Scalar Implicature and Negative Strengthening: What it Takes to be Happy

According to a tacit assumption in the theoretical and experimental literature, scalar implicature is based on a single mechanism and the behavior of one scale generalizes to the whole family of scales (van Tiel et al., 2016). Contrary to this uniformity assumption, experimental research has demonstrated great variability in the rates of scalar inferences across dierent triggers, in part being explained by factors such as boundedness of a scale and semantic distance between scale-mates (Doran et al., 2009; van Tiel et al., 2016).

In this talk, I will present experimental work testing the interplay of scalar implicature and negative strengthening (Horn, 1989; Levinson, 2000; Krifka, 2007) in a balanced set of adjectival scales. The main goal is to show that variation across adjectival scales is more systematic than previously thought and that the scale structure underlying the semantics of gradable adjectives systematically aects inference patterns. Based on our data, we argue that insights concerning the semantics of scales should be integrated in theories of pragmatic strengthening.