Philosophical semantics

The dissertation divides into two general parts: one in which I develop a number of ideas as to the workings of natural language, and a second in which I apply these ideas to various well known philosophical questions related to language, and in particular, questions about the ascription of propositional attitudes. The developmental work is done mostly in Chapters 2 through 4, while the applications come mainly in Chapters 5 and 6. An additional objective of the thesis is to shed light on some of Wittgenstein's later ideas and methods; I attempt this in Chapter 7. The overall goal of the first part of the thesis is to identify and characterize two new (in the sense of heretofore largely unstudied) semantic properties, which I call "topicality" and "semantic openness." Since both properties are connected with a broader semantic feature sometimes referred to (in the philosophical literature) as "content variability," I devote Chapter 2 to content and content variability, explaining as precisely as possible what these properties are supposed to be, and giving some preliminary indication of how I put them to use in subsequent chapters. Chapter 3 is given over to topicality, which is (roughly) a property of certain deictic expressions that vary in content in accordance with variations in the topic(s) that are on the table in the contexts in which they are uttered. It turns out that while many cases of topicality are fairly easy to recognize as such, some topical expressions are difficult to identify as expressions of variable content. This opens up the possibility that various apparently content invariant terms actually vary in content, including some that figure prominently in debates in the philosophy of language. In Chapter 4, I introduce the notion of semantic openness. This is (roughly) a property of certain content variable expressions the uses of which are not strictly rule bound. The burden of this chapter is to give a clear statementof what this unboundedness amounts to. Additionally I give specific examples of semantically open expressions of natural language, and consider some of the ramifications of semantic openness for the study of natural language. I begin to apply the ideas developed in Chapters 2-4 in a chapter addressed to a long-standing puzzle about the behavior of referring expressions within attitudinative sentential contexts. At the most general level, the puzzle is one of reconciling a number of individually intuitive semantic principles that nonetheless have historically defied reconciliation. I argue for a solution to the puzzle based on the claim that psychological verbs (such as "to believe") vary in content indexically, drawing on the results of the first part of the dissertation to develop this idea. In Chapter 6 I turn to a related puzzle also having to do with propositional attitudes. This is what I call the problem of coherent inconsistency, which can be posed in the form of a question: How can a rational individual with full linguistic competence stand in the same propositional attitude to logically incompatible propositions? I argue that the key to answering this question lies in taking ordinary names to be indexical. In defending this suggestion, I appeal to the results of earlier chapters to overcome various objections to an indexical theory of names. Finally, in Chapter 7 I argue for an interpretation of Wittgenstein's notion of family resemblance in terms of semantic openness and topical indexicality, and outline some possible future applications of the ideas developed in preceding chapters, along broadly Wittgensteinian lines.