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

Other Books on This site

On Democide

Lethal Politics: Soviet....

China's Bloody Century

Democide: Nazi Genocide....

Death By Government

On Theory/Conflict/War

Vol. 1: The Dynamic Psychological Field (entire)

Vol. 2: The Conflict Helix (entire)

Vol. 3: Conflict In Perspective (entire)

Vol. 4: War, Power, Peace (entire)

Vol. 5: The Just Peace (entire)

The Conflict Helix: Principles and Practices....

On The Democractic Peace

The Miracle That Is Freedom (entire)

Power Kills

On Methods

Understanding Correlation (entire)

Article: "Understanding Factor Analysis" (entire)


Genocide and Mass Murder Since 1900

By R.J. Rummel

Charlottesville, Virginia:
Center for National Security Law,
School of Law, University of Virginia, 1997; and Transaction Publishers, Rutgers University

We owe respect to the living; to the dead we owe only truth
----Voltaire. Oeuvres Vol. I, p. 15n


Forward (by John Norton Moore)
Figures and Tables
1. Summary and Conclusions [Why Democide? Explanation and Proof]
2. Pre-Twentieth Century Democide


Introduction to Part I
3. Japan's Savage Military
4. The Khmer Rouge Hell State
5. Turkey's Ethnic Purges
6. The Vietnamese War State
7. Poland's Ethnic Cleansing
8. The Pakistani Cutthroat State
9. Tito's Slaughterhouse
10. Orwellian North Korea
11. Barbarous Mexico
12. Feudal Russia


13. Death by American bombing
14. The Gang of Centi-Kilo Murderers
15. The Lesser Murderers


16. The Social Field of Democide
17. Democracy, Power, and Democide
18. Social Diversity, Power, and Democide
19. Culture and Democide
20. The Socio-Economic and Geographic Context of Democide
21. War, Rebellion, and Democide
22. The Social Field and Democide
23. Democide Through the Years


IMPORTANT NOTE: Among all the democide estimates appearing on in this book, some have been revised upward. I have changed that for Mao's famine, 1958-1962, from zero to 38,000,000. And thus I have had to change the overall democide for the PRC (1928-1987) from 38,702,000 to 76,702,000. Details here.

I have changed my estimate for colonial democide from 870,000 to an additional 50,000,000. Details here.

Thus, the new world total: old total 1900-1999 = 174,000,000. New World total = 174,000,000 + 38,000,000 (new for China) + 50,000 (new for Colonies) = 262,000,000.

Just to give perspective on this incredible murder by government, if all these bodies were laid head to toe, with the average height being 5', then they would circle the earth ten times. Also, this democide murdered 6 times more people than died in combat in all the foreign and internal wars of the century. Finally, given popular estimates of the dead in a major nuclear war, this total democide is as though such a war did occur, but with its dead spread over a century.



Figure 1.1. Power and Domestic Democide
Figure 1.2. Democracy, Totalitarianism, and Domestic Democide
Figure 1.3. Trends in Democratic and Totalitarian Democide
Figure 4.1. Population Estimates and Polynomial Fits
Figure 4.2. Population Prediction versus Regime Population Fits
Figure 16.1. Distribution of Total Democide
Figure 16.2. Two Hypothetical Models of Democide Distributions
Figure 17.1. The Political Triangle and Empirical Political Patterns
Figure 17.2. The Democide-Power Causal Nexus
Figure 17.3. Plot of Domestic Democide on Democracy and Totalitarian Scales
Figure 17.4. Domestic Democide within the Political Triangle
Figure 17.5. Plot of the Mean Domestic Democide for Different Levels of Power
Figure 18.1. Plot of Domestic Democide and Diversity
Figure 21.1. The Causal Path of Power to Democide
Figure 21.2. Scatter Plots of War Dead and Rebellion Dead Dead Against Power
Figure 21.3. Area Plots of War Dead and Rebellion Dead Against Power
Figure 21.4. Plot of War-Dead and Rebellion-Dead Means for Levels of Power
Figure 22.1. Plot of Regression Predictions of Democide against Actual Democide
Figure 22.2. Third Order Polynomial Regression of Democide Annual Rate on Power
Figure 22.3. Plot of Democide Annual Rate Against Power Dichotomized
Figure 22.4. Fifth and Seventh Order Polynomial Regressions of Domestic Democide Annual Rate on Totalitarian Power
Figure 22.5. Grouping of Regimes by Discriminant Analysis
Figure 23.1. Yearly Total Democide for All Regimes Versus State Regimes
Figure 23.2. Plot of State Regime Democide and War-Rebellion Dead
Figure 23.3. Democratic and Totalitarian Populations and Democide Dead
Figure 23.4. Total Democide and That By Democracies and Totalitarians


Table 2.1A, lines 1-308 and Table 2.1B, lines 310-750. Pre-20th Century Democide Estimates, Sources, and Calculations
Table 2.2. Wars, War Deaths, and Estimated Democide 30 A.D. to 20 B.C.
Table 3.1. Japan's Democide in China and World War II: Estimates, Sources, and Calculations
Table 4.1A, lines 1a-448 and Table 4.1B, lines 449-845. Cambodian Democide: Estimates, Sources, and Calculations
Table 4.2. Polynomial Fits and Predictions to and from Population Estimates
Table 5.1A, lines 1-436 and Table 5.1B, lines 438-641. Turkish Democide: Estimates, Sources, and Calculations
Table 6.1A, lines 1-472 and Table 6.1B, lines 474-942. Vietnam Democide: Estimates, Sources, and Calculations
Table 7.1. The Reichdeutsch and Ethnic German Democide: Estimates, Sources, and Calculations
Table 7.2. Polish Democide: Estimates, Sources, and Calculation
Table 8.1. Pakistani Democide: Estimates, Sources, and Calculations
Table 8.2. Pakistan Genocide in Bangladesh: Estimates, Sources, and Calculations
Table 9.1. Yugoslavian Democide: Estimates, Sources, and Calculations
Table 10.1. North Korean Democide: Estimates, Sources, and Calculations
Table 11.1. The Mexican Democide: Estimates, Sources, and Calculations
Table 12.1. Russian Democide: Estimates, Sources, and Calculations
Table 13.1. American Democide: Estimates, Sources, and Calculations
Table 14.1A, lines 1-579; Table 14.1B, lines 580-1044; Table 14.1C, lines 1045-1544; Table 14.1D, lines 1545-2086; and Table 14.1E, lines 2087-2560. Centi-Kilo Murdering States: Estimates, Sources, and Calculations
Table 14.2. Centi-Kilo Murdering Quasi-States: Estimates, Sources, and Calculations
Table 15.1A, lines 1-578; Table 15.1B, lines 579-1160; Table 15.1C, lines 1163-1753; Table 15.1D, lines 1754-2255; Table 15.1E, lines 2258-2800; and Table 15.1F, lines 2801-3344. Lesser Murdering States, Quasi-States, and Groups: Estimates, Sources, and Calculations
Table 16.1. Empirical Types of Democide
Table 16.2. Descriptive Statistics on Democide
Table 16.3. Democide Product Moment Correlations
Table 16.4. Component Analysis of Democide By State Regimes
Table 16.5. Patterns and Indicators of State Democide
Table 16A.1. Summary Democide Totals
Table 17.1. Measures of the Political Nature of a Regime
Table 17.2. Component Analysis of Political Measures
Table 17.3. Political Patterns and Indicators of State Regimes
Table 17.4. Correlations of Democide with Political Indicators
Table 17.5. Component Analysis of Democide and Political Indicators
Table 17.6. Interactive Regression of Democide onto Political Indicators
Table 17.7. Contingency Analysis of Domestic Democide and Power Groups
Table 17A.1. Data on Political Measures for 141 Regimes with Democide and 73 Others
Table 18.1. The Social Context of a Regime: Diversity Measures
Table 18.2. Component Analysis of Diversity Measures
Table 18.3. Patterns and Indicators of Social Diversity
Table 18.4. Component Analysis of Democide and Diversity
Table 18.5. Component Analysis of Democide, Political Characteristics, and Diversity
Table 18.6. Component Analysis of Democide and Political Characteristics within Diversity Groups
Table 18A.1. Data on Diversity Measures for 141 Regimes with Democide and 73 Others
Table 19.1. The Cultural Context of a Regime: Cultural Measures
Table 19.2. Component Analysis of Measures of Cultural Variation
Table 19.3. Patterns and Indicators of Cultural Variation
Table 19.4. Component Analysis of Democide and Culture
Table 19.5. Component Analysis of Democide, Politics, Diversity, and Culture
Table 19.6. Component Analysis of Christian and Moslem Regimes
Table 19.7. Component Analysis of Regimes by Region
Table 19.8. Summary Results for Component Analysis within Different Cultures
Table 19A.1. Data on Cultural Measures for 141 Regimes with Democide and 73 Others
Table 20.1. The Socio-Economic and Geographic Context of a Regime
Table 20.2. Component Analysis of Socio-Economic and Geographic Measures
Table 20.3. Patterns and Indicators of Socio-Economic and Geographic Variation
Table 20.4. Component Analysis of Democide, Politics, Diversity and Geography
Table 20.5. Component Analysis of Democide, Politics, Diversity, Culture, and Geography: Rich Vs. Poor
Table 20A.1. Data on Socio-Economic and Geographic Measures for 141 Regimes with Democide and 73 Others
Table 21.1. Average War, Rebellion, and Stimulus to Democide
Table 21.2. Component Analysis of Democide, War, and Rebellion
Table 21.3. Component Analysis of Democide, Politics, War, and Rebellion
Table 21.4. Component Analysis of Democide, Politics, War, Rebellion, and Other Indicators
Table 22.1. Summary of Relationships to Democide and its Patterns
Table 22.2. Regression Analysis of Selected Types of Democide on All Indicators and Selected Interaction Terms
Table 22.3. Frequencies of Actual and Predicted Democide
Table 22.4. Worse Over and Under Predicted Democide from a Regression Analysis Involving Six-Independent Variables
Table 22.5. Canonical Analysis of Democide and Social Indicators
Table 22.6. Discriminant Analysis of Democide Groups and Social Indicators
Table 22.7. Frequencies of Actual and Predicted Democide Groups
Table 22.8. Worse Over and Under Predicted Democide Groups from a Discriminant Analysis
Table 23.1. Serial Regressions of Democide Upon Democracy, Totalitarianism, and War/Rebellion


In the associated volume, Death by Government [1], I described fourteen cases in which since 1900 a regime murdered or is suspected of murdering over 1,000,000 subjects and foreigners. Four of these regimes, the Soviet Union, communist China, Nationalist China, and Nazi Germany, each killed 10,000,000 or more unarmed and helpless men, women, and children.

I also gave some descriptive statistics on these and all 204 other cases of democide (genocide, politicide, massacres, extrajudicial executions, and other forms of mass murder) by state and quasi-state regimes, and non-state groups. These revealed democide's incredible magnitude in this century and well showed the close relationship between the extent of a regime's totalitarian power, or Power in short, and democide. My conclusion was that Power kills, absolute Power kills absolutely.

In 1986 I began this work on democide in order to complete tests of democracy as inherently a structure of non-violence and positive peace. I had shown in previous work that democracies do not make war on each other, and that the more liberally democratic--the more freedom people have--the less their foreign and domestic collective violence.[2] Democracies as a sphere of peace has by now been well established in the literature,[3] as has the negative correlation between democracy and domestic violence, such as riots, coup d'etat, revolutions, and guerrilla warfare.[4] The one controversial finding is that democracy is inversely related to foreign violence. As a side product of this work on democide I will present findings in chapter 21 to further substantiate this relationship.[5]

As of 1986 what remained to be tested was an often asserted negative relationship between democracy and murderous government violence against citizens or foreigners. Moreover, it also made good theoretical sense that the less liberally democratic and more totalitarian a regime, the more people it murders. Unfortunately, good comparative data on democide for testing this did not then exist.

Accordingly, after a preliminary pilot study,[6] I applied for a grant from the Unites States Institute of Peace to pursue the data collection and testing of the relationship between democide and democracy. This I was granted for two years, subsequently renewed for another three.

Collecting data on democide was an horrendous task. I soon was overwhelmed by the unbelievable repetitiveness of regime after regime, ruler after ruler, murdering people under their control or rule by shooting, burial alive, burning, hanging, knifing, starvation, flaying, beating, torture, and so on and on. Year after year. Not hundreds, not thousands, not tens of thousands of these people, but millions and millions. Almost 170,000,000 of them, and this is only what appears a reasonable middle estimate. The awful toll may even reach above 300,000,000, the equivalent in dead of a nuclear war stretched out over decades.

I found that so much of this killing was unknown or ignored by so many that I decided to publish part of the data and case studies of the worst of the megamurderers as separate volumes. Thus I wrote Lethal Politics on the Soviet Union, China's Bloody Century on Nationalist and communist China, and Democide on Nazi Germany.[7] In Death by Government I focused case studies on the lesser megamurderers, such as the Cambodian Khmer Rouge, the Pakistan military in what is now Bangladesh, Japan's military fascists in World War II, and Turkey's Young Turks in World War I. However, space was not available in that book to also present all the estimates, sources, and calculations that underlie the case studies and their democide totals.

That is in part the purpose of this book. Here I do two things. First I list all the relevant estimates, sources, and calculations for each of the case studies in Death by Government, and all additional cases of lesser democide for which I have collected data. This is a total tabulation, with the result that some of the tables are over fifty pages long. The value of this is the listing of each source, its estimate, and comments qualifying the estimate. From these others can check and evaluate my totals, refine and correct them, and build on this comprehensive set of data. These data are presented and annotated in chapter 2 for pre-20th century democide, in Part 1 for the megamurderers, and in Part 3 for the United States and lesser murderers.

The methodological underpinnings for this collection has been given in previous books[8] and I will not repeat this here. I will simply note that I recognize how error full, approximate, and politically biased so many of these estimate may be. I have tried to approach the best overall estimate of democide for a regime, therefore, by determining its best upper and lower bounds, which are given for all estimates tabulated here. The estimates themselves are what appear to me to be the most reasonable or probable within these low-high bounds. Thus my estimate for the Filipinos murdered by the Marcos regime of the Philippines, 1972-1986, is 15,000, which is bounded by a possible low democide of 10,000 and high of 25,000. That is, Marcos is responsible for the murder of 10,000 to 25,000 Filipinos, most likely 15,000.[9]

Second, having finished collecting all these data and completing the major case studies, I finally could systematically test the assumed inverse relationship between democracy and democide. That is the substance of this book. I detail the tests in Part 3 and summarize them in the introductory chapter. My conclusion is that the diverse tests are positive and robust, that the less liberal democracy and the more totalitarian a regime, the more likely it will commit democide. The closer to absolute power, the more a regime's disposition to murder one's subjects or foreigners multiplies. As far as this work is concerned, it is empirically true that Power kills, absolute Power kills absolutely. 


1. Rummel (1994).

2. See Rummel (Understanding Conflict and War, particularly Vol. 4: War, Power, Peace; "Libertarianism and International Violence"; "Libertarianism, Violence Within States, and the Polarity Principle"; and "Libertarian Propositions on Violence Within and Between Nations: A Test Against Published Research Results").

3. See, for example, Ray (1995), Rummel (1994), Russett (1993), Weart (1994, 1996), and the literature referred to therein.

4. For collations of these results, see Rummel ("Libertarian Propositions on Violence Within and Between Nations: A Test Against Published Research Results", 1996).

5. I have given a detailed consideration of this question in Rummel ("Democracies ARE less warlike than other regimes", 1996) and find empirical analyses generally support the proposition that democracies are less warlike than other types of regimes.

6. Rummel (1987).

7. Rummel (1990, 1991, 1993).

8. See Rummel (1990).

9. See table 15.1 [table 15.1E], lines 2687-2747.

For citations see the Statistics of Democide REFERENCES


As I acknowledged in Death by Government, many colleagues, students, and readers of previous chapter drafts contributed to this effort through their ideas, comments and suggestions, recommendation of sources, estimates, or material they passed on to me. I wish to acknowledge their help here also, for they not only contributed to the case studies but in many ways also to the democide estimates. In particular I want to thank Rouben Adalian, Belinda Aquino, Dean Babst, Yehuda Bauer, Douglas Bond, Israel Charny, William Eckhardt, Wayne Elliott, Helen Fein, Irving Louis Horowitz, Hua Shiping, B. R. Immerzeel, Benedict Kerkvliet, Milton Leitenberg, Guenter Lewy, Heath Lowry, J. C. Ramaer, Rhee Sang-Woo, Max Singer, Robert F. Turner, Spencer Weart, Christine White, and J. A. Willinge. I remain particularly indebted to my colleagues Manfred Henningsen and George Kent for their help and support throughout this and the previous work. I absolve them all, however, of any guilt by association with this and the previous volume.

I also am indebted to the United States Institute of Peace for a grant to my project on comparative genocide, of which this book is a part. The views expressed here are mine and do not necessarily reflect those of the Institute or its officers.

And I am especially grateful to Robert F Turner and John Norton Moore for their support and encouragement and for the publication of this work by John Norton Moore's Institute for National Security Law at the University of Virginia.

Finally, as helpmate, friend, critic, my wife Grace is always there. I cannot adequately express my indebtedness to her. Again, thanks sweetheart.

You are the visitor since 11/23/02

Go to top of document