Event

Title: New evidence for an antilocality constraint on ¯A-movement: subject (sub)extraction in Igbo

Abstract:

A frequently used ingredient in more recent structural accounts of subject vs. non-subject extraction asymmetries such as the that-trace effect is antilocality (AL), which imposes a lower bound on the length of a movement dependency. A common argument for an AL-component comes from the observation that adding overt material between the extraction site and the landing site (e.g., an adverb) facilitates movement that would otherwise too local (the Adverb Effect, a.o. Culicover 1993). However, this reasoning is not without problems, as it presupposes, for example, that the intervening material occupies the Spec of an additional projection that is
usually absent. We will therefore provide a novel argument for AL, in particular for Spec-to-Spec AL (Erlewine 2016):
(1) Spec-to-Spec Antilocality: ¯A-movement of a phrase from the Specifier of XP must cross a maximal projection other than XP.
(1) makes two predictions which we will show to be borne out in Igbo (Benue-Congo, Nigeria):
1. Subjects (in SpecTP) cannot undergo ¯A-movement to the local SpecCP (neither as a terminal movement step nor as an intermediate movement step in a long ¯A-dependency)
2. Subextraction from subjects to the local SpecCP is possible (both as a terminal and an intermediate movement step)
The grammatical profile of Igbo allows us to test these predictions because (i) subjects are not islands for subextraction (unlike in most other languages), and (b) various reflexes of ¯A-movement in the language make otherwise string-vacuous local subject movement visible. In the course of the argumentation, we will also briefly address strategies for the formation of long ¯A-dependencies and how they can be distinguished in Igbo (base-generation, movement, prolepsis).