Alison Palmer. Ethnic & Racial Studies. Volume 21, Issue 1. January 1998.
There are many ways in which societies made up of different cultures, ideologies, ethnic groups or social classes can live together in a social polity. These range from accommodation, through public integration (with civic integration and private freedom), different forms of assimilation (which may be voluntary or forced), to expulsion and population transfer. Beyond this lies genocide. Genocide is an extreme method by which a polity eliminates ethnic conflict.
Definition
From its original conception in 1944 by Raphael Lemkin, the term genocide has been problematic both legalistically and scholastically. This was compounded by the definition of the United Nations Convention on Genocide in 1948 which remains the only international standard of genocide. While I am aware of its conceptual and practical problematic definition, I have considered these issues at some length elsewhere and here will merely restate my own definition (Palmer 1994). Genocide is the intention to destroy physically a whole or a substantial part of a group because they are members of that group and whose membership is defined by the perpetrator, regardless of whether or not the whole group is actually destroyed. This includes cases where unintended genocidal consequences of action are noted by the perpetrator, and are continued or increased and where genocide is attempted but fails. The onus of the definition is thus upon the intention to destroy the group, regardless of either any further aims which might be achieved by such destruction or the success of the genocide.
Why do some states choose genocide and why do other states allow genocide? By exploring possible explanations of genocide, I shall point to particular combinations of factors which help to explain the instance and forms of different cases of genocide; demonstrate that genocides of colonialism can differ sufficiently to question the appropriateness of a category of ‘colonial genocide’; point to the importance of distinguishing between state genocides and societal genocides; underline that a general explanation of genocide is not feasible; and show that explanations from sources other than the two paradigmatic cases of genocide—the Holocaust and the Armenian genocide—are critical for understanding genocide in general.
The two cases of colonial genocide here examined are the Aborigines in Queensland, 1840-1897, and the Hereros of South West Africa [SWA], 1884-1906. The two modern cases are the Holocaust, 1939-1945, and the Armenian genocide, 1915-1923. In order to demonstrate that the two colonial genocides help us to understand genocides more generally, they will receive greater attention. It is assumed that readers will have greater familiarity with the Holocaust and the Armenian genocide. The destruction of the Aborigines lasted between approximately 1840 and the late 1880s, was piecemeal and predominantly a frontier activity, claiming between 8,000 and 10,000 lives (Reynolds 1982, p. 122) Both settlers and the Queensland government were involved. Settlers and the Native Police—a government agency—were the main agents of destruction acting with implicit, and at times explicit, state support. Settler killing of individuals and groups was geographically isolated. Methods included poison, large-scale punitive attacks and groups massacres. The genocide of the Hereros occurred between August 1904 and mid-1905, and was a systematically organized military offensive concentrated into a single effort. By 1905 the Herero population was reduced from 80,000 to 16,000, mostly ill and weak children, women and old people. All but 2,000 of these were put into concentration camps where a further 45 per cent died (Bley 1971, pp. 150-51; Katjavivi 1988, p. 10). Documents survive which demonstrate the explicit official intention to commit genocide.
There is no explanation from the victim’s perspective—this is an important lacuna (consider study of the Holocaust without the detailed perspective from Jewish survivors!), yet it was unavoidable because of the absence of survivor testimonies from Aborigines and Hereros. By focusing upon the perspectives of the perpetrator, however, I assume that the perpetrator is responsible for genocide.
Scholars of genocide have suggested a range of economic and sociopolitical factors which may be useful for explaining genocides. While I acknowledge that many fine works have been written on specific cases of genocide, I am concerned to show the relevance for genocides more generally of factors arising from colonial genocides. Subsequently, and unless indicated otherwise, the factors discussed arise from those suggested by others to apply to a category which includes genocides against indigenous peoples (Kuper 1984, p. 5), developmental genocides (Fein 1990, p. 86), utilitarian genocides (Dadrian 1975, p. 102; Smith 1987, p. 25) and genocides to acquire economic wealth (Chalk and Jonassohn 1990, pp. 36-37). I examine these in relation to the four cases.
Victim Group is Excluded from the Universe of Moral Obligation
In all four cases the victim group was practically excluded from the collectivity to whom the state and its associated community considered itself to be responsible. Distinction must be maintained between formal statements and legislation, and their implementation, since in both colonial cases the victim group was officially included within the realm of the state’s responsibility, whereas in practice it was systematically excluded. For example, the Royal Instructions to the Governors of Queensland clearly stated that they were responsible for the welfare of Aborigines (QVP 1860, pp. 451-57). Aborigines were British subjects and equal to whites before the law (Evans 1984, p. 193), yet Aboriginal witness was inadmissible in courts of law and Aborigines were subjected to severe punishments for trivial offences (Bridges 1973, p. 101; Lumholtz 1979, p. 374; Harris 1982, pp. 22-31). Aborigines were not protected and little action was taken to ensure their safety. The Native Police remained the principal government policy for the remainder of the century (QVP 1861, pp. 389-579; QVP 1876, vol. 1, p. 904; Whittington 1964-65, p. 508; Barnett 1975, pp. 9, 22-23; Evans 1988, p. 57). Interracial marriage, whilst not illegal, was abhorred (Trollope 1968, p. 72; Hartwig 1972, pp. 15-17; Loos 1982, p. 171; Evans 1984, p. 200).
Similarly, Hereros were subjected to German protection treaties which stripped them of traditional power and political rights and failed to protect them in inter-tribal wars. Judicial discrimination abounded and racial segregation was ensured by confining the Hereros to tribal areas (Bley 1971, pp. 140-41; Goldblatt 1971, p. 136; Drechsler 1980, pp. 133, 151-52, 168). Formally, the protection treaties respected African traditional laws, yet Article 4 of the treaties stated that any dispute between Germans and Africans would be brought before German courts. In practice, Africans were included within German colonial law and were denied the opportunity to defend their rights using traditional law (Bley 1971, p. 102). Seven African witnesses were required to equal one white witness and it was rare for German abuse of Africans, including rape and lynching, to reach trial (Drechsler 1980, pp. 133, 151-52, 168). Such was the strength of settler support for harsh sentencing of Africans—invariably death sentences—that in 1903 Governor Leutwein protested to the Hereros that he was powerless to intervene (Goldblatt 1971, pp, 135-36; Drechsler 1980, pp. 151-52). Many whites who refused to ill-treat Africans were socially ostracized by the European community (Bley 1971, pp. 111, 123; Drechsler 1980, p. 136).
Importantly, the implication that formal inclusion within the community of moral obligation warrants state protection is not upheld by the Queensland case: the formal inclusion of Aborigines permitted the government both to maintain that the Native Police death squads were acting to protect Aborigines, and to turn a blind eye to the atrocities they committed. On the other hand, a formal proclamation of 2 October 1904 by General yon Trotha, the German authority in SWA, excluded the Hereros from colonial responsibility by removing their status as German subjects (Bley 1971, p. 168). This permitted the Hereros to be labelled as an external threat to the German nation.
Despite similarity, two important distinctions can be made between the colonial and modern cases, based upon the extent to which the victim group is integrated into the perpetrator group prior to their exclusion; and the manner of exclusion (detailed, systematic and individuated legal processes or practical processes of en masse exclusion). Both Jews and Armenians were subjected to processes which officially excluded them from a society of which they had been a part for thousands of years (Bauer 1990, pp. 344-54; Hovannisian 1990, p. 251). They were longstanding members of societies in which there had been intermarriage and religious conversion, and economically, industrially and professionally there was much co-dependence. Perpetrator and victim had the same government (Chalk 1990, p. 324; Hovannisian 1990, p. 251). They had been subjected to long-term discrimination, including violence and massacres, and within the millet system of the Ottoman Empire, Armenians were treated as a second-class minority (Bauer 1990, p. 349; Hovannisian 1990, p. 251). Both were minority groups. The exclusion of the victims from the larger community was a formal, systematic and, for the Jews, a detailed process. In Nazi Germany, each individual right was stripped away by new, specifically designed legislation, and was methodically implemented by a pervasive bureaucracy (Bauer 1990, pp. 344-54). Immediately prior to the Armenian genocide, Armenians were systematically disarmed, demobilized from the army and minority rights and individual liberties were denied (Melson 1990, pp. 268; 280).
In contrast, neither Aborigines nor Hereros had a long tradition of coexistence with the perpetrators in a plural society; and the period of ‘contact’ between the perpetrators and victims had been relatively short. This suggests that longer integration requires detailed, systematic and more individualized formal exclusion. Where there is a weak state and the killing is societal led, it may serve the authorities to maintain a stance of formal inclusion in order to deny overt participation in or responsibility for the killings; to maintain a facade of ‘acting in the best interests of’ the victim group while being thwarted by ‘circumstances beyond their control’, for example, civil disturbances, environment catastrophes, etc; or to deflect external condemnation.
Occupation of the Victims’ Space
In all cases the perpetrators sought to occupy the physical space occupied by the victim group. Occupation may be significant for the form that it takes or as an objective of the perpetrator.
In Queensland, it was both the specific form of occupation which was important—pastoralism in a territory used for hunting and gathering in which cloven-hoofed animals were absent, and large-scale alluvial mining; and the individualized objective of economic gain, as opposed to a collective objective. Pastoral occupation occurred between 1840 and 1884, by which time it had consumed most of the colony south of the 13th parallel (Meston 1895, p. 32; Fitzgerald 1982, p. 146; Rowley 1986, pp. 175, 179). Three main ‘rushes’ of occupation (the early forties, the early 1860s until 1866, and the early 1870s until 1884) swept northwards and westwards. Pastoral legislation resulted in the overstocking of runs, overgrazing and land exhaustion. Together with the belief in an unlimited supply of empty land, this fuelled the common abandonment of runs and the increasing seizure of Aboriginal lands (McKinnon 1940, p. 94; Wright 1981, p. 71).
If the rich alluvial mining for gold in North Queensland from 1867 saved the Queensland colony from economic collapse, perhaps it was this that signalled the end for many Aboriginal people. It not only brought in wealth, but cemented the colony by inaugurating the establishment of roads, townships and ports. It also stimulated sudden massive and unprecedented population influxes in country where there had been little previous contact between Aborigines and Europeans (Loos 1982, pp. 62-63).
Both of these extractive forms of economy required the removal of Aborigines from their land (Brown 1974, p. 88; Hughes 1975, pp. 31-42; Loos 1982, pp. 68-79). Land legislation from 1846 permitted Aborigines ‘free access’ to leased land and to water, flora and fauna necessary to survive (Reynolds 1992, pp. 150-51). Yet most squatters excluded Aborigines from their runs, believing that they would exploit any familiarity with European homesteads to organize violent attack against people and livestock (Bennett 1928, Preface; Rosewarne 1976, pp. 23-24), and that cattle which even smelt an Aborigine would stampede (Reynolds 1974, pp. 155-6) and that Aborigines were unable to resist attacking livestock. Although many pastoralists depended upon Aboriginal labour for the survival of their industries (May 1983, pp. 52, 55; Reynolds 1990, p. 216) among others it was more common to drive them from the runs through use of violence, ‘dispersal’—a euphemism for killing—and the Native Police (Meston 1895, p. 93; May 1983, p. 96). Indigenous fauna were treated as vermin competing with livestock for scarce food and water and were shot or poisoned, while the native flora was decimated by livestock with disastrous effect upon Aboriginal food supplies (Bennett 1928, pp. 22-29; Reynolds 1974, pp. 155-56; Thorpe 1986, pp. 22-29).
Mining was largely a private initiative (Fitzgerald 1982, p. 177) in which individual miners spread over large areas to pan gullies, waterways were extensively used and denied to Aborigines (Brown 1974, p. 88; Kirkman 1984, pp. 263-64) and Aboriginal foodstuff was destroyed causing widespread starvation (Loos 1982, p. 74). Miners’ reliance upon violence to deal with Aborigines was supported by the government who advocated the use of guns and the Native Police to ensure that mining areas were kept clear of Aborigines (Corfield 1923, pp. 90-92; Brown 1974, p. 88; Hughes 1975, pp. 31-32, 42; Loos 1982, pp. 68-70; Kirkman 1984, pp. 243, 295).
The importance of an extractive form of colonial economy is witnessed also in SWA. German farmers sought to settle in Hereroland and develop a pastoral industry based upon the acquisition of Herero cattle. This was achieved through establishing a credit-based system of trading with the Herero for which their cattle were used as security. In 1903 the system collapsed and Herero cattle were seized as repayment (Bley 1971, pp. 137-38; Drechsler 1980, p. 132). Additionally, the Hereros’ fear that the Germans would take over their entire territory prompted the great uprising in 1904 in which the Herero tribes united in an offensive against German settlers and military (Bley 1971, p. 143). The genocidal response of the German military was to prevent the reoccupation of the land by the Hereros.
The extractive form of occupation which required the exclusion of the victim group does not by itself explain why genocide was pursued rather than expulsion. This requires consideration of additional factors: perpetrator resources and the size, density and geographical location of the victim group (particularly whether the group is demographically dispersed or concentrated), in relation to the perpetrator resources. The expulsion of a dispersed or concentrated victim group requires strong perpetrator resources: thus, where a perpetrator has these facilities, both expulsion and genocide are options. (They need not be alternative options: the genocide of the Hereros was a form of expulsion, since they were driven into the Omeheke desert to die of thirst. A military cordon was maintained for over one year to ensure that they could not return to their homelands (Drechsler 1980, p. 156)). Where the victim group is concentrated and the resources of the perpetrator group are weak, both expulsion and genocide are unlikely outcomes. However, where the victim group is dispersed and the perpetrator resources are weak, expulsion remains unlikely while genocide becomes an option. The Queensland case illustrates how a weak state can contribute to genocide as a substitute for expulsion. Insufficient resources available to the colonial settlers and Queensland government, an undeveloped state apparatus, and dispersed perpetrator and victim groups meant that Aborigines first needed to be reduced in number, and amassed at the fringes of white settlement, and for the state apparatus to be developed in structure and manpower, before the survivors of the genocide could be expelled to reserves or off-shore islands at the turn of the century. Genocide does not require the victim group to be collected into large groups as a preliminary procedure, which may be a feature of state led genocides. Expulsion is also more likely to be an option where the group has a definite homeland which is outside the territory claimed by the perpetrators—as was the case with the Pacific Islanders and Chinese in Queensland whose entry was increasingly restricted and who were subjected to legislation of exclusion to encourage their voluntary departure from the late 1880s (Fitzgerald 1982, pp. 248-56; Cronin 1988, pp. 314-16; McConnochie et al. 1988, pp. 75-78).
The modern cases are qualitatively different from the colonial situations on at least two counts. Occupation was based upon ideological demands rather than on economic factors (Weinstein 1980, pp. 117-161; Dadrian 1989, pp. 253-54, 267-72; Melson 1992, pp. 169; 1990, pp. 283). On the issue of homeland, Armenians were removed from an area larger than their homeland (Dadrian 1989, p. 262), a homeland also claimed by Turks (Hovannisian 1990, p. 260); and while the issue of a homeland of European Jewry is vexed and complex, it is difficult to speak of their removal from a homeland in this context.
Victim Group as Obstacle
In all four cases the victim group was perceived to be both an obstacle and a threat to the perpetrator society, and in all cases they were viewed with fear. Yet there are important distinctions in terms of the nature of the obstacle, what they were an obstacle to, and the processes used by the perpetrators to overcome the obstacles.
Aborigines and Hereros were obstacles to European occupation by the fact of their prior occupancy of land and resources and their resistance to European possession. In Queensland, this obstructed the individual economic success of miners and pastoralists (Reynolds 1982, p. 98-113). Colonists’ fear of Aboriginal attack at the frontiers was, at times, a greater obstacle for their success (Brown 1974, p. 93; Reynolds and Loos 1976, pp. 216, 218, 225; Evans 1977, pp. 49-50; Loos 1982, pp. 35, 75, 82, 84, 87; Breslin 1986, pp. 141-42). A dispersed population reliant upon hunting and gathering predisposed Aborigines to a nomadic lifestyle and precluded the possibility of large and sustained attack. Aboriginal clubs and spears could not match European guns. Aborigines had no tradition of fighting to kill or of organized and large-scale warfare with strategies, leadership and supplies (Reynolds 1982, pp. 96-98; Berndt and Berndt 1988, p. 358). Although Aboriginal resistance was successful for short periods, such as at the mining town of Gilberton, 1869-1873, these factors mitigated against lasting success (Brown 1974, pp. 104-105). Colonists dealt with these obstacles in a variety of ways, some of which were intentionally fatal for Aborigines, for example, individualized and collective deadly punishment of Aborigines, the Native Police, large-scale organized retributive attacks (Corfield 1923, p. 91; Reid 1980-81, pp. 70-73; Reynolds 1982, pp. 78-82), pushing Aborigines from their land and denying Aborigines access to waterholes.
The Herero resistance in 1904 was, on the other hand, both real and formidable and posed an obstacle to the colonial administration. It opposed German discriminatory policy towards Africans which undermined traditional power structures, encouraged tribal enmities and sought to restrict them to two reserves. The uprising clearly challenged the initial colonial practice of negotiation backed up by armed force (Bley 1971, pp. 52-53, 61,143;.Drechsler 1980, p. 93).
While the perception of the Hereros and Aborigines as a threat was based predominantly upon practical issues, ideology played a far greater role in the Jewish and Armenian situations. Indeed, the ‘final solution’ for European Jews was kept secret from them during the first ‘concentrating’ period (Bar-on 1971, p. 229). The extreme form of Nazi antiSemitism identified Jews as the ultimate biological and political threat to the German race and nation, to the Aryan ‘race’ and to the rest of humanity (Bauer 1990, pp. 339-40). This is vastly different to the perception of Hereros as a possible threat to the German nation because they attacked a few hundred civilians settled in a distant land claimed by Germany. The Armenians fall between these two, since both ideology and practical factors culminated to portray them as threatening and feared obstacles to the Turkish government. The Young Turk government sought to build up its crumbling identity and empire in a new form as a homogenous, ‘Ottomanized’, Turkish-speaking nation from which minority groups would be expelled (Dadrian 1989, p. 253; Melson 1990, pp. 276-83). As long as they remained within Turkish territory, Armenians were an obstacle to this overriding objective. Further trouble came from the Armenians’ geographical location adjacent to and spreading into Russia, from where irredentism was being urged. The Young Turk’s ‘fear’ that Armenians might either side with Russia against Turkey in the 1914-1918 war or seek secession was unfounded, but both possibilities threatened their nationalist plans which viewed the Eastern provinces as essential Turkish territory (Melson 1990, p. 283).
Additionally, while Aborigines represented obstacles to individuals and economic success, the Armenians, Jews and Hereros were seen as obstacles to the whole perpetrator group and to state authority. This may be a further distinction between societal and state genocides.
Fein shows how exclusion and obstacles might combine to facilitate a genocidal outcome of group relations (1984, pp. 11-13). Genocide is a likely response in situations where two or more groups are stratified, and the subject group is denied citizenship or effective participation in the polity, or a formal structure is imposed which underlines inequality between the groups, such as colonial official and native subject. These situations invariably give rise to increased enmity between the groups and the consolidation of intra-group identity in opposition to the other group,is Any violent or non-violent challenge to the continued domination of the governing group is perceived by these members as a threat to the whole group, which has the potential to ultimately lead to their total loss of power in a zero-sum situation. These challenges warrant punishment, which trigger counterchallenges; the process continues in cyclical momentum, in which punishment and threats are randomized and held collectively by the groups. Fein maintains that these retributive genocides are usually distinguished by a conscious intention to get rid of the threat, through annihilation, expulsion or a mixture of both, and by the supervisory role of the state in executing the @collective punishment’ by the governing class.
The value of this argument is twofold. It points to the increasing perceptions by the dominant group that the threat is collectivized – in terms of assuming both a unity among the subordinate group and that the threat applies to the whole of the dominant group – as is the punishment, which is meted out to any of the subordinate group. Secondly, it points to a process in which genocidal relations might emerge.
Yet is does not explain why annihilation, rather than expulsion, is seen as the solution. The collectivized and randomized punishment by the dominant group must be to end the challenge of the victim group. At what point, then, is this punishment successful? Why, in some instances, does retribution lead to genocide? The answer lies partially in the changing salience of the perpetrator belief system (which may or may not adhere to a zero-sum scenario); the extent to which the perpetrators continue to believe that the challenges continue, which in turn depends significantly upon the nature of the challenge and the perceptions of the perpetrator (that is, perceptions of violent challenges may have nothing to do with the actual actions of the subordinate group); and the extent to which the punishment is state executed or enacted by civilians. Within the process of punishment, the intention of the perpetrators may change as the symbolic or real threat posed by the subordinate group decreases. In a protracted process of punishment, a climax might arise in which the collective annihilation of the victim group is seen as the best solution, yet this might rescind as the threat to the dominant group’s power declines. This scenario would apply to the Queensland case where the intentional killing of Aborigines decreased as the colony became more established. Alternatively, the other three cases represent situations in which the annihilative climax was intensified by a strong perpetrator state which maximized the possibility of quick annihilation. A strong state permits a rapid and much more total solution to a perceived zero-sum situation, whereas a weak state permits a protracted solution in which societal led genocide may represent the climatic point at which the situation is perceived to be zero-sum, but which can accommodate other, non-fatal, solutions.
Perpetrator Ideologies and Popular Perceptions
In all four cases, ideologies or beliefs were present which were exclusivist and totalizing, seeking to expel certain elements in order to homogenize specific societies. A totalizing view of society encompassed within an ideology or popular perceptions may endorse or guide policies and actions which exclude a potential victim group from the perpetrator community. In order for an ideology or popular beliefs to move from exclusion to annihilation, they must assert certain additional elements. They must advocate that conflict with the victim group is necessary; that the perpetrator’s victory and the victim group’s extirpation is predetermined by a suprahuman force, be it a divinity, science, history or nature; and that both issue and outcome are moved into a simplistic sphere of morality (for example, good and evil, conflict is the enactment of human destiny or is necessary for the advancement of all humankind, including the victim group) which perpetuates the conviction that the perpetrators are acting in a fully justified manner that is of service to a larger community. Genocide, then, is seen not as the selfish act of the specific perpetrators but is ‘just’. Ideologies and popular perceptions also portray the victim group as being outside or on the borders of humanity, that is, it dehumanizes them, thus easing the potential moral pain of the perpetrators.
From the 1850s and well into the twentieth century, social evolutionism was the most influential and popular racist doctrine in Queensland (Reynolds 1989, p. 114). It combined a number of strong contemporary views: belief in the notion of social development, progress and the law of natural selection as well as the ethnocentrism of Western colonialism. Competition over resources, or with nature, resulted in conflict between different cultures: the law of nature determined that the people with the highest culture, white Western civilization, would inevitably succeed, whereas those with the lower cultures would inevitably fall away and become extinct (Hartwig 1972, pp. 15-17) center dot H. K. Rusden demonstrated this view in 1876:
… when exterminating the inferior Australian … race. … The world is better for it; and would be incalculably better still, were we loyally to accept the lesson thus taught by nature, and consistently to apply the same principle to our practice; by preserving the varieties most perfect in every way, instead of actually promoting the non-survival of the fittest by protecting the propagation of the imprudent, the diseased, the defective, the criminal. Thus we surely lower the average of, and tend to destroy, the human race almost as surely as if we were openly to resort to communism (cited in Hartwig 1972, p. 16).
Aborigines were commonly viewed by whites as ‘vermin’, ‘like animals’ and ‘ferae natura’ (The Queenslander, 12 June 1880, p. 27; Evans 1984, pp. 201-202). As part of the hostile environment which not only threatened the physical survival of Europeans but also contained the potential to reverse European conceptions of nature, history and progress, Aborigines were considered a most dangerous biological specimen warning of the stagnating effects an impoverished environment could have upon a people (Smith 1960, pp. 240-41; Griffiths 1996, pp. 10-12).
Likewise, the German settler community sought to emphasize its racist superiority in relation to the Africans in SWA and to maintain its separateness from them. Some settlers likened Africans to baboons (Drechsler 1980, p. 167); they were considered inferior, lazy, ‘a barrier to progress’ and those who refused to submit unconditionally to colonialism ‘were not entitled to exist’ (Stoecker 1986, pp. 209-12). In both countries, the racist ideology and popular perceptions militated against full assimilation: racial interbreeding and children of mixed parentage were thought dangerous to the dominant, ‘superior’ white race. They were ‘useless hybrids’ who led to the deterioration of the superior qualities ascribed to whites (Bolt 1971, pp. 9-10; Stoecker 1986, pp. 211-12).
The territorial and cultural nationalism of the Young Turks and the nationalistic racism of the Nazis similarly located Armenians (Hovannisian 1990, p. 261; Dadrian 1993, pp. 190-94) and Jews outside the dominant political communities. Nazism portrayed Jews as the ultimate threat to the German nation, to the Aryan race, to humankind in general and to Nazi plans for political dominance. In contrast to Aborigines and Hereros, they were not dehumanized to the level of animal or environment, but to that of disease and parasite whose continued existence would kill the German host upon which they lived (Bauer 1990, p. 348). Enormous lengths were taken to define who was or was not Jewish and to implement these definitions (Hilberg 1990, pp. 359-61). Turkish nationalism considered history to be a ‘ceaseless struggle of one nation over another … which might end in the physical or cultural annihilation of ‘foreign bodies’ who lacked the same moral and cultural qualities of Turks. National objectives were infused with moral rectitude (Melson 1992, pp. 165-69).
There are some important distinctions to be made between the four cases. First, the earlier ideologies or beliefs were directed at peoples who were obviously outside the perpetrator society: in almost every aspect of their lives they were different, and they were geographically located in a land far from the perpetrators’ homeland. Ideologies and popular attitudes could build upon these psychological, physical and cultural differences. The later ideologies, on the other hand, were directed at a people who were integrated within the perpetrators’ own society, differences in culture and language were not so great, some Jews and Armenians had converted from their old faiths, and there were few overt physical distinctions.
Greater integration may well warrant more extreme ideologies to segregate and disempower the victim group, as is argued by Dadrian (1974, pp. 123-35). This is the second distinction between the earlier and later cases, which were far more concerned with defining the conditions necessary for the nation’s survival.
Of the four cases, the Queensland case warrants special distinction on two counts. One cannot speak of ideologically driven killing of Aborigines, nor even of the presence of a strong ideology. Rather, the perpetrators acted within a compote of popular beliefs and pragmatic concerns of survival in a harsh environment. Secondly, the perpetrators in Queensland did not act to save the nation—nationalism was distinctly absent during this period. This contrasted to SWA where the rise of the particular brands of German nationalism imbued SWA and the Herero rebellion with specific and new significance. Nationalism elevated the armed forces to the position of divinely appointed protectors of the nation (Snyder 1935, pp. 60-63), located the true German spirit within the romanticized peasant farmers now seeking to perpetuate the German crusade against barbarism in a distant land which, of all the colonies, was intended for German settlement, (Gann and Duignan 1977, pp. 43, 49; Snyder 1978, p. 138), and claimed that ‘Wherever a German man lies under the sod, fallen in the faithful discharge of his duty to the Fatherland … that land … is German and will remain German’ (Snyder 1935, p. 65). If one concedes that nationalism is a primary loyalty which binds a people to a state, then the absence may be significant to explain the lesser involvement of the Queensland state and British forces, and the lesser identification of Aborigines as an insidious threat to the colonial collective, distinct from a more individualized process.
The International Context and Victim Group Vulnerability
The international context in which genocide occurs is vital for its understanding. Particularly, it helps to explain the vulnerability of the victim which may be judged by the extent to which there is support for the victim group from outside the state (Harff 1987, pp. 43-44) or the possibility of external sanctions or retribution against the perpetrator should it engage in genocide (Dadrian 1974, p. 131). A situation of war is particularly useful for perpetrators of genocide as a facilitating condition, since it serves to obscure the visibility of its genocidal operations; it obstructs external powers from imposing sanctions against genocide; and it allows genocide on the basis that the victims are described as enemies of the state (Fein 1984, p. 7). War also concentrates power in the hands of a centralized authority that for Dadrian is a necessary condition for genocides, which are aimed at the total obliteration of a group—optimal genocide, including the Holocaust and the Armenian genocide (1975, p. 102).
Britain’s apparent lack of concern for Aborigines can partially be explained by her desire to maintain economic and naval dominance among other European states. After the bitter and expensive loss in the American Wars of Independence, Britain sought to avoid both the loss of Queensland’s economy and conflict with a colony seeking more and more independence, and so from 1859 increasingly relinquished control over matters internal to the colony, including relations with Aborigines (Bailey 1933, p. 398; Melbourne 1933, pp. 277, 295; Gill 1972, pp. 17-18; Loos 1980, p. 165). This contrasts with Britain’s readiness to intervene on behalf of Chinese and Pacific Islanders working in Queensland, to maintain important economic relations with China and to pay lip-service to Exeter Hall’s claim of paternalist concern of ‘native’ subjects where this was not too costly (Robinson 1972, pp. 72-78; Loos 1980, pp. 162-63). London received little information on the fate of Aborigines, reducing the need for any action. Occasional despatches requiring information on allegations of violence, etc. (OVP 1875, vol. 1, p. 621; Evans 1971A, p. 27; Gill 1972, p. 15) were not effectively followed up, suggesting that the Colonial Office had little incentive or interest in intervention.
Germany’s willingness to engage in genocide in SWA might be viewed as a demonstration to European states of her military power and intolerance of threats to the German nation. Germany’s colonial expansion in the late nineteenth century demonstrated the international strength and dignity of the new nation-state (Gann and Duignan 1977, p. 42; Snyder 1978, p. 90). Hence it is possible that the military crushing of the Hereros in 1904-1905 had instrumental value to demonstrate to other colonial powers, particularly Britain, that Germany possessed both the power and ability to deal effectively with any threats to its nation and to protect its own interests.
In neither colonial case did war serve to obscure genocide or to obstruct international sanctions to end genocide. In SWA, the war was genocide, and the Hereros were proclaimed to be enemies of the German nation (Drechsler 1980, p. 173).
These factors show that events in the international scene preoccupied states which could have intervened and allowed them to turn a ‘blindeye’ to the atrocities; and that changing relations between the major international powers appear to have facilitated the genocides. So also with the modern cases. International concern in 1878, 1895 and 1913 over the lack of reforms to end discrimination against Armenians was deeply resented and largely unheeded by the Turkish government (Dadrian 1989, p. 252); while the crumbling of the Ottoman Empire and the loss of large amounts of territory and population to European powers or to newly independent states threatened the power and identity of the Turkish state, and placed the Armenians in a vulnerable position. Fear that Armenians would ally with Russia and relinquish Turkish control over the Eastern provinces propelled the Turk’s Armenian policy (Melson 1990, p. 283; Melson 1992, p. 157). German support for Turkish nationalist plans gave the genocide added momentum (Lang 1988, pp. 13-15; Hovannisian 1990, p. 259). The failure of Allied intervention to prevent the Nazi slaughter is well known (Hilberg 1990, p. 365).
Strong Political or Military Structure of Victim Group
Three factors did not fit the cases. The suggestion that the possibility of genocide will be reduced where the dominant group recognizes within the subordinate society a political structure with clear forms of leadership with whom they can negotiate (Howe 1977, pp. 4-7) was not borne out. The strong Herero political organization actually contributed to their emergence as a formidable obstacle to German colonization, and required drastic solution. Aborigines did not have a state or central leadership. However, the colonizers’ belief that the land was unused or uncultivated (terra nullius) and the particular form of their racism against the Aborigines precluded the possibility of recognizing any Aboriginal political organization or leadership and so also the possibility of ethnic accommodation through negotiation. The use of Jewish leaders in the process of selecting who would be sent from the ghettos to the camps makes mockery of this as a preemptive possibility (Bar-on 1971, pp. 234-36). Jews were also a diaspora with no unified representation.25 Armenian leaders were singled out for individualized deaths (Dadrian 1989, p. 266). The strength of this factor for prevention of genocide more generally must lie predominantly in the military imbalance between the dominant and subordinate groups, in conjunction with other factors: clearly a stronger military structure and a strong international ally would make a potential genocidal perpetrator think twice before pursuing a final solution.
Victim Group as Labour
No case supported the argument that genocide will not occur where the dominant group needs the subordinate group as necessary labour (Price 1950, p. 14; Kuper 1981, p. 46; Ternon 1989, p. sadta; Bodley 1990, pp. 42-46). In the colonial situations the genocides continued despite this need being widespread. Aboriginal labour as guides, shepherds, stockworkers and drivers was crucial for the development of the colonial economy of Queensland throughout the second half of the nineteenth century (May 1981, p. xi; May 1983, p. 55; Thorpe 1985, p. 35; Reynolds 1990, pp. 19, 204-12, 217-24),26 suggesting that societal genocide need not be exclusive of other forms of (exploitative) ethnic relations. German farmers in SWA urged that the Hereros be punished but not exterminated on the basis that their labour was necessary for the future development of the settler economy, a need validated by an acute post-genocidal labour shortage (Goldblatt 1971, pp. 147-48; Drechsler 1980, p. 162).
Absence of Organized Religious Intervention
In none of the four cases did intervention by religious organizations of the perpetrator group act to prevent genocide. Church intervention was minimal and ineffective in both Queensland and SWA. In Queensland, missionary work was scarce, under-resourced and unsupported by the colonial government until the late 1880s, that is, largely after the genocide, when missions served-to maintain racial segregation and a pool of labour for white employers (QVP 1896, vol. 4, p. 734; Evans 1972, pp. 28-29, 32; Maguire 1984, pp. 58-59). German Lutheran missionary activity in SWA predated colonial rule: between 1870 and 1890, eighteen missions had been established specifically to work with Hereros. Again, they received little support from the colonial or imperial authorities. Missionaries acted as mediators in attempts to establish reserves in the 1890s and condoned the use of force against Hereros when it served their own interests (Bley 1971, pp. 114-115, 170; Drechsler 1980, p. 114). While they protested that the main cause of the 1904 Herero uprising was the unprovoked German assaults, they were unable to prevent the subsequent genocide (Goldblatt 1971, pp. 39-40; 74; 140). There was no significant organizational preventative Christian and Moslem intervention for the Jews or Armenians. Although Armenians could convert to Islam, few were saved, while exemptions for Armenian Catholics and Protestants were seldom applied (Moranian 1992, p. 115).
Factors Arising from Modern Genocides
Just as attempts have been made to locate genocide of indigenous peoples within a single category, so have similar attempts been made for modern genocides. The aim of such categorization must be to aid explanation through comparison, yet comparative explanations for ‘modern’ genocides as a general category are thin and undeveloped. I turn now to factors which are attributed specifically to modern genocides. Their brief examination shows how they are of limited usefulness for understanding why genocides occur.
The Role of the State
It is commonly argued that genocide is authorized by the state. Kuper (1981, p. 161) and Fein (1984, pp. 25-26) maintain that this is particularly so in the twentieth century, since the doctrine of state sovereignty implicitly allows the state to pursue any form of ethnic control within its own territory, thereby denying the implementation of the nineteenth-century doctrine of humanitarian intervention. However, others point out that in certain circumstances the state may not be overtly involved in genocide but might give tacit consent to the genocidal conduct of local authorities. These circumstances include the local use of retributive violence against a segment of a minority group which is challenging the majority group the scale of the killing is limited because the victim group is not perceived as a high threat, the perpetrators have limited resources and the victim group has power in other locations (Dadrian 1975, p. 101); or instances of ‘xenophobic genocide’, where genocide is a private initiative, based upon an ideology which stresses that a particular minority group needs to be annihilated (Harff and Gurr 1988, p. 368).
While the Holocaust, the Armenian genocide and the German genocide in SWA were explicitly authorized by the state, in Queensland alone the state government gave tacit consent to local authorities—many of whom were prominent frontier figures whose own financial interests were at stake (Barlow 1987, pp. 61, 66)<SUP>—civilians (Kirkman 1984, p. 295) and the Native Police. The minority-majority question here remains open to contention—European settlers became a numerical majority only in the 1860s (Caldwell 1987, p. 26; Price 1987, p. 2)—and the weight accorded to ideological motivation in Harff and Gurr’s ‘xenophobic genocide’ is not upheld. The crucial distinction in this instance would appear to be between state authorized and instituted genocides and societal led genocides.
Totalitarian State
It has been suggested that genocide is more likely where political power is concentrated in a centralized authority, and where there are few or no mechanisms by which it might be restrained (Horowitz 1980, p. 51; Fein 1984, p. 7; Harff 1987, pp. 41, 50; Bauman 1989, p. xiii; Rummel 1994, pp. 5-13). Thus, genocide is more likely in a totalitarian society and less likely in a democratic one. This distinction does not apply to the colonial cases: neither was totalitarian, and the colonial state structure of a tenuous administration system was extremely undeveloped. Aside from the need to distinguish why some totalitarian states use genocide whereas others do not, and what precipitates their genocidal policies to occur when they do, this strengthens the argument that a totalitarian state structure is not a necessary condition for genocide per se. More significantly, it substantiates the argument that democracies such as Queensland (albeit here in embryonic form) can engage in genocide.
Crisis of the State
A further proposed necessary structural condition is a period of upheaval, such as successful coups, revolutions and anti-colonial rebellions, in which the political structure of the state is changed or weakened (Harff 1986, p. 183; Harff 1987, pp. 43-59). State genocidal policy may be used to ensure obedience to the new political elite (see Harff and Gurr 1988, p. 360). A crisis of state may reinforce the importance of a state ideology as the basis for reconsolidating national integrity, distinguishing groups who are then placed outside of the realm of state or community obligation as potential victims of genocide (Fein 1979, p. 92; Fein 1984, pp. 4-5).
Neither colonial case experienced a pre-genocidal political crisis of such magnitude that it seriously challenged the state authority, or required a new political elite using a state ideology to consolidate its power. The challenge of the Hereto uprising was not sufficient to precipitate such a crisis. Although it was a political crisis to the civilian administration, it was easily crushed by the German military and did not threaten the. political authority of Germany.
Conclusions
While one wishes to avoid making claims of generalization in seeking to understand why genocides occur, it is equally important to avoid arguing that each case is unique and so beyond comparison, since this denies the opportunity of either searching for explanations of why genocides occur and, importantly, whether and how they might be prevented in the future. This discussion serves to highlight a number of important points.
First, it demonstrates that explanations which have not emerged from the modern, European cases of genocide are vital for further understanding the emergence of genocides both generally and throughout history, since they focus upon a different range of factors and their various conjunctures. It shows that understanding colonial genocides is important for shedding light upon our understanding of modern genocides. It also points to a number of factors (exclusion of victim group from universe of perpetrator’s moral obligation, perpetrator occupation of victim’s space, perpetrator perception of victim group as obstacle, presence of exclusivist and totalizing perpetrator ideologies or beliefs, and absence of a significant external ally) which may apply to all genocides—their substantiation requires consideration of a larger number of cases than those discussed here. These do not necessarily suggest a general theory of genocide, but may be useful in identifying conditions and processes which appear to fit in a wide range of cases, providing bases for comparison and contrast to other genocidal and non-genocidal cases of ethnic conflict. They might also provide a basis for ordering high and low possibilities of groups at risk. They may provide a structural backbone or skeleton which is substantially fleshed out in each case. Importantly, they may provide a means to answer two basic questions: what can be learned from individual cases of genocide, and how can one breach the gap between recognizing the uniqueness of each case and attempting to understand genocide as a specific form of ethnic, political, religious, economic, etc. conflict?
There are some differences between the two colonial and two modern cases. These are the extractive economic occupation of the victims’ land and the underdeveloped perpetrator state structure in the victims’ territory. These do not constitute a basis for conflating all cases of colonial genocide within a single category. The latter point, in particular, had different consequences for the two victim groups. I have suggested that the weak Queensland state structure permitted the private killings to continue not only unabated but also with government encouragement, as the colonial state itself lacked sufficient resources to ensure the safety of those members of its community prioritized for protection, that is, whites. The absence of effective British intervention was due to lack of interest, the international context and the absence of sufficient symbolic investment by Britain in Australia. In the African case, the inability of the colonial administration to cope with the Herero rebellion triggered the arrival of strong military force. In contrast to Queensland, the settlers did not wish to engage in widespread killing of the Herero who were deemed important for their labour. This was not a society involved in genocide. The explanation for the military genocide lies, in this case, with the international context within which existed relations between Germany and her colonies, and the symbolic meaning invested in SWA by the particular forms of German nationalism at that time.
State structures thus remain a critical factor for understanding genocide and this discussion crucially points to the important distinction between societal genocides (Queensland) and state genocides (Holocaust, SWA , Armenians). Secondly, it underlines the importance for genocide of ideologies and beliefs, over and above state structures. In particular, genocide requires the perpetrators to adhere to a totalizing ideology or collection of beliefs in which the victim is not only outside, but a dehumanized enemy, imbued with the potential to threaten the survival of the perpetrator; and in which the perpetrator victory in inevitable conflict is assured and morally justified. This emphasizes the necessity of applying this line of investigation to other cases of genocide.