'Sign languages, or iltyem-iltyem angkety, are in daily use in Arandic speaking communities of Central Australia. They are a form of communication used alongside other semiotic systems, including speech, gesture and drawing practices. Whereas sign languages used in deaf communities operate without any connection to speech, these 'alternate' handsign languages are used in various contexts by people who also use spoken language. They are culturally valued and highly endangered, yet there has been little or no systematic documentation of Arandic sign since Kendon (1988). In this paper we describe a pilot program to record Arandic sign languages, conducted by a community language team, funded by the Maintenance of Indigenous Languages and Records (MILR) program and by the Endangered Languages Documentation Program (ELDP), and auspiced by the Batchelor Institute (BIITE). Research into various aspects of multimodal communication brings with it many theoretical and practical challenges. New technologies and the ever-expanding potentials of data annotation systems create a plethora of choices and huge volumes of recorded material. Whereas the use of film in language documentation has recently become de rigueur, at least in some circles, it is often only as an adjunct to studies of spoken language. When the visual is foregrounded, as it is in sign and gesture research, additional layers of complexity are added that impact on all aspects of the documentation process. How, for example, do we balance the desire for naturalistic visual data with the need for visually 'clean' images? What lessons can linguists learn from ethnocinematographers (Dimmendaal 2010)? What kinds of resources will benefit the community and a range of users (scholarly, archival, educational etc), as well as satisfying community aspirations for medium and long-term engagement with their audio-visual language materials? How do we ensure that our methodologies are robust enough to allow comparisons between primary sign language corpora and alternate sign language ones?
' We discuss these issues and various others encountered in our research, including our field methodologies, annotation of film data, community consultations and ethical considerations, and issues that have arisen in designing an interactive sign language website for use as a teaching/learning resource in Arandic schools. Although the creation and management of digital archives for primary sign languages have been documented before (see Johnston & Schembri 2006), 'alternate' sign languages have received little attention.' (Publication abstract)