Rethinking the distribution of English finite clausal complements: evidence from complementiser-how clauses

This paper advocates a new conception of the properties which determine the distribution of finite clausal complements (FCCs) in English. I argue against the orthodox view that FCCs are selected by matrix predicates on the basis of their interpretive type (Grimshaw 1979; Rizzi 1997; Ginzburg & Sag 2000), and propose that distribution rather depends on the specification of the FCC in terms of the syntactically encoded properties [+/-wh, +/-factive]. This proposal is motivated by new distributional patterns which emerge when the typology of English FCCs is expanded to take into account complementiser-how clauses (CHCs) (Legate 2010; Nye 2012). CHCs have their own unique interpretation, yet, strikingly, have exactly the same distribution as embedded exclamatives. This is unexpected under the traditional view of FCC selection, but is explained if CHCs and exclamatives are selected on the basis of a common [+wh, +factive] syntactic specification