"Vagueness is not just loose talk in ordinary language but an essential feature of all communication, even in mathematical and scientific languages. If we say that a building is tall, you can understand me even though neither of us has a clear idea of what counts as tall. This fact challenges the view that successful communication consists in the listeners recognition of the proposition that the speaker is asserting. The problem posed here is a special case of the more general problem of "felicitous underspecification": context-sensitive expressions that lack unique semantic values. Neither traditional theories of vagueness nor diagonalization-a technique for analyzing logical paradoxes-provides a solution to this problem. John MacFarlane argues that the mental states expressed in vague sentences constrain the ethical beliefs we can have as well as our practical border setting. He then applies this view to cases commonly associated with vagueness: bivalence (that a statement can only be true or false), the sorites paradox (that vague terms such as "tall" above have no clear line of demarcation between what counts and what does not), higher-order vagueness (vagueness about the concept of vagueness), and vague thought (thinking that is not based on a clear understanding of what the concepts involved mean). He concludes that an expressivist account, unlike other explanations, can tell us why vague language is useful-in particular, in consensus-building conversations"-- Provided by publisher.