I identify an empirical limitation in Chomsky’s (2013, 2015) labeling theory and propose a theoretical modification that solves that problem. Chomsky argues that a labeling indeterminacy blocks movement of wh-subjects from Spec,TP into Spec,CP. However, I give evidence that wh-subjects can move from Spec,TP to Spec,CP. To solve the labeling problem, I propose (a) that labeling applies to the transferred constituent (normally the phase head complement), rather than the full phase, and (b) that labeling can be delayed by one application of Transfer. This proposal preserves core aspects of labeling theory, without requiring a costly complementizer deletion operation for wh-subject extraction.

This content is only available as a PDF.
You do not currently have access to this content.