Below I offered a rough assessment of areas of greatest LTTE and JVP control based on 1988 Presidential election turnout. In the absence of reliable event-level data, it’s a crude but hopefully useful proxy. The 1971 JVP uprising, by contrast, produced an official report that includes a listing (Appendix 3) of police stations damaged, police stations attacked but not damaged, and areas that “which went under Insurgent control either completely or substantially” (this overlaps heavily with the damaged police stations list). Obviously the report must be taken with a grain of salt (its assessment of civilian casualties seems bizarrely low) but it provides a level of specificity lacking from the 1987-1990 period.
My amateur hour Google Map below shows the geography of the JVP rebellion. Red are damaged police stations, yellow are attacked but undamaged police stations, and the blue boxes are areas the report identified as being under some sort of insurgent control.
As we can see, the efforts at seizing areas of control in and around Colombo and the heavily populated west coast failed, with a clear clustering pattern being instead seen in the deep south and the parts of central, north-western, and north-central Sri Lanka. The ethnic dimension of the revolt is clear, as is its regional and rural concentration. Future work will consider both the spatial characteristics of this revolt and its area of overlap (and non-overlap) with the late 1980s. The 1963, 1971, and 1981 censuses, as well as election returns, will be getting some heavy use in the months and years to come.