The study of child sign language has emerged from the growing interest in cross-linguistic comparisons of language development, stimulated greatly by the early work of Dan Slobin and colleagues (e.g. Newport & Meier in Slobin 1985). However the modality sign language is produced in has made comparisons with other languages difficult (including other sign languages). This has been due in part to a lack of an agreed normative transcription system and the difficulty in representing child forms of fluid three-dimensional linguistic packages. This paper deals with these two issues by describing transcription methods for three levels of sign language analysis: 1. Whole discourse 2. Narrative episodic and 3. Verb phrase, in a corpus of child and adult BSL collected in Bristol and London.
At each level of analysis the transcription process used is described with examples from child BSL data. The usefulness of these methods for cross-linguistic comparison and data presentation is discussed. In the conclusion of this paper a method for using an adapted transcription system based on the CHILDES programme alongside the 'Sign-Stream' data analysis tool is proposed.