The project explores epistemic, theoretical, and methodological ways of recalibrating African linguistics. The main hypothesis of the research is that the current epistemic values, virtues, theories, and methods of African linguistics significantly prevent us from recognizing, let alone understanding, the social and cultural dynamics of which African linguistic diversity is a revelator. This represents a structural gap in the global knowledge system. By considering language as an extensible archive of formal and relational materialities, the project will help to rethink the epistemic spaces constructed by colonialism. These spaces are created by nomenclatures of linguistic classification and contribute to a social geography that emphasizes linguistic diversity in Africa while underplaying the fluidity that characterized pre-colonial spaces. The project aims to bring to light the linguistic fluidity inherent in African social spaces. This will be achieved by assessing the strengths and limitations of colonialism and missionary linguistics. It proposes an open and networked perspective for understanding and representing African linguistic reality. The project reuses data from the World Atlas of Language Structures (WALS), the Automated Similarity Judgment Program (ASJP) databases and primary data from the Cameroon Linguistic Atlas project. Deep learning methods, in particular word embeddings, are used to model the fluidity of both the formal and relational materialities of language in a vector space.
Project
(2024)
Rethinking Linguistics from an African Perspective
- Emmanuel Ngue Um