Questions about whether the noun-verb distinction is continuous or discrete can be about three sorts of linguistic entities: roots, whole words, or syntactic elements. I claim that nouns and verbs qua syntactic elements are discretely different, whereas the noun-verb distinction qua morphological roots may not be. Having clarified this, I focus on an empirical domain in which nouns and verbs do have discretely different behavior: verbal constructions allow object arguments that have structural case (accusative) whereas nominal constructions do not. I show that this asymmetry follows from the conjunction of two independently motivated ideas: (i) nominal projections do not allow specifiers whereas verbal projections do, and (ii) accusative case is assigned to the lower of two nominals within the same local domain. Unlike accusative case, ergative case does often extend from clauses to noun phrases, where it is used to mark the possessor of the noun. I show that this difference between accusative languages and ergative languages also follows naturally from the fundamental structure of nominals together with the idea that ergative is a dependent case, like accusative but its converse. Finally, I consider the fact that "verbal nouns" (gerunds) often do have accusative case objects despite behaving like nouns in other respects. This follows from the fact that verbal noun constructions consist of an ordinary verb phrase embedded under an ordinary noun head, with the verb and the noun simply merged together into a single word on the surface.
|Title of host publication
|How Categorical are Categories?
|Subtitle of host publication
|New Approaches to the Old Questions of Noun, Verb, and Adjective
|Number of pages
|Published - Aug 17 2015
All Science Journal Classification (ASJC) codes
- General Arts and Humanities
- General Social Sciences