It is a long-standing observation that languages differ in the amount of overt structure that they exhibit in a particular part of the clause. Some languages spread certain features over multiple heads, while others bundle those same features on one head, resulting in cartographic and non-cartographic effects, respectively. This paper looks at two clause-types in Wolof (Niger-Congo), which differ in the number of functional projections that they contain in the CP-TP part of the clause—one clause-type has only one projection (CTP), with one specifier position, while the other one exhibits the more commonly seen separation into two projections (CP and TP), with two specifier positions. It is argued that the features traditionally associated with C and T start out on one head, and can in the course of the derivation be spread over two heads. The main goal of this paper is to introduce the mechanism of Head-Splitting, which employs strictly ordered feature-checking and partial head-reprojection as a result of feature-checking failures to derive two heads from one. This proposal allows us to capture the observation that there is cross-linguistic variation in the correspondence between the number of functional features and heads in a derivational manner, without positing parametric variation, thus offering a way to derive other phenomena in which a certain element is sometimes instantiated in one position, and other times in another one, in one and the same language.
Proceedings of the 34th West Coast Conference on Formal Linguistics
edited by Aaron Kaplan, Abby Kaplan, Miranda K. McCarvel, and Edward J. Rubin Table of contents
ISBN 978-1-57473-471-3 library binding
vii + 598 pages
publication date: 2017
published by Cascadilla Proceedings Project, Somerville, MA, USA