A Bayesian Networks Analysis of the Duhem-Quine Thesis (original) (raw)

This paper provides an optimistic analysis of using Bayesian methods to confront the challenges posed by the Duhem-Quine thesis, which argues against the objective revision of scientific theories in light of disconfirming evidence. By investigating the interplay between scientific theories and auxiliary beliefs, the authors show that the commonly assumed probabilistic conditions can be navigated effectively. The findings indicate that problematic applications of Bayesian confirmation theory are limited and context-dependent, leading to a more nuanced understanding of the capabilities and limitations of Bayesian approaches in scientific theory revision.