The Life Cycle of an Argument

This dissertation is an investigation into the syntax and lexical semantics of argumenthood. The goal is to shed light on what I call the life cycle of an argument: how an argument is realized syntactically, what syntactic positions it may occupy, and what behavioral properties it acquires as a result of occupying those syntactic positions. I explore these issues using the Russian language as the primary source of data.

 

In a sense, argument realization is the study of the beginning of an argument's life; its goal is to discover how, exactly, a verb and its arguments come to be associated with a syntactic structure. This question is intimately tied to the dispute concerning the division of labor between the syntactic component of grammar and the lexicon. I address this issue via a study of Russian intransitive verbs that display variable unaccusative/unergative behavior, meaning that they behave at some times as though their sole argument is realized in a high syntactic position and at other times as though their sole argument is realized in a low syntactic position. I propose a model of the relationship between real-world events, verb meaning, and syntactic representation that allows me to account for variable unaccusative/unergative behavior in Russian and beyond. Under this model, certain verb roots are compatible with more than one event schema, leading to their association with more than one syntactic structure. Examining the pieces of this complex relationship and how they fit together helps delineate the links between extralinguistic reality, the lexicon, and the syntactic component of grammar. 

 

Studying the syntactic life of an argument, including what positions an argument can occupy and what properties it has due to its presence in those positions, bears on issues central to syntactic theory. Among them---what is the core structure of a clause? And how are dependencies between elements within the clause expressed? I examine Russian variable word order, the ability of the major constituents of a clause to be arranged in almost any order. By looking at the behavioral properties borne by arguments when they appear in specific orders, I propose an analysis of the derivational path (i.e. the path of movement) of arguments that accounts for the availability of certain word orders. One of the behavioral properties of interest here is first conjunct agreement, which occurs when the verb agrees with the leftmost nominal embedded inside a larger conjoined phrase that appears postverbally. I show that the surface pattern of first conjunct agreement in Russian deserves a separate analysis from the same surface pattern that appears in other languages. The unique distribution of first conjunct agreement in Russian provides clues to language-specific requirements on the movement of arguments and the positions in which arguments can be pronounced.