Aspects of scientific modeling: A philosophical analysis of idealization, abstraction and representation

Modeling is currently a lively topic in philosophy of science, though interest in it peaked during the late nineteen sixties. Until then many philosophers of science had busied themselves for decades with the structure of scientific theories. Especially, the Vienna Circle positivists and their loyal followers devoted their intellectual energies primarily to the reconstruction of physical theories in an axiomatic way. With such a positive attitude towards theories, models were inevitably thought of as a side issue for the philosophy of science. It appears that such a theory-centric attitude is no longer a valid approach to characterizing science. The dominant status of logical positivism is long since challenged, and nearly all philosophers of science are inclined to agree with the view that the modern practice of science is based on modeling. Although there are various types of models, the mathematical ones stand out among the others. It is the pervasiveness of mathematical models that led this research into different aspects. This dissertation studies a mathematical model, specifically the Milankovitch Model of ice ages, by thoroughly focusing on its construction and application processes, and also its success in explaining and predicting the relevant phenomena. This case study also illustrates how a mathematical model works in solving a scientific puzzle. It seems perhaps important to emphasize that the purpose of the case study is not merely to demonstrate how a mathematical model can successfully explain and predict its target phenomenon. By relying on Milankovitch's model, I argue that any account of scientific representation should take idealization and abstraction into account. I also provide reasons for thinking that de-idealization is epistemically a fruitful strategy. These reasons subsequently became the basis of my distinction between idealization and abstraction. In this context, I argue that the reverse processes of idealization and abstraction give adequate reasons to distinguish one strategy from the other.