Idealization, Inference to the Best Explanation, and Scientific Realism

In this chapter we will finally make good on the promises made in Chapters 2 and 3. In effect it will be argued here that scientific realism can be saved from both of Cartwright’s idealization-based attacks. This goal will be achieved by showing specifically what is wrong with the argument from unconfirmability and with the argument from misrepresentation. To this end it will first be demonstrated that the presence of idealizations in the sciences neither forces us to concede the empirical nature of science nor to conclude that idealized theories are unconfirmable. Second, it will be argued that a variety of forms of scientific realism are comfortably compatible with the theory of idealization presented in Chapters 1 and 2. Since these arguments constitute the main result of the book, we will take nothing for granted in order to make the case for the compatibility of scientific realism and the use of idealization in the sciences. So, some of the discussion that follows will involve concepts and ideas with which the more sophisticated reader will already be familiar, but this will help to prevent any potential misunderstandings. Let us then begin the defense of scientific realism by turning our attention to the argument from unconfirmability. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.