Experimental evidence for variation across exclusive modifiers

Abstract

Listen

Translate article

This paper is an investigation of variation across the English exclusive modifiers only, just, and merely—a domain that has received ample attention in the theoretical literature but has thus far not been subjected to experimental testing. Using scalar diversity as a testing ground, we report on two experiments: Experiment 1 tests the robustness of exclusionary inference calculation (e.g., merely intelligent → not brilliant), and Experiment 2 directly tests whether a rank-order (e.g., not brilliant) or complement-exclusion (e.g., not ambitious) reading is preferred with different exclusives. Our findings reveal that 1) just excludes less robustly than the other two exclusives, and 2) while only allows both complement-exclusion and rank-order readings, just has a weak and merely a strong preference for rank-order ones. These results bear on previous theoretical observations about exclusives, and they are also informative about the robust by-scale variation in inference calculation, i.e., scalar diversity. Lastly, we argue that the methodological success of our experiments opens up avenues for further examination of more precise predictions of competing theoretical accounts.