With practical examples taken from real-world data spanning a wide range of languages and issues, this book introduces a novel theory of language processing, the 'Discriminative Lexicon Model'. It is essential reading for academic researchers and students in a wide range of linguistic fields, including phonetics and computational linguistics.