Death tolls and their limits
Deadliest Wars in History: Ranked and Mapped
Every list of the deadliest wars in history is really a list of three different things stacked together: a few wars with reasonably firm records, a larger group with reconstructions that vary by a factor of two, and a handful of pre-modern conflicts whose famous totals come from the collapse of tax registers rather than from any count of the dead. This page keeps those categories visible instead of averaging them into a tidy ranking.
Why the ranking is unstable
Four decisions change the order of any such list, and different compilers make them differently.
First, what counts as a death caused by the war. Combat deaths are one number; add deaths from war-induced famine and epidemic and totals for the same conflict can double or triple. The Russian Civil War is the clearest case: fighting killed hundreds of thousands, and the famine and typhus that accompanied it killed millions.
Second, what counts as one war. The Taiping Rebellion is two records in this atlas’s catalogue, split in 1860 where the source data changes category. The Second World War and the Sino-Japanese war that preceded it overlap. Whether the Chinese Civil War is one conflict from 1927 to 1949 or several separate wars changes its position on any list.
Third, whether genocide and mass killing inside a war are counted as war deaths. The Holocaust, the Armenian genocide and the Rwandan genocide of 1994 each occurred within or alongside a war, and each is included in some totals and excluded from others.
Fourth, whether the underlying demographic data exists at all. For the pre-modern entries it generally does not.
Sources
White, M. (2011), Atrocitology: Humanity’s 100 Deadliest Achievements. CanongateClodfelter, M. (2017), Warfare and Armed Conflicts: A Statistical Encyclopedia, 4th ed. McFarlandHuman Security Report Project (2010), Human Security Report 2009/2010 — The Causes of Peace and the Shrinking Costs of War. Oxford University Press
The deadliest wars by estimated deaths
Ordered by the midpoint of the published range. Treat the order as approximate: several of these rows overlap so heavily that any of them could swap places, and the last group cannot honestly be ranked at all.
| War | Dates | Estimated deaths | Why the range is what it is |
|---|---|---|---|
| Second World War | 1939–1945 | 70–85 million | Military records are good; the range comes from civilian famine and disease deaths in China, the Soviet Union and Southeast Asia |
| Mongol conquests | 1206–1368 | 20–40 million | No death records exist. Figures are inferred from regional population decline over more than a century and should be read as speculation |
| Taiping Rebellion | 1850–1866 | 10–30 million | Based on census differences across the period; most deaths were from famine and displacement rather than fighting |
| Qing conquest of the Ming | 1618–1683 | ≈25 million | A demographic estimate spanning sixty-five years of war, famine and epidemic |
| First World War | 1914–1918 | 15–19 million | Military dead of 8.5–9.5 million are reasonably firm; the civilian range is not |
| Second Sino-Japanese War | 1937–1945 | 15–20 million | Overlaps the Second World War row above and is usually counted inside it |
| An Lushan Rebellion | 755–763 | Unknown | The traditional figure of 36 million comes from a fall in registered taxpayers, not a count of deaths. Modern scholars regard it as unusable |
| Russian Civil War | 1917–1922 | 5–9 million | Combat deaths were a small fraction; famine and typhus were the rest |
| Thirty Years’ War | 1618–1648 | 4–8 million | Well studied for its period. Some German regions lost more than a third of their population |
| Napoleonic Wars | 1803–1815 | 3.5–6 million | Reconstructed from muster rolls and hospital returns; disease dominated |
| Dungan Revolt | 1862–1877 | 8–12 million (disputed) | Commonly quoted at this level, but the figure rests on regional census collapse and includes mass displacement |
| Chinese Civil War | 1927–1949 | 2–6 million | Estimates vary very widely and depend on whether the 1937–45 period is separated out |
| Second Congo War | 1998–2003 | 1.5–5.4 million | Survey-based excess-mortality estimates. The upper figure has been challenged over its pre-war baseline |
| Korean War | 1950–1953 | 2.5–3.5 million | Military records for the UN side are good; North Korean and Chinese civilian and military figures are not |
| Vietnam War | 1955–1975 | 1.3–3.8 million | Retrospective mortality surveys give the higher figures; record-based counts give the lower |
| American Civil War | 1861–1865 | 650,000–850,000 | A census-based reconstruction replaced the older estimate of 620,000 |
Sources: Clodfelter’s statistical encyclopedia for most rows; Platt on the Taiping; Durand on the Chinese census series; Coghlan et al. and the Human Security Report for the Congo dispute; Obermeyer et al. for Vietnam; Hacker for the American Civil War; Wilson for the Thirty Years’ War.
Sources
Clodfelter, M. (2017), Warfare and Armed Conflicts: A Statistical Encyclopedia, 4th ed. McFarlandWhite, M. (2011), Atrocitology: Humanity’s 100 Deadliest Achievements. CanongatePlatt, S. R. (2012), Autumn in the Heavenly Kingdom: China, the West, and the Epic Story of the Taiping Civil War. KnopfDurand, J. D. (1960), "The Population Statistics of China, A.D. 2–1953". Population Studies 13(3)Coghlan, B. et al. (2006), "Mortality in the Democratic Republic of Congo: a nationwide survey". The Lancet 367Human Security Report Project (2010), Human Security Report 2009/2010 — The Causes of Peace and the Shrinking Costs of War. Oxford University PressObermeyer, Z., Murray, C. J. L. & Gakidou, E. (2008), "Fifty years of violent war deaths from Vietnam to Bosnia". BMJ 336Hacker, J. D. (2011), "A Census-Based Count of the Civil War Dead". Civil War History 57(4)Wilson, P. H. (2009), Europe’s Tragedy: A History of the Thirty Years War. Allen LaneMay, T. (2012), The Mongol Conquests in World History. Reaktion BooksMitter, R. (2013), China’s War with Japan, 1937–1945: The Struggle for Survival. Allen Lane
The per-capita ranking is a different list
Absolute death tolls track population size as much as intensity. A war that kills two million people in China of 1860 and a war that kills two hundred thousand in Paraguay of 1865 are not comparable experiences, and the second may have been the more destructive by any measure a contemporary would recognise.
By proportion of population, the strongest candidates are the Paraguayan War of 1864–1870 and the Thirty Years’ War. For Paraguay, estimates of the share of the pre-war population that died range from under ten per cent to nearly seventy — one of the widest disagreements in the entire casualty literature, caused by an unreliable pre-war census and a post-war one taken under occupation. For central Europe in the seventeenth century, the overall German population decline is usually put at fifteen to twenty per cent, with some territories losing more than a third.
The Second World War killed about three per cent of the world’s population. Measured that way, it is not close to the top of the list, which is a fact worth holding alongside the absolute figure rather than instead of it.
Sources
Wilson, P. H. (2009), Europe’s Tragedy: A History of the Thirty Years War. Allen LaneClodfelter, M. (2017), Warfare and Armed Conflicts: A Statistical Encyclopedia, 4th ed. McFarlandWhite, M. (2011), Atrocitology: Humanity’s 100 Deadliest Achievements. Canongate
What this atlas can and cannot show you
The catalogue behind the map contains 740 war records from 1816 to 2014, coded by the Correlates of War project according to its own inclusion rules, including a battle-death threshold. It is a strong basis for asking when and where wars happened. It is not a casualty database, and this atlas deliberately does not draw arrow widths or marker sizes from death tolls.
It also stops. The inter-state, extra-state and non-state releases run to 2007 and the intra-state release to 2014. After those dates the map is empty because the data ends, not because the wars did. Absence from this map is never evidence of peace.
Most of the wars at the top of the list above are older than 1816 and therefore outside the catalogue entirely. They appear on this page as sourced editorial context, and they are not on the map.
Sources
Correlates of War — war datasets and codebooks ↗Sarkees, M. R. & Wayman, F. W. (2010), Resort to War: 1816–2007. CQ PressDixon, J. S. & Sarkees, M. R. (2016), A Guide to Intra-State Wars: An Examination of Civil, Regional, and Intercommunal Wars, 1816–2014. CQ Press
Related reading
- Second World War
World War II Map & Timeline: 1939–1945
A year-by-year map and timeline of the Second World War, from the invasion of Poland to the Japanese surrender, with the sourced dates and casualty ranges behind each phase.
- First World War
World War I Map & Timeline: 1914–1918
A map and timeline of the First World War: how the fronts formed in 1914, why the western front then barely moved for three years, and the sourced casualty ranges for the whole war.
- United States, 1861–1865
American Civil War Map & Timeline: 1861–1865
A map and timeline of the American Civil War: the eastern and western theatres, the blockade, the campaigns of each year, and why the death toll was revised upward from 620,000 to about 750,000.
- Korea, 1950–1953
Korean War Map & Timeline: 1950–1953
A map and timeline of the Korean War: the four changes of hands of Seoul, the Pusan Perimeter, Inchon, the Chinese intervention, and the two years of static fighting before the 1953 armistice.
- Catalogue
Every war on the map, 1816–2014
All 740 records in date order, with participants, regions and recorded combat periods.