REASONS AND THAT‐CLAUSES†

Abstract

What are reasons? For example, if you’re aware that your secretary plans to expose you, and you resign to avoid a scandal, what is your reason for resigning? Is your reason the fact that your secretary plans to expose you? If so, what kinds of facts are eligible to be reasons? Can merely possible facts be reasons (for actual subjects)? Can merely apparent facts? Or are reasons rather attitudes? Are your reasons for resigning your belief that your secretary plans to expose you, and your desire to avoid a scandal? Or are reasons propositions? If so, which propositions? Only propositions that the subject believes and desires? Only propositions that are true? (On some views, those will be facts; on other views, not.) It’s hard to know how to choose between these ontologies. 1 The usual view is that the ontology of reasons won’t matter much to normative disciplines like epistemology and ethics. My sympathies are with this usual view. But we shouldn’t underestimate the opposing evidence. Different theorists at least talk differently about reasons. Humeans identify a subject’s reasons with her beliefs and desires. They have an easy time accounting for talk like “John resigned because he believed his secretary plans to expose him. ” They have to work harder to explain talk like this: “There were two good reasons for John to resign: first, his secretary plans to expose him…” For Davidson, it’s important that our reasons for acting be certain kinds of causes of our acting; so he will also tend to identify our reasons with our attitudes. Some moral realists take the notion of objective reasons, out there in the world, as their primitive. As they see it, the fact that your glass contains petrol is a reason for you not to drink, even if you think it contains gin. 2 The realists have an easy time accounting Work by Jennifer Logan on that-clauses, and papers by Jeff King and Tobias Rosefeldt, stimulated this research. Thanks to Sarah Moss and Tom Kelly for their

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 45 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.