Program
Paul Egré (IJN)
Vagueness and Uncertainty: from Signal Detection Theory to Degrees of Clarity
Vagueness is exemplified by the existence of borderline cases, namely cases for which it is unclear whether a property should apply or not. My aim in this talk is to discuss the notion of unclarity typical of vagueness from the perspective of signal detection theory, a decision-theoretic framework used to model uncertainty and imperfect sensory discrimination, applicable to a wide range of psychophysical tasks. One interest of signal detection theory is that it allows us to refine the margin for error model introduced by T. Williamson in his account of vagueness as epistemic. In particular, it can be used to articulate the intuitive idea that clarity and unclarity come in degrees. A further interest concerns the discussion of the uncertainty associated to higher-order vagueness, in particular the idea that unclear instantiations of a property need not be ``clearly unclear". I plan to discuss experiments by Smith, Shields and Wahsburn that relate uncertainty responses to difficulty in categorization, and to use insights from their account in order to discuss the phenomenon of higher-order vagueness. I will discuss the idea, outlined in recent work with Denis Bonnay (see our "Vagueness, Uncertainty and Degrees of Clarity"), that while clarity can be clear, unclarity is typically unclear in turn.
