Islam's 'Peripheries': Digital Humanities, Algorithmic Analysis, and AI in West Africa and Central Asia

This research project (2026-2027), co-directed with Aksana Ismailbekova, uses artificial intelligence (AI) to unlock valuable historical collections about Islamic communities in West Africa and Central Asia. These archives contain thousands of documents in multiple languages (including Russian, Arabic, Hausa, and Tajik) that have remained largely inaccessible due to their volume and complexity.

The project focuses on two often-overlooked regions: West Africa (post-1960s) and Central Asia (colonial/early Soviet era and documents from the Tajik civil war, 1992-97). By comparing these regions, we explore how Islamic discourse and responses to modernity developed in different contexts.

Three Innovations

1. From raw documents to structured data. Current multimodal LLMs (notably Gemini) now outperform specialised OCR/HTR systems, even on low-resource languages and diverse scripts (Hausa, Arabic, Cyrillic, Old Tatar, Tajik). Combined with context-aware named entity extraction, these workflows transform raw documents into structured, interconnected data.
2. AI-mediated access via MCP and a chatbot. A Model Context Protocol (MCP) server provides a standardised interface for any AI assistant to query both collections, while an AI "skill" encodes curatorial expertise (search strategies, transliteration variants, bias documentation). A researcher can ask a single question in natural language and interrogate both archives simultaneously — no need to master eight languages. The planned deliverable is a conversational assistant on the project website, powered by an open-source LLM.
3. Equitable, accessible, sustainable. Open-source, low-resource AI models run locally without expensive infrastructure. A human-in-the-loop approach (e.g. the AI-NER-Validator) ensures domain experts retain final authority. All code, data, and publications released as open access (FAIR principles).

This approach will reveal previously hidden connections and demonstrate how diverse Islamic communities navigated major historical transitions. All findings and tools will be made freely available, benefiting researchers worldwide, especially in West Africa and Central Asia.

Unlocking "Peripheral" Archives

Our work focuses on two unique, multilingual archives housed at Leibniz-Zentrum Moderner Orient (ZMO):

• The Islam West Africa Collection: 14,500+ items (newspapers, Islamic publications, 9,315 minutes of Hausa audio, photographs) covering Islam and Muslim communities in six West African countries from the 1960s onward.
• The Reinhard Eisener Collection: 1,546 documents across 50 archival boxes in eight languages (Russian, Uzbek, Tajik, Persian, Turki, English, German, French), documenting the Emirate of Bukhara (1917–30), early Soviet governance, and the Tajik civil war (1992–97).

By combining expertise in regional studies, history, and AI, this project will uncover important historical insights while creating new methods for studying complex archives globally.