27 million lives lost Meduza takes a closer look at the Soviet Union’s official death toll in World War II
During talks with China’s Xi Jinping on the eve of Victory Day, Vladimir Putin once again underscored the scale of Soviet losses in World War II. “The Soviet Union gave 27 million lives,” he said. “It laid them at the altar of the fatherland and the altar of victory.” Though this is the standard estimate used by Russian officials, the number 27 million often comes up in academic works, too. To this day, the question of the “cost of war” remains extremely painful, politically charged, and quite confusing from a scientific standpoint. That said, even a crash course in the principles on which this estimate is based can give you a much deeper understanding of what defeating Nazi Germany cost the people of the USSR. Meduza breaks down the origins of the oft-cited Soviet death toll of 27 million people.
How are war deaths classified, and why do these categories matter?
The study of human losses in war falls at the intersection of military history and demography. As a result, researchers interpret the term “losses” differently, leading to disagreements about how to collect and measure data on conflict deaths.
Military historians typically look at “irrecoverable losses,” a concept used to assess the strength of the different parties involved at a particular stage in a given war. Irrecoverable losses include not only soldiers killed on the battlefield, but also those who died from injuries, went missing in action, were captured as prisoners of war, were discharged due to injury, or deserted. Irrecoverable losses can also be divided into combat and non-combat losses, with the latter including those who succumbed to illnesses or injuries, or died by suicide. However, this classification does not include civilians, even if they were killed during a battle. As such, irrecoverable losses don’t offer a complete picture of the scale of the damage caused by a war.
To estimate the real “cost of war” in terms of human lives, researchers Bethany Lacine and Nils Petter Gleditsch suggest breaking down the statistics into three distinct categories: combatant deaths, battle deaths (including both soldiers and civilians), and war deaths (that is, all deaths caused by the war).
In the case of World War II, only “combatant deaths” and “war deaths” can be estimated with any degree of accuracy. (Estimating “battle deaths” requires distinguishing between combatant and non-combatant losses, while also taking into account relevant civilian casualties — an almost impossible task for such a war.)
Tallying Soviet military losses and total war deaths relies on two fundamentally different methods: (1) the number of Soviet soldiers who died can be determined from wartime military records, including casualty reports and other documents from the Soviet Defense Ministry’s archives; and (2) civilian and irregular combatant losses (such as partisans), who aren’t included in personnel lists, can only be estimated by comparing pre- and post-war population sizes.
Despite the extensive academic literature on Soviet losses in World War II, tallying the number of deaths in these two categories — military and total losses — ultimately relies on two key primary sources: archival military records and demographic population estimates.
How do we know the number of Soviet military casualties?
The Soviet Union began tallying its war losses in 1943, setting up an Extraordinary State Commission tasked with investigating Nazi war crimes and collecting data on civilian deaths in the occupied territories. The commission’s findings were presented at the Nuremberg trials, but were not widely discussed in the USSR. Moreover, these findings did not include actual military casualties. Until the USSR’s collapse, only the country’s top leadership could speak publicly about the Red Army’s losses.
In November 1941, Soviet leader Joseph Stalin announced that 350,000 Red Army soldiers had been killed in the first four months of the war. Though it was a significant underestimate, Stalin’s announcement sparked a whole series of statements about Soviet military losses from the country’s top leaders, for political and propaganda purposes. Today, these numbers are only of historical significance, since there was nothing to back them up.
The first widely accepted estimate of Soviet military losses backed by records was made during the Perestroika years. This study, based on declassified archival documents from the Soviet Defense Ministry, was conducted by a group of military historians under the leadership of Colonel General Grigory Krivosheev. Their findings were published in the 1993 book Soviet Casualties and Combat Losses in the Twentieth Century, which became the basis for all future debates about Soviet military losses in World War II. Though some dispute the data, the book’s findings continue to be cited by Russia’s Federal State Statistics Service and Western historians alike.
According to Krivosheev’s team, the Soviet army’s total losses in World War II amounted to more than 8.6 million people. (Though the primary documents originally pointed to some 11.4 million military casualties, the researchers had to subtract 2.77 million people to account for returned prisoners of war and those who were called up multiple times.)
The main drawback of this estimate is that it lacks an independent baseline for comparison. It was calculated by military historians based on Defense Ministry records and speaks directly to that same agency’s effectiveness during the war. This raises concerns about objectivity, though the very nature of military losses means alternate methods would rely on the same records.
Where does the total death toll of 27 million come from?
The number 27 million was first mentioned by Mikhail Gorbachev 35 years ago, in a May 8, 1990, report dedicated to the 45th anniversary of the Soviet Union’s victory over Nazi Germany. To be precise, the death toll he gave was 26.6 million people, and it came from another official study. This time, the team of researchers included representatives from the Soviet Union’s State Statistics Committee, the Moscow State University Academy of Sciences, and the Soviet Defense Ministry. Their findings were later published in a book, The Population of the Soviet Union: 1922–1991.
How did they get to this number? To put it simply, the researchers came up with their estimate by subtracting the Soviet Union’s post-war population from its pre-war population, and then adding the estimated number of excess deaths among children born during the war years.
This is similar to the approach used for calculating excess mortality, which allows us to estimate the losses from the Russia–Ukraine war, for example.
However, there’s a key difference: excess mortality is calculated using existing data sets, revealing sharp spikes in the number of deaths. Therefore, to take this approach, you need access to well-maintained death and birth records. In the case of World War II, when huge stretches of Soviet territory came under occupation and millions of people were displaced, such records are unavailable. As such, the research team working under Gorbachev employed a simpler demographic balance method. Here’s how they did the math:
- The research team estimated the USSR’s population as of mid-1941 at 196.7 million people (notably, this estimate included the residents of territories that didn’t come under Soviet control until 1945).
- To account for those who died of battlefield injuries and prisoners of war who returned home after May 8–9, the demographers used the end of 1945 as their end date, estimating the USSR’s total post-war population at 170.5 million people.
- Of these 170.5 million people, only 159.4 million were born before the USSR entered the war in mid-1941.
- So, 196.7 million – 159.4 million = 37.2 million people. This figure accounts for those born before the start of the war who “disappeared” between mid-1941 and the end of 1945, including people who were killed in action or died of injuries, hunger, wartime hardships, or even natural causes. Notably, it also includes those who fled abroad during the war and did not return home.
- To estimate how many of those 37.2 million died as a direct result of the war, the researchers subtracted the expected number of deaths that would have occurred in peacetime (this is where the standard method for calculating excess mortality comes in). According to their calculations, if mortality rates had remained the same as in 1940, there would have been 11.9 million natural deaths during the war years.
- So, 37.2 million – 11.9 million = 25.36 million excess deaths of people born before mid-1941.
- Calculating the number of deaths among children born during the war years was a difficult task: the Soviet Union hardly kept birth records during this period, and many infant deaths went unrecorded. To account for this fact, the researchers turned to retrospective demographic surveys of women conducted in the 1960s, which included data on birth rates and infant mortality during the war years.
- This data revealed 16.53 million births during the war years, as well as 1.3 million excess infant deaths (calculated based on 1940 infant mortality rates).
- And so, the researchers’ calculations produced the following estimate: 25.36 million + 1.3 million = 26.6 million war deaths (a figure that also includes a small number of citizens who emigrated from the USSR during the war).
The majority of these deaths, more than 20 million, were among men. The sharp increase in male mortality affected those aged 15 to 50. For women, the war increased mortality in a narrower and younger group — those between the ages of 15 and 40.
The group that suffered the most deaths during the war was men between 24 and 35 years old. A Soviet man who fell in this age group as of 1941 had, on average, a 61 percent chance of surviving the war. In other words, more than a third of this demographic died. Moreover, this is the average for the entire USSR, meaning it takes into account regions that didn’t come under occupation or see military action.
What are the limitations of this calculation?
The number 26.6 million is a statistical estimate, not an exact figure. We can’t say for certain that each individual death was a direct consequence of the war, and it doesn’t discern the different causes of increased wartime mortality (lumping together deaths from shelling, disease, famine, psychological trauma, and even domestic repressions). Moreover, this estimate fails to account for the complex wartime migration processes that took place between mid-1941 and late 1945.
Some researchers even argue that it’s not entirely correct to equate excess deaths in wartime with deaths caused by the war, but this is more a question of terminology and consensus than of actual calculations.
The main limitation of this estimate is the quality of the census data on which the 1941 and 1945 population figures were based. The last pre-war census took place in 1939, right before the Molotov-Ribbentrop Pact vastly expanded the USSR’s borders and added no less than 20 million people to its population. The officials in charge of the 1939 census were also under unprecedented ideological pressure; they were expected to show that state socialism had succeeded in raising both birth rates and life expectancy, which inevitably would have distorted the census results. (After the 1937 census delivered lower population figures than anticipated, the organizers were punished and the country’s chief statistician was shot.) The first post-war census, meanwhile, was conducted in 1959 — almost 15 years after the war ended.
Nevertheless, the research team managed to eliminate artificial distortions from the census results, check the data by indirect means, and reconstruct the Soviet Union’s pre-war and post-war population sizes with a fair amount of accuracy. The bulk of their work was devoted to resolving these very problems. Their findings were later verified by scholars Michael Ellman and Sergei Maksudov, historian Mark Harrison, and other Western researchers who confirmed the accuracy of their approach and the overall estimate of 26–27 million war dead.
Though the Soviet Union’s estimated losses in World War II remain the subject of revisionism, no alternative estimates proposed in the last 35 years have achieved widespread expert consensus. As such, 27 million can be considered a reasonably reliable estimate, but we must be clear about whom it includes.
Meduza’s Razbor (“Explainers”) team