UCLA

It is standardly believed that some occurrences of expressions designate singularly, while other occurrences of expressions designate plurally. For instance, the singular expression "the student" may be used on an occasion to talk about one particular student, while the plural expression "the students" may be used on an occasion to talk about several different students.But it is rarely appreciated that several occurrences of expressions may together designate in a manner that is both plural and structured. Hence, the expressions "the students" and "their essays" may be jointly used on an occasion to talk about some students and some essays in a way that relates each student with the sub-collection of the essays that they authored. In other words, several occurrences of expressions may together designate a system of objects that may involve a non-trivial relation among them.

The first part of this dissertation investigates the semantic role that variables play in first-order languages. Variables are the most notable type of expression that partakes in this form of structured plural designation. I present a view according to which variables are interpreted by sui generis intensional entities. More specifically, I claim that variables express special concepts for thinking about the objects that make up a domain of quantification. These concepts differ from other more familiar intensional entities precisely because they provide ways of thinking about structurally dependent systems of objects.

The second part of this dissertation investigates the semantic role that pronouns play in natural languages. Pronouns are commonly believed to be the closest natural language analog to variables. I substantiate this view by arguing that the same intensional entities used to interpret variables should also be used to interpret pronouns. In particular, I claim that some occurrences of expressions are governed by use-conditions that require a speaker to have one or more appropriate concepts of this kind in mind as intended subjects of discourse. Those communicative intentions are then used to fix the content of suitable subsequent occurrences of pronouns. After arguing for this view, I conclude by briefly sketching how it may be compositionally implemented in a standard semantic architecture with a few modest adjustments.