![]()
Part 1 Theories![]()
Chapter 1
From Definition to Process
The Effects and Roots of Genocide
Benjamin Lieberman
No other branch of history or field of inquiry centered on historical events is so dependent on a definition as genocide studies. Studies of war, politics, wealth and poverty, society, culture, men and women, and a host of other topics have all given rise to detailed analysis of terms and definitions; but no other field depends for its very existence upon the invention and definition of a single term. Extermination, mass killings, and massacres were all well known long before the creation of the term “genocide,” but there was no field of historical study devoted to examining and comparing such massacres and killings from different historical epochs. Histories, whether of the Roman Empire, the Mongols, the Pennsylvania frontier, or the French Revolution, often discussed killings or massacres, but there was no field of history dedicated to massacres or mass killing as such.
Genocide studies only became possible with the search by the legal scholar Raphael Lemkin for a single concept “to encompass the phenomenon of existential killing.”1 In the interwar period, Lemkin began to investigate past instances of mass killing, but he did so without a unifying term to describe his research. Only during the Second World War did he coin the word “genocide” to capture the meaning of a phenomenon that he had already noted. Many other concepts of historical analysis, including race, nation, class, and gender, cause intense debates over proper definition, but none of them was invented by a scholar seeking to describe something already perceived. Multiple books, volumes of scholarly journals, seminars, conferences, and encyclopedias debate the terms “nation” or “nationalism,” for example, but neither nation nor nationalism were defined on their very first appearance on the historical stage.
Lemkin logically sought to define the term he had invented. Genocide, he explained, meant “the destruction of a nation or an ethnic group.” Specifically, he referred to “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.”2 Within a few years, the newly-created United Nations had anchored the crime of genocide in international law in the 1948 UN Convention on the Prevention and Punishment of the Crime of Genocide. The Genocide Convention defined genocide as a set of acts “committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group as such.” The short list of five acts began with the most obvious, “killing members of the group,” but also included the infliction of severe physical and mental harm, along with “conditions of life calculated” to destroy the group, impeding the reproduction of the group, and transferring children of the group.
As research into genocide proliferated from the 1980s onward, debate over the term’s definition also intensified. A host of authors elaborated on the UN definition, many offering their own amended definitions. This lengthy and often fruitful discussion cannot be recapitulated here. In general, scholars focused on the flaws of either an overly elastic or an overly restrictive definition. The list of protected groups under the Genocide Convention struck some observers as too narrow: was there no room for the destruction of political groups? To take just one such critique, Totten and Parsons “believe that both political and social groups should be included in any definition of genocide.”3 A revised definition could also include groups defined by sexual identity and gender, though some elements of the original UN Convention cover some targeting according to gender (such as the prevention of births), and international case law developed by such tribunals as the ICTY (International Criminal Tribunal for the former Yugoslavia) has refined and added detail to our understanding of this aspect.4
A definition-driven analysis of genocide creates a common strategy for research. Whatever their methodology or approach, genocide scholars typically begin with a definition. They next identify a sample of cases that meets that definition, then describe and analyze the diverse cases, and conclude by tracing similarities among those case studies.
Individually and collectively, comparative works on genocide have advanced knowledge of violence and victimization; but their shared approach also creates deep and well-worn paths that obscure key conceptual problems and occlude new directions for inquiry. Such definition-driven research tends toward predictable conclusions. We learn that perpetrators dehumanize victims; that genocide may be motivated by hatred of targeted groups; that the effort to destroy a race in whole or in part often reveals intense racism; that perpetrators – when they can – use the instruments of state power to achieve their ends; that genocide in modern states displays key features of modernity, while genocide in the ancient or early modern world does not. None of these conclusions is wrong, though some may be incomplete; but they tend repeatedly to confirm the original hypotheses. At the same time, definition-driven research has spawned practices and assumptions that at the very least require greater scrutiny. These include issues of sampling, historical approach, images of perpetrators and victims, and methods for preventing genocide. Above all, genocide research assumes that genocide as an outcome lends unity to the field, without actually investigating how such a starting point might shape conclusions, or demonstrating that shared outcomes mean that genocide is a single phenomenon.
Sampling and Boundaries
Definition-driven genocide research typically confronts sampling problems. From the start, most discussions of genocide have recognized that genocide existed long before the term was invented and defined. Describing the concept in his 1944 book Axis Rule in Occupied Europe, Lemkin called genocide a “new word … to denote an old practice in its modern development.” The United Nations Convention of 1948 similarly stressed the antiquity of genocide, stating “that at all periods of history genocide has inflicted great losses on humanity.” In other words, genocide predates the invention and definition of the crime of genocide by centuries if not millennia, and the field of genocide studies by and large takes this as a given. Even works that focus on the twentieth century make this point, as Eric Weitz does in observing that “Genocides have occurred since the earliest recorded history” – though he suggests they became more deadly during the last century.5 The debate over whether the Holocaust was unique seldom leads to the extremist and counterfactual position that genocide has only occurred once. Historians – and for that matter, poets, playwrights, and novelists – have long known and lamented that history is full of dreadful instances of massive killing. This means that any comparative analysis of genocide confronts an immense diversity of examples drawn from different time periods, geographical locations, and cultures, employing different techniques in widely varied political and social settings. The writer who chooses to compare any other subject of historical inquiry from ancient Rome to modern Europe, Latin America, Asia, and Africa would confront an extraordinary task; but this is exactly what historians of genocide undertake, if they choose to see the destruction of Carthage as genocide, and expand the analysis to the Holocaust along with mass violence in Guatemala, Cambodia, Rwanda, and Darfur, to take only some of the many examples.
Facing this bewildering array of potential cases and examples, some studies of genocide seek to reduce the sample set. The most common approach is to take a particular time period, such as the twentieth century, or to concentrate on only the most important examples. Only a few enterprising works attempt a comprehensive approach. Either approach shapes likely conclusions. The more expansive the sample in geographical space and historical time, the harder it is to find a convincing common theme or thread. Interpretations of genocide based on broad samples run the risk of isolating the peripheral or highlighting the very general. Two of the most impressive recent comprehensive histories, Ben Kiernan’s Blood and Soil and Mark Levene’s Genocide in the Age of the Nation State, highlight these challenges. Kiernan considers four major themes, including “the preoccupation of perpetrators with race, antiquity, agriculture and expansion.”6 But factors that prove significant in one case of genocide, such as a cult of antiquity, may have little obvious connection to mass killing in another instance. Where did such a cult rank among all the important causes of the Armenian genocide, for example? And a desire for agricultural land would hardly explain why Nazis targeted Jews for extermination in areas of Eastern Europe where, for the most part, they were proportionally less likely to be landowners than other ethnic and religious groups. Levene, for his part, notes the development of “what we think of as the specific phenomenon of genocide primarily through the historical transformation of human societies worldwide as a politically and economically interacting and universal system of modern – mostly nation – states.”7 The point is a reasonable one: a territorially-bounded state that represents a single nation may endanger populations within those borders that do not belong to the state’s chosen nation. But the connection to genocide is still notably indirect. Many such states exist, but only a small minority of such states carries out genocide.
The use of a narrower sample, on the other hand, introduces risks of another kind. Isolating a small number of cases creates a hierarchy of genocide, with those deemed essential at the top. While he mentions other cases, Manus Midlarsky in The Killing Trap, for example, focuses on the Holocaust, the Armenian genocide, and the Rwandan genocide, on the grounds that these three cases “meet the criteria of state-sponsored systematic mass murder of a targeted ethnoreligious group and of non-combatant status of the victims.”8 Presumably, any broadly authoritative study of comparative genocide would include these cases. But such sampling brings with it a danger of circular logic. The principles used to select the essential cases of genocide will most likely turn up in whatever conclusions are drawn about the core common factors, causes, and effects of genocide. If state sponsorship of mass killing is a key criterion for selecting the sample, we would expect the analysis to confirm the importance of such state sponsorship. The criteria for selection derived from “essential” cases also lead to the exclusion of cases that are widely recognized as genocide. Thus, Midlarsky excludes the mass killing of Herero in German Southwest Africa “because of the absence of state policy authorizing genocide, and the status of virtually all Herero including women and older children at least as combat associates,” though he does refer to “genocidal behavior.”9 Stressing such criteria lends a veneer of science to the decision to leave out the Herero case, but the criteria chosen seem almost arbitrary, given the actual effects of the assault on the Herero. If criteria derived from a total of three “essential” cases can so easily produce such results, then perhaps the criteria themselves require revision, or alternatively should be derived from a larger sample; but even a larger sample would still likely betray a circular logic.
Another approach to comparative genocide potentially avoids some of the problems of arbitrary sampling by focusing on genocide within a given time period – most often the twentieth century. Such studies still rely on examples chosen according to a particular definition, but at least they offer a common rationale for inclusion. More than one study has referred to the twentieth century as a “century of genocide.” However, even with this chronological focus, surveys of twentieth-century genocide share the same difficulty in locating common factors for killings carried out in extremely varied settings. If one identifies the twentieth century with genocide, it is hardly surprising to discover that powerful modern states are responsible for genocide. The conclusion is plausible; but is this a real finding, or again the result of initial assumptions shaping conclusions? Take the case of Rwanda, cited in all comparative studies of twentieth-century genocide written after 1994. It is possible to find elements of modernity in the Rwandan events: killers rifled through official records of identity to sort out Tutsis from Hutus and find victims, though they sometimes killed Hutus as well as Tutsis, and propagandists employed the modern technology of hate radio to whip up animosity. At the same time, how many of the authors who stress the modernity of killing in Rwanda would have cited the country, one of the poorest and least-developed in the world, as emblematic of modernity prior to the genocide of 1994? Does a society only become modern after it has s...
