**Restricting the fourth reading** *Ido Benbaji, MIT* **Time**: Poster Session 2 -- Saturday, 1:30pm-3:30pm Eastern Time **Overview**: Szabó (2010, 2011) argues that in addition to the three commonly accepted readings of DPs in intensional contexts, there is a fourth, specific opaque reading, in which a DP’s determiner scopes above an intensional operator while its restrictor is nevertheless interpreted opaquely. The existence of this reading poses a potential problem to both the world-pronoun framework (Percus 2000) and the modified scope theory of intensionality (Keshet 2008), as in both wide quantificational scope implies transparent interpretation (though not vice versa). Szabó takes the existence of such readings as evidence that natural language, like logical ones, allows for bare quantification over the entire domain, and posits a mechanism that splits quantifiers from their restrictors at LF. Contra Szabó, I argue that fourth readings do not involve bare quantification, but are syntactically restricted to relativization environments and are generated by invoking independently motivated mechanisms for NP reconstruction into relative clauses (RCs). On this account, the fourth reading is obtained when a DP dominates a raising RC with an intensional operator. The DP’s determiner originates clause externally, and therefore must take wide quantificational scope relative to the intensional operator, while the NP that appears to occupy the DP’s restrictor position originates clause internally and can therefore reconstruct to a position below the operator to receive an opaque interpretation (as proposed by Bassi and Rasin 2018 for data presented in Grosu and Krifka 2007). This analysis predicts the fourth reading to be unavailable when a raising structure for the RC is independently blocked; a prediction that is shown to be borne out with Hebrew resumption, extraposition, and condition C.