Agree Versus Attract: A Relativized Minimality Solution to a Proper Binding Problem

I examine an extraction paradigm rst discussed by Kroch and Joshi (1985), which Lasnik (2002) takes to provide an argument for feature movement. I show that Lasnik’s solution is problematic on both conceptual and empirical grounds. I offer an alternative approach that in contrast to previous solutions does not rely on the Proper Binding Condition (whose status in minimalism is far from clear), but instead deeply implicates Relativized Minimality. Although I take issue with Lasnik’s analysis, I agree with him that the Kroch-Joshi paradigm bears on the nature of feature checking, and agreement. In particular, I show that an Agree-based account handles the facts straightforwardly.