Abstract
The range of estimates of excess deaths under Pol Pot's rule of Cambodia (1975–79) is too wide to be useful: they range from under 1 to over 3 million, with the more plausible estimates still varying from 1 to 2 million. By stochastically reconstructing population dynamics in Cambodia from extant historical and demographic data, we produced interpretable distributions of the death toll and other demographic indicators. The resulting 95 per cent simulation interval (1.2–2.8 million excess deaths) demonstrates substantial uncertainty over the exact scale of mortality, yet it still excludes nearly half of the previous death-toll estimates. The 1.5–2.25 million interval contains 69 per cent of the simulations for the actual number of excess deaths, more than the wider (1–2 million) range of previous plausible estimates. The median value of 1.9 million excess deaths represents 21 per cent of the population at risk.
Supplementary material for this article is available at: http://dx.doi.org/10.1080/00324728.2015.1045546
Notes
1 Patrick Heuveline is at the Department of Sociology, University of California Los Angeles, 285 Haines Hall, PO Box 951551, Los Angeles, CA 90095–1551, USA. E-mail: [email protected]
2 The author received no grant to conduct this research, but he made use of the facilities and resources of the California Center for Population Research, UCLA, which is supported by infrastructure grant R24HD041022 from the Eunice Kennedy Shriver National Institute of Child Health & Human Development. The author wishes to thank Mark Handcock, Adrian Raftery, Rob Mare, Ron Brookmeyer, Patrick Gerland, Sander Greenland, Matthew Salganik, and Seth Sanders for their insightful suggestions and Andrew Hicks for implementing model simulations in Stata (release 12).