Economy of structure and information: Oddness, questions, and answers

We examine two conflicting perspectives on oddness: Magri (2009, 2011)’s theory, which derives oddness from blind inferences that clash with common knowledge, and Spector (2014)’s theory that derives oddness from trivial alternatives. Building on these works, we offer a third alternative, one that relies on a discourse condition that says that a good assertion is one that provides a good answer to a good question. A remaining difficulty is the persistence of oddness when the relevant sentences are embedded in environments that are predicted to satisfy the proposed appropriateness conditions.