Introduction

Since there is no general notation for phonological representations, theories of phonological representation are proposed and defended using theory-internal notations. Classical phonemic phonology used strings of lower-case romanic symbols for its notation, Firthian Prosodic Phonology used Greek and upper-case romanic symbols, with subscripts and superscripts. Transformational-generative phonology used two-dimensional matrices of distinctive features, and non-linear phonological theories make use of two-and three-dimensional graphical notations to portray phonological relations such as constituency, dependency and association.

Faced with two apparently different phonological representations from two different theories, it is not possible to determine by inspection alone whether they are notational variants which denote the same phonological object in different notations, or whether they denote different phonological objects. In order to understand how theories of phonological representation work, it is helpful to understand first how phonological notations work. In this chapter I shall examine phonological notation from a cross-theoretical perspective.