This paper develops an account of scientific objectivity for a relativist theory of evidence. It briefly reviews the character and shortcomings of empiricist and wholist treatments of theory acceptance and objectivity and argues that the relativist account of evidence developed by the author in an earlier essay offers a more satisfactory framework within which to approach questions of justification and intertheoretic comparison. The difficulty with relativism is that it seems to eliminate objectivity from scientific method. Reconceiving objectivity as a function (...) of the social character of science, rather than of individually practiced methods, allows us to claim that science is objective even if relativism is true, and provides a more realistic account of scientific objectivity than is possible on either the empiricist or the wholist accounts. (shrink)

The thesis that certain competing scientific theories are incommensurable was introduced by Kuhn and Feyerabend in 1962 and has been a subject of widespread critique. Critics have generally taken incommensurable theories to be theories which cannot be compared in a rational manner, but both Kuhn and Feyerabend have explicitly rejected this interpretation, and Feyerabend has discussed ways in which such comparisons can be made in a number of his writings. This paper attempts to clarify the incommensurability thesis through the examination (...) of a number of examples, both scientific and nonscientific, of cases in which incommensurable points of view compete. There follows a discussion of Kuhn's analogy between paradigm change and gestalt shifts, the ways in which incommensurable theories can be compared, and one key reason why the incommensurability thesis has been so often misinterpreted by its critics. (shrink)

In this book Martin Bunzl considers the prospects for a general and comprehensive account of explanation, given the variety of interests that prompt explanations in science. Bunzl argues that any successful account of explanation must deal with two very different contexts - one static and one dynamic. Traditionally, theories of explanation have been built for the former of these two contexts. That is to say, they are designed to show how it is that a 'finished' body of scientific knowledge can (...) be put to explanatory use. But finished sciences are few and far between. Real 'explanation' also occurs in a dynamical context in which questions are asked and answers are given as theories are in the process of being constructed. Here, Bunzl argues that attending to explanation produced under these dynamic circumstances undermines prominent features of the theory of explanation produced in the traditional static context. (shrink)