Seeing sentence boundaries : the production and perception of visual markers signalling boundaries in signed languages

Current definitions of prosody present a problem for signed languages since they are based on languages that exist in the oral-aural modality. Despite this, researchers have illustrated that although signed languages are produced in a different modality, a prosodic system exists whereby a signed str...

Full description

Bibliographic Details
Main Author: Fenlon, J. J.
Published: University College London (University of London) 2010
Subjects:
419
Online Access:http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.564983
Description
Summary:Current definitions of prosody present a problem for signed languages since they are based on languages that exist in the oral-aural modality. Despite this, researchers have illustrated that although signed languages are produced in a different modality, a prosodic system exists whereby a signed stream can be structured into prosodic constituents and are marked by systematic manual and non-manual phenomena (see Nespor & Sandler, 1999; Wilbur, 1999, 2000). However, there is little research examining prosody in British Sign Language (BSL). This thesis represents the first serious attempt to address this gap in the literature by investigating the type and frequency of a number of visual markers at intonational phrase (IP) boundaries in BSL narratives. An analysis of 418 IP boundaries shows linguistic visual markers are not frequently observed. The most frequent marker observed were single head movements (46%) followed by holds (30%) and brow movement (22%) and head nods (21%). This finding suggests that none of the visual markers included in this study can be considered a consistent marker to IP boundaries in BSL narratives. As well as examining the production of markers at IP boundaries, the perception of boundaries by different groups in a series of online segmentation experiments is investigated. Results from both experiments indicate that boundaries can be identified in a reliable way even when watching an unknown signed language. In addition, an analysis of responses suggests that participants identified a boundary corresponding to a discourse level (such as when a new theme is established). The results suggest that visual markers (to these boundaries at least) are informative in the absence of cues that can only be perceived by native users of a language (such as cues deriving from lexical and grammatical information). Following presentation of results, directions for future research in this area are suggested.