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

Statistics of Democide

Contents | Figures | Tables | Preface

Chapter 1: Summary and Conclusions [Why Democide?...]
Chapter 2: Pre-Twentieth Century Democide
Chapter 3 Japan's Savage Military
Chapter 4: The Khmer Rouge Hell State
Chapter 5: Turkey's Ethnic Purges
Chapter 6: The Vietnamese War State
Chapter 7: Poland's Ethnic Cleansing
Chapter 8: The Pakistani Cutthroat State
Chapter 9: Tito's Slaughterhouse
Chapter 11: Barbarous Mexico
Chapter 12: Feudal Russia
Chapter 13: Death American by bombing
Chapter 14: The Gang of Centi-Kilo Murderers
Chapter 15: The Lesser Murderers
Chapter 16: The Social Field of Democide
Chapter 17: Democracy, Power, and Democide
Chapter 18: Social Diversity, Power, and Democide
Chapter 19: Culture and Democide
Chapter 20: The Context of Democide Socio-Economic and Geographic
Chapter 21: War, Rebellion, and Democide
Chapter 22: The Social Field and Democide
Chapter 23: Democide Through the Years

Other Democide Related Documents On This Site


  • What is democide?
  • "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"
  • "The Holocaust in comparative and historical perspective"
  • Graduate Syllabus on Repression and Democide


  • "Power kills: genocide and mass murder"
  • "Power predicts democide"


  • Lethal Politics
  • China's Bloody Century
  • Democide
  • Death By Government

    Chapter 10

    Statistics Of
    North Korean Democide
    Estimates, Calculations, And Sources *

    By R.J. Rummel

    From 1948 through 1987 the Democratic People's Republic of Korea was ruled by Kim Il-sung, an absolute communist dictator who has turned his country into an Orwellian state. People were so tightly controlled in all their activities, and those visitors that were allowed in were so managed, that comparatively little independent information about the regime's purges, executions, and concentration and forced labor camps filtered out of the country. Nonetheless, through defectors, escapees, agents, Korean War refugees, and analyses of Korean publications and documents, a hazy picture emerges of systematic democide little different than that carried out in the first decades of the Soviet Union or early communist China.

    Perhaps from 710,000 to slightly over 3,500,000 people have been murdered, with a mid-estimate of almost 1,600,000. But these figures are little more than educated guesses. In this case Kim's thought control over all his people and their foreign and domestic communications has protected him and his party from nothing more than deep suspicion about having committed democide so enormous as to be megamurder. But given the nature of his society and what bits and pieces have come out about his purges, labor camps, and executions, there is enough evidence to at least indict him and his party for this crime against humanity.

    Table 10.1 provides the estimates and my calculations on North Korean democide. Much of it is taken up with estimates of the Korean War dead (lines 1 to 96 in the table). This was a very bloody war in which probably close to 2,550,000 people were killed on all sides (line 96), not counting the war-time democide (line 95).

    As a testament to Kim's absolute control over the North, I could find no overall estimates of executions or deaths in labor or concentration camps, purges, land reform, or other mass campaigns. That the communist party killed thousands of people for political reasons is indisputable. But the aggregate numbers are missing from the literature available to me.

    There are some estimates, however, of specific massacres or killings and these are given in Table 10.1. As North Korean troops advanced into South Korea during the Korean War and were followed by communist officials, they systematically massacred former South Korean government officials, anti-communists, and others deemed hostile to the communists; and such killing was intensified as North Koreans retreated from the South. We do have some estimates of the dead, as for Taejon (lines 103 to 105) and Wonju (lines 106 to 107). There is one overall estimate of the minimum number of South Koreans that were murdered, which is from the South Korean Overseas Information Agency (line 111).

    How many Republic of Korea (ROK) POWs were killed by the North Koreans is difficult to pin down. This is because the communists claimed that they had captured 70,000 soldiers overall but they only returned near 8,000 of them.1 We do know they killed near 5,500 ROK POWs and may have impressed into their military another 50,000.2 From this it seems that North Koreans killed from 5,000 to 12,000 ROK POWs (line 121), which is consistent with their murder of 5,000 to 6,000 American POWs (line 141).

    Besides illegally impressing POWs, the North also forced 400,000 South Koreans into their army. They are therefore responsible for their deaths. Given that the army often ordered these people to do the most dangerous tasks or combat and that the North Korean army suffered around 350,000 killed throughout the war (line 13), almost two-and-a-half times the army's original strength (lines 3 to 4), a range of one-third to two-thirds of the impressed/conscripted killed in battle seems conservative. This means a North Korean democide of around 225,000 (line 128).

    Altogether, during the war the North Korean communists probably killed near 500,000 Koreans (excluding at least 6,000 killed by the South-line 152), including their own citizens (line 95). With a probable 1,500,000 civilians killed in the war (line 81), this democide seems, if anything, an underestimate and the true figure may be closer to the high democide calculation of almost 775,000 dead (line 95).

    Aside from the war, North Korean domestic democide has few estimates (lines 161 to 192), and none overall. For two mass campaigns I have tried to estimate a minimum (lines 166 and 173), but the biggest calculations and the ones that most help make the indictment of North Korea as a megamurderer are for camp and hard labor deaths. For labor and concentration camps there are some figures on recent camp populations (lines 176 to 182), but none for years before 1982. Almost uniformly for other communist systems, the labor/concentration camp populations was by far the greatest in the earlier years of the regime, but North Korea may have broken this pattern. In any case, without further information I extrapolated the population for earlier years as about the same and fixed the annual average accordingly (line 183).

    As to how many of these prisoners died per year, there is nothing accept anecdotal accounts from defectors to the South. From these I judge the camps to be as lethal as those early camps in China under Mao, but not as bad as the Soviets ones under Stalin. Applying to the population averages the Chinese death rate I have developed elsewhere,3 I get a range of 71,000 to 707,000 people killed or that otherwise died in the North Korean camps; probably 265,000 (line 186). Over the near thirty-nine years of the regime (to 1987), this means an annual toll of about 6,700 prisoners a reasonable estimate for this absolute totalitarian system from what we now know of her still or former communist neighbors.

    An even more significant and also questionable democide calculation is for the hard labor the party forced on millions of citizens. There was that corvée labor (traditional in Asia) in which the party would order thousands and some times tens of thousands of citizens to leave for months at a time to work on a remote dam, irrigation canal, bridge, or other project. Living and labor conditions were often poor, with food, shelter, and medical care inadequate. Many people would perish as a result, in addition to those who were executed for laziness or anti-party behavior. Then there was that permanent hard labor that the party forced on those it considered "hostile," for which I have no reason to assume that the conditions were any less inimical than for the corvée labor. For both types of laborers I assume the toll was 20, 25, or 30 percent of the camp death rate (line 191), which over some 39 years gives a mid-democide of almost 1,000,000 hard laborers (line 186), or a little over 25,000 a year. Given the large number of "hostiles" alone at hard labor, possibly 4,000,000 in 1989 according to human rights groups,4 this is less than .6 percent a year surely a conservative calculation.

    All these calculations add up to a probable North Korean domestic democide of 1,293,000 men, women, and children (line 199). This is, as is clear from the above, a very uncertain figure. For this reason I have compared it to what the toll would be were the killing carried out in North Korea with the same annual intensity as in the Soviet Union (line 206), China (line 209), and Vietnam (line 212). As can be seen, that domestic democide for North Korea calculated here is about what it would be were the killing on par with the average (line 215) for these three communist systems, although more than that for China and Vietnam. In evaluating these comparisons, keep in mind that these models are based on democide from the inception of the regimes through 1987, and that for each and particularly China and the Soviet Union, the worst democide by far was concentrated in the early years under Stalin and Mao, while in North Korea the killing has continued since 1948 and if anything, has tended to increase in recent years. Moreover, for the more totalitarian North Korea we might expect that ßthe level of democide would be higher than for the Soviet Union, China, and Vietnam. The comparisons show that my calculations here, if anything, seem to underestimate the domestic democide.

    Still, the uncertainties involved in these calculations, the paucity of estimates, and the origin of what few estimates there are being the South Korean government, invites considerable caution about my domestic democide total. And the same holds true for the foreign democide figures, which when added to that of citizens totals given a probable 710,000 to 3,549,000 people killed, or a mid-value of 1,663,000 (line 226). This appears to make the North Korean regime and in particular Kim Il-sung a megamurderer, but for the reasons given I argue that we can do no more than indict North Korea for these killings. 


    * From the pre-publisher edited manuscript of Chapter 10 in R.J. Rummel, Statistics of Democide, 1997. For full reference to Statistics of Democide, the list of its contents, figures, and tables, and the text of its preface, click book.

    1. Lawson (1964, p. 128).

    2. Leckie (1962, p. 331).

    3. Rummel (1991, TableII.A1, line 378)

    4. Maass (1989).

    For citations see the Statistics of Democide REFERENCES

    You are the visitor since 11/25/02

    Go to top of document