ABSTRACT
The underdetermination argument in the realism debate runs as follows: since the available observable evidence always supports at least two logically incompatible scientific theories, and since our only reason for believing our scientific theories to be true is the observable evidence on which they are based, we can never have any epistemic reason to choose one of these theories over the others. In this chapter, I first discuss the premises underlying the underdetermination argument before moving on to recent work dealing with specific cases and sciences. I argue that all of these discussions suggest that there is no general, in-principle argument in favor of underdetermination but that its real threat can only be assessed on a case-by-case basis.
