Captain Swing: Did the Riots Follow the Machines? A Parish-Level Look at the Swing Riots of 1830 to 1832

In 1830 farm labourers across southern and eastern England broke threshing machines, burned ricks and sent letters signed "Captain Swing". Historians have long debated whether the rising was about the machines or about poverty, wages, enclosure and the poor law. Using only the open replication data of Caprettini and Voth (2020), which place 3,117 Swing incidents and 641 recorded threshing machines in the 10,699 parishes of 1851 England and Wales, I first reproduce their published descriptive statistics and main coefficients exactly, and then offer a descriptive, visual reading of the same data. Parishes with a recorded machine saw unrest in 27.0% of cases against 13.8% without; against machine-free neighbours within 10 km the gap is 9.8 percentage points (95% interval 6.1 to 13.6). Yet 90% of parishes with unrest had no recorded machine. Enclosure shows almost no gradient (16.3% to 17.7% across fifths), whereas poor relief per head in 1801 (14% to 34%) and winter unemployment reported in 1834 rise steeply, in the smaller sets of parishes where they are measured. Unrest was concentrated within 150 km of London, and the machine gap was larger far from industrial towns. The riots followed the machines, but not only the machines. A short, clearly bounded section asks what this early revolt against labour-saving technology can and cannot say about such technology in general. Built entirely on the open replication data of Caprettini and Voth (2020), American Economic Review: Insights 2 (3), https://doi.org/10.1257/aeri.20190385, data https://doi.org/10.3886/E117044V1 (CC BY 4.0). Their causal results are theirs; the descriptive analysis is mine. Interactive version: https://tristan-hotham.github.io/captain-swing/. Data and code: https://github.com/Tristan-Hotham/captain-swing. Blog post: https://blog.tristanhotham.com/2026/10/28/captain-swing/.