OUP user menu

‘Ethnic cleansing’ bleaches the atrocities of genocide

Rony Blum , Gregory H. Stanton , Shira Sagi , Elihu D. Richter
DOI: http://dx.doi.org/10.1093/eurpub/ckm011 204-209 First published online: 18 May 2007


Genocide has been the leading cause of preventable violent death in the 20th–21st century, taking even more lives than war. The term ‘ethnic cleansing’ is used as a euphemism for genocide despite it having no legal status. Like ‘Judenrein’ and ‘racial hygiene’ in Nazi medicine, it expropriates pseudo-medical terminology to justify massacre. Use of the term reifies a dehumanized view of the victims as sources of filth and disease, and propagates the reversed social ethics of the perpetrators. Timelines for recent genocides (Bosnia, 1991–1996, 200 000; Kosovo 1998–2000, 10 000–20 000; Rwanda, 1994, 800 000; Darfur 2002–2006, >400 000) show that its use bears no relationship to death tolls or the scale of atrocity. Bystanders’ use of the term ‘ethnic cleansing’ signals the lack of will to stop genocide, resulting in huge increases in deaths, and undermines international legal obligations to acknowledge genocide. The term ‘ethnic cleansing’ corrupts observation, interpretation, ethical judgment and decision-making, thereby undermining the aim of public health. Public health should lead the way in expunging the term ‘ethnic cleansing’ from official use. ‘Ethnic cleansing’ bleaches the atrocities of genocide, leading to inaction in preventing current and future genocides.

‘ “Ethnic cleansing” just did not seem to be hitting the mark … . We queried New York if … Rwanda could be labeled genocide … . Little did I realize the storm of controversy this term would invoke … in the capitals of the world. To me it [genocide] seemed an accurate label at last ….’

Lt. Gen. Roméo Dallaire (with Major Brent Beardsley), Shake Hands With The Devil: The Failure of Humanity in Rwanda, Carrol and Graf NY 2004: 333.


Twelve years after the Srebrenica massacre of 1995, the public health community needs to examine the strange history of the term ‘ethnic cleansing’, a euphemism for genocide, which has appeared in a proposed UN resolution against mass atrocities.1

Slobodan Milosevic, who died while on trial for crimes against humanity in the United Nations International Criminal Tribunal for the former Yugoslavia (ICTY), was the first politician to use the term in April 1987 to characterize Kosovar Albanian commanders’ violence towards Serbs. ‘Ethnic cleansing’ became the euphemism, first used by the perpetrators, and later by bystanders, to describe individual and mass killings, arbitrary extra-judicial executions, mass rapes, starvation, destruction of residences and religious institutions and expulsions.2 Before Srebrenica, Serbian commanders used the military code-words: ‘etnicko ciscenj’ (‘cleansing of the region’) and ‘ciscenje prostor’ or ‘terena’ (‘cleaning the territory’) for leaving nobody alive.3

From July 1991, journalists and politicians began adopting the term ‘ethnic cleansing’4 which gradually penetrated the official language of diplomacy and international law—with the implication that it applied to scenarios which somehow could not satisfy the legal requirement for proof of intent to commit genocide.5 The United Nations referred to the ‘new term’ of ‘ethnic cleansing’ in 1993, (la purification ethnique, nettoyage ethnique, or épuration ethnique in official French translations), using it in seven subsequent Security Council Resolutions.6 It is ironic that the UN itself adopted a euphemism invented by Milosevic, an accused perpetrator of genocide, despite its never having been formally defined or recognized as a term with specific legal status and mandated obligations, as genocide has been since the 1948 Genocide Convention.

Learning from modern history

Twentieth century genocide emerged from a lethal combination of social Darwinism, racist genetic theory and nationalism.7 Beginning with the Armenian genocide in the Ottoman Empire in 1915, newly emerging xenophobic totalitarian regimes committed genocide and political mass murder on a scale unprecedented in history.

The genocides of the last century have shown that propagation of an in-group exterminatory exclusivity based upon myths of hygiene or purity, and dehumanization of the other group, are warning signs of imminent genocide. The Young Turk regime in 1915 called the genocide of Armenians the eradication of ‘dangerous microbes’.8 The Nazi term ‘Judenrein’ in Western Europe, which means ‘Jew-free’, as well as the Russian term for ‘purges’ (‘chitki’),—now ‘ethnic cleansing’ (‘chischenie’)9—first stigmatized the victim group as a carrier of filth and disease, and then reified it as the disease to be eradicated. Hitler called the Jews ‘parasites, plague, cancer, tumour, bacillus, bloodsucker, blood poisoner, lice, vermin, bedbugs, fleas and racial tuberculosis’ on the German body that would supposedly be killed with the ‘Jewish disease’.9,,10 From 1937 to 1949, Stalin and Beria used the term ‘purge’ when deporting (‘korenizatsiia’) over two million members of ethnic minorities to slave labour camps in Siberia, at the cost of hundreds of thousands of lives: Soviet Koreans, Ingrian Finns, Karachays, Kalmyks, Chechens, Ingush, Balkars, Crimean Tatars, Crimean and Black Sea Greeks, Meshetian Turks, Kurds, Khemshils and the Jewish community to a Soviet-invented ‘Jewish homeland’ in Birobijian.11 In 1988, the Soviets used the term ‘ethnic purge’ (‘etnicheskie chistki’) to describe expulsions of Azerbaijanis from Nagorno-Karabakh. In Rwanda, Hutu radio in 1994 used the term ‘cockroaches’ (‘inyenzi’) to incite mass murder of Tutsis by machete-wielding militias.12

Totalitarians, whose power depended upon incitement, exploited ethically flawed and misconceived pseudo-medical theories to instigate forced sterilizations, selection, massacres and genocide.13 Prestigious northern European medical institutions fostered eugenic massacres that preceded World War II. Anthropologists and geneticists were the misclassifiers, public health experts were the mistaken theoreticians and coordinators, and physicians, especially psychiatrists, were the death ‘selectors’.14 These perpetrators disavowed their ethical responsibility for genocide at Nuremburg, where they claimed to have been mere ‘technical experts’, not decision-makers. The world medical community has been reluctant to recognize that many of its illustrious members legitimized massacre as a public health measure.15 Charismatic xenophobic leaders characterized vulnerable victim communities as ‘filth’ and ‘disease’ to motivate young male perpetrators. The collective delusion of their heroic role in sanitizing society increased group belonging while lessening personal responsibility.16,,17 Inflicting violence on resented communities while distributing plundered goods reinforced in-group complicity and commitment to this collective delusion.18 The role of ‘reverse jargon’ in reversing ordinary social ethics has been crucial to the genocidal agenda of the perpetrators and to sustaining in-group self-esteem.19 The term ‘cleansing’ ‘normalizes’ the delusion that massacres are measures to promote ‘hygiene’.19–20

The ICTY21 concluded that what happened in Bosnia was genocide, while Kosovo was described as ‘ethnic cleansing’ despite 11 000 dead in 529 mass graves, a systematic campaign to burn or destroy bodies of the dead to obliterate the evidence, the destruction of 1200 cities and towns, commander-organized rapes, castrations, violation of medical neutrality, enslavement, imprisonment in concentration camps, torture, enforced prostitution, slaughter of leaders and elites, and persecution on political, ‘racial’, and religious grounds.22 Recurrent motifs of sanitation and disease also surfaced as Albanian commanders called dark-skinned Roma ‘majupi’ (‘lower than garbage’), as hundreds of Roma were dying23 —only 50 years after a quarter to half a million of Roma perished from genocide in World War II.24

Death tolls and use of the terms ‘ethnic cleansing’ and ‘genocide’

Use of the term ‘genocide’ does not necessarily guarantee intervention. But does use of the term ‘ethnic cleansing’ signal less will or action to recognize genocide, and stop the perpetrators, than does the term ‘genocide’? To examine this question, we tracked trends in the use of the terms ‘ethnic cleansing’, ‘genocide’, and both terms simultaneously in New York Times articles in years 1990–2005.25 These 15 years included the genocide in Bosnia 1990–1995, with 200 000 dead, Kosovo 1998–1999, with 10 000–20 000 dead, Rwanda, 1994, with 800 000 dead, and Darfur 2003ff., now with more than 400 000 dead.26 We also tracked data on time trends in citation of these terms in the international legal literature,27 UN press releases,28 and statements by Amnesty International and Human Rights Watch.29,,30 (Spread sheets for references 25–29 are at www.genocidewatch.org.26, op cit)

Our premise is that the number of times the terms were cited (separately and together) roughly indicated the level of interest, but that the ratio between the terms—‘genocide’ and ‘ethnic cleansing’—measures the will for emergency response. We surmised that when both terms were used together, the term ‘ethnic cleansing’ conveyed the same sense of urgency associated with ‘genocide’.

Figure 1 presents data from word searches of the New York Times. From 1991, when Serbs began attacking Bosnians and Croats, the term ‘ethnic cleansing’ alone appeared with increasing frequency up to 1993,—3.5 times more than the term ‘genocide’. From 1994 and thereafter, through the Srebrenica massacre, until NATO bombing and the Dayton Peace Accords halted the genocide, use of the term ‘genocide’ greatly superceded that of ‘ethnic cleansing’—but not until 3 years elapsed and most of the 200 000 dead had already been killed in Bosnia. As the quote from General Dallaire noted, in Rwanda, there was a shorter, but more catastrophic, delay before UN officials and agencies recognized that the 3-4 months organized massacres were genocide. In contrast, the first recognition of genocide in Kosovo in 1998–1999 occurred more rapidly, with much lower death tolls. Thereafter, citation of the term ‘ethnic cleansing’ rose and fell sharply, but remained substantially less than citation of ‘genocide’. Indeed, intervention in Kosovo followed the use of the term ‘genocide’ by United States Ambassador for War Crimes Issues, David Scheffer,31 in 1999—a characterization for which he was criticized.32

Figure 1

Use of terms ‘Genocide’ and ‘Ethnic Cleansing’ in relation to cumulative death tolls in four genocides, New York Times: 1990–2006

In summary, (i) New York Times citations of the term ‘ethnic cleansing’ fell, while that of ‘genocide’ rose; (ii) the terms were usually mutually exclusive, (iii) use of the term ‘ethnic cleansing’ did not lead to, or promote, earlier recognition of genocide or shorter response times in former Yugoslavia and (iv) lower death tolls in Kosovo were associated with more rapid dominance of the term ‘genocide’ vis-à-vis ‘ethnic cleansing’. But in Darfur, with its far larger tolls of deaths and other mass atrocities, a more ambiguous response emerged. (see subsequently)

Figure 2 shows that citation of ‘genocide’ in the international legal literature (n = approximately 600) progressively increased in relation to citation of the term ‘ethnic cleansing’ (n = approximately 150) in 2004, (ratio 4 : 1). But use of both terms fell to 500 and 100, respectively as the Darfur genocide toll progressively increased in 2005.

Figure 2

Citations for ‘Genocide’ and ‘Ethnic Cleansing’: International Law Review Articles: 1990–2005

In UN press releases on Darfur,28 there were 438 citations of ‘genocide’ and ‘ethnic cleansing’ in 2004 and 2005—less than one half the total for Kosovo in 1998 and 1999 (n = 991), despite cumulative death tolls some five times greater than Bosnia and some 20-fold greater than Kosovo. Ratios of citations for ‘genocide’ in relation to ‘ethnic cleansing’ in United Nations release press releases throughout years 2002–2005 actually rose from 2.9 to 3.9 and then fell back to 3.2 and then 2.3. The ratio has been approximately the same during the Darfur years as during the Kosovo genocide in years 1998–1999 (3 and 2.1, respectively), Even so, during a period of peak interest in Darfur (April–Dec 2004), the total number of citations for ‘genocide’ and ‘ethnic cleansing’ was 181, or a mere 38% the number of citations for Kosovo in December–March 1998 (n = 478), and the ratios for citations for ‘genocide’ in relation to ‘ethnic cleansing’ were not appreciably different—3.1–1, and 2.8–1 for Darfur and Kosovo, respectively.

For human rights organizations, data on citations for the terms ‘genocide’ and ‘ethnic cleansing’ in relation to Darfur show a 5:4 ratio (153 to 128) for Amnesty International.29 For Human Rights Watch official reports on Darfur, the corresponding ratio was 4:9, (363–451) as compared with a 30-fold preference (27 000–898) for the term ‘genocide’ over ‘ethnic cleansing’ in all articles on its website.30 As late as November 2006, Human Rights Watch prominently featured the term ‘ethnic cleansing’ as the headline in large fonts on its home page illustration about Darfur, Sudan.33 Both Human Rights Watch and Amnesty International have failed to officially call the widespread massacres in Darfur ‘genocide’, thereby appearing to acquiesce in the official tolerance for the increased scale of atrocities (see subsequently).

These findings on the ambiguous response of the UN and human rights organizations in Darfur contrast with the increasing number of citations for ‘genocide’ in both the NY Times and international legal literature. In short, both the frequency, and ratio, in use of both of the terms by the UN and human rights organizations bears no relationship whatsoever to death tolls and the scale of atrocity.

The prominent genocide legal scholar William Schabas5 has asserted that ‘ethnic cleansing’ can never be genocide because the intent of ‘ethnic cleansing’ is to drive out a population, whereas the specific intent of genocide is to destroy it. But this distinction ignores the fact that genocidal massacres often have both intents. They intentionally destroy a substantial part of an ethnic group, the specific intent necessary to prove genocide, and also have the intent to terrorize a population into flight or forced deportation. The findings on the relatively greater use of the term ‘ethnic cleansing’ by the UN press releases and human rights groups suggest the hypothesis of possible UN intent—with the possible acquiescence of human rights groups—to stymie public awareness of genocide. In Darfur, this approach has relieved governments of their obligations under the Genocide Convention to stop the genocide.

A pattern of documentary evidence suggests that sometimes the UN has tried to sway global public opinion to deny the genocide supported by the Government of Sudan—which was the conclusion of the Coalition for International Justice investigation reported by Secretary of State Powell to the United Nations in September 2004. The State Department investigators concluded that the pattern of deaths, rapes, expulsions and pillaging indicated that acts of genocide were taking place.34 The subsequent UN investigation did not accept the conclusion of the US State Department investigation, thereby neutralizing the impact of Secretary of State Colin Powell's use of the term ‘genocide’ for triggering effective intervention. Totten provides evidence to indicate that major methodological flaws in the UN report were both contributory causes and consequences of prior intent to manufacture doubt about the conclusions of the US State Department report. This report had a rigorous study design, based on a sampling strategy for interviewing refugees from different parts of Darfur, a standardized protocol for questionnaires, training of interviewees, and a specified statistical analysis for producing a range of estimates of the number of deaths from violence and other causes, and their regional distribution. It also collected reports by victims that the Janjaweed had used racial derogatory epithets—an indicator of genocidal intent. In contrast, the UN study had none of the above requirements for a valid epidemiologic investigation. Totten reported that the director of the UN investigation seemed to be planting the notion in the minds of the investigators that he did not expect them to discover an organized pattern of targeted violence or mens rea (proof of intent).34

In Darfur, the use of the term ‘genocide’ in the press and legal literature has not until now ensured effective intervention to prevent genocide, perhaps because the events described by Totten suggest that a UN decision not to use the term virtually ensured non-intervention. The Report of the International Commission of Inquiry on Darfur35 erroneously concluded that the Sudanese government's motive of expulsion of insurgents relieved it of ‘intent’ to commit genocide. The Report was erroneous for two reasons. First, the unprovoked murder, enslavement and rape of abandoned women and children continued well after all insurgents had left the villages. Second, the Commission's legal smokescreen was directly contrary to the ICTY decision in The Prosecutor vs. Radovan Krstic, which reaffirmed that intent to commit genocide may be found in destruction of only part of a group, and does not require intent to destroy a whole group. The Krstic case also established that genocide can occur during civil war. The two are not mutually exclusive.36

Furthermore, use of the term ‘ethnic cleansing’, by the UN and human rights groups, even if well intentioned, such as in an essay recommending a peacekeeping force by Samantha Power37 in The New Yorker, reinforced a systematically misleading and mutually exclusive alternative term for genocide. The term, is associated with denial and delay in bystander response, and an increase in death tolls, seems to have produced a numbing effect, with ever-larger numbers of violent deaths, rapes, forced starvation, enslavement, plunder and a campaign of expropriation against black African communities. The term renders collective expulsions, even when accompanied by the above atrocities as ‘not genocide’ both in usage by the United Nations Security Council and the ICTY, first with, and then without, quotation marks, as noted by the genocide legal scholar William Schabas.5, op cit

In May 2006, there were more than 8 000 000 citations for ‘ethnic cleansing’ on Google, up from 221 000 citations in January 2006, and 76 000 in September 2005. The term, often used without quotation marks, has already penetrated the medical literature38,,39 (36 citations), including The Lancet.40 We ask: what would happen if a peer-reviewed article in a medical journal would have used the word ‘Judenrein’ without quotation marks just once as part of an objective technical description of the killing and expulsion of Jews from the Warsaw Ghetto during World War II?

Genocide and the ethical import of delay: Rwanda and Darfur

In Bosnia, Kosovo, Rwanda and Darfur, while diplomats debated, the perpetrators raped, pillaged and murdered. In Rwanda, avoidance of use of the term ‘genocide’ at the outset of the killing by official United Nations agencies was the diplomatic excuse for inaction that allowed the genocide of 800 000 people to occur over a period of 3 months in 1994.41 In Darfur, Sudan, in September 2004 there were 1.6 million crisis-affected people and an estimated 70 000 deaths by late 2004,42 that, as a result of the inaction noted earlier, has probably exceeded 400 000 deaths.43

Had the media, decision-makers and NGOs concerned with human rights immediately employed the terms ‘genocide’, ‘butchery’ instead of ‘ethnic butchery’ in Bosnia, Kosovo, Rwanda and Darfur, would there have been earlier intervention to save tens of thousands of lives? In Darfur, two epidemiological assessments inferred genocidal intent from a pattern of mass killing across space and time, the ethnic and racial backgrounds of victims, and statements made by perpetrators after mass rapes and atrocities against women and children were committed.44,,45 Neither used the term ‘ethnic cleansing’ even once, whereas ‘genocide’ appeared once in the former article, and 31 times in the latter.

The challenge to the public health community is to initiate and carry out rapid epidemiologic assessments when first reports of mass atrocities surface, no matter how fragmentary, using the models derived from epidemiologic investigations of epidemic disease. False negative reports and catastrophic delay46 result from the repression and suppression biases inherent when the evidence is buried e.g. hidden mass graves, destruction of evidence of identity, residence and ownership.37

Parties to the 1948 Genocide Convention undertake to prevent and punish genocide in keeping with Article I.47 In public health terms, prevention implies early warning systems, effective interventions, and punishment to help deter future perpetrators of genocide. Genocide in Rwanda, Bosnia and Darfur bespeak the catastrophic failures in applying the UNCG. Use of the term ‘ethnic cleansing’ conceals the failure to investigate, collect and report the evidence of genocide, and worse, to prevent it. The UN's failure to properly collect and report the evidence and the catastrophic consequences in Darfur—represents the most extreme example of a false negative.

These consequences state the case for action for applying the Precautionary Principle to expedite intervention to prevent or stop genocide. This Principle states that when there is uncertainty concerning the possibility of the occurrence of a major catastrophic event, the human costs of inaction prevent the outcome far outweigh those of preventive action. It shifts the burden of proof from those suspecting a catastrophic risk to those denying it. The ethics of public health lead us to conclude that invoking an approach based on this principle is preferable to invoking a euphemism that either promotes or excuses delay.48,,49

Genocide or prevention

In the 20th century, the death toll from genocide, massacres, forced starvation, expulsions and other atrocities is estimated to have exceeded 170 million.50,,51 The proportion of non-combatant deaths in wars has increased from 5% of the total death toll in World War I to 60% during World War II, to 80% in the civil wars of the 1970s and 1980s, and a large majority of the current 20 million refugees from war are women and children.52 In all instances of genocide since 1948, there has been shameful delay in response by the UN, regional alliances, and major powers to first reports of genocidal acts, despite immediate media attention.

Our data on the use of the term ‘ethnic cleansing’ relative to the use of ‘genocide’, shows that it bears no relationship to death tolls. This fact alone belies the claim that its use is for legal rigour. Use of the term ‘ethnic cleansing’ by the media, legal community, politicians, diplomats, NGOs and even medical experts, obscures perception of this alarming assault on human life and public health, and indeed, may well have become one more tactic to preempt public recognition of genocide.


We call on the medical world to lead the way in expunging the term ‘ethnic cleansing’ from use by the media, national and international governmental agencies, diplomats, legal bodies and human rights NGOs. Professional medical ethics should forbid the borrowing of dehumanizing euphemisms from perpetrators of genocide. The medical community must particularly reject a term that implies that genocidal mass atrocities ‘cleanse’ society of filth and disease.53

The term ‘ethnic cleansing’ adopts the distorted conceptions of its perpetrators as our own. Therefore, the case for expunging this term is more imperative than it was for ‘Judenrein’ and ‘racial hygiene’, since history has already forewarned us. The term ‘ethnic cleansing’ corrupts observation, interpretation and ethical judgment and decision-making, and lacks official legal status. It is inimical to the aim of public health. ‘Ethnic cleansing’ bleaches the atrocities of genocide and its continuing use undermines the prevention of genocide.


We thank Tamar Berman for editorial assistance and Professor Elliot Berry for advice and encouragement, and the anonymous reviewer for his helpful notes.

Key points

  • The term ‘ethnic cleansing’ has been used as a euphemism for genocide, despite its having no official legal standing.

  • As with Nazi medicine, the term expropriates pseudo-medical terminology to propagate the perpetrators' dehumanized view of the victim population as a source of filth or disease.

  • Timelines for recent genocides show the term's use bears no relationship to death tolls.

  • ‘Ethnic cleansing’ bleaches the atrocities of genocide and should be expunged from official use.


  • This paper is dedicated to the memory of the late Eric Markusen, who, shortly before succumbing to cancer, contributed thoughtful suggestions concerning the sequence of events regarding the failure to prevent genocide in Darfur


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 26.
  28. 27.
  29. 28.
  30. 29.
  31. 30.
  32. 31.
  33. 32.
  34. 33.
  35. 34.
  36. 35.
  37. 36.
  38. 37.
  39. 38.
  40. 39.
  41. 40.
  42. 41.
  43. 42.
  44. 43.
  45. 44.
  46. 45.
  47. 46.
  48. 47.
  49. 48.
  50. 49.
  51. 50.
  52. 51.
  53. 52.
View Abstract