Embedded scalar diversity

This paper is an experimental investigation of embedded scalar implicatures in the context of scalar diversity. We test whether a sentence such as Every student read some of the books leads to the implicature that No student read all of the books, and similarly whether Every soup was warm leads to the implicature that No soup was hot—across 42 different lexical scales. We find 1) that embedded implicatures arise; 2) that there is across-scale variation in embedded implicatures, paralleling scalar diversity among global implicatures; and 3) that properties of alternatives (namely, semantic distance and boundedness) that predict global scalar diversity predict variation at the embedded level too. It is argued that these findings are most compatible with an account of embedded implicatures that builds on alternatives, such as the grammatical theory (i.a., Chierchia 2004; Chierchia, Fox & Spector 2012), a modified neo-Gricean account such as Sauerland (2004), or the “neo-Gricean uncertainty” version of the Rational Speech Act with lexical uncertainty account (RSA-LU, Potts, Lassiter, Levy & Frank 2015). They are, however, less compatible with the “unconstrained uncertainty” RSA-LU model (Bergen, Levy & Goodman 2016; Potts et al. 2015), which leaves unexplained (without further assumptions) why the same alternative-driven variation should occur both in global and embedded implicatures.