The Corpus of Greek Aphasic Speech: Design and compilation

The paper presents the design and compilation of the Corpus of Greek Aphasic Speech, a new resource for the study of aphasia in Greek, and discusses its possible applications. The aims and design of the corpus and the methods followed for its compilation are presented. A pilot corpus, including two texts (spontaneous speech and picture description) from the spoken output of 20 patients was first created. On the basis of this, a classification of paraphasias or speech errors has been attempted and some preliminary findings have been gathered, while the target corpus is planned to include texts from 120 patients of about 50.000 words. It is argued that computer language corpora can offer a new perspective to the linguistic study of aphasia, especially in Greek, by providing systematic evidence for patients’ linguistic profiles, drawn from language use rather than from isolated examples of lack of competence.