Experimenting with every American king

Area of Science:

• Semantics and Pragmatics
• Formal Semantics
• Natural Language Understanding

Background:

• Standard semantic theories predict sentences with empty restrictors (e.g., 'Every American king') are true.
• However, many find such sentences odd, particularly with determiners like 'every' compared to 'no'.
• Empirical investigation into the perception of empty restrictors is lacking.

Purpose of the Study:

• To empirically investigate how people judge the truth and oddity of sentences with empty restrictors.
• To examine whether determiner type influences these judgments.
• To test existing semantic accounts against empirical data.

Main Methods:

• Three experimental surveys were conducted.
• Participants judged the truth and/or oddity of sentences containing 'every' and 'no' with empty restrictors.
• Data were analyzed to identify patterns in judgments based on determiner type and truth value assignment.

Main Results:

• No significant difference was found in the tendency to judge sentences as odd based on determiner type ('every' vs. 'no').
• Assessors who assigned a truth value to 'every' sentences with empty restrictors overwhelmingly assigned 'false'.
• Existing semantic theories do not straightforwardly account for these findings.

Conclusions:

• The empirical results challenge standard semantic predictions for 'every' with empty restrictors.
• The tendency to assign 'false' to such sentences requires new theoretical explanations.
• The study highlights the need for semantic theories to align with empirical judgments of language users.