Abstract
With the goal of advancing research on sign language corpora, the present paper proposes an integrated learner corpus of Japanese Sign Language that not only offers solutions to conceptual problems that have until now remained unsolved, but also contributes to Japanese Sign Language linguistics and education. The corpus has four main characteristics. First, the corpus covers linguistic abilities in the linguistic domains of generating and understanding. Second, the corpus enables examination of linguistic abilities with regard not only to the linguistic output generated by language learners, but also how linguistic output is generated. Examination of linguistic abilities with regard to these aspects reveals both the precision and process efficiency of linguistic output. Third, the corpus enables examination of basic linguistic abilities such as the manipulation of phonological-phonetic, morpho-syntactic, and semantic-discourse properties, because the designed language use situation prevented the use of advanced linguistic abilities including stylistic and rhetorical techniques, as well as non-linguistic abilities in specific academic or business domains. Finally, the corpus allows the development of linguistic abilities to be examined, as it can be regarded as quasi-longitudinal data. The corpus also provides data on the proficiency of language learners measured by a norm-referenced test that yields a scaled score. The proposed learner corpus is expected to serve as a useful language resource for research on Japanese Sign Language.
