Repository URL:
Sebastian Lutz
preprint description
Elliott Sober has suggested his contrastive criterion of testability as an improvement over previous criteria of empirical significance like falsifiability and the standard Bayesian criterion of empirical significance. I argue that the criterion fails to meet four of the conditions of adequacy for a criterion of empirical significance that follow from Sober’s position or are presumed in his arguments. I suggest to define empirical significance as empirical non-equivalence to a tautology, because this definition does meet the conditions of adequacy. Specifically, it is equivalent to the standard Bayesian criterion of empirical significance whenever all probabilities are defined and contains falsifiability as a special case. This latter feature is important because those conditions of adequacy that apply to criteria of deductive empirical significance single out falsifiability.

This preprint has 0 Wikipedia mention.