Research Group
in Analytic Philosophy

Retraction and disagreement data have been used to motivate relativist accounts of certain types of claims. Among these figure epistemic modals, knowledge attributions, or personal taste claims. On the relativist proposal, sentences like “the ice cream might be in the fridge” or “Pocoyo is funny” only get assigned a truth-value relative to contexts of utterance, indices of evaluation, and contexts of assessment. The novelty are the contexts of assessment. Retractions play a crucial role in the argument for assessment relativism, because a retraction is supposed to be obligatory. Assessment relativism would accommodate this: A speaker must retract a previously made assertion if it is not true at the context of assessment. If retractions were not obligatory, there would be no normative difference between assessment relativism and non-indexical contextualism. This paper has to goals. The first is to offer a review of several objections to the obligatoriness of retractions. A second, is to suggest how a contextualist (indexical or not) can accommodate permissible retractions. This will satisfy a further aim, which is to undermine the claim that assessment relativism is normatively distinct.