Military history


Fourteen million is the approximate number of people killed by purposeful policies of mass murder implemented by Nazi Germany and the Soviet Union in the bloodlands. I define the bloodlands as territories subject to both German and Soviet police power and associated mass killing polices at some point between 1933 and 1945. They correspond closely to the places were the Germans killed Jews between 1941 and 1945. In the east, more or less of Soviet Russia might have been included; but the existing line allows the consideration of the main German killing sites of the war as well as the western Soviet lands disproportionately struck by earlier Soviet terror. Though I discuss the western lands of today’s Poland, which belonged until 1945 to Germany, I do not include them in the bloodlands. This is to respect the difference between mass killing and ethnic cleansing. Hungary might arguably have been included, since it was occupied briefly by the Germans late in the war, after years as a German ally, and then occupied by the Soviets. After Polish and Soviet Jews, Hungarian Jews were the third-largest group of victims of the Holocaust. Romania, too, would have a kind of claim to belong to the bloodlands, since many of its Jews were killed and the country was occupied at the end of the war by the Soviet Union. Romania, however, was also a German ally rather than a victim of German aggression, and the murder of Romanian Jews was a Romanian rather than a German policy; this is a related but different history. Yugoslav citizens suffered many of the fates described here, including the Holocaust and mass reprisals; but the Jewish population of Yugoslavia was very small, and Yugoslavia was not occupied by the Soviet Union.

These matters of political geography are debatable on the margin; what is not is the existence of a zone in Europe where Soviet and German power overlapped and where the tremendous majority of the deliberate killing of both regimes took place. It is indisputable, to state the point differently, that the contiguous area from central Poland to western Russia where Germans killed Jews covers the regions where all of the other major German and Soviet policies of mass killing had already taken place or were concurrently taking place—if not completely, then in very significant part. The purposeful starvation of Ukraine took place within the zone of the Holocaust. The purposeful starvation of Soviet prisoners of war took place within the zone of the Holocaust. Most Soviet and German shootings of Polish elites took place within the zone of the Holocaust. Most German “reprisal actions” took place within the zone of the Holocaust. A disproportionate amount of the shooting of the Stalinist Great Terror took place within the zone of the Holocaust.

I use the term Molotov-Ribbentrop line to signify an important boundary running north to south through the bloodlands. This line (which appears on some of the maps) is the German-Soviet border as agreed in September 1939 after the joint invasion of Poland. It was significant for Polish citizens, since it marked the division between German and Soviet occupation policies. This line took on another meaning after the Germans betrayed their allies and invaded the Soviet Union in 1941. To its west, Germans were holding Jews in ghettos; to its east, Germans began to shoot Jews in very large numbers. The Holocaust began east of the Molotov-Ribbentrop line with shooting actions, and then shifted west of the Molotov-Ribbentrop line, where most victims were gassed.

In the historical literature on the Holocaust, people east of the Molotov-Ribbentrop line are often referred to as “Soviet Jews,” and people to the west as “Polish Jews.” This is an inaccurate description of the people in question: more of the people killed east of the Molotov-Ribbentrop line had Polish than Soviet citizenship in 1939, when the war began. Referring to these people as “Soviet Jews” also tends to confirm an account of the war in which the Soviet invasion and occupation of its western neighbors are marginalized or overlooked entirely. If these people were “Soviet Jews,” then their homeland must have been the Soviet Union, and the war must have begun with the German invasion of the Soviet Union. In fact, the war began with the German-Soviet alliance that destroyed Poland and left such Jews in an expanded Soviet Union. The use of the term Molotov-Ribbentrop line, though it may seem awkward at first, allows us to see a very special zone of Europe, whose peoples suffered three rounds of occupation during the Second World War: first Soviet, then German, then Soviet again.

On the wall of a Gestapo prison in Warsaw, a Polish prisoner wrote: “It’s easy to talk about Poland. It’s harder to work for her. Even harder to die. And hardest of all to suffer.” With a few exceptions, this is a study of the dying rather than the suffering. Its subject is policies that were meant to kill, and the people who were their victims. In a deliberate mass killing operation, mass death is the desired goal of policy. It is an end in itself or a means to some other end. The count of fourteen million is not a complete reckoning of all of the death that German and Soviet power brought to the region. It is an estimate of the number of people killed in deliberate policies of mass murder.

I therefore generally exclude from the count the people who died of exertion or disease or malnutrition in concentration camps or during deportations, evacuations, or flight from armies. I also exclude the people who died as forced laborers. I am not counting people who died of hunger as a result of wartime shortfalls, or civilians who died in bombings or as a result of other acts of war. I am not counting soldiers who died on the fields of battle of the Second World War. In the course of the book I do discuss camps and deportations and battles, and provide figures of those killed. These are not, however, included in the final figure of fourteen million. I also exclude acts of violence carried out by third parties that were clearly a result of German or Soviet occupation, but not German or Soviet policy. Sometimes these brought very significant numbers of deaths, as with the Romanian murder of Jews (some three hundred thousand) or the Ukrainian nationalist ethnic cleansing of Poles (at least fifty thousand).

This is a book about deliberate mass murder more than a book about abuse. It is a book about civilians (and prisoners of war) rather than a book about soldiers on active duty. By all of these distinctions and exclusions I do not mean to suggest that such people were not victims, direct or indirect, of the Nazi and Soviet systems. I do not wish to minimize the horror of German and Soviet concentration camps, or the murderous character of ethnic cleansing, or the repressive nature of forced labor, or the awful total of war death. I wish to test the proposition that deliberate and direct mass murder by these two regimes in the bloodlands is a distinct phenomenon worthy of separate treatment, by writing a history whose subject is the purposeful murder of fourteen million people by two regimes over a short time and in certain parts of Europe.

Fourteen million, after all, is a very large number. It exceeds by more than ten million the number of people who died in all of the Soviet and German concentration camps (as opposed to the death facilities) taken together over the entire history of both the Soviet Union and Nazi Germany. If current standard estimates of military losses are correct, it exceeds by more than two million the number of German and Soviet soldiers, taken together, killed on the battlefield in the Second World War (counting starved and executed prisoners of war as victims of a policy of mass murder rather than as military casualties). It exceeds by more than thirteen million the number of American and British casualties, taken together, of the Second World War. It also exceeds by more than thirteen million all of the American battlefield losses in all of the foreign wars that the United States has ever fought.

The count of fourteen million mortal victims of deliberate killing policies in the bloodlands is the sum of the following approximate figures, defended in the text and notes: 3.3 million Soviet citizens (mostly Ukrainians) deliberately starved by their own government in Soviet Ukraine in 1932-1933; three hundred thousand Soviet citizens (mostly Poles and Ukrainians) shot by their own government in the western USSR among the roughly seven hundred thousand victims of the Great Terror of 1937-1938; two hundred thousand Polish citizens (mostly Poles) shot by German and Soviet forces in occupied Poland in 1939-1941; 4.2 million Soviet citizens (largely Russians, Belarusians, and Ukrainians) starved by the German occupiers in 1941-1944; 5.4 million Jews (most of them Polish or Soviet citizens) gassed or shot by the Germans in 1941-1944; and seven hundred thousand civilians (mostly Belarusians and Poles) shot by the Germans in “reprisals” chiefly in Belarus and Warsaw in 1941-1944.

In general, these numbers are sums of counts made by the Germans or the Soviets themselves, complemented by other sources, rather than statistical estimates of losses based upon censuses. Accordingly, my counts are often lower (even if stupefyingly high) than others in the literature. The major case where I do rely upon estimates is the famine in Soviet Ukraine, where data are simply insufficient for a count, and where I present a total figure on the basis of a number of demographic calculations and contemporary estimates. Again, my reckoning is on the conservative side.

With such a subject, care must be taken with the use of terms, and with their definitions. There is a notable difference, usually not noted, between Final Solution and Holocaust. The first was the general term that the Nazis used for their intention to eliminate the Jews from Europe. For much of the time that it was used, it indicated one of four deportation plans, all of which were eventually discarded. At some point in the second half of 1941, Hitler endorsed mass killing as the method by which Jews were to be eliminated from Europe, and made this widely known that December. At that point the Final Solution was understood to mean the murder of all Jews. The term Holocaust was introduced after the war and, by the 1990s, was generally (although by no means always) understood to mean the mass murder of the Jews by the Germans.

In this book the term Holocaust signifies the final version of the Final Solution, the German policy to eliminate the Jews of Europe by murdering them. Although Hitler certainly wished to remove the Jews from Europe in a Final Solution earlier, the Holocaust on this definition begins in summer 1941, with the shooting of Jewish women and children in the occupied Soviet Union. The term Holocaust is sometimes used in two other ways: to mean all German killing policies during the war, or to mean all oppression of Jews by the Nazi regime. In this book, Holocaust means the murder of the Jews in Europe, as carried out by the Germans by guns and gas between 1941 and 1945.

I refrain from using the term Holodomor for Stalin’s deliberate starvation of Soviet Ukraine, not because the term is less precise than Holocaust but simply because it is unfamiliar to almost all readers of English. I use Great Terror to mean the Soviet mass shooting and deportation actions of 1937 and 1938, of which the most important were the kulak and national operations.

I prefer mass killing to genocide for a number of reasons. The term genocide was coined by the Polish-Jewish international lawyer Rafał Lemkin in 1943. Through a miracle of energy and persistence, he managed to encode it in international law. By the terms of the Convention on the Prevention and Punishment of the Crime of Genocide, adopted by the United Nations General Assembly in 1948, genocide involves “acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such.” It lists five ways in which genocide is committed: by “killing members of the group”; “causing serious bodily or mental harm to members of the group”; “deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part”; “imposing measures intended to prevent births within the group”; and “forcibly transferring children of this group to another group.” This legal instrument has allowed for prosecutions, if only recently. As a guide to historical and moral interpretation, however, the term genocide has limitations.

The term genocide gives rise to inevitable and intractable controversies. It relies upon the intention of the perpetrator in two places: “intent to destroy” a certain group “as such.” It can be argued that policies of mass killing were not genocide, because rulers had some other “intent,” or because they intended to kill someone, but not a specified group “as such.”

Though the term genocide in fact has wide application, it is often thought to refer only to the Holocaust. People who associate themselves with victims will wish to define past crimes as genocide, thinking that this will lead to recognition of the kind awarded to the Holocaust. Meanwhile, people associated with states that perpetrated a genocide resist the term with great energy, because they believe that its acceptance would be tantamount to acceptance of a role in the Holocaust. Thus, for example, Turkish governments resist the classification as genocide of the mass killing of a million or more Armenians during the First World War.

A final problem arises from a known political modification of the definition. The Soviets made sure that the term genocide, contrary to Lemkin’s intentions, excluded political and economic groups. Thus the famine in Soviet Ukraine can be presented as somehow less genocidal, because it targeted a class, kulaks, as well as a nation, Ukrainians. Lemkin himself regarded the Ukrainian famine as genocide. But since the authors of the policy of starvation edited his definition, this has been controversial. It is remarkable that we have the legal instrument of genocide; nevertheless, one must not forget that this particular murder statute was co-drafted by some of the murderers. Or, to put the matter less moralistically: all laws arise within and reflect a certain political setting. It is not always desirable to export the politics of that moment into a history of another.

In the end, historians who discuss genocide find themselves answering the question as to whether a given event qualifies, and so classifying rather than explaining. The discussions take on a semantic or legalistic or political form. In each of the cases discussed in this book, the question “Was it genocide?” can be answered: yes, it was. But this does not get us far.

The peoples of the bloodlands functioned in a world of daunting linguistic complexity. Most of the victims described here knew or had regular contact with two or more languages, and many of them were bilingual or trilingual. Broader histories of Europe involve problems of transliteration, since words from languages written in non-Roman characters must be rendered in Roman characters in English. The problem arises not with languages such as French, German, Polish, or Czech, which are written in Roman characters with more or less familiar diacritical marks. But Yiddish and Hebrew are written in Hebrew characters, and Belarusian, Russian, and Ukrainian are written in Cyrillic alphabets.

For each of these languages there are competing systems of transliteration, each with its advantages and disadvantages. To make matters worse, many of the people and places discussed in this book were known by different names in different languages written with different characters. So there might be a dozen legitimate ways, for example, to spell the name of a Jew from Ukraine in English. For many readers, the people and places discussed in this book will be unfamiliar enough without the burden of the more elaborate solutions to the transliteration problem. The risk with too much precision is that people and places become more rather than less exotic.

I have generally given names in simpler and more familiar forms rather than those that correspond perfectly to transliteration. I spell Russian surnames ending with a double ii just that way, partly so that readers can tell the difference between Russian surnames (ending with ii), Ukrainian surnames (ending with yi), and Polish surnames (ending with i). I usually give city names in an English form if one is familiar, so Warsaw rather than Warszawa and Kiev rather than Kyïv. All of the cities were spelled differently by different inhabitants at any given time—and at different times fell under different rulers with different official topographies. I tend to favor current English spellings, although I make exceptions when it would seem absurd not to do so. Cities in the Soviet Union had a tendency to change names; I use the names under which they were known at the time. Thus famine and terror strike Stalino (not Donetsk) in the 1930s and the German Sixth Army is destroyed at Stalingrad (not Volgograd) in 1942. I try to refer to cities by the same name throughout the book, although in certain cases where boundaries moved and populations were transferred (Lwów/Lviv, Wilno/Vilnius) this would have been too distorting. In citations I give the name of the author as it appears in the cited work, even though this means that some authors’ names will be spelled different ways in different citations. In the notes and bibliography I transliterate more precisely than in the main text, following simplified versions of the Library of Congress guidelines. These are all matters of judgment; readers with experience in these problems will understand the impossibility of avoiding certain compromises. Translations, except where otherwise noted, are my own.

If you find an error or have any questions, please email us at Thank you!