HomeDocuments on SitePersonalDemocratic PeaceDemocide20th C. DemocideMegamurderersLesser MurderersWhy DemocideDimensionsConflictMethodsTheoryPolicyLinks PHOTOS OF DEMOCIDEGalleries

Other Democide Related Documents On This Site


"Democide vs genocide. Which is what?"

"War isn't this century's biggest killer"

"How many did communist regimes murder?"


"Democide in totalitarian states: mortacracies and megamurderers"

Graduate Syllabus on Repression and Democide


"Power kills: genocide and mass murder"

"Power predicts democide"


Lethal Politics

China's Bloody Century

Democide: Nazi Genocide and Mass Murder

Death By Government

Statistics of Democide (entire)



By R.J. Rummel

A massive amount of research has been done on the Holocaust, the most extensive, best organized, thorough, and unlimited case of genocide in the modern age. The second most studied genocide has been that of the Armenians in Turkey. But little research has been done on other genocides per se, and virtually no systematic historical or comparative research has been done on genocide in general. There are collections of studies on different genocides.1 There are exemplary lists of genocides.2 But until the publication of my Death By Government 3 and availability of its auxiliary work, Statistics of Democide,4 the field has been lacking a comprehensive collection of all the genocides and mass murders to have occurred.5 Moreover, there are few attempts to compare the occurrence of genocide to other forms of mass murder or to fit genocide within a larger context of mass killing. Finally, there has been virtually no systematic attempt to assess the underlying conditions and causes of genocide.6 Here I will present and describe results as yet unpublished that may help fill this void.7 And in the process I will try to save the idea of genocide to mean that for which we badly need an exclusive concept--the murder of individuals by virtue of their ethnicity, race, religion, language, or nationality.

At its the core there is no doubt as to what genocide is--all recognize that the Nazi program to kill all Jews was genocide. Nor is there any doubt that the Bosnian Serb massacre of Bosnian Moslems and vice versa, or the slaughter of Hutu by Tutsi and Tutsi by Hutu in Rwanda was genocide. But was genocide also the recent massacre of helpless villagers in the Sudan by government forces fighting a rebellion, the 1965-1966 Indonesian army purge of communists, the 1948 assassination of political opponents by the Nationalist government on Formosa, the 1949-1953 "land-reform" executions of landlords in communist China, or the 1975-1980 rapid death of inmates in Vietnamese re-education camps? What about non-killing which has been called genocide, such as the absorption of one culture by another, the disease spread to natives by contact with colonists, the forced deportation of a people, or African slavery?

Let me remind the reader that in international conventions and the professional literature, genocide was initially defined in part as the intentional destruction of a people because of their race, religion, ethnicity, or some other indelible group membership. As now well known, the origin of the concept is the 1944 work by Raphael Lemkin on Axis Rule in Occupied Europe:

New conceptions require new terms. By "genocide" we mean the destruction of a nation or of an ethnic group. This new word, coined by the author to denote an old practice in its modern development, is made from the ancient Greek word genos (race, tribe) and the Latin cide (killing), thus corresponding in its formation to such words a tyrannicide, homicide [sic], infanticide, etc. Generally speaking, genocide does not necessarily mean the immediate destruction of a nation, except when accomplished by mass killings of all members of a nation. It is intended rather to signify a coordinated plan of different actions aiming at the destruction of essential foundations of the life of national groups, with the aim of annihilating the groups themselves. The objectives of such a plan would be disintegration of the political and social institutions, of culture, language, national feelings, religion, and the economic existence of national groups, and the destruction of the personal security, liberty, health, dignity, and even the lives of the individuals belonging to such groups. Genocide is directed against the national group as an entity, and the actions involved are directed against the individuals, not in their individual capacity, but as members of the national group.8

Of course this was written at the height of the Jewish Holocaust, a clear case of a regime trying to exterminate a whole group, its intellectual contributions, its culture, and the very lives of all its people. There was an immediate need for some way of conceptualizing this horror and "genocide" did it. During the Nuremberg trials of the Nazi war criminals and in the post-war discussion and debate over how to prevent such killing in the future, "genocide" became commonly used. And in incredible little time, it passed from Lemkin's pages into international law. In 1946 the United Nations General Assembly recognized that "genocide is a crime under international law which the civilized world condemns, and for the commission of which principles and accomplices are punishable." Then two years later the General Assembly made this concrete. It passed the Convention on the Prevention and Punishment of the Crime of Genocide. This international treaty, eventually signed by well over a majority of states, affirms that genocide is a punishable crime under international law, and stipulates the meaning of genocide to be

any of the following acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such:
(a) Killing members of the group;
(b) Causing serious bodily or mental harm to members of the group;
(c) Deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part;
(d) Imposing measures intended to prevent births within the group;
(e) Forcibly transferring children of the group to another group.

Note that the Convention is consistent with Lemkin's definition and elaboration. Relevant here, the gravity of both is that genocide is the intent to destroy in whole or part a group. One way of doing this is to kill members of the group, but also genocide includes the intent to destroy a group in whole or in part by other means, such as by preventing births in the group or causing serious mental harm. That is, by both definitions, genocide does not necessarily include killing group members.

In the early years of its use "genocide" was applied almost entirely to the Jewish Holocaust and then, especially through the work of Armenian scholars, to the mass murder of Armenians by the Young Turk regime during World War I. However, scholars increasingly have come to realize that restricting the killing aspect of the concept to those murdered by virtue of their group membership does not even account for the millions of non-Jewish Poles, Ukrainians, Yugoslavs, Czechs, Frenchmen, and others, the Nazis wiped out. How then do we conceptualize the purposive government killing of protesters or dissidents, the reprisal shooting of innocent villagers, the beating to death of peasants for hiding rice, or the indiscriminate bombing of civilians? How do we conceptualize torturing people to death in prison, working them to death in concentration camps, or letting starving them to death, when such killing is done out of revenge, for an ideology, or for reasons of state having nothing to do with the social groups to which these people belong?

Because of such questions some scholars have generalized the meaning of "genocide." In some cases it has been extended to include the intentional killing of people because of their politics or for political reasons,9 even though this has been explicitly excluded from the Genocide Convention. Some scholars also have extended the definition of genocide to cover any mass murder by government whatsoever;10 some have even stretched the concept much further, such as to characterize the unintentional spread of disease to indigenous populations during European colonization, including that in the American West.11 To all these scholars the critical aspect of "genocide" is intentional government killing.

All this is confusing. Because of the non-killing aspect of "genocide" and the need to have a concept covering other kinds of government murder, all the following have been called genocide: the denial of ethnic Hawaiian culture by the American run public school system in Hawaii, government policies letting one race adopt the children of another race, African slavery by Whites, South African Apartheid, any murder of women by men, death squad murders in Guatemala, deaths in the Soviet gulag, and, of course, the Jewish Holocaust. The linking of all such diverse acts or deaths together under one label has created an acute conceptual problem that begs for the invention of new concepts to cover and be limited to intentional government murder. Thus, both Barbara Harff12 and I have independently developed the concept of politicide for a government's premeditated killing of people because of their politics or for political reasons. But this new concept is still not sufficient, since many mass murders by government cannot be so labeled either, such as the working of POWs to death by the Japanese army in World War II or the killing of Black Africans that resisted enslavement.

Clearly, a concept is needed that includes all intentional government killing in cold blood and that is comparable to the concept of murder for private killing.

The killing of one person by another is murder whether done because the victim was Black or White, refused to repay a loan, or hurled an insult. It is murder if the killing was a premeditated act or the victim died because of a reckless and wanton disregard for their life. Nor does it matter whether the killing is done for high moral ends, for altruistic reasons, or for any other purpose, it is murder under Western and most other legal codes (unless officially authorized by government, as for judicial executions or military combat). And as a crime murder is limited by definition to intentionally taking the life of another in some way. Although we use murder metaphorically, as in someone "murdering" the language, it is not the crime of murder to hurt someone psychologically, to steal their child, or to rob them of their culture.

As an analogous concept for public murder, that intentionally done by government agents acting authoritatively, I offer the concept of democide. Its one root is the Greek demos or people; the other is the same as for genocide, which is from the Latin caedere, to kill. Democide's necessary and sufficient meaning is that of the intentional government killing of an unarmed person or people. Unlike the concept of genocide, it is restricted to intentional killing, and does not extend to attempts to eliminate nations, races, or religions by means other than killing members of the group. Moreover, democide is not limited to genocide (that aspect involving the killing of group members), nor to politicide, mass murder or massacre, or terror. It includes them all and also what they exclude, as long as such killing is a purposive act, policy, process, or institution of government. In short democide is government murder.

Since much killing takes place during wartime, I must be absolutely clear on what then constitutes democide. War related killing by military forces that international agreements and treaties directly or by implication prohibit is democide, whether the parties are signatories or not. That killing explicitly permitted is not democide. Thus, the death of civilians during the bombing of munitions plants in World War II is not democide. Nor is the death of civilians when through navigation or bombing errors, or the malfunction of equipment, bombs land on a school or hospital, unless it is clear that the bombing was carried out recklessly in spite of a high risk to such civilian buildings. Nor is the death of civilians in a bombed village democide when beneath it has been built enemy bunkers. Nor is the death of civilians caught in a cross fire between enemy soldiers democide, or those civilians killed while willingly helping troops haul supplies or weapons. Seldom is it easy to make these distinctions, but the aim here must be clear. In the findings to be described below I discriminate between democide in time of war and war-deaths. The latter are those of the military and civilians from battle or battle related disease and famine. The former are those victims (which may include the military, as when POWs are massacred) of internationally prohibited war-time killing, what may be called war-crimes or crimes against humanity. Such was the Holocaust.

Pulling all this together, a death constitutes democide if it is the intentional killing of an unarmed or disarmed person by government agents acting in their authoritative capacity and pursuant to government policy or high command (as in the Nazi gassing of the Jews). It is also democide if these deaths were the result of such authoritative government actions carried out with reckless and wanton disregard for the lives of those affected (as putting people in concentration camps in which the forced labor and starvation rations were such as to cause the death of inmates). It is democide if government promoted or turned a blind eye to these deaths even though they were murders carried out "unofficially" or by private groups (as by death squads in Guatemala or El Salvador). And these deaths also may be democide if high government officials knowingly and purposely allowed conditions to continue that were causing mass deaths and issued no public warning (as in the Ethiopian famines of the 1970s). All extra-judicial or summary executions comprise democide. Even judicial executions may be democide, as in the Soviet show trials of the late 1930s. Judicial executions for "crimes" internationally considered trivial or non-capital, as of peasants picking up grain at the edge of a collective's fields, of a worker for telling an anti-government joke, or of an engineer for a miscalculation, are also democide.

Genocide (in its killing aspect) is then a type of democide that involves the government murder of people because of their ethnicity, race, religion, language, or nationality.

With the understanding of both genocide and democide, what can we empirically say about their general occurrence, patterns, causes and conditions?

I have collected data on this century's democide by all state regimes, quasi-state regimes (e.g., the communist soviet enclaves in Nationalist China or the White army territories in Russia during the civil war in 1920), and group regimes (such as the Palestine Liberation Organization). The largest of the resulting estimates, including that for genocide, are listed in Table 1 and graphed in Figure 1. These are for this century's megamurderers--those states killing in cold blood, aside from warfare, 1,000,000 or more men, women, and children. These fifteen megamurderers alone have murdered over 151,000,000 people, almost four times the almost 38,500,000 war-dead for all this century's international and civil wars up to 1987.13 The most totalitarian regimes, that is the communist U.S.S.R., China and preceding Mao guerrillas, Khmer Rouge Cambodia, Vietnam, North Korea, and Yugoslavia, as well as Nazi Germany, account for nearly 128,000,000 of them, or 84 percent. In addition to this democide by these megamurderers, 203 lesser murderers have killed near 17,700,000 more people.

These figures on democide are new to students of the Holocaust and genocide. They are based on almost 8,200 estimates of genocide, politicide, massacres, terrorism, extrajudicial executions, and other relevant types of killing. These estimates were recorded from over a thousand sources, which include general works, specialized studies, human rights reports, journal articles, and news sources.14

Of course estimates of democide are very uncertain15 and often propagandistic. Therefore I generally calculated a low to high range of probable democide, the low being the sum of the lowest estimates across events for a regime and the high being a similar sum. In this way I tried to bracket the most probable figure, which I then judged or calculated based on the central thrust, objectivity, and quality of the estimates. However, many of the figures in Table 1 will seem so precise as to belie this cautious approach. The reason for this apparent over precision lies in the method by which they were determined, which often involved calculations on dozens and sometime hundreds of estimates. The democide I give here for, say Cambodia, was then the outcome of all these calculations, including polynomial regressions of estimates of her population for each year from the early 60s to late 1980s.

In addition, much of this democide occurred during wartime and may appear to be confused with war-deaths. I have tried to separate battle-dead or those dying in the wake of war from genocide and mass murder. The Holocaust during the Second and genocide of the Armenians during the First World War are easy cases of this separation. So is the reprisal killings of Czechs or Yugoslavs by the Nazis, or those who died in Soviet labor camps during the Second World War. Some cases are not so easy, as of American and British indiscriminate bombing of urban populations during the Second World War, American bombing in Vietnam and Cambodia, or the British food blockade of the Levant in the First World War which caused many deaths from starvation and malnutrition. I have followed this approach in classifying those killed or dying in war as either war-dead or democide. If these deaths would be considered a crime against humanity or a war crime, if they are now internationally outlawed by the Geneva Conventions and their 1977 Protocols, they are counted as democide.

Finally, to make sure I understood the democide estimates and could qualitatively evaluate them, I did case studies on democide by the Soviet Union,16 Chinese regimes 1900-1987,17 Nazi Germany,18 Cambodian regimes, Vietnamese regimes, Turkey's regimes 1900-1923, North Korea, Russia 1900-1917, Mexico 1900-1920, Pakistan, Yugoslavian regimes 1941-1987, and Japan 1936-1945.19

With this in mind consider again the total democide of near 170,000,000 given in Table 1. This figure is incredible, indigestible, and unimaginable. One simply cannot comprehend how many people these are. It surpasses the 1987 population of all but six nations in the world. If without stopping one were to have this many people come in one door, walk at three miles per hour across a room with three feet between them (assume generously that each person is also one foot thick, naval to spine), and exit an opposite door, the time it would take for all to pass through the room would be over four years and ten months. If all these dead were laid out head to toe, and assuming each is an average five feet tall because of the many children, they would reach from Honolulu, Hawaii, across the vast Pacific and then the huge continental United States to Washington D.C. on the East coast, and then back again over sixteen times.20

What about genocide deaths? As can be seen from Table 1, near 39,000,000 people have been killed in genocide, or near 23 percent of this toll. This itself is more than all the war-dead of all this century's international and civil wars, including World Wars I and II, the Korean and Vietnam Wars, the Russian and Mexican Revolutions, and the Spanish and Chinese Civil Wars. These genocides not only involved the Holocaust and the killing of the Armenians, the best known of this century's genocides, but also the lesser known genocide of Gypsies by the Nazis and of Greeks by the Turks. But then there were also the many genocides by other regimes, such as Stalin's deadly deportations of the Volga Germans, Greeks, Koreans, Chechens, and Crimean Tatars, and other nations groups; Kaiser Germany's almost total annihilation of the Herero in Namibia ; Pre-Revolutionary Mexico's genocide against its Indians; post-World War II Poland's, Yugoslavia's, and Czechoslovakia's killing deportation and genocidal treatment of their ethic and Reich Germans; Croatia's World War II genocide of their Jews, Gypsies and Serbians and the subsequent genocidal treatment of Croatians by the Tito partisans and then new post war Tito regime; Indonesia's post-coup 1965-1967 slaughter of ethnic Chinese (as a side-show to their massacre of communists) and in later years of East Timorese after their invasion of the country; Communist Chinese genocide of Tibetans and Nationalist Chinese of Formosans and both of Muslims; Rwanda's genocide of Tutsi and Burundi's of Hutu; East Pakistan's mass genocide of Bengalis in former West Pakistan (now Bangladesh); the Cambodian Khmer Rouge genocide of Buddhists, Chan (Muslims), ethnic Vietnamese, and ethnic Chinese; and on and on for a total of 141 regimes committing genocide. In no way, however, does listing these genocides or lamenting over their toll demean the importance, horror, and uniqueness of the Holocaust. For of all these genocides, the Holocaust is the only one in which a regime, as a matter of public policy, aimed to exterminate all members of a specific religious group--the Jews--root and branch, where ever they could be found, whether in Germany or some occupied country, and the Nazis even prepared plans to kill them all in countries not yet defeated, such as in Great Britain. In this sense the Holocaust is unparalleled among genocides.

What now can be said about the conditions and patterns of genocide (again, understood as the killing aspect), including the Holocaust? How does genocide empirically relate to other forms of democide? How does it correlate with socio-economic, cultural, and geographical conditions and assumed causes? What are the best predictors of genocide? In order to answer these questions, I will present in summary fashion the results of a multivariate analysis of 214 state-regimes, including all 141 of them committing some kind of democide in this century, 1900-1987. A state-regime is a particular kind of government, such as a military dictatorship, a monarchy, or communist system. A country may have had several regimes during the century. Russia, for example has had three up to 1987, that of the Czar, then the brief Kerensky government, followed by the Bolshevik coup and communist rule. The Czar and communist regimes are two that were analyzed among the 214 regimes. For Germany there were the regimes of the Kaiser, Weimar Republic, Hitler, communist East Germany and democratic West Germany. All, with the exception of the Weimar Republic, were included in the analysis. Some countries, such as the United States, Canada, and Great Britain had only one regime through this century. In total 432 regimes have existed 1900-1987. The focus is on the regime rather than the state, since it is the regime that commits democide and at issue is whether certain types of regimes are more or less disposed to murder their citizens or foreigners. As to the analysis, this is not the conference within which to present the actual methods, correlations, coefficients, and the like; and they are given elsewhere.21 Technical material, where possibly useful, will be confined to the footnotes. Now for the results.

The first question has to do with whether genocide correlates with other forms of democide. That is, does genocide comprise a general empirical pattern in state murder? Now such an empirical pattern would be a distinct and observable intercorrelation among different kinds of killing, such as genocidal murders and nongenocidal massacres, extrajudicial executions, and assassinations. And intercorrelation means (if positive) that when a regime commits genocide it also commits such other killing, and when it does not commit genocide it also does not commit these other kinds of killing. Many would argue, I am sure, that genocide is a basic and pervasive pattern among regimes, that genocide reflects wide-scale murder by regimes and is a central indicator that general democide is occurring. And that therefore to focus on genocide is to deal with the central and most basic state murder.

Yet, surprisingly, I have found that this is not so. Rather, genocide is a pattern of democide independent of other empirical democide patterns. That is, genocide is largely uncorrelated with other kinds of democide. For all 432 state regimes in this century, 1900-1987,22 I determined the empirical patterns among fourteen different types of democide, including those killed in genocide, deportations, massacres, terror, forced labor, concentration camps and prisons, man-made intentional famines, indiscriminate bombing, and the killing of POWs; and also including total democide, domestic democide, foreign democide, and the annual rate of domestic democide.23 The major and statistically independent patterns comprise domestic democide, foreign democide, the annual rate of domestic democide, indiscriminate bombing, and genocide (which is highly intercorrelated only with massacres).

Genocide itself is therefore a distinct empirical pattern of democide . This means that in the history of a regime it may or may not have committed genocide and massacres regardless of what other types of democide it has engaged in. Moreover, one cannot predict from the amount of democide that has been committed or the lethality of regime, as measured by the annual domestic democide rate, that genocide or massacres will or will not occur. Nor will the extent of a regime's foreign democide indicate that it will commit genocide. All this means that the immediate causes and conditions of genocide are different than those for other types of democide or democide overall. Nonetheless, at a higher and more basic level there still may be causes and conditions that encompass genocide and other patterns of democide. And there is one that I will now point out.

The more totalitarian and less democratic a regime the more democide, the more genocide, and the greater the annual rate of democide that it commits. That is, although the independent patterns of domestic democide, foreign democide, genocide, and the others, are not correlated, together they are accounted for by a regime's totalitarian power.24 Power is the means through which a regime can accomplish its goals or whims. When a regime's power is magnified through its forceful intervention in all aspects of society, including its control over religion, the economy, and even the family, then when conjoined with an absolutist ideology or religion, mass killing becomes a practical means of achieving its ends. Thus we have the megamurderers shown in Table 1, such as the totalitarian USSR, communist China, Nazi Germany, and Cambodia under the Khmer Rouge. And thus, when the regime finds for whatever reason that the continued existence of a social group is incompatible with its beliefs or goals, totalitarian power enables it to destroy that group. Genocide follows. On the other hand, democratic elites generally lack the power to, and democratic culture anyway opposes, the outright extermination of people or social groups for whatever reason.

Pictures speak louder than words and Figure 2 well displays this relationship.25 The vertical dimension in the figure is domestic democide, which in this relationship also includes the genocide pattern. The two other dimensions define the two scales, one for totalitarian power and other for democratic power, which together predict democide (and genocide). The figure also shows that as a regime has greater totalitarian power its overall domestic democide in general and genocide in particular are likely to increase exponentially.26

Power is the basic explanation and empirical correlate of genocide and other kinds of murder by the state. But there is also a related characteristic that is intrinsic to power. The more power a regime has the more it is likely to commit foreign violence and to have rebellions against it. The empirical evidence on this is overwhelming.27 The least warlike regimes are democratic, the most are totalitarian. Indeed, democratic regimes do not make war on each other at all while warfare between totalitarian regimes, such as the Soviet Union and Nazi Germany, are the most deadly of all. Using the number killed in war or rebellion as the indicator of such violence and connecting this now to genocide, I find that the likelihood of genocide by a regime increases significantly the greater the characteristic number of its people killed in war and rebellion. The more a regime has or will suffer dead from involvement in war or rebellion, the greater its foreign democide and genocide.28 Clearly, war or rebellion provide an excuse and cover in the fog of war for a regime to eliminate those social groups it finds objectionable. But also, the results show that over the life of a regime the more disposed it is to be involved in deadly foreign and domestic wars, the more likely it will commit democide, whether or not carried out during these wars. This is because totalitarian power not only underlies democide and genocide, but also because such power underlies as well the occurrence and intensity of war.

But, many would ask, what about racial, ethnic, and religious diversity and accompanying hostility? What about antisemitism and Nazi Germany? Turkey and the Armenians? Pakistan and India and the Hindus and Moslems. Rwanda and Burundi and Hutu and Tutsi. And all the other ethnically, racially, or religiously diverse societies in which their regimes have systematically carried out genocide. Surely such diversity is correlated with genocide.

But it is not. The social diversity of a nation is not correlated with nor does it predict its regime's overall domestic or foreign democide or in particular the regime's genocide. This is the most difficult to accept but the case studies and quantitative analyses are consistent. A nation's ethnic, religious, racial, linguistic, or national divisions, the relative size of such minorities or the nation's overall social diversity are uncorrelated with its domestic or foreign democide or its genocide. This is true even when various controls are introduced for the level of power, involvement in war or rebellion, education and level of economic development, or the nature of its culture.29 In other words, some regimes whose societies are riven with social diversity will commit little genocide and some regimes with little diversity will commit much genocide; and some with much diversity will also commit much genocide and those with little diversity will have little genocide. And this lack of correlation is apparently not caused by any intervening or masking conditions.30

For domestic genocide to occur, of course, there must be some social diversity and such usually will exist even in apparently homogenous nations. For example, Japan is looked at as highly homogenous, yet its pre-militarist regime allowed 2,600 to 11,000 Korean residents to be massacred in Japan after the 1923 Yokohama earthquake (they were accused of poisoning public water, hoarding food, and starting fires). It is not diversity that predicts to genocide, but a regime's power. Minorities have been massacred in authoritarian or totalitarian states while in democracies very large minorities usually are secure in their lives, as in Switzerland or Belgium.31

Perhaps it is not a question of diversity but of culture. Possibly some cultures are more disposed to genocide than others and there are those that would pin such infamy on Western cultures; others might point to African cultures or Asian. Still we might even be more specific and say that Christians are less disposed to genocide than non-Christian or Moslem societies. Many other cultural distinctions might be made and I have tried to include measures of them in my analyses.

No matter. Whatever the cultural distinction, the nature of a regime's culture is uncorrelated with and does not predict to its overall foreign or domestic democide or its genocide. This is almost as hard to accept as the lack of correlation with diversity, but the analyses are consistent for this also.32 The variation among regimes in the degree to which they are Christian or Moslem, or influenced by English culture, or anti-women, or even whether they are located in Africa, Europe, Asia, and so forth, does not predict to a regime's overall domestic or foreign democide or its genocide. As with diversity this is generally true for genocide even if one introduces various socio-economic controls.

Aside from diversity, perhaps the most popular solution to genocide has been education. It is often assumed that the more educated a population, the less likely its regime will commit, or be allowed to commit, genocide. This is the belief that with greater education comes a greater understanding of other groups, of the horror of genocide, and of a willingness to compromise. In line with this some add to this that economic development is also necessary. They assume that an educated and prosperous society has no reason to destroy minorities--that the mass frustrations and deprivation attendant upon poverty and that can be organized and unleashed upon out groups by elites no longer exists. I wish it were true, especially about education, but the data deny it. The level of education or economic development of a nation is uncorrelated with and does not predict to the foreign or domestic democide or the genocide of its regime.

This finding may be no surprise to those who realize that just before World War II Germany was considered one of the most developed and educated nations in the world. Moreover, Japan was the most educated and developed nation in Asia at the same time it was carrying out mass extermination campaigns in China. The megamurders by Nazi Germany and militarist Japan alone should caution those who believe that improving national education and wealth will decrease the likelihood of genocide and mass murder. The results for all democides confirm this in general. There is no meaningful correlation of these socio-economic characteristics and regime's overall democide, or genocide specifically. This is also true even when various political controls or a regime's involvement in war and rebellion is taken into account.33

What does this say in particular about the "other" as a threat and demonization, a central topic of this conference. I have not done systematic comparative research on this question, but the various case studies I have published are helpful in answering this. First, demonization and perception of the other as a threat appears a general process in war, whether international or domestic. We all know that in war enemies dehumanize each other, publicize each other as threats to humanity, civilization, and the Good, and thus justify their mutual destruction. Thus in World War II the Japanese were treated in the American media as monkeys, unfeeling and inscrutable, savage and barbaric, and a threat not only to Asia and the United States, but to Western civilization.

But aside from national enemies in time of war, what about internal groups? Is there a relationship between demonization, the perception of threat, and genocide. Here I must deal with elite opinion, particularly that of those in power, for there is little information on what the mass of people perceived preceding one or another democide. Now, we do know well that in some genocides the victims have been perceived by the regime as a threat and publicly characterized as less than human, as apes, pigs, cockroaches, vermin, and the like. The Nazi view of the Jews well exemplifies this. Not only were they the lowest of humanity, if at all seen as human, but they were believed to be a direct genetic threat to the master race of Aryans and a pollutant of the good German society and culture.

The Armenians genocide by the Young Turk regime is another example. In build up to this genocide during World War I the Armenians were treated as bloodsuckers, aliens, greedy, unpatriotic, anti-Turk, Pro-Russian, and a direct threat to the security of Turkey in the East where its forces confronted massed Russian armies. That the Armenians were a distinct ethnic, national and Christian subgroup in Muslim Turkey and dominated commerce, crafts, and professions, gave substance to these claims. However, the real threat of the Armenians was to the desire of the Young Turks to purify Turkey of non-Turks and to recreate the ancient glory of the Turk. In particular, the Armenians had been protected in the past by the intervention of Britain, Germany, and Russia, and thus were perceived as a continual threat to true Turkish independence. Once the Armenian protectors were engaged in war with each other and turkey allied with Germany, then this alien group and threat to Young Turk designs could be exterminated.

Another example is of the Bengalis in East Pakistan. They were already ethnically, linguistically, and geographically separated from the governing majority in West Pakistan, and although also Moslem, their beliefs were considered by Moslems in the West as vulgar. They were not a threat, however, until they won a majority in the national legislature were thus in position to achieve the desired political independence of East Pakistan. The dehumanization of the Bengalis by the governing military elite and the resulting genocide soon followed.

Indeed, I am sure that demonization and the elite perception of threat from the outgroup was a general part of the process of genocide in this century. But this seems almost axiomatic. After all, genocide is by definition (again, in its killing aspect) the attempt to eliminate a social group. By definition, therefore, the concept of genocide only applies to those who a regime has killed by virtue of their membership in a distinct group. For such killing to take place, therefore, a group as such must be singled out for the killing. And it hardly conceivable that , as in war, such killing would not be preceded by a media blitz dehumanizing and demonizing the group and its members.

A broader question is whether democide in general involves such demonization. And I believe the answer is no. Much of the nongenocidal killing took place because the victims opposed a regime, criticized it, were killed as examples to deter others from opposition or sabotage (as in hanging ten subjects selected at random in retaliation), were of the wrong class (as of a landowner), did not work hard enough, violated a minor rule, were disrespectful (as in hanging one's coat on a bust of Lenin), or were worked to death. Many were simply worked to death, as in the German, Soviet, and Chinese forced labor camps, and had done nothing more, if anything (and people were often arrested for nothing but to supply slave labor) than violate a petty law or rule, or come under suspicion of being an enemy of the state or people. Tens of millions of people were killed simply as expendable bricks and lumber in the building of a utopia, as in Cambodia under the Khmer Rouge, Vietnam, or communist China.

Near 40,000,000 people died or were killed in the Soviet slave-labor system alone, a number that exceeds all the genocides of this century. And although once within the system political prisoners were systematically dehumanized as "enemies of the people" and were treated by guards and true criminals (those that had committed murder, burglary, and the like) as worse than scum, they may have been before imprisonment highly respected members of society. Many, in fact, were former communist party members themselves. Therefore, I argue, demonization and seeing or treating the other as a threat is not a necessary preliminary to democide in general. It is, however, an intrinsic part of the process of genocide.

In any case, demonization is an handmaiden of power. Where civil liberties and political rights exist, where regimes are democratic, where power is thus balanced, checked, and accountable, some demonization of outgroups may exist, but genocide is most unlikely. Where the opposite it true. Where a few or one dictator holds all power and such power is arbitrary, neither controlled by law or publicly responsible, then demonization is a technique, a means of eradicating some group that may be perceived as a threat to power, an evil presence, or a block to creating utopia.

In sum, the bottom line of this research is that power kills and the more power the more killing. The degree of a regime's power along a democratic to totalitarian scale is a direct underlying cause of domestic democide, including genocide. Moreover, acting through war and rebellion it is an indirect cause of foreign democide as well. The more power a government has, the more it can act arbitrarily according to the whims and desires of the elite, the more it will make war on others and murder its foreign and domestic subjects. The more constrained the power of governments, the more it is diffused, checked and balanced, the less it will aggress on others and commit democide. At the extremes of power, totalitarian governments have slaughtered their people by the tens of millions, while many democracies can barely bring themselves to execute even serial murderers. 


* Paper delivered to the Conference on "The 'Other' as Threat--Demonization and Antisemitism," June 12-15, 1995, at the Hebrew University of Jerusalem. Published in Albert J. Jongman (Ed.). CONTEMPORARY GENOCIDES: CAUSES, CASES, CONSEQUENCES, Leiden, Netherlands: Interdisciplinary Research Program on the Root Causes of Human Rights Violations (PIOOM), 1996

1. There are of course many collections of case studies and qualitative analyses, such as Charny (1984), Fein (1992), Porter (1982b), Veenhoven (1975-1976), and Wallimann and Dobkowski (1987). There are also excellent overviews and analyses of genocide and mass murder. See in particular Chalk and Jonassohn (1988), Kuper (1981, 1985), and Glaser and Possony (1979). For bibliographic reviews, see Charny (1988, 1991a).

2. See, for example, Porter (1982a).

3. Rummel (1994).

4. Rummel (1995 [published in 1997).

5. Harff and Gurr (1988) presents a list of genocides and politicides since World War II; Rummel (1987, 1988) gives a preliminary list of genocides and mass murder in this century. These are the only two lists of which I am aware that are meant to be comprehensive and are presented in a comparative context, but neither focuses on or presents a list of genocides per se and which is genocide versus politicide or mass murder is not always clear.

6. The only application of quantitative analysis to genocide that I have seen is Fein's (1979) use of multiple regression.

7. These are mainly presented in Rummel (1995 [published in 1997).

8. Lemkin (1944, p. 79).

9. See, for example, Fein (1984); Kuper (1981) and Porter (1982a).

10. See, for example, Chalk and Jonassohn (1988); Charny (1991b).

11. See Stannard (1992).

12. See Harff and Gurr (1988).

13. Battle-dead up to 1980 is from Small and Singer's (1982) compilation of foreign and domestic war battle-dead. I added to this my own estimate for the years 1981-1987).

14. The estimates, sources, and calculations for the Soviet Union are given in Rummel (1990); for the Chinese Warlords, Nationalist regime, communist guerrillas, and communist regime, see in Rummel (1991); and for Nazi Germany, see Rummel (1992). All other sources and estimates are given in Rummel (1995 [published in 1997).

15. After decades of scholarly research in the German archives, study of reports and official documents of other involved countries, and interviews with participants and survivors, the best estimates of the Holocaust dead still vary by over 40 percent.

16. Rummel (1990).

17. Rummel (1991).

18. Rummel (1992).

19. Those case studies not footnoted are published in Rummel (1994).

20. Back and forth, over 4,838 miles one way, near sixteen times? This is so incredible that I would not believe the calculation and had to redo it several times.

21. Rummel (1995 [published in 1997).

22. For only this analysis I was able to do it on all the 432 regimes; all other analyses had to be limited to 214 regimes.

23. This was done through component analyses with varimax orthogonal and oblique rotation of all 432 state regimes existing during 1900-1987. What I am calling an empirical pattern is a dimension (component, factor) defined by orthogonal rotation.

24. This is based on many different canonical, regression, and component analyses of various subsets of variables from a set of over eighty democide, political, socio-economic, cultural, and geographic variables for 214 state regimes.

25. The inverse squared distance technique used to draw the surface shown in the table is not based on regression, but interpolates domestic democide logged (the Z height of the surface at a XY point) as the weighted average of the totalitarian and democratic (X and Y) scales. The squared Euclidean distances across the totalitarian and democratic scales comprise the weights.

26. This is also verified through regression analysis.

27. This is not the place to go into this evidence in detail. See Ray (1993, 1995), Russett (1993), and Weart (1994, 1995 [actually published in 1998]).

28. This is clear from a regression analysis of genocide on a variety of characteristics, including a regime's war dead and rebellion dead.

29. In this context, "controls" means that these variables were held constant. Their influences were partialled out of the correlations between diversity and democide, and still the correlations between these two, or with genocide, were near zero.

30. Because of the importance of this finding, a variety of data were analyzed in many ways. For example, various component analyses were done with genocide and other types of democide and a variety of measures of diversity, and redone with indicators of diversity and various political, socio-economic, cultural and geographic indicators. Genocide also was regressed alone on diversity measures and then on diversity indicators plus the other indicators and several interaction terms. The multiple R was .52, with only the political indicators and war and rebellion-dead being significant.

31. Note that Rwanda and Burundi are not really diverse, less so than the United States, Canada, Great Britain, or many other European, Latin American, or Asian countries. In Rwanda and Burundi the majority Hutu comprise about 85 percent of the population and around 70 percent of the population are Christian in Rwanda and over 60 percent in Burundi.

32. This is based on component, canonical, and regression analyses.

33. This is based on component, canonical, and regression analyses.


Chalk, Frank, and Jonassohn, Kurt (1988). THE HISTORY AND SOCIOLOGY OF GENOCIDE: ANALYSIS AND CASE STUDIES. New Haven: Yale University Press.


Charny, Israel W. (Ed.) (1988) GENOCIDE: A CRITICAL BIBLIOGRAPHIC REVIEW. New York: Facts on File Publications.

Charny, Israel W. (Ed.) (1991a) GENOCIDE: A CRITICAL BIBLIOGRAPHIC REVIEW: Vol. 2, London: Mansell.

Charny, Israel (1991b). "A Proposal of a New Encompassing Definition of Genocide: Including New Legal Categories of Accomplices to Genocide, and Genocide as a Result of Ecological Destruction and Abuse." Invited Address to the first Raphael Lemkin Symposium on Genocide, Yale University Law School, February.


Fein, Helen (1984). "Scenarios of Genocide: Models of Genocide and Critical Responses." In TOWARD THE UNDERSTANDING AND PREVENTION OF GENOCIDE: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON THE HOLOCAUST AND GENOCIDE, [edited] by Israel W. Charny. Boulder: Westview Press, pp. 3-31.

Fein, Helen (Ed.) (1992). GENOCIDE WATCH. New Haven: Yale University Press.

Glaser, Kurt and Stefan T. Possony (1979). VICTIMS OF POLITICS: THE STATE OF HUMAN RIGHTS. New York: Columbia University Press.

Harff, Barbara and Ted Robert Gurr (1988). "Toward Empirical Theory of Genocides and Politicides: Identification and Measurement of Cases since 1945." INTERNATIONAL STUDIES QUARTERLY, 32: 359-371.

Kuper, Leo. (1981). GENOCIDE: ITS POLITICAL USE IN THE TWENTIETH CENTURY. New Haven: Yale University Press.

Kuper, Leo. (1985). THE PREVENTION OF GENOCIDE. New Haven: Yale University Press.

Lemkin, Raphael (1944). AXIS RULE IN OCCUPIED EUROPE: LAWS OF OCCUPATION, ANALYSIS OF GOVERNMENT, PROPOSALS FOR REDRESS. Washington, D.C.: Carnegie Endowment for International Peace.

Porter, Jack Nusan (1982a) "Introduction: What is Genocide? Notes toward a Definition." GENOCIDE AND HUMAN RIGHTS: A GLOBAL ANTHOLOGY, [edited] by Jack Nusan Porter. Washington, D.C.: University Press of America, pp. 2-32.

Porter, Jack Nusan (Ed.) (1982b). GENOCIDE AND HUMAN RIGHTS: A GLOBAL ANTHOLOGY, [edited] by Jack Nusan Porter. Washington, D.C.: University Press of America.

Ray, James Lee.(1995). "Wars between democracies: rare, or nonexistent?" INTERNATIONAL INTERACTIONS 18 (No. 3, 1993): 251-276.

Ray, James Lee (1995) DEMOCRACY AND INTERNATIONAL POLITICS: AN EVALUATION OF THE DEMOCRATIC PEACE PROPOSITION. Columbia, South Carolina: University of South Carolina Press.

Rummel, R.J. (1987). "Deadlier than War." IPA REVIEW (Institute of Public Affairs Limited, Australia) 41 (August-October ): 24-30.

____________.(1988). "As Though a Nuclear War: The Death Toll of Absolutism." INTERNATIONAL JOURNAL ON WORLD PEACE, 5 (July-September):27-43.

___________ (1990) LETHAL POLITICS: SOVIET GENOCIDE AND MASS MURDER. New Brunswick, New Jersey: Transaction Publishers.

___________ (1991) CHINA'S BLOODY CENTURY: GENOCIDE AND MASS MURDER SINCE 1900. New Brunswick, New Jersey: Transaction Publishers.

___________ (1992) DEMOCIDE: NAZI GENOCIDE AND MASS MURDER. New Brunswick, New Jersey: Transaction Publishers.

___________ (1994) DEATH BY GOVERNMENT: GENOCIDE AND MASS MURDER SINCE 1900. New Brunswick: Transaction Publishers.

___________ (1995 [published in 1997) STATISTICS OF DEMOCIDE. New Brunswick, New Jersey: Transaction Publishers

Russett, Bruce. (1993). GRASPING THE DEMOCRATIC PEACE: PRINCIPLES FOR A POST-COLD WAR WORLD. Princeton, New Jersey: Princeton University Press.

Small, Melvin and J. David Singer (1982). RESORT TO ARMS: INTERNATIONAL AND CIVIL WARS 1816-1980. Beverly Hills, California: Sage Publications.

Stannard, David E. (1992). AMERICAN HOLOCAUST: COLUMBUS AND THE CONQUEST OF THE NEW WORLD. New York: Oxford University Press.

Veenhoven, Willem A., and Crum Ewing, Winifred (Eds.) (1975-1976). CASE STUDIES ON HUMAN RIGHTS AND FUNDAMENTAL FREEDOMS: A WORLD SURVEY, 5 v. The Hague: Nijhoff.

Wallimann, Isidor and Michael N. Dobkowski (Eds.) (1987). GENOCIDE AND THE MODERN AGE: ETIOLOGY AND CASE STUDIES OF MASS DEATH. New York: Greenwood Press.

Weart, Spencer. "Peace among democratic and oligarchic republics." JOURNAL OF PEACE RESEARCH (1994).

Weart, Spencer (1995 [published in 1998]) NEVER AT WAR: WHY DEMOCRACIES WILL NOT FIGHT ONE ANOTHER. New Haven: Yale University Press. .

You are the visitor since 11/27/02

Go to top of document