The Central Question
"How can we build spatial data infrastructure that puts slavery and war precisely in place — across time, across the world, and across the gaps left by erasure?"
Leverhulme Centre Training Day — Dr James Williams, 15 April 2026
Four Systems, One Programme
Each system addresses a distinct challenge in the spatial data infrastructure for slavery and war research, designed to complement one another.
“How can a centralised data infrastructure be designed to integrate, standardise, and enable analysis of heterogeneous data on slavery and war across space and time?”
A unified query layer over dispersed, heterogeneous datasets with a four-step wizard and five-layer architecture spanning Classification, Time, Space, Actors, and Events.
“Can a neural network learn the character of a city neighbourhood from map data alone — and find its twin on the other side of the world?”
Computes ~120 morphological features per H3 cell from OpenStreetMap, then embeds each cell into a 64-dimensional space. Applied across 6 cities on 4 continents.
“How can contextual AI improve evidence geocoding in slavery, displacement, and war scenarios?”
A five-stage pipeline running nine parallel geocoding backends with seven-component scoring, six structured failure modes, and a Spatial Coherence Score.
“How do we study places whose physical or nominal existence has been deliberately altered or destroyed?”
Reconstructs destroyed places from stratified archival, cartographic, satellite, and oral evidence. Treats erasure itself as a spatial phenomenon.
How the Systems Fit Together
Each occupies a distinct niche. Together they form a complete spatial data infrastructure for slavery and war scholarship.
Indexes events in places. Unified query over dispersed slavery and war datasets.
Encodes the morphological character of places. Finds urban twins across continents.
Resolves ambiguous place references in evidence. Contextual geocoding at scale.
Reconstructs places that no longer exist. Treats destruction as spatial data.
Topodex resolves ambiguity for recent places (2000s+). MORPHEME resolves similar places across continents. CDISaW indexes events that happened in places. There is no system that treats erasure itself as a spatial phenomenon — where the object of study is a place whose physical or nominal existence has been deliberately altered.
About the Programme
Slavery and War in Place is a research programme developed by Dr James Williams (Research Fellow, University of Nottingham) in association with the Leverhulme Centre for Research on Slavery in War. It builds spatial data infrastructure for the historical and contemporary study of slavery, conflict, and their geographies.
The four systems are designed to work in concert, each solving a distinct problem in the chain from data discovery to spatial analysis of slavery and war evidence.
- Build out CDISaW: full data pipeline, classification matrix, and query wizard.
- Publish MORPHEME methodology paper and apply to slavery and war geographies.
- Deploy Topodex geocoding pipeline against live evidence corpora.
- Develop Erasure framework for probabilistic reconstruction of destroyed places.
Leverhulme Centre Training Day — 15 April 2026
Dr James Williams, Research Fellow, University of Nottingham
As part of my role at:
Hero video: Street Traffic Cars City Night by Pixabay