On Location: The Structure of Case and Adpositions

Date of Completion

January 2010


Language, Linguistics




The general goal of this dissertation is two-fold: first, I provide a unified structure for spatial expressions (local cases and adpositions) and second, I propose a novel approach to vocabulary insertion and generation of portmanteau morphemes. ^ I propose a novel structure for local case affixes, based on data from 114 languages and argue that spatial expressions always have two components: locational and directional. I base my argument for the geometric organization of local case suffixes on the evidence from the distribution of portmanteau morphemes and implicational universals. I also show that adpositions and local cases have essentially the same structure, which can be derived by head movement and rebracketing. I also argue that languages differ with respect to how much functional structure they have in their PPs, which is reflected in their syntactic properties of PPs. I use three diagnostics (the availability of measure phrases in PPs, the possibility of pronouns co-referential with clausemate subject, and the availability of PP-internal quantifier float) to determine the amount of functional structure in PPs of a given language. ^ The investigation of the structure of local cases involves an issue of cumulative exponents in morphology, which is usually dealt with by appealing to the post-syntactic operation of Fusion. I propose a new approach to portmanteau morphemes which relies on a new approach to vocabulary insertion, the Vocabulary Insertion Principle, which allows vocabulary insertion at non-terminal nodes, and which is tested against the data from local case morphology (114 languages) and Tense-Aspect-Mood morphology (200 languages). I also show that the Vocabulary Insertion Principle along with the morphological operation of rebracketing correctly captures the distribution of Tense-Aspect-Mood portmanteaus. ^