ABSTRACT
Assertion is normatively constrained by logical relations, in particular by consistency and by logical consequence. This gives inference a role in the critical evaluation of others' assertions allowing us to escape Harman's challenge that all reasoning is a matter of reasoned change in view. However, Harman's distinction between reasoned change in view and logical consequence is crucial to the evaluation of certain patterns of discourse involving indicative conditionals, patterns that are significant for the determination of the semantics of the indicative conditional of natural languages. Inferentialism says that the meaning of an expression in a language or the content of a concept is determined by some among its inferential relations to other expressions/concepts. The logic tradition's account of when indicative conditionals are true is completely at odds with what psychologists of reasoning tell us about how probable, how likely the conditional statements are.
