After 1910 or so, the attention of most philosophers and physicists shifted from the foundational uncertainties of the classical physics of continua-that is, continuously extended, flexible objects such as wires, plates, or fluids-to the novel puzzles raised by quantum mechanics and relativity. For various extraneous reasons, twentieth-century physicists began to see the special subject of point masses as comprising all of `classical mechanics,` although this anti-continuum view would not have been shared by most of their nineteenth-century forebears. Compared to continua, point masses present relatively few conceptual problems, and the modern shift in the meaning of `classical mechanics` has led many contemporary philosophers and historians to misunderstand the conceptual problems with which the nineteenth century struggled. At the same time, there has been much ongoing work on the behavior of classical extended bodies in this century, conducted largely by engineers and applied mathematicians. These investigations have taught us much about the structure and range of validity of classical ways of thinking that were completely unknown in the nineteenth century. Unfortunately, although philosophers of science continually cite `the lessons to be learned from classical theory,` most philosophers have not kept current with new knowledge on the topic. The purpose of this project is to reexamine, in the light of twentieth-century research on mechanics, a number of central historical episodes that continue to play pivotal roles in philosophical argumentation today, including the reasons why so many classical physicists believed that science needs to be inherently `idealized` or otherwise `anti-realistic` in its procedures; Duhem's motives in articulating the famed Quine-Duhem thesis; the considerations that motivated his rejection of atomistic modeling; whether it is reasonable to expect that a single set of axioms can adequately cover all of what we usually regard as `classical mechanics.`
