Whamit!

In this speculative presentation, I’ll examine some constraints on movement, from the perspective of the nominal domain. I’ll argue that certain asymmetries in extraction from a given DP, versus a second DP embedded within the first, are indicative of a theory in which certain extractions from a phase require that phase to be “Unlocked” by Agree (Rackowski & Richards 2005, Halpert 2015, 2018, Branan 2018). 

 

Taking DPs to be phases, a phi-probe on v can target and Unlock a given DP, but subsequent DPs embedded within the first will be too distant for v to target. Thus an A-bar probe on v can subsequently probe into and extract from the highest DP, but lower DPs within remain “Locked” for A-bar extraction. I’ll suggest that this sort of approach makes some correct predictions for Chichewa, English, and Russian, though in the latter two, the phi-agreement involved in Unlocking is not overtly expressed.