Genocide is the most extreme denial of the right to self-determination. It is a twentieth-century expression for an old and odious practice, namely, the partial or comprehensive extermination of a distinct people. Genocide is relevant to self-determination controversies for at least four major reasons: the international convention describing the crime of genocide establishes the right to existence of groups protected under the convention, who, in turn may make self-determination claims (Thornberry: 1991 59-85); national, ethnic or religious groups that have suffered genocide, or claim to have done so, may invoke the crime to justify a remedial right to autonomy or secession; thirdly, neighboring, regional, or great powers, or the UN Security Council,  may justify humanitarian intervention in response to evidence of genocide, overturning the presumptive sovereignty of the accused host state; and, lastly, the perpetrators of genocide sometime invoke both self-determination  rationales against groups whom they believe threaten the security of their peoples.

      Raphael Lemkin coined the word genocide (Lemkin: 1944; Lippman: 1985). His neologism combined the Greek noun γενοσ, “genos,” meaning tribe or race, and the Latin verb, caedere, meaning “to kill.” Lemkin, a Polish jurist, had been appalled by the treatment of the Armenians by the Young Turks. He lobbied hard for the global outlawing of similar crimes. He was the key agent behind the adoption of the United Nations Convention of 1948 that outlawed genocide in international law (see Schabas: 2000 14-101 for a full legal analysis; and Power: 2002 Chs. 1-5).

      Article II of the Convention on the Prevention and Punishment of the Crime of Genocide, which entered into force in 1951, defines the crime as, “any of the following acts committed with intent to destroy, in whole or in part, a national, ethnical 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.”

      This wording is not easy for the layperson to interpret. Any of the named acts from (a) to (e) severally or jointly is an act of genocide, provided that the relevant specific intent is present. No other acts are named in the convention as genocidal, e.g. ethnic expulsions, war crimes, crimes against humanity, or gross human rights violations. The convention specifies other crimes, however, notably conspiring to commit genocide; inciting it; attempting it; and of being complicit in it. It also places an obligation on states to prevent and punish the crime.

      The intent to destroy the group “as such” confines the charge to those to whom such precise intent can be convincingly attributed. By legal logic, the execution of any or all the acts named in (a) to (e) without this proven intention does not constitute genocide. A government, for example, might plea that it was targeting rebels not an ethnic group “as such”–even if the rebels’ support group was confined to one ethnic group.  Also contrary to a widespread popular belief, the perpetrators of genocide need not display the ambition to destroy the whole group wherever its members may be found on the planet; the convention leaves open to judges and others to determine the threshold at which genocidal acts begin.

      Those protected through Article II are reasonably clear, namely, national, ethnic or religious groups. There are debates, however, about how such groups may be legally defined (objectively or subjectively). But what is vital to understand is that in international law it is not genocidal to kill large numbers of people who are not covered by these protected categories. Those who belong to a language community, to a social class, or who share specific political beliefs, cannot successfully claim that campaigns to exterminate them are acts of genocide. Governments pursuing linguistic assimilation, e.g. France and Turkey, and governments which did not wish to be charged with killing class or political enemies, such as the USSR under Stalin, successfully maneuvered to restrict the scope of the convention’s protected groups. They did so even though it was clear from the beginning that the genocide convention would not have retrospective application (a guarantee that was essential to ensure the passage of the Convention through the United Nations). Again, it is important to avoid misunderstanding: it is almost certainly unlawful, both in domestic and international law, to destroy or to aim to destroy large numbers of individuals belonging to groups which are not national, ethnical or religious, but such actions are not genocide according to the 1948 convention.

      The Convention’s definition of genocide, which however flawed, constitutes international law on the matter,  has been criticized from two quite different directions (see inter alia Kuper: 1981 19-39; Chalk: 1989; Jonassohn: 1992; Horowitz: 1997 (1980); Jones: 2006). On the one hand it has been accused of being too restrictive: as just shown, class, political and other actual or alleged identity groups (e.g. homosexuals) are not expressly protected. Under the relevant provisions it is also not obvious that slavery is intrinsically genocidal, even though the actions entailed in measures (b) through to (e), as well as periodic killings, have accompanied all cases of large-scale exploitation of slave labor. Genocidal acts may well accompany enslavement, but slavery is only genocidal under the Convention’s provisions if the victims belong to one of the protected groups, and are being targeted for partial destruction with deliberate intent because of their membership of one of the protected groups. Some writers object to including slavery within the category of genocidal acts because masters wanted to keep their slaves alive (e.g. Ignatieff: 2001); others observe slavery’s sustained and destructive effects on distinct groups.

      The exclusiveness of the definition (and of the groups protected under the Convention) has led some authorities to argue that genocide should be defined as “a form of one-sided mass killing in which a state or other authority intends to destroy a group, as that group and membership in it are defined by the perpetrator” (Chalk and Jonassohn: 1990 23; a similar argument was advanced earlier by Drost: 1959). This approach would incur its own problems as a legal definition because it would appear to make many wars, at least toward their conclusions, and many episodes of battle, liable to the charge of genocide. It would also appear to exclude the possibility of reciprocal genocidal ambitions among conflicting parties. There is, however, widespread agreement among non-lawyers that genocidal destruction has to include systematic mass-killing, even if such killing may have spontaneous origins, or past precedents in pogroms and deadly ethic riots. There is much less agreement on whether the goal has to be to destroy the relevant group in whole or in part. There is, however, convergence on the importance of “asymmetry.” During genocidal actions the victims are at least temporarily vulnerable, i.e. defenseless, poorly armed, disarmed, disorganized, ghettoized or incarcerated. There is also widespread agreement that genocide may occur through systematic indirect physical destruction of a vulnerable collectivity through the deliberate termination of the conditions which permit its (biological and cultural) reproduction.

      On the other hand the Convention has been accused of being too inclusive. It may be difficult to establish objective standards to assess serious “mental harm” to a group, when the other elements of (a) through to (e) are absent.  It has been suggested that the Convention conflates attempts at annihilation with collective punishments, and that it conflates “coercive assimilation,” i.e. when people are obliged to acculturate or fuse with others, with physical annihilation. Such critics want to confine genocide to mass killing. In extremis, their position would be that if the people survive, but not their culture, that there has been no genocide. Lemkin would likely have argued to the contrary; he considered the distinctiveness of the crime to be the destruction of a distinct group, culturally as well as physically.  The United States was very late to ratify the UN convention and incorporate it within its domestic law in part because Senators (usually from the Southern states) feared that the segregation and assimilationist schooling of Native Americans rendered the US liable to being charged with the crime (Power: 2002).

      Debate also occurs over whether the crime of genocide overlaps with war-crimes. Insurgencies which target members of certain protected groups may look on the initial face of the  evidence to be genocidal; certainly such claims are made against counter-insurgencies which target the national, ethnic or religious communities presumed to support insurgents. The prosecutor of the International Criminal Tribunal has indicted the President of Sudan for genocide; a warrant has been issued for his arrest. The Khartoum government maintains, however, that it is conducting a counter-insurgency in Darfur (for different perspectives on this case see De Waal and Stanton: 2009). Critics of “modern” warfare, argue that aerial bombardments, which foreseeably generate large-scale civilian casualties, are genocidal. The same claim has been made about the use of nuclear weapons; some argue that they should be outlawed precisely because their use would abolish any distinction between combatant and non-combatant targets.  Article 1 of the Convention commits the contracting parties to regard “genocide, whether committed in time of peace or in time of war,” as “a crime under international law which they undertake to prevent and to punish.” It therefore insists that genocide can occur within war, though it does not specify whether war includes rebellions, or revolutions. Governments running counter-insurgencies typically claim that civilians are being killed as collateral damage in the restoration of order, not deliberately, and they blame rebel groups for making civilians into vulnerable hostages. Insurgents frequently claim that governmental authorities are committing genocidal atrocities. 

      A litany of related concepts have been inspired by the genocide convention. “Politicide,” is defined as mass killing for political reasons; “ethnocide,” is used to describe the eradication of a group’s culture while leaving its members’ bodies and minds intact; “linguicide” describes the deliberate attempt to eradicate the use of particular languages; and “democide” has been coined to cover the killing of “citizen bodies” and to encompass genocide, politicide and mass murder (Rummel 1997: 31-42). Rudolf Rummel distinguishes genocide (see Figure 1) as the mass killing of peoples by governments who assume that the relevant peoples have “indelible” features (Rummel: 1997: 31).

      Scholarly and polemical debates surround the incidence and causation of genocides. High caliber research on genocide began with the work of Raul Hilberg (2003, first published in 1961), which used careful archival work and Franz Neumann’s model of totalitarianism (1944) to show how the Nazis organized the systematic destruction of European Jews. His work has inspired many claims and counter-claims. One school of thought argues that genocide is distinctively modern. Zygmunt Bauman is perhaps the best-known advocate of understanding genocide in this way. He emphasizes the role of bureaucratic ‘rationality,’ science and technology, in the emergence of what is now widely known as the Holocaust, i.e. the destruction of European Jewry (Bauman: 1989). Critics suggest that Bauman misreads and over-generalizes Hilberg’s work; fails to examine long-run historical evidence; and neglects the importance of national, ethnic or religious hatred (see e.g. Bauer: 2001). Others observe that the genocidal killings of Tutsis by Hutu (and vice versa) in both Burundi and Rwanda did not rely on modern science and technology–though radio broadcasts were important to incite and organize killings  (Lemarchand: 1970; Lemarchand: 2009). Other “modernists” link the incidence of genocide to gruesome processes of exclusionary nation-building, which they also think differentiates our times from that of the ancients. This argument figures heavily in accounts of the genocide of the Armenians by the governments of the Committee of Union and Progress (the young Turks) in late Ottoman and early Republican Turkey (see e.g. Akçam: 2006; Dadrian: 1995; Melson: 1992). A variation on this theme interprets genocides and ethnic expulsions as the “dark side of democracy” (Mann: 2005; for a reply see Laitin: 2006), which again is seen as a modern development. Others, by contrast, insist that, “democratic states do not commit genocides because, if they did, we would not call them democratic” (Jonassohn: 1998 135). Critics of those who take too rose-tinted a perspective on democracies observe that colonial genocides against indigenous peoples in the United State and Australia occurred under parliamentary governments (see e.g. Brantlinger: 2003; Lindqvist: 1996; Moses: 2000). It is true, however, that many of the mass killings and genocidal acts against indigenous peoples were executed by settlers without governmental authorization.

      The drafters of the Convention, by contrast, assumed that genocide and related crimes have occurred throughout all phases of human history, in all inhabited continents, at all times, a position supported by many contemporary scholars (e.g. Chirot and McCauley: 2006). There is powerful evidence to support this position (Chalk and Jonassohn: 1990 xvii, Parts II and III; Jonassohn: 1998 passim; Kiernan: 2007). The Holocaust executed by Nazi Germany was certainly distinctive in its horrors, but was not without recent precedent (Friedländer: 1997). Hitler infamously asked, “Who remembers the Armenians?” The Tasmanian population had long since been thoroughly exterminated. Historians emphasize major massacres in history–for a pioneering effort to catalogue and estimate humanity’s worst 100 hundred atrocities see White (2011). However, the “uniqueness of the Holocaust” is a scholarly and political controversy that shows little sign of abating (Katz: 1994; Rosenbaum: 2001). The topic is a subject in comparative genocide studies, but not for those who insist on its alleged incomparability.

      A subtler question is more difficult to assess, partly because of the inadequacy of the historical record, and partly because of the attendant difficulties in agreeing comparative data about particular cases. The issue is the intensity through time of the incidence of genocide. Have genocides occurred at approximately the same frequency throughout world-history? Have genocides and related mass killings become more frequent because of the greater technical capabilities available to governments and militias, or because of certain other attributes of capitalism and modernity (e.g. Davis: 2001)? Or, are genoicdes falling in number because of the defeat or dissolution of the major Nazi and communist mass-killing regimes of the twentieth century, and the end to large-scale inter-state wars (which make killing less easy either to hide or justify)? Steven Pinker (2011) emphasizes the slow albeit uneven success of civilizing, enlightening, liberalizing and democratizing processes in reducing our collective propensity toward mass violence. But it will be some time before others share his confidence that the first half of the twentieth century was the last major peak in human violence toward other humans.

      One survey by Barbara Harff in the American Political Science Review describes the following places as sites of genocides or politicides between the end of the Holocaust in 1945 and 2002 (Harff: 2003):

• China, in Tibet, especially in 1959;

• Indonesia, which targeted Chinese people as well as Communists in 1965;

• Guatemala, between 1978 and 1996, when the Maya, and rebels, were killed en masse;

• Uganda, in 1972-79, and again during 1980-86, witnessed numerous ethnic atrocities, halted by the overthrow of Amin’s regime by Tanzania in the first case, and in the second case by the consolidation of the Museveni regime;

• Cambodia, under the Khmer Rouge regime (1975-79);

• Burma, especially in 1978; its governments have frequently carried out atrocities against non-Burmans;

• South Sudan, under assault from the Khartoum government and its sponsored militias after 1983;

• Iraq, especially during 1986-91, when the Baath regime attacked Kurds and Shiite Arabs;

• Burundi, in 1988 and again during 1993-94, with previous precedents;

• Bosnia and Herzegovina during 1992-95; &

• Rwanda in 1994

        It is possible to argue that Harff’s list should be expanded, to include, for example, the destruction of the Marsh Arabs after 1991 in Saddam’s Iraq; and the counter-genocide during and after 1994, executed against Rwandan Hutus, both in Rwanda and the Congo.  Subsequent to Harff’s study, other genocidal massacres have been highlighted both in the Congo and in Sudan’s Darfur region (for an unpersuasive rebuttal see Mamdani: 2010). Many small indigenous peoples have suffered genocide or, minimally, gross human riots violations since 1945, but usually their fate does not enter into studies focused on large numbers, and partly because reliable data may not be available. That said, no one could credibly deny the cases that Harff highlights.

        Developing a risk model and testing it against statistical data drawn from worldwide regimes in the latter half of the twentieth-century Harff argues that certain factors predispose places to genocide and politicide. Notable among these are previous genocides in the relevant place; previous major political upheavals; exclusionary ideological orientations among the ruling elite; undemocratic regimes; and dominance by an ethnic minority (e.g. the Sunni Arabs of Iraq). Separately, she finds that places with economies with more open trade are at reduced risk of genocide and politicide, though no clear causal mechanism is suggested to explain the latter correlation. Harff’s large N study may be contrasted with the case-derived arguments of Helen Fein who maintains, driven largely by the cases of the genocides of the Armenians and the Jews, that a necessary condition of genocide is that the target group be defined as outside the moral universe of obligation of the dominant group. She suggests that a severe loss of national prestige is important in triggering genocidal ambitions among powerful agents: perpetrating states may have been recently humiliated by defeat in war, or diminished by internal strife, generating a quest for vengeance and scapegoats. New elites may rise with new formulae to justify the dominant nation’s domination or expansion. The calculus of the costs of exterminating the likely victims changes if the perpetrators instigate a war with antagonists who have earlier protested, or might be expected to protest against, the persecution of the victim-group. Under the cover of war the planning of genocidal crimes is less visible, and fears about sanctions recede in importance (see e.g. Fein: 1993; Fein: 1993; Fein: 2001).