This presentation aims to address, and hopefully contribute to clarify, some of the major theoretical and methodological issues a researcher faces when confronted with the task of transcribing, analysing and coding sign language acquisition data. The observations I will present are based primarily on work conducted within our group on adult and child Italian Sign Language, but I will try to demonstrate how most, if not all of the problems encountered in the analysis of child (and adult) LIS are similar to those encountered in research on other sign languages, hence can be considered relevant from a crosslinguistic perspective.
I will first consider how the transcription problems encountered with child language data are essentially the same as those encountered in the transcription of any text segment of adult LIS (and of sign languages in general). These problems stem primarily from the fact that, to date, none of the exhisting sign languages communities have developed autonomous writing or notation systems for their languages, hence all sign languages can be considered akin to "oral" languages which do not have a written tradition.
I will then briefly review a number of systems that have been proposed for the written notation of signs and sign discourse (e.g. Stokoe-derived representations, Hamnosys, SignFont, SignWriting), and point out what are, in my view, the main reasons underlying the apparent failure of the notation or writing systems currently available to solve and/or even to appropriately address the representational issues at stake.
Finally, drawing directly on LIS acquisition data, I will examine the theoretical and methodological problems posed by transcribing and coding child and adult signs via so-called glosses, or written labels given in, for, example, Italian, English or any spoken language with a written tradition. These glosses can be enriched with particular coding or notation devices that attempt to capture the particular spatial-temporal features proper of the morphology and syntax of signs in discourse, and can be applied to represent either the global meaning, or even some of the morphological components of the signs. In particular, I will demonstrate how gloss-based analysis of manual signs need to take into account specific morpho-phonological features of the signs marked by their places of articulation and by the use of one vs. two hands in the production of individual signs and/or sign sequences.
Taken together, the observations I will summarize in this presentation point out the severe limitations of all the notation and/or coding systems currently employed -- for child as for adult sign language data --, and underscore the need of a major research effort to address more appropriately the problems posed by the written representation of visual-gestural linguistic signs.