Experimenting with Degree

Semantic theories differ in the role they assume for degrees in the interpretation of gradable adjectives, and in the assumptions they make about the nature of degrees and the structure of the scales they comprise. We report on two experiments investigating speakers' use of gradable adjectives across varying contexts, with the goal of gaining insight into the nature of the degree ontology underlying their semantics. We find that the truth conditions for the positive form must be stated in terms of degrees rather than rankings of individuals, and further that the relevant scale structure is one where distances between scale points are meaningful, and not an ordinal scale derived from an ordering relation on a comparison class. We also find no evidence that scale structure depends on the presence or absence of a corresponding system of numerical measures