Learning and Diachronic Laws for Partial Blocking

Abstract

In this talk we introduce a complete system of diachronic laws for predicting partial blocking. The laws get their justification from an underlying learning model, and the system is complete in the sense that all possible meaning shifts predicted by the learning model can be accounted for by using diachronic laws instead. We propose them as an alternative to Horn's division of pragmatic labour and the principle of weak optimality.