Introduction

As explained in Chapter 1, a central aspect of RCVP is that the basic building blocks of phonology are not phonetically defined features, whether innate or not. Views on the innateness of features have shifted over recent years. The central idea of RCVP is that phonetic features/properties are correlates of abstract phonological categories that impose a categorisation on phonetic spaces, according to general principles, such as the Opponent Principle, the Combinatorial Principle, the Binarity Principle, the Head-Dependency Principle, the SAA, and Monovalency. The ultimate units in the phonological structure are the elements C and V, which are correlated with phonetic implementations (both acoustic and articulatory; see § 1.3.1) that more or less resemble the phonetically defined features that phonologists have assumed for several decades. The actual occurrence of the basic elements in the structure emerges from successive splitting during language acquisition, based on the prior recognition of phonetic categories, which are a statistical reflection of contrastive segments in the ambient language (see § 9.8). This means that there is no overall innate phonological structure for segments in any specific language. The nature of the structure that is needed is dependent on which phonetic properties in the phonetic space are used contrastively in a given language. The preceding chapters have discussed in detail how the principles of RCVP lead to a set of contrastive categories in spoken languages. However, the procedure was never meant to be specific to the spoken modality and it therefore carries over to what we can say about phonological structure in another modality: the manual-visual modality. If RCVP is a truly a-modal theory, its principles must also be appropriate for constructing a phonological organisation for sign languages. In this chapter, I demonstrate that the RCVP approach can be fruitfully applied to sign language phonology. In previous work over the last two decades, I have developed, in collaboration with others, an explicit model of sign language phonology. In that previous work, I did not always emphasise how an appropriate structure for signs can be derived from the principles of RCVP, using more ‘descriptive’ phonetic labels for contrastive specifications, although in van der Hulst (2000b) I make an explicit comparison between the two modalities in terms of the RCVP approach.