David J. Stump

Duhem’s concept of “good sense” is central to his philosophy of science, given that it is what allows scientist to decide between competing theories. Scientists must use good sense and have intellectual and moral virtues in order to be neutral arbiters of scientific theories, especially when choosing between empirically adequate theories. I discuss the parallels in Duhem’s views to those of virtue epistemologists, who understand justified belief as that arrived at by a cognitive agent with intellectual and moral virtues, showing how consideration of Duhem as a virtue epistemologist offers insights into his views, as well as providing possible answers to some puzzles about virtue epistemology. The extent to which Duhem holds that the intellectual and moral virtues of the scientist determine scientific knowledge has not been generally noticed.

In april 1872, with the caisson at a depth of seventy-odd feet and still no bedrock, two men died. The strain for Roebling was nearly unbearable, as his wife later said. On May 18, a third man died, and that same day Roebling made the most difficult and courageous decision of the project. Staking everything — the success of the bridge, his reputation, his career - he ordered a halt. The New York tower, he had concluded, could stand where it was, at a depth of 78 feet 6 inches, not on bedrock, but on “hardpack” sand. From examinations of the strata he had determined to his own satisfaction that no movement had occured at the level since the time of deposit millions of years in the geologic past; so, he said, it was “good enough to found upon”

I trace the development of arguments for the consistency of non-Euclidean geometries and for the independence of the parallel postulate, showing how the arguments become more rigorous as a formal conception of geometry is introduced. I analyze the kinds of arguments offered by Jules Hoüel in 1860-1870 for the unprovability of the parallel postulate and for the existence of non-Euclidean geometries, especially his reaction to the publication of Beltrami’s seminal papers, showing that Beltrami was much more concerned with the existence of non-Euclidean objects than he was with the formal consistency of non-Euclidean geometries. The final step towards rigorous consistency proofs is taken in the 1880s by Henri Poincaré. It is the formal conception of geometry, stripping the geometric primitive terms of their usual meanings, that allows the introduction of a modern fully rigorous consistency proof.

Poincare’s arguments for his thesis of the conventionality of metric depend on a relationalist program for dynamics, not on any general philosophical interpretation of science. I will sketch Poincare’s development of the relationalist program and show that his arguments for the conventionality of metric do not depend on any global strategies such as a general empiricism or Duhemian underdetermination arguments. Poincare’s theory of space, while empirically false, is more philosophically sophisticated than his critics have claimed.

Poincaré's claim that Euclidean and non-Euclidean geometries are translatable has generally been thought to be based on his introduction of a model to prove the consistency of Lobachevskian geometry and to be equivalent to a claim that Euclidean and non-Euclidean geometries are logically isomorphic axiomatic systems. In contrast to the standard view, I argue that Poincaré's translation thesis has a mathematical, rather than a meta-mathematical basis. The mathematical basis of Poincaré's translation thesis is that the underlying manifolds of Euclidean and Lobachevskian geometries are homeomorphic. Assuming as Poincaré does that metric relations are not factual, it follows that we can rewrite a physical theory using Euclidean geometry as one using Lobachevskian geometry and express the same facts.