Henri Poincaré and Pierre Duhem: Conventions and scientific reasoning
By the middle of the nineteenth century the experimentalist approach to scientific method that we find in Bacon and Mill had begun to lose some of its power and influence. One reason for this was that the term 'science' had become more circumscribed and more specialised. It signified, primarily, physical science, and its previous associations with systematic, demonstrative knowledge in general were no longer as strong as they had been. 'Scientists' - a word coined by Whewell in 1834 - in this new sense are not really philosophers, or even natural philosophers. The poet, Samuel Coleridge, declared that he was 'half angry' with his friend Humphry Davy for 'prostituting the name of Philosopher . . . to every Fellow who has made a lucky experiment' (quoted in Yeo 1991: 178). Experiment was indeed important, in varying degrees, to the physical sciences, but it was no longer a distinctive feature of them. Many other kinds of enquiry could boast of being experimental. What was distinctive of the physical sciences was their progressive character, and a proper understanding of scientific method should show how this had come about. Moreover, science in this new sense had become a profession. It was no longer the preserve of wealthy amateurs but of people who depended on its success for their livelihood. Scientists needed for practical purposes an understanding of scientific method; they needed, that is, guidance on how to pursue science successfully. New educational institutions were recognising and meeting the demand for trained scientists, and those responsible for their education saw it as one of their tasks to disseminate sound methodological principles. Not surprisingly, there was a feeling that the most reliable accounts of scientific method were those produced within the profession. Mill had indeed taken trouble to provide his readers with apt and scientific examples of his methods, as well as with a sophisticated analysis of the way they worked, but he was not a professional scientist; he was 'ignorant of science'. Because of this, his views, among scientists at least, counted for less than those of his adversary Whewell.