Outside the Discipline, Inside the Data: A Retrospective Account of an Undocumented Tunisian Language Corpus in an Extractivist Research Context

A previously undocumented corpus of Tunisian language interviews, collected during a five-day fieldwork period in 2022, has been retroactively deposited and made reusable. This corpus—comprising 152 structured interviews with residents of the Medina of Tunis buffer zone—was unintentionally generated by an architect researcher working outside traditional linguistics or data science disciplines. The interviews were conducted, transcribed in Tunisian, and translated into French but lacked formal documentation, deposition, or citation as a dataset. Now archived on Zenodo, this work serves as both a case study of extractive research practices and a cautionary account for humanities and built-environment researchers who may inadvertently collect valuable linguistic data without recognizing its potential for broader reuse.

Stiti, K. (2026). Outside the Discipline, Inside the Data: A Retrospective Account of an Undocumented Tunisian Language Corpus in an Extractivist Research Context. Journal of Open Humanities Data, 12, 66. https://doi.org/10.5334/johd.525