Our aim in writing this book has been to introduce the reader to some of the issues in the representation of the structure of the basic units of phonology. We have approached this by first, in Chapters 1 and 2, considering the ways in which the smallest phonological units, features, characterise the structure of sounds, or, more technically, segments. Chapters 3 and 4 are concerned with larger phonological units, in particular syllables and feet. As the title of the book suggests, we do not consider the representation of phonological units larger than the word, and therefore pay little attention to topics such as intonation.

Most of our analyses are formulated in terms of what has come to be referred to as non-linear phonology, as opposed to the ‘linear’ theories of phonological representation manifested in work in the tradition of Chomsky and Halle (1968). The term ‘non-linear phonology’ does not refer to a single coherent theory of the representation of phonological structure – whether segment-internal or suprasegmental – rather, since the early 1980s, work in phonology which has been concerned with enriching the structural properties of linear models has dealt with different aspects of these models, so that various apparently distinct theories have grown up. Two of the most familiar of these are metrical phonology, originating in the work of Liberman (1975) and Liberman and Prince (1977), and autosegmental phonology, which finds its first exposition in Goldsmith (1976). However, in recent years it has become apparent that many of the claims made in the various models are not in fact independent of each other, and that claims made within the framework of one approach are often restatements of those made elsewhere.