Charting New Territory: Digital Humanities and AI in African Studies

Charting New Territory: Digital Humanities and AI in African Studies is the website for a scoping workshop that Vincent Hiribarren and I organised at the Xplanatorium Herrenhausen in Hanover from 18 to 20 February 2026, funded by the VolkswagenStiftung. It gathered 26 researchers from 16 countries across Africa, Europe and beyond for three days of structured discussion rather than presentations.

I designed and built the site. It announced the workshop in advance and now documents its outcomes. The event itself is listed under conference activity, and it belongs to a wider research agenda.

Programme and participants

Tabs split the three-day programme by day and sync to the URL, so any session can be linked to directly. A searchable directory covers all 26 participants: browse them by thematic group, or see where they work on an interactive map of their institutional affiliations.

References

The reading list gathered to prepare the workshop is now a standing bibliography of over 200 works on digital humanities, artificial intelligence and African studies: journal articles, books, chapters, theses, blog posts, videos and web pages, in English, French and German.

It filters by type, language and year, and by any of 943 keywords covering data sovereignty, digital colonialism, African language technologies and AI ethics. Every entry carries an abstract and a link to the work itself, and any selection exports to BibTeX or RIS, so readers can pull the bibliography into Zotero instead of only reading it on the page.

Concept map

Reading 215 works produces notes; the harder question is what they collectively say. I read and annotated each reference in Obsidian and captured recurring themes as interlinked concept notes.

Fifty-five foundational concepts seeded the map, matched to the workshop’s thematic groups. The wiki-links between notes then surfaced connections nobody had planned, and the network grew to 105 concepts joined by 863 relationships. A Python script converts that graph to JSON and a D3 force layout draws it: node size follows connection density, seeds are marked, and clusters emerge around archives, epistemologies and decoloniality, infrastructure and governance, and language technologies.

The map is a reading of a corpus, drawn from annotation rather than from memory. The script that produced it is in the repository. On the live map, Ctrl+scroll to zoom, drag the background to pan, and drag a node to reposition it.

Interviews and photographs

Six short interviews, filmed by Calum Houston, let participants set out their own positions: Emmanuel Ngue Um on African language resources, Duncan Money on archival digitisation, Karen Byera Ijumba on restitution data, Fallou Ngom on Ajami manuscripts, Menno van Zaanen on digital language resources, and Albrecht Hofheinz on Arabic manuscript analysis. A photo gallery covers all three days.

Position paper

The workshop’s main outcome is a co-authored position paper on the ethical integration of digital humanities and AI in African studies, written for specialists and policy readers. A drafting committee has been working on it since the workshop closed, and it will appear in the ZMO Programmatic Texts series. The site carries a page for the paper and will publish the details on release.

How it’s built

The site is a static SvelteKit build (Svelte 5 runes, TypeScript, Tailwind CSS) deployed to GitHub Pages, with D3 for the concept map and Leaflet for the participant map. Two standard-library Python scripts do the data work: one fetches and normalises the references, the other extracts the concept graph from the Obsidian vault. The interface has a light and dark mode, meets WCAG 2.1 Level AA, and publishes JSON-LD structured data.