Abstract

This dissertation is a defense of a form of platonic realism about properties, relations and propositions via a discursive survey of a number of issues surrounding what has traditionally been termed the problem of universals. ;In chapter one I lay out what I take the problem of universals to be and the principal tenets of my view. ;Chapter two is an examination of one type of argument for nominalism about a certain type of property, namely, sortal properties. Specifically, I examine Stephen Schiffer's argument for the claim that there is no genuinely language-independent, nonpleonastic property of being a dog. I argue that Schiffer does not establish this claim, and that we do in fact have good reason for thinking there is such a property. ;Chapter three examines a number of critical features of D. M. Armstrong's so-called scientific realism, in which Armstrong defends an immanent realism about universals which is also a posteriori--it is left to total science to determine what universals there are. I argue, among other things, that Armstrong's account fails to solve the multiple-location problem for universals. ;Chapter four is an examination of a popular nominalistic alternative to realism--trope theory. I focus on the theories given by D. C. Williams and Keith Campbell. After examining a variety of trouble areas with each, I argue that both fall victim to the same principal difficulty. ;In chapter five I argue that semantic considerations should influence our ontological investigations. I also urge that David Lewis is correct in assessing Armstrong's "sparse" account as inadequate for doing semantics, but that this criticism does not apply to my platonic realism. ;Chapter six is an assessment of my view versus that of Lewis in analyzing the related notions of resemblance and duplication. I argue that resemblance claims of the form 'a resembles b' and of the form 'a resembles b more than a resembles c' always involve an ineliminable reference to our interests, which provides for the fact that our assessments of resemblance are subject to change