The Villages That Vanished: Size, Setting and Timing of England's Deserted Medieval Villages

England has thousands of deserted medieval villages, but most studies of them are local. I use the open database Beresford's Lost Villages (University of Hull), which digitises Beresford and Hurst's 1968 gazetteer, to look at all 2,222 listed sites at once, and I match them to Domesday Book through the Hull Domesday Project data. Three findings stand out. First, desertion was concentrated: in Lincolnshire 20% of located Domesday places later appear as deserted villages or hamlets, against almost none in Cornwall, Devon and Shropshire. Second, the doomed places were already smaller in 1086: a median of 13 households against 17 for places that survived, and 72% sat below the median of their own county. The risk of later desertion peaked for middling places of 6 to 20 households (11%) and was lowest for places of more than 40 (2.5%). Within counties, their height, relief, distance to water and distance to Roman roads were typical. Third, the presumed dates of depopulation recorded for 299 sites in the 14 fully documented counties put only 20% of the dated weight in the century after the Black Death and 74% after 1450, with the busiest decade in the 1500s. The written accounts mention enclosure or conversion to pasture, emparking and monastic landholders more often than plague. The data are a gazetteer, not a census, and the fullest records cover a minority of counties, so these are descriptive results with stated limits. Interactive version: https://tristan-hotham.github.io/lost-villages/. Data and code: https://github.com/Tristan-Hotham/lost-villages. Blog post: https://blog.tristanhotham.com/2026/10/27/lost-villages/.