Abstract

This study criticizes one prominent attempt at the mereologization of classes and proposes several alternate routes to an adequate mereologization of classes which do not suffer similar problems. It offers an examination of David Lewis's claim that the parts of a class are all and only its subclasses, suggests that there is insufficient reason given to adhere to such a claim, criticizes Lewis's identification of the null set, examines his claims concerning the ontological innocence of mereology, and argues that the claims are unfounded. Finally, an argument is given to show that mereology is not ontologically innocent. ;Next the project develops a mereologized class theory satisfying the following criteria: the project does not take as primitive the operation which carries a member into its singleton, it offers some insight into the ways in which members enter into classes, and it is fruitful in that it allows us to develop a set theory adequate to meet the demands of most normal mathematics. The development avoids the set theoretic paradoxes, it respects our intuition that any thing whatsoever can be a member, and it has available to it a defense of the set theoretic axioms which is no weaker than the typical defenses of these axioms offered by either the iterative or the limitation of size conceptions of sets