The exponence of caseless NPs in Moksha

Abstract

In this paper, I argue that the difference between indefinite- and definite-declension genitives in Moksha (Uralic) is not in the semantic or referential characteristics of the nominals they mark (as the name suggests), but rather in the syntactic size of the nominals. I show that there is no one-to-one correspondence between the definiteness of a noun phrase and its marking. In particular, indefinite-declension genitives can mark non-specific indefinite as well as specific definite noun phrases. At the same time, indefinite-declension genitives are number-neutral, while definite-declension genitives are specified for number. In contrast to indefinite-declension nominals, definite-declension ones also trigger possessive agreement on the head noun. I analyze indefinite-genitives as NPs and definite-genitives as DPs. Based on what is generally known about DPs, I make several predictions regarding the distribution of the genitives; two definite-declension genitives should be incompatible within one enclosing DP, while there should be no such restriction on the co-occurrence of an indefinite-declension genitive with either another indefinite-declension genitive or a definite-declension genitive. These predictions are borne out, and these new findings enrich the discussion of the ban on the co-occurrence of two DPs which are too close in the structure. More generally, my analysis supports the view that nominals of different structural size can coexist within one language.