Open text collections

• Applicant: Mandana Seyfeddinipur
• Subject Area: General and Comparative Linguistics, Experimental Linguistics, Typology, Non-European Languages
• Term: since 2023
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 517860213

This project will provide freely available text collections of low resource languages in a well-defined format with well-defined interfaces. Thereby, text collections hidden in digital archives and inaccessible formats will become discoverable, accessible and usable. With these collections linguistic inquiry and understanding of human diversity will improve allowing the testing of hypotheses developed from grammar or dictionary data, and the generation of new hypotheses either by humans accessing the content in book form, or by machine tools doing quantitative analyses of the structured data. Text collections together with grammatical descriptions and dictionaries constitute the so-called “Boasian Trilogy” in language documentation and description (Himmelmann 2002). All three components are needed to get a good understanding of an oral and hitherto not well-described language or dialect in the context of linguistic research. They not only rely on each other, but they cross-fertilize each other. Lack of prestige and academic merit, lack of simple and efficient workflows creating publishable text collections, and lack of visibility of curated collection across research areas are the three main obstacles preventing field linguists from the creation and publication of digital and open text collections. This project provides addresses and provides solutions for the three obstacles so that the third member of the Boasian Trilogy will thus live up to its potential and impact.

• DFG Programme: Science Communication, Research Data, eResearch (Scientific Library Services and Information Systems)