How Many People Died in World War 1?
The best modern estimates for the First World War, the reason the civilian figure is so uncertain, and where the 1918 flu fits
World War I killed an estimated 15 to 22 million people in total, made up of roughly 9 to 11 million military deaths and roughly 6 to 13 million civilian deaths, and the range is that wide because the military dead were counted by armies while the civilian dead largely were not. Careful scholarly tallies of the military figure converge more tightly than the headline band suggests: the reference work 1914-1918-online arrives at about 10.06 million military deaths across all combatants, split into roughly 5.65 million for the Allies and 4.41 million for the Central Powers. Estimates that quote 8.5 to 10 million military dead are drawing on the same evidence with tighter inclusion rules.
The civilian figure is where the uncertainty lives. It ranges from about 6 million to about 13 million, a spread of more than double, because it has to absorb the Armenian genocide, the Allied blockade of Germany, famine in Russia, the Ottoman Empire, and Persia, and epidemic disease across a continent whose public health systems had collapsed. The author of the 1914-1918-online survey is blunt that the civilian figures are largely suppositions rather than counts.
One thing must be stated separately and plainly, because it is the most common source of inflated numbers. The 1918 influenza pandemic is not the same event as the war. It killed an estimated 25 to 50 million people worldwide between 1918 and 1920, with published estimates running from 17 million to 100 million, and it is normally treated as a separate catastrophe. Some national casualty tables do fold flu deaths among soldiers and civilians into their war totals, which is one reason the war's own upper bound climbs toward 22 million, but the pandemic's global toll should never simply be added to a war total.
Military deaths: the firmest figure on the page
Armies in 1914 kept records. Soldiers were enlisted, paid, posted, and, in most of the belligerent forces, formally reported when they were killed or went missing. That is why the military death figure holds together across independent tallies while the civilian figure does not.
The number usually given is 9 to 11 million, with the detailed reconstruction in 1914-1918-online reaching about 10.06 million: roughly 5.65 million Allied and 4.41 million Central Powers military deaths. Narrower statements of 8.5 to 10 million are common in general reference works and generally reflect a stricter definition, for example counting only deaths during the war years themselves and excluding those who died later of wounds.
Not all of those men died in battle. Disease accounted for roughly a third of total military deaths, which is easy to forget in a war remembered for artillery. Typhus, dysentery, and in 1918 influenza killed on a scale that trench warfare narratives rarely convey. Deaths in captivity add further tens of thousands, and Russian and Ottoman military records are the least complete of the major powers, which is where most of the remaining slack in the range sits.
Civilian deaths: why the range doubles
The civilian estimate of 6 to 13 million is not one uncertain number, it is several. The Russian Empire's civilian dead are given anywhere from about 1 million to 2 million depending on the estimate, and the Ottoman Empire's civilian toll is the most disputed of all, with published figures differing by millions according to how the Armenian genocide is counted. Germany's civilian deaths are estimated at roughly 300,000 to 760,000, mostly attributed to the effects of the Allied blockade on food supply. Serbia, which suffered proportionally more than any other combatant, is given a civilian range of about 450,000 to 800,000.
Each of those bands rests on a demographic reconstruction rather than a count, and the reconstructions disagree about what to include. Deaths from malnutrition are the hardest case: a blockade or a requisition does not kill people directly, it kills them through a weakened immune system months later, and no death certificate says so.
There is also a category problem at the edge of the war. The Armenian genocide, the Russian famine, and the collapse of the Ottoman food economy were all set in motion by the war and all continued past the armistice. Where a tally draws its end date changes the civilian figure by millions, which is the single largest reason the published totals for the war run from 15 million to 22 million rather than clustering.
The 1918 influenza pandemic, counted separately
The influenza pandemic of 1918 to 1920 killed an estimated 25 to 50 million people worldwide, with the full spread of published estimates running from about 17 million to about 100 million. It is one of the deadliest events in recorded history in its own right, and on most estimates it killed more people than the war it overlapped.
The pandemic and the war are entangled but they are not the same thing. Mass troop movements, crowded camps, and the censorship regimes of the belligerent states plainly accelerated the spread, and the pandemic's popular name comes from the fact that neutral Spain reported it honestly while the combatant powers suppressed the news. But the virus reached neutral countries, colonies, and remote regions with no connection to the fighting, and India alone lost millions.
The practical rule we use is this. Do not add the pandemic's global toll to the war's. Do check, when reading a country-level table, whether influenza deaths among soldiers and civilians have been folded into that country's war figure, because some tables do include them and some do not. Historians disagree openly about whether they should, and that disagreement is one of the named reasons the war's civilian totals cannot be pinned down.
Why the First World War's numbers differ from the Second's
In the First World War, military deaths were the majority. In the Second, civilian deaths outnumbered military deaths by roughly two to one. That inversion is the clearest single statistic distinguishing the two conflicts, and it reflects a real change in how the wars were fought: strategic bombing of cities, deliberate starvation as policy, and industrialised genocide were features of the second war on a scale the first did not reach, terrible as the Armenian genocide and the blockade were.
It also means the First World War's total is better known in relative terms, because the well-recorded component is the larger one. A range of 15 to 22 million for the First World War is a band of about 7 million around a total of roughly 18 million. A range of 70 to 85 million for the Second is a band of 15 million.
The proportional losses tell yet another story. Serbia is usually reckoned to have lost the largest share of its population of any combatant, on the order of a sixth or more. France and Germany each lost a very large fraction of their young men, which shaped both countries' politics for a generation. The United States, which entered in 1917, lost around 116,000 military dead, a figure often quoted precisely because American records for the period are unusually complete.
Using these figures honestly
If you need one number, use 15 to 22 million and call it an estimate, or say roughly 17 to 20 million if you want a tighter working figure and are willing to note that you have excluded the widest civilian estimates. For military deaths alone, 9 to 11 million is the safe statement and about 10 million is the best single approximation.
Avoid the two most common errors. The first is quoting a total that silently includes the global influenza toll, which produces figures in the 50 million range for a war that killed nothing like that many. The second is quoting a precise total to the nearest thousand for the civilian dead, which no source can support.
It is worth saying why this matters beyond pedantry. The First World War's death toll is a political object. It has been raised to justify reparations, lowered to minimise responsibility, and reshaped by the successor states that inherited the Ottoman and Russian records. Giving the range, and saying who produced each end of it, is the only version of the answer that survives contact with the sources.
Frequently asked questions
How many people died in World War 1?
An estimated 15 to 22 million people in total, comprising roughly 9 to 11 million military deaths and roughly 6 to 13 million civilian deaths. The detailed reconstruction in 1914-1918-online gives about 10.06 million military deaths, split into about 5.65 million Allied and 4.41 million Central Powers.
Why do World War 1 death toll estimates differ so much?
Because the civilian figure is an estimate rather than a count. It has to absorb the Armenian genocide, famine in Russia and the Ottoman Empire, deaths caused by the Allied blockade of Germany, and epidemic disease, and different tallies disagree about which of those to include and where to end the count. The military figure is far more stable.
Are the 1918 flu deaths included in the World War 1 death toll?
Usually not, and they should be stated separately. The 1918 to 1920 influenza pandemic killed an estimated 25 to 50 million people worldwide, with published estimates ranging from 17 million to 100 million. Some national casualty tables do fold flu deaths among soldiers and civilians into their war totals, so it is worth checking any table you use, but the pandemic's global toll should never simply be added to a war total.
How many soldiers died in World War 1?
Roughly 9 to 11 million, with about 10 million the best single approximation. Around a third of those deaths were caused by disease rather than combat, including typhus, dysentery, and the 1918 influenza.
Which country suffered the worst losses in World War 1?
In absolute military deaths, Germany and Russia lost the most. Proportionally, Serbia is usually reckoned to have lost the largest share of its population of any combatant. The Ottoman Empire's civilian figure is the most uncertain of any country: published estimates differ by millions, and the difference is largely whether and how the Armenian genocide is counted.
Further reading
1. The First World War — John Keegan (1998)
The First World War was a tragic and unnecessary conflict whose course was driven less by grand strategy than by the failure of commanders to adapt to a technology of defence that outran their ability to manoeuvre.
The standard one-volume military history in English, and it is unusually careful about the difference between reported casualties and actual dead.
Pick this if: Readers who want one clear account of the whole war. (Level: Intermediate)
2. The Sleepwalkers: How Europe Went to War in 1914 — Christopher Clark (2013)
The powers of 1914 did not stumble into war blindly nor did one of them plan it; they were sleepwalkers, watchful but unseeing, each taking rational-seeming steps toward a catastrophe none intended.
The best modern account of how the war began, which is the necessary context for why so many people died in it.
Pick this if: Readers who want the causes rather than the campaigns. (Level: Advanced)
3. The Guns of August — Barbara W. Tuchman (1962)
The first month of the war set the pattern for the following four years, and the decisions of August 1914 locked Europe into a conflict its leaders could no longer stop.
Still the most vivid narrative of how the killing started, and the book that shaped how generations understand the war's opening.
Pick this if: Readers who want narrative history at its best. (Level: Introductory)
Prefer listening? History is Audible’s strongest catalog — an annual Premium Plus membership is the best value, or compare the monthly plans.
Related guides
Sources consulted
- Wikipedia, 'World War I casualties'
- 1914-1918-online International Encyclopedia of the First World War, 'War Losses'
- Wikipedia, '1918 influenza pandemic (Spanish flu)'
- Open Library record, John Keegan, The First World War
Researched with AI and verified against mainstream scholarship and the cited sources on this page. See our methodology and corrections policy.
As an Amazon Associate, AskHistoryAI earns from qualifying purchases. Books are chosen editorially.