Introduction

This chapter summarises the contributions of RCVP to our understanding of segmental and syllabic representations. I will outline strengths, as well as weaknesses, and point to future research which will contribute to the advancement of models in this domain. The RCVP model was developed largely as a ‘private’ enterprise that helped me understand that the set of features needed for contrast, as well as their grouping and correlations between them, is neither random nor a direct mirror of articulation. RCVP was never part of mainstream Generative Phonology, a fate (or blessing) that it shares with the various models that it is most closely associated with (such as DP and GP). Nevertheless, its goals and most of its results are compatible with many proposals that have been made in mainstream models, or at least better-known models (see especially Chapter 11). In this chapter I will limit myself to some general remarks about the model, its basic properties and directions for future research.

Goals and basic principles

The most important goal of the enterprise that I report on in this book is to develop a theory of phonological structure which is in accordance with, or follows from, a set of basic ‘first’ principles and some further assumptions.

To start with the assumptions (see also § 1.3), I assume that there are phonological segments, which are cognitive categories that correspond to the informal notion of ‘speech sound’ (or ‘sign’). These segments are not atomic, but can be broken up into basic primes that are organised in a hierarchical structure that corresponds to the phonetic dimensions and phonetic categories that are available for a given modality. I also assume that segments are grouped into syllables, with a basic division between onset and rhyme. A requirement for being admitted into the set of phonological categories is the possibility of occurring contrastively in some language, expectedly in more than one. Segments are minimally specified with the goal of capturing contrast. I also assume that phonology is substance-based, which means that all units and structures correlate with cognitive representations of physical events in the domain of psychoacoustics and articulation.