their affiliation with Saussure than with Duhem. Pierre Duhem was openly Catholic and claimed that physical science was not incompatible with Catholicism. However, in this area, he did not impose any dogmatism and believed in a form of independence of science from metaphysics: "What I have said about the method by which Physics proceeds, about the nature and scope that must be attributed to the theories that it constructs, does not prejudge the metaphysical doctrines or religious beliefs of the person who accepts what I have said. In the progress of physical science, as I have tried to define it, the believer and the unbeliever can work together". Physical science thus appears to be autonomous from the ideological beliefs of physicists. Such a conviction can naturally be criticised, in particular by those who consider the troubled interplay between religion and ideology in general and the unconscious, and thus by all those who are attentive to Marxist and Freudian discourse. However, by proposing a structure for scientific theory, Duhem intends to show that the difficulties encountered by science are not due to external problems, but that the very nature of science is to be inhabited by a doubt for which the scientific structure gives us the reasons.

Why read this book today?

At a time when we deplore a certain lack of interest in science among the 'elite' (it's better to go to business school than science school, and if you have the bad idea of studying science, you'll have to finish your studies with an MBA), which is leading to a certain cretinisation of the ruling classes, it's worth asking how a scientific theory works. It serves a dual purpose: it tells you how a theory is put together and it allows you to criticise it in a more relevant way. One of the points that seems most interesting is that of understanding experience. We usually make a distinction between everyday experience and scientific experimentation. And Pierre Duhem remains on this line. However, he considers that scientific experience is also an interpretation. Scientific experiment consists of asking nature a question. That is the generally accepted expression. It means entering into a dialogue that may or may not be understood as inquisitorial. As we know, in dialogue, the questioner is the master and the responder is the servant. Of course, this does not mean that the servant is deprived of resources. It just means that it's not up to him to come to a conclusion. But how does one conclude? We conclude, Duhem tells us, by interpreting.  Scientific experimentation is an interpretation. The use of a term with religious connotations may come as a surprise. Yet Duhem insists on the 'faith' character inherent in scientific interpretation. To interpret an experiment in physics is to refer to a theoretical whole that is an extremely tangled network of propositions that hold together. The problem of the whole and the part is thus clearly posed: in a crucial experiment that doesn't work, what happens? "By condemning this system en bloc, by declaring that it is flawed, experience does not tell us where this flaw lies". In the whole or in the part? In the whole theory or in one element? Duhem's structuralism prefigures the problems of the psychology of form.