• Genocide;
  • displacement;
  • extermination;
  • elimination;
  • mortality;
  • migration


  1. Top of page
  2. Abstract
  3. Introduction
  4. Explaining extermination and elimination in Darfur
  5. Explaining state-led elimination by displacement in Darfur
  6. Displacement and the atrocities documentation survey
  7. Findings from survey interviews and proportional hazard models
  8. Conclusions
  9. Bibliography

The millions of survivors who fled from attacks to Sudanese-controlled displacement camps and the refugee camps in Chad are the living ghosts of the Darfur genocide. The 1948 Genocide Convention incorporates extermination by mass killing and elimination through forced migration as two distinct elements of genocide. Genocide scholars and public discourse emphasize extermination by killing, but they give far less explanatory attention to the elimination processes that the Genocide Convention describes as ‘deliberately inflicting on the group conditions of life calculated to bring about its physical destruction’. (Article II United Nations 1948.) In Darfur, understanding the latter processes requires theoretical attention to the history of food, water, and famine and detailed methodological attention to temporal processes of displacement. We demonstrate how intentional state-led attacks on food and water massively dislodged Black Africans in Darfur from February 2003 to August 2004. The political leadership of the Sudanese state dehumanized and forcibly displaced Black Africans from their homes in Darfur to camps where they largely remain, not only through mass killings and rapes, but also by destroying life-sustaining access to food and water, leading to the genocidal elimination of group life in this region.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Explaining extermination and elimination in Darfur
  5. Explaining state-led elimination by displacement in Darfur
  6. Displacement and the atrocities documentation survey
  7. Findings from survey interviews and proportional hazard models
  8. Conclusions
  9. Bibliography

There is increasing convergence and growing confidence in estimates of the mortality and forced migration associated with the conflict that began in 2003 and is still ongoing in Darfur. The estimates are that from 200,000 to 400,000 Darfurians have died (Hagan and Palloni 2006; Degomme and Guha-Sapir 2010) and that from two to three million Darfurians have been involuntarily displaced from their homes (UN 2005). The death toll speaks to the issue of the partial extermination of groups in Darfur, while the displacement numbers address the prospect of their elimination from Darfur (de Waal 2005). The International Criminal Court [ICC] identifies the Black Africans who are the predominant victims of this death and displacement as members of the Zaghawa, Masalit, and the Fur ethnic groups, while this Court identifies the perpetrators as heads of Arab Janjaweed militias and leadership figures in the Government of Sudan, including President Omar al-Bashir and then Deputy Minister Ahmad Harun (Office of the Prosecutor 2008).

The 1948 Genocide Convention defines genocide as ‘acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group’ (United Nations 1948: Article II) including in addition to killing, ‘deliberately inflicting on the groups conditions of life calculated to bring about its physical destruction in whole or in part.’ (United Nations 1948: Article II) This definition identifies both mortality and forced migration – death and displacement – as key elements of extermination and elimination constituting genocide. The inclusion of extermination in the Genocide Convention definition is an obvious consequence of the context of its drafting in the shadow of the Nazi Holocaust. The further reference to elimination is a potentially farther reaching element of this convention that makes the meaning of genocide relevant to a wider range of conflicts than a singular reference to extermination would allow. The element of elimination expands the meaning of genocide beyond immediate wholesale killing to include the intentional creation of physical and social conditions leading to the destruction of individual communities, as well as in multiple communities and whole nations.

Even though the ICC Chief Prosecutor, Luis Moreno-Ocampo, resisted until 2008 calling Darfur genocide, he none the less had already remarked that ‘this strategy has been seen before’ (Office of the Prosecutor 2005: 93). He explained what he meant with an example from the conflict in the former Yugoslavia, noting that,

In March 1995, President of Srpska Radovan Karadzic . . . , issued Directive 7. It specified that the Republika Srpska was to ‘by planned and well-thought out combat operations, create an unbearable situation of total insecurity with no hope of further survival or life for the inhabitants of Srebrenica’ The parallel to Darfur is clear.

The International Criminal Tribunal for the former Yugoslavia (UN 2001) had previously found that genocide had occurred in the former Yugoslavian town of Srebrenica in the mid-1990s. This judgment is a precedent for Moreno-Ocampo's analysis by citing both evidence of the selective killing of young adult (‘fighting age’) men and the forced displacement of women and children, which made it ‘impossible for the Bosnian Muslim people of Srebrenica to survive.’ (Office of the Prosecutor 2005) The killing combined with forced displacement to eliminate Bosnian Muslims from Srebrenica, ending an era of group life in this community. Prosecutor Moreno-Ocampo did not seek a genocide charge against President Al-Bashir of Sudan for the massive crimes in Darfur until 2008, and the Pre-Trial Chambers of the ICC only finally authorized a warrant for the arrest of Al-Bashir to stand trial for genocide in 2010, more than five years after the crimes took place.

Analysts of genocide in general, as well as public and political discourse, far more often focus on death and extermination than on displacement and elimination, and this has been true in Darfur as well as elsewhere (but see Physicians for Human Rights. 2006). De Waal (2005: xix) underlines the particular importance of this point in context of Darfur by noting that

What is happening in Darfur is not Genocide (capitalized) in this sense of the absolute extermination of a population. It does, however, fit the definition contained in the Genocide Convention, which is much broader and encompasses systematic campaigns against ethnic groups with the intention of eliminating them in whole or in part.

It is important to note that de Waal does not conclude that the ICC should charge President Al-Bashir with genocide; he instead argues that ‘an effective response to Darfur's crisis will be complicated, comprehensive, and long’ (2005: xix).

There is a further paradox to the focus on death and extermination in recent genocide research in Darfur. As convergence has emerged in the estimated size of the mortality in Darfur – in a range around 300,000 deaths – the means of calculating this number has increasingly revealed the importance of better understanding the unfolding of the genocide. Thus, the most recent estimation of deaths, published in Lancet (Degomme and Guha-Sapir 2010), does not actually include any reflection of the first major wave of killing in Darfur in the summer of 2003. This report indicates that by its criterion of inclusion, this first period of death and destruction is ‘not included in any retrospective survey, and mortality data should therefore be estimated with other techniques’ (2010: 296). On the other hand, this report is quite important in indicating that high levels of mortality have continued through 2008. Thus, both for what it includes and excludes, this report is important in pointing to the need to better understand genocide as an unfolding process.

This paper is concerned with the unfolding of both extermination and elimination in Darfur, but it gives greater attention to the latter element of forced displacement. The importance of this focus on elimination and displacement is highlighted by the millions of Darfurians today still stranded in Sudanese internal displacement camps and refugee camps in Chad, with poor prospects for ever returning to their homes, and with a whole generation of affected groups disconnected from the livelihoods that nurture and sustain their ways of group life. The genocide in Darfur was an anti-livelihood crime. The complex causes and consequences of group elimination and forced displacement are understudied and poorly understood, as well as obviously important in their own right.

Explaining extermination and elimination in Darfur

  1. Top of page
  2. Abstract
  3. Introduction
  4. Explaining extermination and elimination in Darfur
  5. Explaining state-led elimination by displacement in Darfur
  6. Displacement and the atrocities documentation survey
  7. Findings from survey interviews and proportional hazard models
  8. Conclusions
  9. Bibliography

The roots of the current Darfur conflict date at least to the middle of the last century when Sudan gained its independence from Britain. Independence was played out in the context of a competition between Egypt and Great Britain for ongoing influence on Sudan's emergent nationalist and subsequent governing elite (Sharkey 2003). This elite group, known as the effendis, was largely educated at Gordon College in Sudan, an Eton-esque school for the offspring of prominent Sudanese families who traced an Arab lineage and expressed a cultural affinity for Egypt. These newly emerging elites combined their Gordon College Anglophile education with their Arab and Egyptian identities to provide a dual foundation from which they could play off their political interests in Sudan (Kalema 2010).

The nationalist effendis were ‘double agents’ in ways that went beyond their mixture of British and Arab backgrounds. They on the one hand further saw themselves as a group apart and separate from ‘native’ Sudanese, but on the other hand they presented themselves as speaking for and protecting their fellow citizens. In the transition to independence, this involved endorsing a policy of ‘Sudanization’ that replaced British with Sudanese civil servants. Sudanization was advanced as a kind of nationalism in which the country's educated elites regarded themselves as ‘guardians’ of the general population, including African tribal groups in Darfur.

The idea of Sudanization allowed for a merging of some Arab with African identities and included a limited prospect of upward mobility. O'Fahey (2004) notes, as an example of this possibility, that until the latter part of the last century in Darfur, when a successful Fur farmer obtained a certain number of cattle, he could develop an identification with the Arab Baggara, and in a few generations his descendants could boast of an ‘authentic’ Arab genealogy. A lasting influence of this aspect of this version of Sudanization is still today perhaps reflected in the resistance even among journalists, as contrasted with Darfurians, to identify ‘non-Arab’ groups as ‘African.’ And it is still the case that some small groups identified by others in Darfur as African, such as the Gimir people, have in recent decades come to self-identify as Arabs and have fought alongside Arab groups against other groups who identify themselves as Black Africans in Darfur.

Despite examples of mixed groupings, however, and the claims and even intentions of Sudanese elites at times to advance inclusion along with their development policies, the contemporary history of Darfur is more persistently one of marginalization and neglect if not exploitation of Black African groups by the Sudanese government. The prospect of a benign Sudanization policy is not currently plausible, nor is any widespread fluidity of identity transformation that such a policy might once have promised or implied. Perhaps the origins of Sudanization as an elite movement made all of this foreseeable. In any case, what remains is a cleavage along Arab/non-Arab lines of lineage, livelihood and language, combined with attention to subtle perceived differences in skin tone that are today identified in racial terms.

By the mid-1980s, the intertwined processes of desertification and famine aggravated disputes over land and water and intensified the socially constructed, racially tinged division between Arabs and other Africans. The causes of the conflict in Darfur are clearly a mixture of environmental and political forces. Understanding the interconnection of these forces and how they are played out is central to explaining state-led genocide through elimination of non-Arab groups in this African setting.

Desertification is an environmental hazard and challenge caused both by natural climate change and overgrazing and farming. Migration patterns were intensified by the severe drought and famine that plagued Sudan from 1980 through 1984. The UN (UNEP 2007) has reported that the reduction in rainfall alone has changed the natural environment, apart from and regardless of human factors.

Major clashes led to the hundreds and then thousands of deaths in the late 1980s, when violence intensified between Arab nomadic herders and non-Arab farmers (Flint and de Waal 2005, 2008). The disputes involved access for the Arab herders to grazing lands cultivated by the non-Arab farmers, with the crimes increasingly involving crops and livestock (Young et al. 2005). North and West Darfur were overwhelmed by the famine, and Libya's Muammar al-Gaddahi intervened and exploited the crisis by bringing guns to the Arab herders in Darfur. Gaddafi's goal was to create an ‘Arab belt’ across sub-Saharan Africa. Armed Arabization replaced more benign aspirations to political sudanization. The election of Sadiq al-Mahadi as prime minister of Sudan in 1986 added to the process of group polarization with his push for an ‘Arab and Islamic Union’ (Prunier 2005).

Gaddafi later adopted a more pan-African ideology, and al-Mahadi did not entirely exclude non-Arab groups from his regime, but Omar al-Bashir, who seized the presidency with a military coup in 1989, more brutally excluded non-Arabs from his government. Al-Bashir subsequently allowed the landless Arab pastoralists, who were growing ever more desperate for access to water and pastures, into the government's Popular Defense Forces [PDF]. Group polarization increased as non-Arab groups were excluded from participation in the PDF. Although both groups were predominately Muslim, the Black Africans were less likely to speak or understand Arabic. Leaders increasingly linked differences in livelihood and language to variations in skin tone and collectively framed these perceived differences in racially defined terms (Hagan and Rymond-Richmond 2008).

The beginning of the recent waves of mass atrocities in Darfur is often traced to the government's response to scattered rebel attacks on government forces in the early months of 2003 (e.g., Power 2004). There were two intervals of particularly intense attacks on Black African villages. The two intervals were from June through August of 2003 and from December 2003 through March 2004 (US Department of State 2004). Several events during these intervals played major roles in the onset and then reduction of violence in Darfur.

The first interval or wave of attacks began with the return to Darfur of Musa Hilal, who is often identified as a leading Janjaweed militia leader. Hilal repeatedly articulated his intent to eliminate Black Africans from Darfur. He did this in public speeches and by personally leading attacks on settlements in Darfur (Hagan and Rymond-Richmond 2009). The Office of the Prosecutor (2007: 60–1) of the international Criminal Court identifies Hilal and the Government of Sudan Deputy Minister, Ahmad Harun, as leading figures in the Darfur attacks. For example, the Prosecutor's brief documents their appearance together at a very public gathering in West Darfur in July 2003.

On that day, Harun's speech was preceded by that of the notorious Militia/Janjaweed leader Musa Hilal. Hilal's speech was characterized by the witness who heard it as ‘very racist.’ Hilal was enthusiastic about unifying to fight the enemy and characterized the conflict as a ‘holy war.’ Hilal's remarks were followed by Harun's announcement that the President had handed him the Darfur Security Desk and that he had the power and authority to kill and forgive whoever in Darfur.

Harun was responsible for recruiting large numbers of militia members and also distributed weapons and money for training camps. The Prosecutor's brief indicated that Deputy Minister Harun controlled an ‘unlimited and unaudited budget’ for these purposes.

Harun, in his role as a government minister with resources, encouraged Hilal and other militia leaders to begin intense recruitment of members in June 2003. The subsequent first interval of attacks ended with a negotiated ceasefire in September 2003 that promised government disarmament of militias. This ceasefire proved unenforceable by the late fall of 2003. The second interval began with a December 2003 vow by Sudanese President Al-Bashir to ‘annihilate’ Darfurian rebels. It ended after Al-Bashir's premature announcement of the end of ‘major military operations’, followed by the anguished warning of the UN's representative to Sudan, Mukesh Kapila, about the genocidal parallel between the mass atrocities in Darfur and Rwanda, where Kapila had served earlier.

While desertification, death, and displacement continue today in Darfur, the pace of the death and displacement, at least for the moment, has slowed. Yet, the millions of Black Africans who have now been in Sudanese displacement and Chadian refugee camps for more than five years make genocide a continuing reality in Darfur. The tasks of explaining this persistent reality and enumerating its genocidal scale persist.

Explaining state-led elimination by displacement in Darfur

  1. Top of page
  2. Abstract
  3. Introduction
  4. Explaining extermination and elimination in Darfur
  5. Explaining state-led elimination by displacement in Darfur
  6. Displacement and the atrocities documentation survey
  7. Findings from survey interviews and proportional hazard models
  8. Conclusions
  9. Bibliography

As we have noted, explanations of genocide in general, and specifically of the genocide in Darfur, have focused on death and extermination more than on displacement and elimination. Yet, there is even greater consensus about the massiveness of the numbers of displacements than deaths, and the elimination of Black African group life as it previously existed in Darfur arguably is even more comprehensively imposed by the displacements than by the deaths. Explanations need to take this reality into account.

Explanations of state-led elimination by displacement in Darfur must attend to the motive and intent of the perpetrators and the vulnerability, and especially the food and water insecurity, of the victims. It can be argued that, when it comes to crime, vulnerability is opportunity and that opportunity itself provides the motivation that shapes intent (cf., Gottfredson and Hirschi 1990: 24). As we have already noted, the environmental pressures of desertification are an exogenous constant in Darfur, and in this sense it is correct to say that possession of arable land is everything in the Darfur conflict.

Black African groups have long possessed arable farm land in Darfur, but they did not individually or collectively possess sufficient arms or military means to protect their farms and villages, and this vulnerability presented an opportunity for landless Arab groups. While a number of Arab herding groups also have had recognized claims to land in Darfur, several northern Rizeigat groups have not, and these groups are key participants in the Janjaweed militia accused of attacking the Black African farming groups. Libya's Muammar al-Gaddahi helped arm the militias, who were already arming themselves and were fed, uniformed, and further armed by the Sudanese government, which has recruited the militias into the PDF.

The intent to take land from Black African farming groups played an explicit part in the mobilization and training of the militias. Musa Hilal asserted that the Black African groups had settled and farmed land that originally belonged to Arab groups. In the context of growing desertification and an increasingly desperate need for grazing land and water, Hilal's claims and ensuing threats became a salient ‘crisis framing’ of the situation in Darfur (Hagan and Rymond-Richmond 2008).

There are many accounts of how this crisis framing was expressed. A Washington Post journalist, Emily Wax (2004), gained access to the training camps and reported that the militia prepared for attacks by singing war songs proclaiming, ‘We go to the war. We go to defeat the rebels. We are not afraid of war. We are the original people of the area.’Samantha Power (2004: 9) reported in the New Yorker an interview with a defector who described men in a camp parading around singing songs challenging local Africans with claims that ‘We are lords of this land. You blacks do not have any rights here.’

The racial characterization of the conflict in Arab-African terms is contested if not controversial, and it is important to emphasize that our focus is on the use of race as a socially constructed motivational tactic in Darfur. De Waal (2005: xiv) emphasizes the contemporary and externally driven origins of this social construction:

Darfur's Arab-African dichotomy is an ideological construct that has emerged very recently, largely as a result of events outside the region. Arab supremacism in Darfur was born in 1987 along with the region's ‘Arab Alliance,’ which owes more to Khartoum and Libya than to any realities in Darfur. This is turn led Fur and Masalit militants to adopt the label ‘African,’ emphasizing a common political identity with Southerners and the Nuba.

The tracing of the racialization of the conflict to Khartoum speaks in particular to the issue of state leadership.

Prunier (2005: 162) links the racial motivation and intent in Darfur to its land based environmental foundation, noting that the 1984 famine sharpened the divide between the nomadic herders and farmers and that now this dichotomy is superimposed on an Arab versus African dichotomy, with state-led agency. He concludes that, ‘This marked the beginning of years of low-intensity racial conflict and harassment, with the “Arab” Centre almost automatically siding against the “African Periphery”.’ The Sudanese government defined ‘Arab’ as good, and ‘African’ as bad.

In Darfur, state-authorized agents, such as Ahmad Harun, integrated the local Janjaweed militia with the PDF and local police (Office of the Prosecutor 2007: 40). Locally-organized indoctrination included instruction in ‘us’ and ‘them’ distinctions that escalated from demeaning and degrading to dehumanizing characterizations. These included attributions of subordinate, slave, and sub-human statuses. Racial epithets constituted the hooks for the dehumanization leading to elimination and extermination. The dehumanizing logic and intention was clear, for as Dower (1986: 89) brutally explains and our analysis will further assess, ‘it is . . . easier to kill animals than humans.’ Of course, not only killing but displacement and elimination are made easier by dehumanization as well.

Our thesis is that the racial epithets heard during attacks in Darfur were transformed into motive and intent and expressed in an eliminationist frenzy to drive the Black African groups in Darfur from their lands. Attackers shouting racial epithets undertook ground assaults on African villages. These epithets in Darfur involve tropes of slavery and sub-humanity:

They called her Nuba [a derogatory term for Black Africans], dog, sons of dogs, and we came to kill you and your kids.

You donkey, you slave; we must get rid of you.

You blacks are not human. We can do anything we want to you. You cannot live here.

We kill our cows when they have black calves – we will kill you too.

All the people in the village are slaves; you make the area dirty; we are here to clean the area.

You blacks are like monkeys. You are not human.

Black prostitute, whore; you are dirty – black.

We will kill any slaves we find and cut off their heads.

The government has ordered anyone black to be killed – even the black birds.

These words and phrases shouted by the perpetrators are explicit evidence of dehumanizing motivations and intentions during attacks on Black African villagers. Prunier (2005: 165) captures the significance of this racialization when he concludes that

since Darfur had been in a state of protracted racial civil war since the mid-1980s, the tools were readily available; they merely needed to be upgraded. It was done and the rest is history.

Thus, the government of Sudan helped train and joined its forces with the Janjaweed militia in organized attacks on Black African groups. However, it is also essential to understand that the perpetrators took special advantage of the vulnerability and insecurity of the latter groups by systematically attacking their food and water supplies, so that these groups would no longer be able to sustain their lives in their farms and villages. This vulnerability, of course, intensified in the context and against the backdrop of the increasing desertification of Darfur.

De Waal's (2005) classic research on famine previously has revealed the adaptiveness of Black African survival strategies during the mid-1980s in Darfur. One of the most remarkable findings of this work was that members of these groups would store meager amounts of grain and seeds through even the last painful stages of starvation: they would actually die with remaining hidden and unconsumed amounts of grain and seeds. Thus, Darfurians practiced especially well-developed and interdependent coping strategies within their villages that they had learned during droughts and famines. For example, villages would build communal wells that were operated both by hand and with pumps, where they could afford them. There were also private wells, and, in the rainy season, water was also drawn from the wadis around which the villages were often built.

Preserving access to food and water as a risk-management strategy of survival was understood as among the highest priorities in these villages, even in the face of death. The Sudanese government and the attacking Arab groups recognized this vulnerability and insecurity, and they targeted food and water supplies with scorched earth tactics aimed at overwhelming these communal survival strategies. They recognized that if they could dislodge and displace villagers from their homes and communities, they would in addition be highly vulnerable to the scorching heat and wind in the desert, and therefore also vulnerable to starvation, dehydration, and disease.

Attacks on food and water supplies were powerful weapons for displacement and genocidal elimination, which underlines the overlapping explanatory importance of famine and genocide to one another. The joined forces of the government and the Janjaweed militias coordinated their attacks not only to kill and rape, but also to systematically burn all homes and crops, steal and kill all livestock, poison and destroy all wells, seek out and destroy all food stores, and uproot and kill all trees.

These were attacks designed to eliminate as well as exterminate Black African groups in Darfur. These attacks might be called famine crimes as well as crimes of genocide (de Waal 1997). De Waal (2005: xii) observes that ‘Ethnographers of famine and genocide have much to learn from one another. . . . In Darfur today, where much violence is directed at destroying livelihoods . . . the convergence is evident.’

Displacement and the atrocities documentation survey

  1. Top of page
  2. Abstract
  3. Introduction
  4. Explaining extermination and elimination in Darfur
  5. Explaining state-led elimination by displacement in Darfur
  6. Displacement and the atrocities documentation survey
  7. Findings from survey interviews and proportional hazard models
  8. Conclusions
  9. Bibliography

The US State Department's Atrocities Documentation Survey [ADS] (see Howard 2006) was conducted in July and August 2004 in preparation for Colin Powell's testimony before Congress and the UN Security Council. The Survey was conducted in and around the refugee camps in Chad and provides a unique opportunity to study displacement processes in an ongoing genocide. The ADS team proportionately sampled in relation to size and ethnicity within sectors of the camps and informal refugee villages. The survey cost nearly a million dollars to conduct and includes 1136 respondents. This analysis is based on 932 of these respondents who fled from 22 originating village clusters (henceforth called ‘villages’) that had 15 or more respondents each included in the survey.

A limitation of the ADS for our purposes is that it includes only displaced Darfurians who ultimately fled to Chad. However, we focus on incidents of reported displacement, including displacements prior to fleeing to Chad. As well, population pyramids from displacement camps within Darfur are similar with regard to age and gender to those in Chad. Probably as a result of targeted killing, both Chadian and Darfurian camps have a disproportionate absence of fighting-age men (aged 18–29 years). There are no indications that the Darfur refugees in Chad differ in significant ways from internally displaced Darfurians in the bordering areas.

The descriptive characteristics of this sample are shown in Table I. The average refugee in the sample was 37 years old. The sample was about 60 per cent female and 40 per cent male, reflecting the reduced representation of fighting age men. About half were Zaghawa (52 per cent), about one quarter Masalit (27 per cent), and about one twentieth Fur (5.5 per cent), with the remainder from other ethnic groups (14 per cent).

Table I. Descriptive statistics
Respondent attributes  
Ethnic group  
Reported/rebel presence  
 Rebels in town0.1720.130
 Rebels nearby0.0320.177
Attacks by perpetrator group  
 Janjaweed alone1.8803.229
 Combined attack6.6105.221
Racial intent  
 Racial epithets0.4060.543
Violent crimes  
 Killing/missing persons2.2901.264
 Bombing or poison gas1.6962.564
 Rape or sexual violence0.4260.777
 Other severe violence1.1181.232
 Threat of attack0.1380.443
Property crimes  
 Targeting of food/water1.8270.955
 Arson: home1.7090.992
 Arson: other buildings0.2480.535
 Other property crimes1.0620.756
N – subjects918

Unique strengths of the survey for our purposes were that it was conducted as the second of the two major waves of attacks was abating, that the survey collected specifically dated information about 35 kinds of criminal victimization linked to displacements – including the burning of homes, killings, and attacks on food and water supplies, and that the survey included reports of hearing racial epithets during the attacks and information about the attackers, as well as information about rebel activity in the area of the attacks. The attackers were coded as belonging either to government forces or Janjaweed militia by categorization of clothing and equipment reported by interviewees.

It has been argued that traumatic events and other significant life events can be reliably reported up to ten years (Burt, Kemp, and Conway 2001). The maximum recall required for this analysis was 18 months, dating from March 2003. It is especially important that the ADS included dated information about the attacks. This allows our analysis to consider the genocide as an unfolding process. Genocide is, of course, a dynamic rather than a static phenomenon, even when it occurs in an explosive rather than a more drawn out process, as it has in Darfur.

Findings from survey interviews and proportional hazard models

  1. Top of page
  2. Abstract
  3. Introduction
  4. Explaining extermination and elimination in Darfur
  5. Explaining state-led elimination by displacement in Darfur
  6. Displacement and the atrocities documentation survey
  7. Findings from survey interviews and proportional hazard models
  8. Conclusions
  9. Bibliography

Our analysis is based on estimations of standard semi-parametric hazard models and examples drawn from the survey interviews. Readers will note clearly recurring patterns in the examples we present from the interview narratives. These patterns are reflected and anticipated by the proportional event-history models.

We first examine observed patterns from March 2003 to August 2004 of the risk of being attacked by Government of Sudan [GoS] forces and/or Janjaweed militias. To conserve space, we summarize the main findings for this preliminary part of our analysis here and in Figure I (tabled results are available on request). This Figure is based on a model that controls for age and gender of the refugees.


Figure I. Risk of being attacked by perpetrator group*

Download figure to PowerPoint

The curves in Figure I, representing risks of attack by the different perpetrator groupings, are consistent with the two intervals of attacks in 2003 and 2004 described above. They also reveal that the risks of attacks were greater in the second wave than during the first wave. The lowest curve indicates that the risk of attack by the Janjaweed alone increased across most of the months and peaked late in the second wave, but was lower than for the other groupings.

The risk of attack from the GoS troops acting alone by air (i.e., bombing) or by land (i.e., ground assaults) were higher than by the Janjaweed on their own, and these also peaked in the second wave. GoS air attacks often preceded ensuing ground attacks:

A 71 year old male Zaghawa living in North Darfur reported that his town had been bombed repeatedly for three months before the day of the attack that led to his displacement. GoS soldiers came to the town market the day before the attack that led to the final displacement. On the day of the attack, GoS planes bombed his village more than 20 times. This day the attack included GoS trucks and Janjaweed on horses. The trucks had mounted machine guns. Many villagers were killed as the attackers shouted ‘Nuba, Nuba’ and ‘we want to kill the Blacks and take the land.’ He estimated that the military were driving about 100 trucks and that about 300 Janjaweed were on horses and camels. He fled when the shooting started and people were falling around him. He broke his hip. The attackers took their cattle, burned the village, and poisoned their wells, with the water changing color and smell. (from survey interviews)

The most significant finding in Figure I is that the highest risk of attack through both waves and most of the months, and especially during the second wave of attacks, involved the GoS and the Janjaweed acting together. Figure I is consistent with attributing leadership responsibility to the GoS in the genocidal attacks in Darfur. An eyewitness account describing the joint organization of one such attack follows:

A 43 year old male Masaleit living in a village near Masteri in West Darfur reported an attack in mid-February of 2004. The attack began at about four in the morning and involved about 600 GoS soldiers and militia. An Antinov aircraft circled the area and the soldiers arrived in trucks armed with mounted automatic weapons and on camels and horses. They sprayed bullets across and into the village, killing children, women, men, and elderly persons. While some of the soldiers were shooting, others looted the homes and set fire to the huts. The attackers shouted to one another, ‘don't leave anything’, as they called to each other to loot and burn the huts. His wife and uncle were killed as they tried to run away while carrying a child. His wife was shot in the doorway as she tried to escape. The attackers took all the livestock (camel, sheep, goats, cattle, and horses) and food that they could carry and transport, and then they burned the rest. He tripped and broke his collarbone and lay still pretending to be dead for the three hour duration of the attack. All the wells were poisoned with what smelled to him like DDT. Birds and animals that drank from the wells died. He counted 22 dead bodies after the attack, including seven young girls. A few scattered huts remained after the attack, but when other villagers tried to return a week later, they found that the attackers had returned before them and burned even these. (US Department of State 2004)

We next analysed patterns from March 2003 to August 2004 involving the hazard of hearing racial epithets during the attacks using the Cox proportional hazard function. We estimated competing risk models to determine the relative hazards or risks of hearing racial epithets when attacked by Government of Sudan [GoS] forces and/or Janjaweed militias. These models thus represent three different risks at any given time: the risk of hearing racial epithets during an attack by Janjaweed militias, the risk of hearing racial epithets while under attack by GoS forces, and the risk of hearing epithets while being attacked by both in combination. The results of the competing-risk models are presented in Table II. These hazards or risks are again estimated in simple models that include only age and gender of the refugees. The logistic regression coefficients in this and following models are exponentiated, which allows us to interpret the estimated hazards or risks in terms of percentage increases or decreases.

Table II. Cox proportional hazard of hearing racial epithets
 JanjaweedGoSJanjaweed & GoS
  • Notes:

  • Exponentiated coefficients; Standard errors in parentheses

  • ** 

    * p < 0.05, p < 0.01,

  • *** 

    p < 0.001.

Respondent attributes      
LR Test2.3120.52325.318***
N – events306430643064
N – subjects918918918

Of the three competing risk models that we estimate, only the one involving attacks by the combined GoS forces and Janjaweed militias yields a statistically significant likelihood-ratio test (LR = 25.318, P < 0.001, indicating that this model predicts hearing racial epithets far better than do the models where GoS forces or Janjaweed militia act alone. These results are consistent with earlier findings that collective racial intent is present almost exclusively when GoS forces attack simultaneously with Janjaweed militia (Hagan and Rymond-Richmond 2008). In these attacks the risks were greater for male and younger refugees, who were probably more likely to hear and understand the racial epithets because they were more likely to know Arabic (i.e., boys were more likely to be sent to school, especially the Islamic schools).

Figure II shows the relative risks over time of hearing racial epithets by each of these perpetrator groups. This Figure indicates that when the GoS attacked together with the Janjaweed, the risks of hearing racial epithets were much greater, and that this risk of racialization was especially high during the second wave of attacks.


Figure II. Risk of hearing racial epithets by perpetrator group*(Cox Proportional Hazard)

Download figure to PowerPoint

Several examples from the survey interviews illustrate the context of these joint attacks that featured racial epithets.

A 16 year old Fur woman in North Darfur reported that the attack began at four in the morning. GoS and Janjaweed attacked together and they began by looting food stores and burning the houses. The attack included one white Antinov airplane and five black helicopters. On the ground, the attackers shouted that ‘They will destroy all the people with black skin’ and that they ‘want to kill all the Black people and clean the ground.’ Her brother saw the soldiers pour poison in the wells, and they told him not to drink the water.

A 30 year old Zaghawa female from North Darfur reported that her village was attacked in the early morning with Antinov planes, helicopters, and soldiers in vehicles. The attackers were in uniforms and were a mix of GoS forces and militia. They looted the homes, set them on fire, and poisoned the wells. The entire village was burned by the bombing and shelling. During the attack a soldier yelled, ‘You dirty servants, we killed your husbands and should take you to be servants for our wives.’. They looted the sheep and killed the villagers who protested.

A 61 year old Masaleit male from a medium sized town in West Darfur reported an attack from September 2003. He was working on his farm outside of town when he heard shooting. He ran toward his home and was stopped by soldiers. They demanded to know where all the animals were kept and beat him with sticks. The attackers wore military uniforms and arrived in vehicles and on horses and camels. Planes flew overhead. The attackers were shouting ‘Nuba, Nuba.’. About 65 people were killed. He was able to hide in the undergrowth of the Wadi. Before the attack, government soldiers were in the town and Janjaweed came during the night on horses. The next day someone went out to draw the water and found that oil had been put in the well. They could not drink the water. His mother was beaten and died of the injuries. After burying her he fled. (US Department of State 2004)

We turn finally to our estimation of the Cox proportional hazard models of displacement in Table III. Model 1 in Table III shows that, with gender and age held constant, the risk for being displaced was about 18 per cent higher for the refugees who heard racial epithets. Racial epithets become statistically non-significant in Model 2, which now includes the identities of the victim and perpetrator groups. We interpret the reduction in the significance of the racial epithets as reflecting a mediation process in which these epithets were uniquely influential in motivating the GoS forces and Janjaweed militias in their combined targeting of the Masaleit and Fur for displacement. Indeed, previous analyses have shown that racial epithets were almost exclusively present in attacks on these ethnic groups and when GoS forces and Janjaweed militia cooperated in combined attacks (Hagan and Rymond-Richmond 2008).

Table III. Cox Proportional Hazard of Displacement
 Model 1Model 2Model 3Model 4
  • Notes:

  • Exponentiated coefficients

  • p < 0.05,

  • ** 

    p < 0.01,

  • *** 

    p < 0.001.

  • a 

    ‘Other ethnicity’ is included but not listed.

Respondent attributes        
Ethnic groupa        
 Fur  0.9800.1290.8900.1210.8880.121
 Masaleit  1.318***0.1051.237*0.1131.226*0.112
Perpetrator group        
 Janjaweed alone  1.2720.1660.9760.1350.9660.134
 Combined attack  2.105***0.1811.314**0.1181.301**0.117
Racial intent        
 Racial epithets1.177**0.0730.9650.0650.868*0.0600.873*0.060
Violent crimes        
 Killing/missing persons    1.875***0.2841.871***0.284
 Bombing or poison gas    1.0720.0971.0760.097
 Rape or sexual violence    1.0260.0831.0210.083
 Other severe violence    0.9250.0680.9330.069
 Threat of attack    1.0850.1181.0780.117
Property crimes        
 Targeting of food/water    2.289***0.3212.288***0.321
 Arson: home    1.225*0.1231.227*0.124
 Arson: other buildings    1.0660.0931.0780.095
 Other property crimes    0.9800.0810.9820.081
Reported rebel presence        
 Rebels in town      1.0060.261
 Rebels nearby      0.7850.143
LR Test12.872**122.595***259.985***261.881***
N – events2182202020202020
N – subjects918880880880

Besides predicting the presence of collective racial intent, the combination of GoS forces and Janjaweed militia are also highly predictive of forced displacement. The results in Model 2 indicate that the risk of displacement was about 110 per cent higher when the GoS and Janjaweed joined together in the attacks than when GoS forces attacked alone and about 85 per cent higher than when Janjaweed militia attacked alone. Figure III shows the disproportionate displacement risks associated with each perpetrator group with all other variables in the final model set at their mean values. According to this figure, the unfolding of attacks that led to displacement again includes two peaks associated with the two waves of attacks, with the second wave larger than the first. The risk of displacement is highest during these waves of attacks and when the GoS and Janjaweed attacked together. Model 2 also indicates that Masaleit groups had approximately a 32 per cent higher risk of being forcibly displaced than did Fur or Zaghawa groups.


Figure III. Risk of displacement by perpetrator group*(Cox Proportional Hazard)

Download figure to PowerPoint

When we include the different tactics involved in the attacks, as in Models 3 and 4, the effects of the combined GoS and Jajanweed attacks are reduced (but none the less significant) because of the relationship between the perpetrating groups and the severity of the attacks. The estimates from these Models indicate that attacks resulted in displacement about 23 per cent more often when they involved burning and destroying homes, about 87 per cent more often when they involved killings and missing persons, and about 129 per cent more often when they involved the targeting of food and water supplies. Figure IV illustrates the displacement risks associated with the most influential forms of attack, again with the other variables set at their mean values. Figure IV also shows the effects of the two waves of attacks and the step-like increase in risks associated respectively with home arson, killings, and attacks on food and water. Thus, as general explanations of genocide would expect, killings and destruction of homes were significant factors in predicting displacement. Yet, attacks that involved theft or destruction of food supplies, theft or killings of livestock, and poisoning or destruction of water sources were even more influential, as the single, most powerful factor in provoking forced displacement.


Figure IV. Risk of displacement by type of attack*(Cox Proportional Hazard)

Download figure to PowerPoint

The Sudnese government argues that the attacks in Darfur were justified as simply a result of counter-insurgency against rebel forces. Model 4 introduces further controls for the presence of rebels in the attacked village or in the surrounding area and villages. Neither of these rebel measures is significant. Together with previous non-significant findings about nearby rebel presence (see Hagan and Rymond-Richmond 2008; Hagan, Rymond-Richmond, and Palloni 2009), these findings discount such explanations.

Two examples from the survey interviews provide final illustrations of the ways in which the elements of the attacks combined to create a recurring scenario in North and West Darfur villages:

A 36 year old Zaghawa male who lived near Kornoi reported GoS bombing along with Janjaweed wearing uniforms attacking on the ground. They said ‘We came here because we want to kill all the Black people.’ He said that if he had been close enough to hear more, he probably would have been killed. They took 18 cattle, 24 camels, and about 400 goats. His home was burned to the ground. They took 52 men from the village, forced them to their knees, and executed them, each with a gun in his mouth. The entire village was destroyed. He reported that ‘We passed through one village where there was nothing left at all. They even ruined the well water because there was something in it that smelled bad and we were told we couldn't drink it.

A 40 year old Zaghawa woman who lived near Kornoi reported that the GoS and Militia attacked her village in January of 2004. They killed many people and poisoned the water well. They poisoned the well by killing a donkey and throwing it and other dead animals into the well. They also took food stocks. They separated the men from the women and killed the men (US Department of State 2004)

There is a recurring pattern to these accounts that is further illustrated in a description by Human Rights Watch (2004: 26) of coordinated, joint GoS-Janjaweed attacks in the Masteri area of West Darfur in January and February 2004. The pattern included coordination of the attacks by time and place, a close working relationship of the GoS and Janjaweed forces, their arrival and departure together, and a recurring pattern that included the killing of fighting-age men and the destruction of food and the poisoning of wells. The attacks around Masteri systematically cleared the Black African settlements and left mostly Arab groups as the remaining presence. GoS planes not only bombed these settlements beforehand, they circled the settlements for days after to make sure the villagers left and did not return. When Human Rights Watch visited the Masteri area after the attacks, ‘the only civilian life encountered was a terrified group of some fifteen people – men and women, and pitifully thin – who were attempting to reach their former village to dig up buried food stores.’


  1. Top of page
  2. Abstract
  3. Introduction
  4. Explaining extermination and elimination in Darfur
  5. Explaining state-led elimination by displacement in Darfur
  6. Displacement and the atrocities documentation survey
  7. Findings from survey interviews and proportional hazard models
  8. Conclusions
  9. Bibliography

The evidence presented in this paper is of a pattern of racialized, state-led attacks on food, livestock, and water supplies, indicating the intent by the political leadership of the Government of Sudan to eliminate the collective livelihoods of Black African groups in Darfur. The widespread and systematic poisoning of wells is perhaps the most striking evidence of the intent of jointly attacking GoS and Janjaweed forces to dislodge and displace Black African groups from their villages and farms, but the further evidence of killing, looting, burning, and, more generally, of completely destroying these villages with scorched-earth tactics fills out the picture of the intent to exterminate and eliminate these groups.

Most analyses of the conflict in Darfur have focused on the killings and have involved contentious efforts to establish the level of mortality involved. This work is important and is now producing a convergence regarding the hundreds of thousands of deaths resulting from this conflict. This evidence speaks to the 1948 Genocide Convention's definition of genocide as destruction of protected groups ‘in whole or in part’, and the extent to which this criterion of extermination is met in Darfur. However, the Geneva Convention also emphasizes ‘deliberately inflicting on a group conditions of life calculated to bring about its physical destruction in whole or in part’, which is a criterion of genocide that focuses more on the elimination of protected groups. This criterion addresses the concern in Darfur that, beyond widespread killing, the joint attacks of the GoS forces and Janjaweed militias have used forced migration and displacement as a means of permanently eliminating the Black African groups from their villages and farms in Darfur. The fact that these groups have nearly entirely been removed from their land and have been isolated in displacement and refugee camps for more than five years supports the conclusion that they have been the victims of elimination.

We have already noted parallels between the circumstances of extermination and elimination in Darfur and the earlier prosecution at the ICTY of genocide in the town of Srebrenica in the former Yugoslavia. The Holocaust and Armenian genocide also provide historical parallels to the situation in Darfur. A recent report (Physicians for Human Rights 2006: 40) notes that French delegates who contributed to the drafting of the post-World War II Genocide Convention insisted on the inclusion of the provision regarding the creation of conditions that make life unsustainable. They maintained that this was necessary to address circumstances in which members of a group, though not killed immediately, were subjected to conditions calculated to bring about the same result over a prolonged period of time. The French cited as examples the ghettos where Jews were confined in conditions of starvation and illness that led to their extinction in the lead up to the Holocaust. They also noted actions taken by the Turkish government during World War I to deprive Armenian populations of food during forced marches with the intent to undermine the capacity to sustain group life. Still, this tactic of elimination is underdeveloped in applications of the Genocide Convention and in research on the topic of genocide.

A key challenge in developing the international criminal law and research on the use of elimination strategies is the issue of intent. There is abundant evidence of intent in Darfur, both as understood legally and social-scientifically, and this is reflected in our analysis of the ADS data. The direct evidence of this intent includes the dehumanizing racial epithets that we find most prominent in the second wave of attacks in Darfur, against protected groups – especially the Fur and Masaleit – and during combined attacks that include both the GoS forces and Janjaweed militias. The frequently cited Akayesu decision in Rwanda (UN 1998) and the Jelisi decision in Bosnia (UN 1999) both emphasize the importance of spoken language as evidence of genocide. The words and phrases used by perpetrators to dehumanize victims play similar roles in research on genocide and hate crimes (Green, McFalls, and Smith 2001; Horowitz 1980Jenness and Broad 1997). In both kinds of crime, dehumanizing language is an intrinsically important motivational process that diminishes moral and practical constraints on participants and bystanders, as recognized in law and social science research.

There is also strong circumstantial evidence of genocidal intent, which is allowed as proof of specific intent to commit genocide in international law. In the Jeliisic case in Bosnia, the court (UN 1995) held in an appeal decision that,

in the absence of direct explicit evidence, [intent may] be inferred from a number of facts and circumstances, such as general context, the perpetration of other culpable acts systematically directed against the same group, the scale of the atrocities committed, the systematic targeting of victims on account of their membership of a particular group, or the repetition of destructive and discriminatory acts. (1995: Point 4)

All of these circumstantial forms of evidence apply in the current case of Darfur, where the attacking and targeted groups, and the scale and repetition of mass atrocities is well documented through our analysis of the ADS, and is corroborated by other research.

After the intervention of the Appeals Chamber of the ICC, the Pre-Trial Chamber (ICC 2010) in July of this year finally reversed an earlier decision and concluded that there were now ‘reasonable grounds’ to approve a warrant for the arrest of President Al-Bashir for trial on charges of genocide. The reasoning of the judges in this decision was based on the kind of elimination argument we have advanced in our analysis of the ADS data. They concluded that,

the conditions of life inflicted on the Fur, Masaleit and Zaghawa groups [in Darfur] were calculated to bring about the physical destruction of a part of those ethnic groups' and that ‘forcible transfer by resettlement by member of other tribes, [was] committed in furtherance of the genocidal policy. (ICC 2010)

The Prosecutor's Office (OTP 2010) maintained a month later that ‘the information available suggests that genocide continues today.’

Although Sudan's government defends its internal displacement camps in humanitarian terms, it is clear that the personal and food insecurity that the government has created in Darfur is the factor that overwhelmingly keeps the displaced persons in the camps. Almost all of the respondents in a follow-up 2009 survey in the Chad refugee camps indicated a desire to return to their land in Darfur, and most indicated that security-related concerns prevented them from doing so ( 2010, The concerns are explicitly about physical, food, and water security – in other words, the ‘conditions of life’ necessary for survival. The displacement is itself evidence of genocide by elimination.

Evidence of genocide in Darfur continues to mount with report of the mistreatment of persons in the camps in Darfur. Since March 2008, when the ICC issued its first warrant for the arrest of President Al-Bashir, the Sudanese government has followed a policy in Darfur of disrupting the work and expelling the employees of Western humanitarian aid groups. In June of 2009, President Obama's Special Envoy to Sudan, Scott Gration, referred to Darfur as experiencing ‘the remnants of genocide’. The Administration's US Ambassador to the United Nations, Susan Rice, disagreed with this assessment and President Obama has spoken of Darfur as the scene of ‘ongoing genocide’. Evidence of genocide in Darfur continues to mount with reports of the mistreatment of persons in the camps in Darfur. The UN recently reported that constraints on aid organizations were increasing (Osman 2010).

In August of 2010, Sudan reinstated a ban on humanitarian access to Kalma, the largest displacement camp with nearly 100,000 persons in South Darfur. Sudanese Government policies pitting groups against one another in camps such as Kalma have provoked violent confrontations between political rivals and forced many displaced persons from their camps. The Human Rights and Advocacy Network for Democracy [HAND] reported that,

In addition to the ongoing violence there is severe shortage of food, medicine, potable water and other essential service in the IDP camps which also played a major role in accelerating the exodus. (2010: 3)

In July 2010, the majority of the persons in Kalma were believed to have fled the camp. The HAND report indicates that

Some IDP leaders expressed fears that movement of IDPs outside Kalma camp meets an old goal set by GoS to gain full control over the IDP camps in Darfur, split the IDPs on ethnic and political lines, and finally, dismantle the IDP camps all over the region. (HAND 2010)

The conditions of displacement in Darfur continue to reflect their origins in the genocidal intentions of the leadership of the Government of Sudan. These intentions are expressed through processes of extermination and elimination which require further combined documentation in research on genocide. Genocide needs to be understood as a process of displacement and elimination that can presist long after the mass killing and extermination that are more commonly regarded as its hallmark.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Explaining extermination and elimination in Darfur
  5. Explaining state-led elimination by displacement in Darfur
  6. Displacement and the atrocities documentation survey
  7. Findings from survey interviews and proportional hazard models
  8. Conclusions
  9. Bibliography
  • Burt, C.D., Kemp, S. and Conway, M. 2001 ‘What Happens if you Retest’, Autobiographical Memory? Memory and Cognition 29: 12736.
  • Degomme, Olivier and Guha-Sapir, Dearati 2010 ‘Patterns of Mortality Rates in Darfur Conflict’, Lancet 375: 294300.
  • De Waal, Alex 1997 Famine Crimes: Politics and the Disaster Relief Industry in Africa, Bloomington: Indiana University Press.
  • De Waal, Alex 2005 ‘Preface to 2005 Edition’ in Alexde Waal (ed.) Famine that Kills: Darfur, Sudan, 1984–1985, New York: Oxford University Press.
  • Dower, John 1986 War Without Mercy: Race and Power in the Pacific War, New York: Pantheon Books.
  • Flint, Julie and De Waal, Alex 2008 2005 Darfur: A New History of a Long War, London: Zed Books.
  • Gottfredson, Michael and Hirschi, Travis 1990 A General Theory of Crime, Stanford, California: Stanford University Press.
  • Green, Donald, McFalls, Laurence and Smith, Jennifer 2001 ‘Hate Crime: An Emergent Research Agenda’, Annual Review of Sociology 27: 479504.
  • Hagan, John and Palloni, Alberto 2006 ‘Death in Darfur’, Science 313: 157879.
  • Hagan, John and Rymond-Richmond, Wenona 2008 ‘The Collective Dynamics of Racial Dehumanization and Genocidal Victimization in Darfur’, American Sociological Review 73: 875902.
  • Hagan, John and Rymond-Richmond, Wenona 2009 Darfur and the Crime of Genocide, New York: Cambridge University Press.
  • Hagan, John, Rymond-Richmond, Wenona and Palloni, Alberto 2009 American Journal of Public Health 99: 138692.
  • HAND 2010 ‘Disturbance in Darfur's IDP Camps’, Human Rights and Advocacy Network for Democracy (HAND), Geneva, August 13, 2010. .
  • Horowitz, Irving Louis 1980 Taking Lives: Genocide and State Power, New Brunswick, New Jersey: Transaction Books.
  • Howard, Jonathan 2006 ‘Survey Methodology and the Darfur Genocide’, in S.Totten and E.Markusen (eds) Genocide in Darfur: Investigating the Atrocities in Sudan, New York: Routledge.
  • Human Rights Watch 2004 ‘If We Return, We Will Be Killed, Consolidation of Ethnic Cleansing in Darfur, Sudan’, HRW Report, November 2004.
  • ICC 2010 ‘Situation in Darfur, Sudan: In the Case of the Prosecutor v. Omar Hassan Ahmad Al Bashir’, Pre-Trial Chamber, July 12, 2010, International Criminal Court.
  • Jenness, Valerie and Broad, Kendal 1997 Hate Crimes: New Social Movements and the Politics of Violence, New York: Aldine de Gruyter.
  • Kalema, William 2010 ‘The “Graduates”: Sudanese Nationalism and the Anglo-Egyptian Rivalry, 1936–1947: The Role of the Educated Elite’, Senior Honors Thesis, Department of History, Northwestern University
  • Office of the Prosecutor 2005 ‘Seventh Annual Report of the Prosecutor of the International Criminal Court to the UN Security Council Pursuant to UNSCR 1593’, International Criminal Court: Office of the Prosecutor, June 5, 2008.
  • Office of the Prosecutor 2007 ‘Situation in Darfur, The Sudan. Prosecutor's Application under Article 58(7), No. ICC-02/05’, International Criminal Court, February 27, The Hague.
  • Office of the Prosecutor 2008. ‘Situation in Darfur, The Sudan: Summary of the Case: Prosecutor's Application for Warrant of Arrest Under Article 58 Against Omar Hassan Ahmad AL BASHIR. No. ICC-OTP-20080714’, International Criminal Court, July 14, The Hague.
  • Office of the Prosecutor 2010. ‘Genocide in Darfur Continues and Ahmad Harun Remains Free to Attack Civilians, Now in the South of Sudan’, OTP Weekly Briefing, August 17–23 – Issue #51, International Criminal Court.
  • Power, Samantha 2004 ‘Dying in Darfur’, The New Yorker, August 30, p. 58.
  • O'Fahey, R.S. 2004 ‘A Complex Ethnic Reality with a Long History: Darfur’,International Herald Tribune, May 15, 2004, p. 8.
  • Osman, Mohamed 2010. ‘Sudan SAys Will Deport Foreign NGO Workers’, Associated Press, August 19, 2010.
  • Physicians for Human Rights 2006 Darfur: Assault on Survival, Cambridge, Massachusetts: Physicians for Human Rights.
  • Power, Samantha 2004 ‘Dying in Darfur’, The New Yorker, August 20: 112.
  • Prunier, Gerard 2005 Darfur: The Ambiguous Genocide, Ithaca, New York: Cornell University Press.
  • Sharkey, Heather 2003 Living with Colonialism: Nationalism and Culture in the Anglo-Egyptian Sudan, Los Angeles: University of California Press.
  • United Nations 1948 Convention on the Prevention and Punishment of the Crime of Genocide, adopted by Resolution 260(II)A of the United Nations General Assembly on 9 December 1948.
  • United Nations 1998 Judgment: The Prosecutor v. Jean-Paul Akayesu, International Criminal Tribunal Rwanda, Case No. IT-96-T.
  • United Nations 1999 Judgment: The Prosecutor v. Goran Jelisi, International Criminal Tribunal for the Former Yugoslavia, Case No. IT-95-T.
  • United Nations 2005 Report of the International Commission of Inquiry on Darfur to the United Nations Secretary-General, Geneva.
  • UNEP 2007. Sudan: Post-Conflict Environmental Assessment. Geneva: UNEP. Available at
  • US Department of State 2004. Document Atrocities in Darfur. Washington D.C.: Bureau of Democracy. Human Rights and Labor and Bureau of Intelligence and Research.
  • Wax, Emily 2004 ‘In Sudan: “A Big Sheik” Roams Free’, Washington Post, July 18, 2004, p. A1.
  • Young, H., Osman, A., Akhu, Y. and Dale, R. 2005 Darfur: Livelihoods Under Seige, Tufts University, with Agfhad University.