This Departmental Research Seminar is the first in a series of academic seminars on the theme of disability inclusion, taking place taking place across the university around Disability Week, which runs from 16 to 20 May.

This seminar reports on one of the first known formal linguistic studies on the coda constraints of syllables in South African Sign Language (SASL). Coda constraints are common in spoken languages: German, for example, can only have voiceless obstruents in the coda position (Blevins 1995). Most formal sign language research has come from the sign languages of Western countries, such as American Sign Language (ASL). My research serves to contribute to the description of SASL, an under-studied African sign language.  It also contributes to the phonological theory on syllables and has a methodology that allows for cross-linguistic research, strengthening our understanding of sign languages in general.

It is well-known that syllables in spoken languages require a vowel nucleus to be well-formed. Sandler and Lillo-Martin (2006) provide evidence of sign language syllables requiring movement to be considered well-formed: even seemingly stationary signs such as WHO in SASL, which occurs at the chin, will have finger wiggling as some form of movement. It is thus natural to assume that movement is akin to vowels in spoken language syllables (Brentari 1998), and locations are similar to consonants. However, the visual nature of sign allows for simultaneity – multiple features can co-occur at syllable boundaries. Thus handshapes and locations are phonetically complex features to which coda constraints may apply.

The data examined in this study comes from a video dictionary of 175 SASL words in citation form.  From this dictionary, a database of coded locations and handshapes was recorded for both the onset and the coda, using features from Brentari’s (1998) Prosodic Model. From this, a phonological inventory was made and patterns were identified. These patterns were then examined to determine what possible phonological rules were dictating constraints. Since I as a Hearing researcher cannot claim native knowledge of a sign language, the conclusions drawn from the data were tested with the aid of native SASL signers.

The findings of the research suggest that certain onset-coda combinations may be prohibited due to the presence of natural classes of handshape and location. There is a tendency for location and handshape features to match across onsets and codas in monosyllabic and repeated signs. Where there is handshape change, the handshape specified at the coda will be a member of a small class of unmarked handshapes found in asymmetrical two-handed signs. Where there is a location change from onset to coda, the change in both regions and settings shows movement towards the centre of the signer's body. This can be seen as a result of either markedness or centralisation-type sonority. These constraints demonstrate that SASL has a rich phonological structure which warrants much further research.