The Asia Foundation | The Contested Corners of Asia: A Visual Companion



  • Start
  • Deaths/Year
  • Deaths/Total
  • End

Deaths ( in thousands )


Additional Insight

Statistics were derived from the Uppsala Battle Related Deaths dataset. As identifying a reasonably accurate number of deaths in a subnational conflict area can be quite challenging, the Uppsala Battle Related Deaths dataset uses a variety of sources to create an estimate of deaths within a certain range. This estimate is typically considered to be very conservative. For this reason, we have presented the high end of that range for the Contested Corners of Asia report and visualization.

Additionally, the Uppsala dataset does not track when conflicts have a death toll below 25. Since subnational conflicts often have long periods of relative quiet followed by intense violence and then calm again, we do not assume that the conflict is no longer active when conflict areas go through these quiet periods. You can see these periods in the visualization when there is no battle-related death activity. But again, this does not mean that the conflict was resolved, only that there had been relatively low-level violence during that year.

How and when these conflicts begin and end has been measured in different ways. For this report we have referred to the Heidelberg Institute for International Conflict Research to determine the lengths of many of these conflicts. Occasionally there have been discrepancies in the timelines presented in this dataset and the Uppsala Battle Related Deaths. In these cases we have combined the two datasets to determine the length of conflicts. The conflict summary text was derived from the Uppsala Conflict Encyclopedia, so this means there is also the occasional discrepancy between the conflict start and end dates on the visualization map and summary text.

Many conflicts can be defined by state/provincial borders, so we have used these areas to determine conflict locations. A small portion of these conflicts span multiple states/provinces. In these cases we have included all of the affected states/provinces, even though conflict intensity may be confined to a relatively small area within certain states/provinces.