PHILOSOPHY OF SCIENCE QUA EPISTEMOLOGY Forthcoming in Continuum Companion to the Philosophy of Science (eds Steven French and Juha Saatsi) Alexander Bird 1 Philosophy of science between generalism and par- ticularism The relationship between philosophy of science and epistemology as practised at the heart of general philosophy has been variable. Philosophy of science is caught between potentially opposing forces. On the one hand philosophy of science needs to be true to the (at least apparently) distinctive and even arcane practices of actual scientists. This I call the particularist tendency, because it tends to emphasize the particular, special nature of science (and maybe even of the individual sciences). On the other hand philosophy of science needs to relate its account of scientific belief to the entirely general account of knowledge and justification provided by episte- mology. This I call the generalist tendency, because it seeks to place the philosophy of science within a general epistemological framework. Modern philosophy of science emerged in the middle of the nineteenth century. The most significant work of philosophy of science since Bacon’s Novum Organon, William Whewell’s The Philosophy of the Inductive Sciences, Founded Upon Their History (1840) exhibits the particularist tendency. As the title of his book indicates, Whewell’s philosophy of science is built on his earlier work The History of the Induc- tive Sciences (1837). While he does construct his account within a general framework that bears a superficial resemblance to Kant’s, Whewell’s description of science and its processes of reasoning are clearly specific to science and are motivated in part by reflection on the details of particular episodes in the history of science. For example, he uses Kepler’s discovery of the elliptical orbits of the planets as an example of dis- coverer’s induction: the unfolding of fundamental ideas and conceptions permits the ‘colligation’ of the observed facts concerning the orbit of Mars to the conception of those facts as satisfying an elliptical orbit, which is then generalized for all plan- ets. Whewell insisted that a philosophy of science must be inferable from its history. As a consequence Whewell’s account is detailed, and permits multiple routes to dis- covery and knowledge. Discoverer’s induction is itself a multi-stage process, and the colligation component may be achieved by a number of different inferences. Like- wise the process of confirmation can involve three distinct components: predic- tion, consilience, and coherence. Whewell addresses questions that are still much discussed in philosophy of science, such as whether novel predictions have greater 1