30 May 2007, Computational Linguistics Seminar, Reinhard Blutner

Classical truth-functional semantics and almost all of its modifications have a serious problem in treating prototypes and their combination. Though some modelling variants can fit many of the puzzling empirical observation, their explanatory value is seldom noteworthy. Reinhard Blutner will argue that the explanatory inadequacy is due to the Boolean characteristic of the underlying semantics, which only allows mixing possible words but it excludes the idea of superposition crucial for geometrical models of meanings.

For more information, see http://www.illc.uva.nl/LaCo/CLS/ or contact Jelle Zuidema (jzuidema at science.uva.nl).