Linguistic Circle

Speaker: Kajsa Djärv (University of Edinburgh)

Title: Selectional restrictions and complex discourse moves: licensing embedded Main Clause Syntax

Abstract: In this talk, I examine the interpretation of embedded clauses with main clause word order, and the conditions under which such clauses are licensed (Hooper & Thompson 1973, Green 1976, McCloskey 2006, Truckenbrodt 2006, Julien 2009, Krifka 2014, Woods 2016, a.o.). I argue that such clauses are not special types of indirect assertions, as has typically been argued in the literature. Rather, the discourse effects of such clauses are best captured in terms of the statement that they share the conventional discourse effects associated with main clause declaratives; in the sense of Farkas and Roelofsen (2017).

I show that this proposal nicely captures the core discourse effects associated with such clauses, while still allowing for a wide range of attested uses, including cases where the embedded clause is not asserted, but functions as a type of biased question (a type of use which is not captured by accounts that link the availability of such clauses to an assertion operator or feature, such as Krifka 2014, a.o).

This perspective also provides us with a nice way of understanding the role of the matrix predicate in licensing such clauses. Recent work in formal semantics has increasingly challenged the role played by selection in the domain of question-embedding, arguing instead that the selectional properties of a predicate depend on global (compositionally derived) semantic inferences (e.g., Theiler et al 2018). In this talk, I show that certain `selectional' properties of clause-embedding predicates depend on global pragmatic inferences and that what looks like selectional restrictions in out-of-context judgements and in large-scale corpus data is in fact about creating discourse coherence within a given utterance, when the sentence in question represents a complex discourse move.

The talk also discusses consequences of the current account and the data discussed here for theories of clausal complementation and theoretical modelling in inquisitive semantics.

Contact details

Linguistic Circle committee