Based on a new book, Cormac Ó Gráda argues that previous estimates of civilian deaths in the first and second world wars are almost certainly too low.
Berthold Brecht’s classic Mutter Courage is a play about the Thirty Years’ War. Its eponymous heroine, after much suffering in her struggle to survive, complained toward the end: “In general, it is fair to say that both victory and defeat cost us common people dearly”.
Mother Courage survived but several of her family did not; they were all victims of a long-drawn-out war in which it is reckoned one in four of Germany’s population, the great majority of them civilians, perished. Some died during events like the notorious Sack of Magdeburg in 1631, but most were the victims of disease and famine.
You might say that the Thirty Years’ War was exceptional, and that most wars spared civilians because, in the words of one rather bellicose recent US president, “war is an ugly thing, but we had rules in which we made sure that soldiers fought soldiers but did not victimise civilians”. Not so, and especially not so in the cases of WW1 and WW2.
The hidden victims
In a new book, The Hidden Victims: Civilian Casualties of the Two World Wars, I document and quantify the extent of this “not so”. My attempt to produce fair-minded and careful estimates of civilian suffering in its many forms during the two wars means that there is a lot of debating about numbers in the book. However, the book is not – and must not be – just about numbers.
There are tensions about who were civilians and who were victims. In all wars the line between combatants and civilians is unclear, but more so in total wars such as WW1 and WW2. And as for the victims, there is debate about how victimhood should be measured and described. In Germany and elsewhere, describing the civilian victims of aerial bombing and sexual violence and mass expulsion during and in the wake of WW2 as “victims” was controversial a few decades ago, though it is less so now.
Only in the 2000s had there been enough “healing” for the German Federal Government to allow the creation of a permanent exhibition on the mass exodus from the east within the Haus der Geschichte (House of History) in Berlin. And if such people are indeed considered victims – as I believe they must be – there is the issue of Relativierung or moral relativisation; and therefore also with adding together, like apples and oranges, victims from various causes and in widely different contexts.
Data on civilian victims in past wars will always be problematic. The task of providing credible numbers for WW1 and WW2 is daunting, but it has not deterred some in the past. My work builds on those earlier efforts, while not trying to hide the sensitivities and uncertainties about the numbers. My standpoint is that we can never arrive at the full truth, but we must try to provide reasoned estimates in order to counter partisan agendas to deny or exaggerate estimates for political reasons.
Civilian deaths
So what did I find? I found that, taking the two wars together, civilian deaths outnumbered military fatalities by over two to one. The civilian deaths totalled about 16 million during WW1 and somewhere between 44 and 48 million during WW2. During both wars hunger and famine were the single most important cause of death; during WW1 famines accounted for 12.5/13 million civilian deaths (so as many as four-fifths of the total), while during WW2 they accounted for a further 20.5/21.5 million (so somewhat less than half the total).
Aerial bombing counted for little during WW1, while it killed 1.3 to 1.6 million during WW2. Genocides were responsible for up to 2 million deaths during WW1, and 6.5 million and maybe more during WW2. In both wars, civilians, in what became the Soviet Union in 1917, suffered more than anywhere else. The greatest uncertainties surround civilian deaths in Asia and in the Soviet Union. Sometimes deaths, faute de mieux, can only be estimated as a residual.
The famines varied greatly in magnitude and causations. In some places, as in Bengal in 1943-44, in Vietnam in 1944-45 and in Moldova in 1946-47, harvest deficits greatly exacerbated the challenges caused by war and its aftermath. In several others, blockades were partly or mainly responsible: in Lebanon in 1915-16, in Germany in 1917-1919, in Leningrad in 1941-44, in Greece in 1941-43 and in the Netherlands in 1944-45.
Relatively well-off countries were not spared, as the cases of Greece and the Netherlands during WW2 testify. Broadly speaking, the causes of death varied according to how developed the economy was: in Leningrad, Greece, and the Netherlands during WW2, few died of infectious diseases; whereas in Bengal, in Java, Vietnam and China most of the deaths were due to such diseases. Famines killed more males than females almost everywhere; active-age males in Leningrad were an exception, but that is because most of them were elsewhere.
The Holocaust
I include victims of the Armenian genocide and the Jewish Holocaust because it seems unlikely that most of them would have perished had there not been a war. The literature about the Holocaust is already vast, yet looking at the numbers anew raised a few issues worth highlighting.
First, the data recall the great Primo Levi’s The Drowned and the Saved (1986). There he described “Why didn’t you run away… before the borders were closed?” as one of the questions most often put to “the saved”. Yet, in reality, the numbers highlight just how important migration was as a lifeline for German and Austrian Jews before the borders were sealed.
In Germany only 29 per cent of Jews living there in 1933 perished in the Holocaust; in Austria the percentage was 31. The migrations were subject to selection, of course. Males were more likely to escape than females, the young rather than the old, the rich than the poor. So one is left with the paradox that Jews at the epicentre of the Nazi empire were more likely to survive than those in many other places.
Second, the correlation between how anti-Semitic a country was and the proportion of the Jewish population murdered is not strong: a far higher proportion of the Jews of Greece and the Netherlands, countries not noted for their anti-Semitism, perished, than the Jews of Romania, where there was a lot of anti-Semitism at the time.
Third, the only escape open to some Jews in eastern Europe was to the Soviet Union. About 300,000 fled there and survived. They may not have loved the regime, but they tended to not become cold warriors, either: as one of them reminisced, “regardless of the Soviet regime, I owed a debt to that land”.
Survivors and international law
Besides those who perished, there were the millions who survived, wounded, bereaved, and traumatised. Some victims were traumatised for the rest of their lives, some were more resilient: but the true proportions are elusive. There were refugees; scores of millions of them. And there were victims of sexual violence, hardest of all to count.
One thing seems sure: the empirical basis for the much-recycled figure of two million rapes committed by Soviet troops toward the end of WW2 is very tenuous indeed. The true number was undoubtedly high, but no convincing estimate of it exists. Finally, there were the hundreds of thousands of women accused of la collaboration horizontale in Nazi occupied Europe during WW2. Guilty victims perhaps; attitudes towards them have certainly softened since 1944-1945.
Finally, my book speaks to the inability of international law, such as it was at the time, to protect innocent victims. The carnage of WW2 led to the creation of various institutions to protect civilians in future, an implicit admission that some of the acts carried out by even those on the high moral ground were wrong. I leave how effective those institutions have been, and how they might become more effective in future as issues for another day.
For more information, see Cormac Ó Gráda’s book, The Hidden Victims: Civilian Casualties of the Two World Wars (Princeton University Press, 2024). An accompanying lecture based on the book will be held at LSE on 20 February 2025.
Note: This article gives the views of the author, not the position of EUROPP – European Politics and Policy or the London School of Economics. Featured image credit: Davor Flam / Shutterstock.com
Discussion about this post