Groups : a semantic and metaphysical examination

Since the linguistic turn, many have taken semantics to guide metaphysics. By examining semantic theories proposed by philosophers and linguists, I argue that the semantics of a true theory in a natural language can serve as only a partial guide to metaphysics. Semantics will not always lead to determinate answers to questions of the form 'Does theory T carry an ontological commitment to Fs?' Further, semantics will never deliver answers to questions regarding the nature of Fs. If semantics is to be our guide, we must look to our best semantic theories to determine whether a theory carries ontological commitments to Fs. I develop criteria to determine when a semantic treatment is semantically adequate and should be counted amongst our best theories. Given these criteria, there can be more than one empirically adequate semantic treatment of a natural language theory. To determine ontological commitments I appeal to Quine's Criterion, which states that a theory has Fs in its ontology just in case it says or entails that there are Fs. To determine what a theory says and entails, we must appeal to semantic treatments. Since different equally adequate semantic treatments can yield different contents and entailments, Quine's Criterion delivers ontological commitments only relative to a semantic treatment. I then argue for a supervaluationist principle that delivers unrelativized, but possibly indeterminate, ontological commitments of a theory. Next, I apply my methodology to two case studies which exemplify two kinds of answers the supervalutationist principle might deliver concerning ontological commitments. I argue through an examination of data and formal treatments that plural expressions carry indeterminate ontological commitments to summed entities, while collective nouns carry determinate ontological commitments to group-like entities. Finally, I undertake an examination of what groups, things like teams, committees and courts, might be that accords with the minimal verdict delivered by the semantics of collective noun -- that they exist -- but which goes beyond this to examine their nature. I assess and reject the views of groups currently on offer and propose and defend a novel view of groups as realizations of structures.