1st World Congress on Logic, Chance, and Money

Speaker: Roy T. Cook, Dept. of Philosophy, University of Minnesota, Twin Cities

Pierre Duhem, and later W.V.O. Quine, argued that the data we collect in scientific experiments will always be insufficient to specify a single theory with which we can explain said data - a problem called the Underdetermination of Theory by Data.

In this talk we will look at a simple example of this phenomenon constructed using "crazy dice" (dice with odd numbering and sometimes equally odd probabilistic properties). Our examination of crazy dice allow us to provide a simple demonstration that even perfect information about the probability distribution of various outcomes is insufficient to determine either the identity of the "hidden" causes (i.e., the particular dice) or the number of kinds of causes (where "kinds" are shapes and numberings.) In addition, the examples we will discuss are simple enough to be presented in undergraduate classrooms, providing us with a simple, practical pedagogical tool.