Claire Halpert, University of Minnesota

In this talk, I investigate cross-linguistic variation in raising-to-subject constructions.  In particular, I consider the differences between languages that have that have so-called hyperraising (A-movement out of a finite clause) and languages that have the more familiar pattern of A-raising out of nonfinite clauses.  Standard accounts of raising-to-subject in languages like English rule out hyperraising via multiple factors, including Case/Activity and PIC effects (e.g. Chomsky 2000, 2001): nominals cannot raise out of an embedded (finite) CP because they are already Cased/inactive and/or because they cannot escape the embedded CP phase.  Hyperraising constructions in languages like Zulu (Bantu) seem to show the opposite effect: they permit raising out of finite CPs, but not out of nonfinite clauses.  I argue that these different surface patterns are the predictable result of the PIC effect as an A-over-A configuration (Chomsky 1964, Rackowski and Richards 2005).  In particular, raising-to-subject results when matrix T  probes for a phi-goal; in some situations the embedded clause itself may act as an intervenor for the embedded subject it contains.  I propose that the type of raising constructions we find in a given language can be determined by independent properties of CP and TP in the language, including: 1) whether CPs or infinitival phrases are phi-goals and 2) whether T shows an EPP effect (and if so, which elements can satisfy it). I show that variation in these factors can capture a number of different attested raising profiles, including the hyperraising found in Zulu, the lack of raising in Makhuwa (Bantu), and the ``standard'' raising pattern of languages like English.  I further show how this proposal can extend to the raising out of nominalized clauses found in languages like Uyghur (Turkic).  This approach to raising shifts our understanding away from the notion of ``standard'' vs. divergent patterns and instead makes testable predictions about the range and types of raising profiles that we might expect to find cross-linguistically.  In addition, this proposal lends support to the idea of the PIC as an A-over-A effect and extends the idea of PIC obviation as an Agree effect from A-bar to A-constructions (Rackowski and Richards 2005).  Finally, by approaching raising predicates in this way, we gain new insight on how embedded clauses interact with higher elements in the syntax and on variation in how the EPP works across languages.

Go back to Colloquia