Can a car really run on empty?: An experiment testing for within- and between-individual comparison classes in relative and absolute adjectives.

Abstract

This thesis investigates two theories that provide an account for the distinct difference between relative and absolute gradable adjectives. Kennedy (2009) on one hand proposes that the difference is fundamental to the underlying scale structure of the adjective, while Toledo & Sassoon (2011) on the other hand proposes that the difference is fundamental to the type of comparison class that the adjective is assigned through their typical arguments. This thesis reports on an experiment investigating the nature of the type of comparison class for relative and absolute adjectives, as well as their dependency on their scale structure for interpretation. I find that a combination of the two theories must be made in order to fully comprehend the result of the experiment, and further propose that functionality is one pragmatic factor that influences the interpretation of positive forms of gradable adjectives, among other possible factors.