Antwerp Street Stories
±90,000 machine-readable pages of handwritten local police reports (1876-1945)
DOI:
https://doi.org/10.52024/xzejp971Keywords:
handwritten text recognition, police reports, digital history, urban history, social historyAbstract
We present Antwerp Street Stories: a machine-readable dataset of approximately 90,000 pages from the incident books of Antwerp, Belgium, spanning from 1876 to 1945. In these books, local police officers meticulously documented daily incidents on the streets, capturing information on the location, time, and date of each event, along with personal details of the involved citizens. The records offer a unique window into the social dynamics of everyday urban life in a rapidly expanding metropolis. This paper outlines the digitization process, including handwritten text recognition (HTR). It also describes the dataset’s key characteristics, such as its size, temporal and geographical scope, languages, and HTR accuracy. Despite some inherent limitations, such as data gaps and registration biases, the dataset holds significant potential for further research. Antwerp Street Stories supports data-driven approaches as well as in-depth explorations for (urban) histories of all kinds, among others urban crime, police-community interactions, and the lived experiences of lower social groups.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 L. Lefranc, M. Kestemont, I. Van Damme

This work is licensed under a Creative Commons Attribution 4.0 International License.
Licensing information can be found here.
