An Empirical Study on the Axiality of Gradable Adjectives

This study sought to empirically test the hypothesis that one adjective in adjective-adverb combinations acts as a central axis. The research focused on four semantic domains—degree of state and change, physical size, physical properties, and quality and state evaluation—selecting representative adjective pairs from each domain and analyzing how they combine with the adverbs &quotalmost&quot and &quotslightly.&quot Using frequency analysis from the COCA corpus, the findings revealed that it is challenging to generalize that a specific adjective consistently serves as an axis. Instead, the way adjectives pair with adverbs can vary depending on the context. This study underscores the need to consider the complexity and flexibility of adjective usage, indicating that simple binary models may not adequately capture the nuances of adjective-adverb combinations. Future research and language education should take these factors into account.