by  Jo Ellen Fair (University of Wisconsin-Madison) &  Lisa Parks (University of California at Santa Barbara)  
(This paper is a very rough draft. Please do not quote or cite. [the authors])  

Introduction 
Seeing and Constructing the Refugee 
Televisions Flow of Refugees  
Tracking Refugees from the Sky: Satellite and Aerial Images  
Conclusions 
References 

Introduction

 The 1994 Rwandan conflict has its roots not in simmering "tribal hatreds" but in a colonial and post-colonial political and economic order that politicized differences among ethnic groups to allow one group (Tutsis) more power over others (Hutu and Twa primarily).[2] Events that eventually led to the 1994 genocide can be tied to armed conflict that first occurred in October 1990 between Rwandese government forces and the Rwandese Patriotic Front (RFP), a movement comprised largely of Tutsi exiles, along the Rwanda-Uganda border. Despite a number of UN-sponsored cease-fire agreements, conflict resumed in February 1993. Later, in October, the UN Assistance Mission for Rwanda (UNAMIR) was deployed to implement peace agreements between the Rwandan government and the RPF. UNAMIR's efforts to restore peace in the area were unsuccessful. Then, on April 6, 1994, when a plane carrying the presidents of Rwanda and Burundi crashed, the country was thrown into violent disarray, with Tutsis and some Hutus targeted for death by the Rwandan government. By late April, the RPF had launched a new offensive against government forces and successfully had staged a coup in the capital, Kigali. Fearing reprisals from the RPF-controlled government and military forces for the massacres, some 800,000 to 1 millions Rwandans (most of them Hutus) left the country in June and July 1994 and relocated in eastern Zaire/Congo,[3] Burundi, and Tanzania. 

Covering Rwanda's 1994 genocide was challenging and difficult. Logistically, U.S. news media had to scramble to gain entry into a country where conditions were deteriorating rapidly. Organizationally, because few U.S. news groups support regular reporting from Africa and still fewer operate bureaus on the continent, scarce resources for international reporting had to be redirected from regions thought to be of more geopolitical and/or cultural interest to American consumers and, ultimately, to advertisers. Journalistically, U.S. reporters were dropped into a region of the world in which they knew little of the complex political, economic, and social relations that were the basis of the brewing conflict and into a country in which their own government advocated disinterested non-involvement. Hence, the reporting of the Rwandan genocide was far too typical of reporting of conflict in Africa more generally. U.S. journalists relied on stereotypes -- thoroughly tested by news organizations in stories set in other African countries, such as Uganda, Somalia, Ethiopia, and Sudan -- that the genocide in Rwanda was a result of some uncontrollable primordial "tribalism" that drove Hutus and Tutsis to murder that was described as inexplicable. 

While framing news coverage of the Rwandan genocide as a Hutu versus Tutsi conflict made news reports simpler to produce and easier for American television audiences to digest, the Rwandan genocide still was not a "good" news story for many U.S. news organizations. News organizations' attempts to peg events neatly as one "tribe" against another demanded that reporters be able to identify which "tribal" side U.S. audiences should support. The problem with the Rwandan genocide story was that Western reporters were unable to make clear distinctions as to which side was "good" or "bad." As a result of this ambiguity, the genocide story received far less coverage (say how much maybe) than subsequent movements of thousands of refugees into settlement camps in 1995 and 1996 (Minear, Scott, and Weiss, 1996; Murison, 1996; de Waal, 1997). Moreover, following refugees was easier than reporting from conflict zones, and refugees made "good" visuals because they evoked the now familiar images of famine and conflict in places such as Ethiopia and Somalia in the Horn of Africa. 

Although Western news media only reluctantly covered the Rwandan genocide in 1994, they provided extensive coverage of the resulting refugee crisis in eastern Zaire/Congo by comparison. Over the next year following the 1994 Rwandan civil war, as many as 1.5 to 2 million Rwandan refugees entered the neighboring countries of Zaire/Congo, Burundi, and Tanzania (Refugees International, 1996a; Prunier, 1997: 373-389). From October through December 1996, refugees who had settled in eastern Zaire/Congo found themselves trapped between fighting Zairian Armed Forces and an alliance of local "rebels," assisted by the Rwandan and Burundian military. At the request of Zairian/Congolese officials, aid workers helped supervise an exit corridor from eastern Zaire/Congo to Rwanda, facilitating the exodus of more than 500,000 Banyamoulenge (Zairian Tutsi ethnic group) and Rwandan Hutu refugees fleeing violence in the region (Refugees International, 1996a). Despite efforts to relocate people in an organized manner, heavy fighting throughout the region sent refugees in many directions. Most Hutu refugees returned to Rwanda. But others fled west to nearby mountain regions, traveling through toxic lava fields where there was little water or food, because they feared retaliation upon their return home. Many reportedly died there in early November 1996, and others were murdered in mountain regions by Zairian/Congolese militia. In late October and early November aid workers announced that hundreds of thousands of Hutu refugees were missing. After their evacuation from eastern Zaire/Congo on October 22, aid workers called upon the U.S. government to provide satellite images to help locate the "lost refugees" (Refugees International, 1996b). 

 U.S. news media's decision to focus on the Rwandan refugee crisis as opposed to the Rwandan genocide fits into a well-known, conventionalized understanding of Africa as a place where bad happenings occur and where Africans are in constant need of Western intervention and assistance (Girardet, 1996; Minear, Scott, and Weiss, 1996; de Waal, 1997). This paper is an exploration of the visuals and text televised by the four major U.S. news outlets -- ABC, CBS, CNN, and NBC -- and satellite images distributed on the Internet of refugee movements from eastern Zaire/Congo to Rwanda in 1996, with most of our concentration on the October-December period when refugees were forced to leave their encampments. Television crews, satellites, and aircraft tracked the movement of these refugees chronicling their displacement and the precariousness of their lives for media viewers in the United States. We particularly are concerned in this paper with the various ways in which television and aerial cameras alike inspected the bodies of refugees, and in doing so, fit Rwandan refugees into discourses of Western compassion and African complacency. Furthermore, we suggest that television news media represented events in eastern Zaire/Congo in ways that reinforced U.S. officials' suppression of genocide. Rather than reminding viewers that the refugee crisis was an effect of the genocide of hundreds of thousands of Rwandans, reporters instead turned their cameras upon a massive refugee crisis, showing the world hundreds of thousands of barely surviving people with little or no explanation of what has caused their predicament. As a result, their indeterminate migratory status is portrayed not as a result of a specific political and economic context, but as representing the inherent "victim" status of Africans. 

To bear witness to the Rwandan genocide involves recognizing media images of Rwandan refugees as part of the genocidal atrocities that U.S. political officials refused to publicly acknowledge and that U.S. news media tried to avoid. As Donna Haraway suggests, the witness is no longer an authoritative bystander whose gaze verifies or validates an event from a distance. Rather, witnessing is a more situated and embodied practice -- one that involves being accountable to and for that which what one sees. Haraway further suggests that witnessing is about "interrogating critical silences, excavating the reasons questions cannot make headway and seem ridiculous, getting at the denied and the disavowed in the heart of what seems neutral and rational" (1997: xx). In the case of the Rwandan genocide, witnessing would involve examining U.S. media images of Rwandan refugees in relation to the one million people that now lay buried in Rwanda. The paper begins with a discussion of how Western governmental and non-governmental groups, as well as media, see and construct refugees as a special target for assistance. From there, we move into our analyses of broadcast news stories and aerial images of refugees. The final section of the paper describes how these images of refugees circumscribe what we can know of displaced peoples. 

Seeing and Constructing the Refugee

 In one week, in mid-July 1994, nearly a million Rwandans -- some seeking to 

 escape death and some seeking to avoid retaliation for their involvement in the killings -- left the country to find refuge in Goma, Zaire. By the year's end, another million would leave Rwanda to relocate in neighboring Zaire, Uganda, and Tanzania. The displacement of so many Rwandans in the span of just five to six months was one of the fastest and largest cross-border movements of people in recent years (United Nations High 

 Commissioner for Refugees, 1995: 123). Although the genocide was largely overlooked, this vast displacement of people and the ensuing health-related complications of their movement -- for instance, epidemics of dysentery and cholera -- helped to generate greater U.S. media interest in Rwanda (Minear, Scott, and Weiss, 1996; Murison, 1996; de Waal, 1997). Covering genocide meant having to understand politics to assess both domestic and international accountability. By comparison, covering refugees was simple.[4] 

U.S. news organizations' earlier experiences, particularly in Ethiopia and Somalia, demonstrated that at least for a few weeks U.S. news viewers would be content to watch the pain and suffering of others as long as the story remained an uncomplicated one of U.S. benevolence. Simplifying humanitarian stories for domestic audiences in terms of what the West is doing for others allows journalists and news consumers alike to perceive a certain superiority to the extent that crises of such magnitude do not occur (or have not yet occurred) in the West. As a result, the West is made separate and distant from Africa, permitted journalists to ignore questions about how historical and contemporary Western involvements contributed to current problems (Hawk, 1992; Fair, 1996; Girardet, 1996). Speaking of how the U.S. media simplified it Rwandan coverage, ABC's Ted Koppel explained: 

 Maybe it's a natural outgrowth of the age of television, but we do 
 prefer to keep our crises simple, stories with a definable beginning and a 
 predictable end. We like our villains to be foreign and our heroes 
 home-grown. What we do not like are long, open-ended, complicated 
 involvements far from home, in which America's good intentions are 
misunderstood. By those standards, we will not much like even our limited 
involvement in Rwanda. It is one thing to respond with American skill and 
generosity to a human disaster, fly in the food and the medicine, build the roads, 
set up the water purification plans. But at that point, our national attention 
span starts to lag. If people are no longer dying at the rate of 2,000 a day, 
Rwanda slowly, inevitably, begins moving off our radar screens (Nightline, 1994). 

 But refugees are not really individuals in news stories. Rather, they are a human category upon which reporters build their stories. What constitutes a "refugee" and who is allowed to become one (so as to qualify for humanitarian assistance) is determined by a number of complex and fluid international rules and conventions (Barnett, 1997; Pottier, 1997). Generally, refugees are people forced to leave their countries as a result 

 of foreign aggression and/or well-founded fear of persecution.[6] Refugees are groups of people who are deterritorialized -- that is, they are a placeless people, often on the move. As such, they signal a rupture in the order of a nation-state. Shunted from a national geographic entity, refugees pose a challenge to the national order because they no longer belong to any one country. Their bodies in motion demonstrate that it is not just geographic boundaries that are porous but that the categories used to distinguish "national" from "foreigner" are as well.[7] As anthropologist Mary Douglas (1982) has noted, the insecurities and anxieties held about social relations often are inscribed on physical bodies. Hence, refugees, no longer contained within a given nation-state, embody certain insecurities and anxieties on the part of the nation-state to define who does and does "belong." This placelessness and lack of clear national affiliation serves Western Eurocentric notions of Africa and Africans as a monolithic mass, without separate national identities, who need assistance and order that can be provided only by the West. 

Although refugees' movement may expose the constructedness of national boundaries and thereby support postcolonial models of resistance that valorize "in-between" identities and "deterritorialized" spaces, it is important to recognize that there is nothing necessarily liberatory about refugees' mobility because it constitutes a kind of enforced rather than willful nomadicism (Deleuze and Guattari, 1987; Bhabha, 1990). In a world where "belonging" to a nation-state usually affords individuals certain rights and benefits as citizens or residents, the abrogation of this status casts individuals into the netherworld of statelessness. Refugees become liminal beings in the sense that they belong nowhere and are claimed by no nation-state. As stateless, liminal beings, refugees are transients who cross national borders in search of a location to which they can belong (Malkki, 1995; de Waal, 1997). 

 It is their transience and lack of "belonging" to a nation-state that make them invisible as political actors and objects of paternalistic Western benevolence. Indeed, states, international organizations, and the media are quick to identify refugees as "problems" in need of care (Benthall, 1993; Minear, Scott, and Weiss, 1996; de Waal, 1997). Identifying refugees as "problems" encourages largely Western intervention strategies that contain peoples and their cultures to particular national boundaries. Because refugees are displaced, they are no longer rooted in a specific national history, culture, or sense of place, and so pose a threat to national and global order (Escobar, 1995; Malkki, 1995). 

Without a tie to nation, nation-state, and culture, refugees become generic figures who embody the identified problem of massive displacement of stateless people. Writing about the experiences of Burundian refugees, anthropologist Liisa Malkki explained how classification of individuals as refugees inscribes "problem" upon their bodies. Noted Malkki of refugees: 

Not just "ordinary people," they are constituted, rather, as an anomaly requiring specialized correctives and therapeutic interventions. It is striking how the abundant literature claiming refugees as its object of study locates "the problem" not first in the political oppression or violence that produces massive territorial displacements of people, but within the bodies and minds of people classified as refugees (1995: 8). 

 As "problems" of displacement, refugees come to represent a pure and universalized humanity, victimized by political, economic, and historical forces over which they have no control, yet for which they ultimately are to blame. They are objects, rather than subjects, of interventions used by states, aid organizations, and the media to watch, count, categorize, and track them (Escobar, 1995; Malkki, 1995: 1-17). Thus, the process by which refugees are stripped of individuality, as well as history, culture, and place, serves to make them the focal points of observation and inspection. Television news cameras and satellite imagery allow various constituencies -- media audiences, Internet users, state security organizations, diplomats, policymakers, and aid groups -- to produce and circulate knowledges about refugees' experiences. Indeed, the public visibility of refugees raises questions about the ways in which they are "seen" by Western spectators. Donna Haraway (1988) has suggested that scientists' control over technologies of visualization such as photography, sonography, satellite images, and video has positioned Westerners as distant observers of others' problems. For her, such technologies can reinforce power differentials between Westerners and others to the extent that they transform vision into "unregulated gluttony" (1988: 581). As mediated spectacles relayed from Africa, images of refugees call for detached and disembodied looking that relieves viewers from any sense of involvement, responsibility, and accountability. As Haraway suggests, however, witnessing involves a more active engagement with that which one sees (Haraway, 1997; Parks, 1998: 163-239). 

Western media institutions' construction of Rwandan refugees as a "problem" is predicated upon their command of an array of visual technologies. Without television cameras and satellite and aircraft imagery, many relief organization officials have argued that events in Rwanda would have gone unnoticed. One assistance official insisted, "The spread of TV and reporting by satellite have made it possible to provide audiences in industrial countries with graphic images of large-scale human misery" (in Minear, Scott, and Weiss, 1996: 7; see also, Benthall, 1983; de Waal, 1997). The official noted that such images were useful because they fueled Western-led refugee assistance efforts. Yet, as his statement attests, these images of refugees are very much for Western consumption. Although relief workers no doubt have noble intentions, they often promulgate Western assumptions that order the world into beneficent television viewing publics of Western industrialized nations and "problem-ridden" masses of underdeveloped countries (Girardet, 1996). 

By controlling imaging technologies, Western/U.S. news organizations have the power to produce carefully regulated knowledge(s) about refugees, as well as the ability to perpetuate distinctions between Africa and the West. In this production of knowledge, refugees' experiences, made distant from those of ordinary Americans, compel television viewers and Internet users in the privacy of their homes to see the pain of others but not to feel it (Fair, 1996; de Waal, 1997). Refugees' pain is revealed not only in its physical manifestations of disease, malnutrition, and wounds, but also in more abstract, psychological forms suggested by crying, rocking the body, huddling in groups, or hugging oneself. The graphic images of others' dislocation, confusion, and loss forces a redefinition of space between simulated and lived (Virilo, 1996). Private pain is no longer privately experienced by those in pain; rather, it is publicly exhibited via cameras and satellites to thousands, if not millions, of consumers who will observe and comment (perhaps in the privacy of their own homes) on someone else's distant pain. 

 Witnessing someone else's pain is an important means of demonstrating power on the part of the maker of those images and the consumer. The spectacle of the refugee is not just a simple series of images designed to help the viewer understand the situation or feel the need to donate money. Rather, to engage the viewer, the spectacle emphasizes the pain of others, thereby revealing difference between those who are witnessing and those are living as refugees (Scarry, 1985; Chow, 1992). As Guy DeBord would remind us, "The spectacle is not a collection of images, but a social relation among people, mediated by images" (1983: no. 4). In our particular case of Rwanda, television cameras and satellites placed refugees within a field of vision where the consumer of these images could pass judgments about refugees, their plights, their needs, and in doing so, be an active participant in the production of knowledge about refugees. As the spectacle, refugees' voices are most often silenced and accounts of their lived experiences ignored in favor of knowledge produced from the outside by observers (Escobar, 1995: 154-211; Malkki, 1995: 105-152; Thayer, 1998). 

In the remainder of this paper, we explore how U.S. broadcast news media and satellite imagery of Rwandans reproduced knowledge of refugees as an embodied "problem" to be handled by states and relief organizations. We are particularly interested in asking whether it is possible to witness the pain and misery of others without necessarily supporting or reinforcing the Western privilege to gaze upon refugees? In other words, how can American television viewers struggle over the meanings of "objective" news and aerial images? We raise the question of witnessing as a way of trying to intervene in the discursive construction of the refugee. Such a proposition would involve seeing the refugee not as Africa's "problem," but as part of Western colonial history, cultural imperialism and state-sanctioned violence. Genocide and refugees are not "problems" indigenous to Rwanda, rather they are the effects of historical processes and postcolonial tensions. Witnessing not only involves struggles to situate "objective" images, but also a critique of how visual technologies themselves are used and what they make visible and to whom. 

Televisions Flow of Refugees 

Much of the news coverage on American television from October through December 1996 focuses upon Rwandan refugees' movement from camps in eastern Zaire/Congo to Rwanda, their former "homeland." On numerous occasions and on different networks, reporters refer to Rwandan refugees as a fluid body, as a "wave of humanity," a " human river," that "flooded across national borders." Such language not only positions Hutu refugees as a constantly moving mass, but also constructs the refugee crisis as a "natural disaster" beyond human control. By naturalizing Hutu refugees' movement is naturalized as a flooding river, the problem can be understood not as an effect of Western colonial history nor local political conflict, but rather as an inevitable natural catastrophe. This conflation of indigenous peoples with the landscape itself has long been a strategy of Western colonial discourse, and it has the effect of reducing the political agency and autonomy of colonized people to the unpredictable whims of the natural world (McCarthy, 1983; Hickey and Wylie, 1993). 

 Television news coverage not only referred to refugees as pure flow, but also used specific visual techniques to construct refugees as a slowly moving fluid that overruns the borders of nation-states, refugee camps, and relief agency resources. Five distinct visual strategies foreground refugees' movement. First, static panoramic perspectives show refugees continuously passing through the frame one after the other. Second, hand-held tracking shots follow refugees in transit, often showing them from behind as they "return home." Third, close ups of refugees' walking feet graphically reveal the traumatic physical effect of constant movement. Often the feet shown are bare, swollen and injured--with particular emphasis on children's' feet. Fourth, close-ups of refugees faces show them walking toward the camera and only to quickly exit the frame. Such images put a face on the moving refugee, but the line of faces stretches beyond the frame, as if to imply the refugees' infinite replacement. Finally, the editing of such images together has the effect of multiplying the sense of refugees' movement, placing the Western spectator in a vertigo of displacement as refugees are shown moving in different directions. Neither the refugees' movement, nor the screen direction of the editing, is unidirectional. Rather refugees and the shots scatter unpredictably in all directions. 

In contrast to the images discussed so far where refugees are constructed as constantly moving, other sequences in news stories represent refugees as confined to particular spaces such as camps, tents, medical wards, and even stretchers. Enticed by aid agencies that were counting on publicity to further their assistance efforts and to gain publicity for themselves, broadcast journalists arrived ready to report on a story already seemingly written from previous encounters with the "refugee" situated in the "campsite," where s/he now is thought to belong (Girardet, 1996; Pottier, 1996; Rosenblatt, 1996). 

That people who are so labeled as "refugees" are confined to "refugee camps" is fully consonant with cultural assumptions that associate physical displacement with absence of culture, nationality, and history. This naturalized association of people with place is what anthropologist Arjun Appadurai calls the "spatial incarceration of the 'native'" (1988: 36-49). For Appadurai, "natives" of a given locale can be known only by understanding the place in which they belong. Hence, natives are incarcerated by and confined to their places. It is these places to which Westerners come to extract information about (and sometimes from) natives, which is then reproduced as knowledge. 

 In Appadurai's sense of spatial incarceration, Rwandan refugees are doubly incarcerated. First, as Rwandans, they belong to Africa, which invokes a whole set of stereotypes revolving around the "dark continent" as a place where "tribal" warfare occurs. Second, as displaced and stateless persons, refugees belong only to the camps in which they are contained. It is their placement in "refugee camps" that allows them to become the objects of humanitarian aid and the objects of Western knowledge production. In the quest to present refugee life to audiences removed physically and 

 culturally from displaced persons in the camps, journalists often begin their stories with images of people contained by fences, many barbed, or by ropes and posts guarded by soldiers (see, e.g., CNN, May 22, 1996; CNN, Oct. 29, 1996; ABC, Nov. 3, 1996; CNN, Nov. 28, 1996; ABC, Nov. 30, 1996). These barriers literally and symbolically incarcerate refugees to the place in which they belong. The fences and soldiers literally contain any threat of actual movement. At the same time, the fences symbolically remind viewers that refugees constitute a special category of marginalized people to be watched, taken care of, and pitied. 

With some camps the size of small U.S. cities with upwards of 180,000 people, the barriers and people standing behind them were most often shot from a distance. They were shot in such a way to show the great numbers of people and perhaps the potential for threat contained in the camps. When reporters began to discuss need for assistance, cameras then closed in on one individual or a small group to personalize the refugee's plight. Refugees' personal pain is shown for viewers in images of mothers attempting to care for their children, of children crying, of women gathering fire wood or cooking, of teens milling about, of despondent men sitting alone. 

 In particular, women and children are used to evoke sympathy in viewers. Often viewed as innocents, without political attachments and involvement, they embody a sense of pure humanity because being a refugee has made them into the pure victim (Malkki, 1995: 8-17; Fair, 1996). Their pain is closely inspected: Flaccid breasts depict women's inability to feed their children; children have swollen stomachs and have red hair, often a sign of malnutrition. Indeed, in many closing scenes of news stories, viewers are left to look into the sad eyes of a child. For instance, in one story (CNN, May 22, 1996) on crowded camp conditions, the reporter ended his story with a short discussion of whether the United States ought to intervene in the region. As he began his voiceover, "The United States and its allies could justify intervention on humanitarian and moral grounds," the viewer sees a close up of a young boy. Continuing, he said that such intervention might be justified "to solve the long-running refugee problem left by the last round of massacres." At this point, viewers see a distant, wide-angle shot of refugees in both camps and in transit. Concluding his story, the reporter suggested that not only could intervention stop future massacres but intervention was needed "to avoid the impression that the world would stand aside once again until it is too late to prevent another tragedy." The visuals that accompany this final part of the story are of women huddled together as the words "to avoid the impression that the world would stand aside." As the reporter finishes with "until it is too late to prevent another tragedy," the camera shows a close up of the sharp barbs of a wire fence. The image, then, dissolves to refocus on the face of a toddler contained just behind the barrier. 

 Many of the reports close with a single lasting image that asks the television viewer to look into the eyes of an individual refugee, often a child. In attempt to establish connection between the Western spectator and the anonymous and voiceless refugee, news stories often use the eyes of children as points of entry into and sympathy for Africans. What these images frequently accomplish is not to create rapport but to reduce refugees to a part of their body -- their wide-open eyes -- in an attempt to symbolize their plight. In doing so, television viewers may find it easier to view crisis through children's eyes than through adults. But such images also have the impact of constructing Africa itself as an incarcerated space on the far side of the television news camera. In this space, the history, culture, and identity of Rwandan refugees is erased. The effect, then, of such images is to establish a relationship between the refugee "victim" and the American citizen as spectator. In one particularly striking image, two Hutu children are featured through a metal framed window, peering out of the darkness toward the cameras. They are incarcerated by the metal frame of the object that protects them from violence and trapped in the incarcerated space of the Western television camera's caricature of their experience. In the end, by framing their faces, the story attempts to transform them into portraits that convey a condensed tell-all version of the conflict. 

 Perhaps the most common images featured in news stories of campsite were of white, Western relief workers organizing camp life, tending to the medical needs of refugees, and providing food aid to refugees. Relief workers by far were the most frequently used sources for stories. In short, news accounts repeatedly show the good deeds done by relief workers. These images, of course, are nothing new. They are the stock from which Western paternalism has fed itself (Malkki, 1995; Fair, 1996; Girardet, 1996). What these images accomplish is to deny refugee camp dwellers the possibility of acting with political agency. In stories that use images of Rwandans confined to camps, refugees are acted upon by outsiders -- international relief workers, government officials of various states, soldiers protecting one interest or another, and journalists -- who have the power to determine the conditions in which refugees live, the amount of food they receive, and their ability to reside in camps. Yet, as Malkki suggests, these portrayals of refugees who appear to be acted upon by outsiders fail to recognize that camps are often sites where displaced groups dynamically stake out new allegiances, create new histories, and reconstituting cultural identities and consciousness (1995: 232-258; see also, Thayer, 1998). 

 Acknowledging agency among hundreds of thousands of refugees spatially 

 incarcerated into camps is a risky proposition for journalists. Audiences and news managers at home expect their refugees to be passive, nobly resigned to their lot in life, not actively engaged in reformulation of their political, cultural, and historical identities. Journalists' perception that refugees were unable or unwilling to act on their own behalf also is found in a group of stories focusing on the closure of camps in Zaire/Congo and Tanzania in October and November 1996. When Zaire/Congo and Tanzania began to close refugee camps and demand repatriation of refugees to Rwanda,[8] journalists began to describe refugees as groups "trapped" in camps (see, e.g., CNN, Oct. 22, 1996; ABC, Nov. 3, 1996; NBC, Nov. 16, 1996; NBC, Nov. 23, 1996; ABC, Nov. 30, 1996). The stories suggested, of course, that those who were "trapped" would require the assistance of Western humanitarian and possibly military intervention efforts to locate and retrieve them for repatriation. For example, one NBC story -- showing close up visuals of a young child -- concluded that "the job of locating and keeping alive those refugees trapped in Zaire is the world's focus (Nov. 16, 1996). It was assumed that refugees would want to 

 be repatriated. Several stories mentioned the importance of "homecoming" and "going home" (see e.g., ABC, Nov. 6, 1996; CNN, Nov. 16, 1996; ABC, Nov. 18, 1996; CNN, Nov. 28, 1996). As one story noted as the camera followed a man and his child walking through a forest, "These people are considered the lucky ones. After two years in exile, they are returning home" (NBC, Nov. 16, 1996). However, the use of the word "trapped" to describe the refugees' inability to leave camps to go "home" marks a redefinition of the refugee. Earlier, refugees who lived in camps were not "trapped," but were seen needing to be confined to the camps for their survival. Now, when repatriation (even it seems when it is forced) becomes a possibility for refugees, journalists speak about the refugees need to leave the incarcerated spaces of their camps to return to Rwanda, where they may shed their status as refugee to become once more a "Rwandan" national on Rwandan soil. 

 U.S. television news coverage of the refugee crisis in eastern Zaire/Congo, Burundi, and Tanzania reinforced existing popular notions about refugees' displacement by representing them either as constantly moving or as occupying incarcerated spaces. In stories of displaced Rwandans, representation and reality intersect through processes in which selected parts of refugees' reality are put into concrete form, whereby the parts stand for (or are representative of) particular experiences or knowledge in a manner that seems natural or normal. As Malkki (1992) argues, the location of peoples, who often are conceived of as a culturally unitary group, into a place naturalizes associations between people and state, people and (state or national) territory, and people and land. The consequence of these naturalized associations between peoples and place is that they appear as commonsensical and normal, when, in fact, they may be opposed, uncertain, or in transition. Thus, the implication of televised representations is that refugee camps and the long marches back to Rwanda come to represent for U.S. audiences a one-dimensional reality of refugees' lives. 

 As new imaging technologies have emerged over the past several year, altogether different representations of refugee populations have appeared. Not only are reporters able to capture hand-held footage of refugees' conditions, but more distant and wide-scale representation are available. Additionally, both satellite images and aircraft photos have been used to monitor the movements of Rwandan refugees in eastern Zaire/Congo. Although these images did not appear in the television news broadcasts, it is important to consider how they are used to produce knowledges about refugee populations from afar. Satellite images, in particular, are increasingly being made available to media organizations and public agencies (Stix, 1996; Public Eye, 1997; Parks, 1998). 

Tracking Refugees from the Sky: Satellite and Aerial Images 

 On November 18, Refugees International again called for the release of satellite data, claiming "More than ever, good information, including aerial and satellite photos, are needed to better determine where the unaccounted for refugees are" (Refugees International, 1996d). In an attempt to urge President Clinton to "mobilize an all-out international rescue effort" Refugees International President Lionel Rosenblatt (1996b) declared: 

The Zaire humanitarian disaster is unfolding off-camera. As it did on occasion in Bosnia, the U.S. should make public the satellite photos of the fleeing refugees. Ö In World War II and since there have been too many occasions where government photos of atrocities and humanitarian disasters remained in secret archives. Those days should be over and photos of eastern Zaire should be released. 

Rosenblatt recognized the satellite image not only as a mapping tool in relief efforts -- that is, for refugee monitoring and tracking -- but also as a spectacular news image. In a presentation to the House International Relations Committee, Subcommittee on International Operations and Human Rights, Refugee International's Rosenblatt insisted that in Zaire a "reverse form of the `CNN factor'" was at work. He said, "Because the current humanitarian catastrophe in eastern Zaire/Congo is not on television, many don't believe it's happening (or feel that, politically, they can afford to ignore it)." Stating the need for satellite images to reveal the whereabouts of refugees, Rosenblatt (1996c) argued: 

 In order to reach the `lost' refugees with assistance, the international rescue effort must first locate them. Satellite and aerial photography -- which should certainly be continued and intensified -- have yielded basic information about the directions many of the refugees have headed. Ö The photographs that have been collected should be released publicly so that all agencies concerned can be working from the best information possible. 

Despite television coverage of "waves" of refugees in the Great Lakes region, the international community did little to help. 

The satellite image was perceived not only as a useful aerial map that would enable relief workers to locate lost refugees' and render assistance, it also more importantly served as a dramatic visual representation of refugee conditions in the region. Aid workers argued that the public release of satellite images would incite the U.S. public to pressure the government to intervene, and they suggested that such images would force Western news consumers to comprehend the magnitude of the refugee crisis underway in eastern Zaire/Congo. Moreover, the release of U.S. satellite images would provide an angle on this news story that television cameras on the ground could not. Unlike the television news reports that focused on groups of refugees in transit, satellite images could show hundreds of thousands of refugees trapped in conditions of displacement. The satellite image thus served the dual function of pinpointing lost refugees for relief workers and providing a graphic display of refugees' living conditions for Western television viewers. 

This use of the satellite image as a call to political action is part of a broader movement among non-governmental organizations to use satellite technologies in emergency relief efforts. RefMon, a Norwegian-based refugee monitoring organization, has advocated the creation of an agency that will collect satellite images that "can be used to monitor, in near real-time, refugee movements and camp locations, thus rapidly providing information on the magnitude of the emergency"(RefMon, 1996). Human Rights Watch communications director Susan Osnos suggests, encourages the use of satellite images as supporting evidence to eyewitness accounts of human rights abuses, especially in places like Bosnia, and describes them as "powerful advocacy tools"(Stix, 1996). Aid workers' use of satellite images is likely to become more common with the emergence of new remote sensing companies such as Earthwatch, ORBIMAGE, and Space Imaging Systems, which now sell high resolution satellite images in the public sector for relatively low cost. 

 It was not until early December 1996 that the U.S. State Department finally released select satellite images showing refugees in eastern Zaire/Congo. Both UN and U.S. officials examined the satellite images and came up with radically different interpretations. While UN officials located 750,000 Hutu refugees in the pictures, U.S. officials spotted none. The U.S. government's interpretation of these satellite images reinforced the Rwandan Tutsi government's contention that refugee numbers had been inflated, and it allowed the United States to justify its withdrawal from the UN Security Council's plan for military intervention in the region. Thus, even from the seemingly omniscient position of the satellite, the lost refugees could not be "found," for analysts generated divergent interpretations of the same satellite images. U.S. military officials later boasted that the "timely distribution of [aerial] data prevented the unnecessary deployment of a multinational force" (Operation Guardian Assistance, 1998). 

 Although released to the UN, U.S. satellite images of Hutu refugees did not appear in American newspapers, nor were they incorporated within television news reports. Some aerial photographs taken by U.S. Navy aircraft as part of "Operation Guardian Assistance," however, appeared on the homepage of the Federation of American Scientists (FAS), an organization that has initiated a campaign called "Public Eye" in the public interest. Public Eye has gathered image intelligence of areas around the world including Bosnia, Sudan, Korea, Pakistan and made it available on its Internet site. 

In January 1997, six declassified aerial photographs taken by U.S. military aircraft were made available at Public Eye's web site. Such aerial and satellite images are increasingly being incorporated within global media events and they serve as a stark contrast to on-the-ground television coverage of refugee conditions in east Zaire/Congo (Stix, 1996; Public Eye, 1997; Parks, 1998). Where television images represented Rwandan refugees as a moving mass, aerial images represented them as barely visible dots on an high-tech map. A black and white photo taken on November 15, 1996, shows an evacuated refugee camp 22 miles southwest of Sake, Zaire. The image shows several buildings superimposed by arrows and text that reads "snapshot of refugee camp est. pop.: 100,000." There are no refugees visible, however, within the image. An aerial photo taken on November 16 shows a refugee camp near Mumbi, Zaire. Thousands of tiny white specs appear scattered across a vast green space. The white specs, the caption suggests, are refugees' tents, signs of encampment. Once again no refugees are visible in the image. Their presence is coded in the image only by the shelter which relief agencies have provided. A November 20, 1996, image actually shows 16 refugees walking in isolation on a road near Goma, a stark contrast to the "waves" of refugees shown by television news cameras. Two December 2, 1996 images reveal the Kilambo refugees camp in Mitumba mountain areas. One of them features an area of "approx. 100 blue and white tents" and the other of "apron 500+ tents." 

 These aerial images are worth analyzing because increasingly global crises are mediated by satellite images which become part of media events. Not only did U.S. news consumers watch surgical strikes from the cockpits of pilots during the Persian Gulf War, but television viewers also watched global events from the vantage point of the satellite in the Balkans war, during the nuclear experiments in Pakistan, and most recently during the U.S. missile strikes in the Sudan and Afghanistan. Observing from the vantage point of the aircraft or satellite ultimately enables Western spectators to keep themselves distant and safe from any actual conflict. From this position, the Western spectator becomes a participant in the production of knowledge about who should live or die in lands far off from the United States. 

 Although relief workers anticipated using aerial and satellite images to compel American television viewers to political action, they did not anticipate how such images might be read by or interpreted for television viewers. Especially when contrasted to television camera's ground perspectives, such aerial images reduce refugees to barely visible white and blue dots. No longer a set of wide open eyes, refugees are positioned as the opposite extreme, as evacuated and destroyed camps, scattered tents, or isolated travelers walking on a road. In this way, the satellite image does not provide entry into refugee's soul through wide open eyes, but rather allows the Western spectator to monitor refugee conditions from great distance and a position of seeming omniscience and objectivity. The aerial image constructs refugees as a hard to track nationless body, a moving target, a wandering collective rather than as a group of socially situated individuals with distinct histories and interests. The aerial image tends to privilege science over humanitarianism -- historically it has been used by military officials to generate strategic data about refugees and their location rather than to compel citizens to action. The satellite image's scientific data, however, tells little about the embodied circumstances of exile and displacement. In other words, physical pain and the struggles of survival are submerged within the barren space of the satellite image. 

Refugees International was the first non-governmental organization to call for the public release of U.S. satellite images. The organization's leaders recognized that the remote sensing satellite could be used to make visible (and therefore make known) the dire conditions of refugees within the region. But the satellite image, like the television news reports discussed earlier, enables Western viewers to gaze upon refugees' bodies and produce knowledges about them while situated away from the violence, malnutrition and disease that often accompanies social displacement. The satellite image further elaborates the Western spectator's distant positioning by transforming refugee bodies into a blur of dots on a high tech electronic map. Such an image is highly symptomatic, however, of the U.S. position of non-intervention in Rwanda. Although satellite images of refugees in Zaire did not see widespread public circulation, they raise very important questions about how the American government and its citizens "witness" political turmoil abroad. 

Conclusions

Media studies scholars have developed theories of spectatorship to account for the ideological assumptions encoded within news texts, but few have considered how viewers might interpret these images in ways that counteract their dominant meanings. We find the concept of witnessing to offer useful ways of thinking about how Western news consumers might critically engage with media events like the Rwandan refugee crisis of 1996. If witnessing involves accepting responsibility for and developing affinity with that which one sees, then witnessing television news images of Rwandan refugees involves understanding these events as more than as an isolated "natural disaster." Rather, witnessing demands the direct connection of large-scale movement of refugees from eastern Zaire/Congo, Tanzania, and Burundi in late 1996 to the complex political events that allowed state-sanctioned genocide in Rwanda in 1994 and permitted the U.S. government until only recently to refuse public acknowledgment of "genocide." Witnessing Rwandan refugees' images is to see refugees not purely as victims to be inspected, but as individuals who are part of a larger social and political body engaged in struggles for autonomy, place, and identity. To witness Rwandan refugee images means that U.S. media consumers must recognize publicly the pain and trauma of enforced displacement and demand culpability for it. Witnessing, then, calls for engaged media viewing rather than apathetic and cynical detachment. Only active spectators can force television news images and reconnaissance photos to bear witness to events that our political leaders refuse to "see." 

References

Anderson, Benedict (1991) Imagined Communities: Reflections on the Origin and Spread of Nationalism. London: Verso. 

Appadurai, Arjun (1996) Modernity at Large: Cultural Dimensions of Globalization. Minneapolis: University of Minnesota Press. 

Balakrishnan, Gopal (ed., 1996) Mapping the Nation. London: Verso. 

Balibar, E. and Wallerstein (1991) Race, Nation, Class: Ambiguous Identities. New York: Verso. 

Barnett, Michael (1997) "The UN Security Council, indifference, and genocide in Rwanda". Cultural Anthropology, 12(4) 551-578. 

Beigbeder, Yves (1991) The Role and Status of International Humanitarian Volunteers and Organizations: The Right and Duty to Humanitarian Assistance. London: Martinus Nijoff Publishers. 

Bennet, James (1998) "Clinton declares U.S., with world, failed Rwandans". New York Times, March 26, 1998, A1, 10. 

Benthall, Jonathan (1993) Disasters, Relief, and the Media. London: I.B. Tauris. 

Bhabha, Homi (ed., 1990) Nation and Narration. New York: Routledge. 

Chow, Rey (1992) "Postmodern automatons". In Judith Butler and Joan Scott (eds.) Feminists Theorize the Political. New York: Routledge, 101-117. 

Deleuze, Gilles and Guattari, Felix (1987) A Thousand Plateaus: Capitalism and Schizophrenia. Minneapolis: University of Minnesota Press. 

DeBord, Guy (1983). Society of the Spectacle. Detroit: Black and Red. 

De Waal, Alex (1997) Famine Crimes: Politics and the Disaster Relief Industry in Africa. Oxford: James Currey. 

Destexhe, Alain (1995). Rwanda and Genocide in Twentieth Century. Trans. Alison Marschner. New York: New York University Press. 

Douglas, Mary (1982) Natural Symbols. New York: Pantheon Books. 

Escobar, Arturo (1995) Encountering Development: The Making and Unmaking of  the Third World. Princeton: Princeton University Press. 

Fair, Jo Ellen (1996) "The body politic, the bodies of women, and the politics of famine in U.S. television coverage of famine in the Horn of Africa". Journalism and Mass Communication Monographs, no. 158. 

Fair, J.E. and Chakravartty, P. (forthcoming, 1998) "`Touring disaster': U.S. television coverage of famine". In Kaarle Nordenstreng and Michael Griffin (eds.) News Monitoring. Cresskill, N.J.: Hampton Press, 145-161. 

Girardet, Edward R. (1996) "Reporting Humanitarianism: Are the New Electronic Media Making a Difference?" In Robert Rothberg and Thomas Weiss (eds.) From Massacres to Genocide. Washington, D.C.: Brookings Institute, 45-67. 

Haraway, Donna (1997) Modest_Witness@ Second_Millennium.FemaleMan_Meets _OncoMouse: Feminism and Technoscience. New York: Routledge. 

Haraway, Donna (1988) "Situated knowledges: The science question in feminism and the privilege of partial perspective". Feminist Studies, 14(3) 575-599. 

Hawk, Beverly (1992) "Introduction: Metaphors of African coverage". In Beverly Hawk (ed.) Africa's Media Image. New York: Praeger, 3-14. 

Hickey, Dennis and Wylie, Kenneth (1993) An Enchanting Darkness: The American Vision of Africa in the Twentieth Century. East Lansing, Mich.: Michigan State University Press. 

Hobsbawm, Eric J. (1991) Nations and Nationalism since 1780: Programme, Myth, and Reality. Cambridge: Cambridge University Press. 
Hutchinson, Sharon (1996) Nuer Dilemmas: Coping with Money, War, and the State. Berkeley: University of California Press. 

Keane, Fergal (1996) Season of Blood: A Rwandan Journey. London: Penguin. 

Lamb, David (1994) "Rwanda war isn't just horrific, it's telegenic". Los Angeles Times, July 12, 1994: Section World Report, 2. 

Lemarchand, Rene (1970) Rwanda and Burundi. London: Pall Mall Press. 

Lemarchand, Rene (1996) Burundi: Ethnic Conflict and Genocide. Washington, D.C.: Woodrow Wilson Center Press. 

Liebes, Tamar (1998) "Television's disaster marathons: A danger for democratic processes?" In Tamar Liebes and James Curran (eds.) Media, Ritual, and Identity. London: Routledge, 71-84. 

Loescher, G. and Loescher, A.D. (1994) The Global Refugee Crisis: A Reference Handbook. Santa Barbara: ABC-CLIO. 

Malkki, Liisa (1992) 'National Geographic: The rooting of peoples and the territorializing of national identity among scholars and refugees'. Cultural Anthropology, 7(1), 24-44. 

_____. (1995) Purity and Exile: Violence, Memory, and National Cosmology Among Hutu Refugees in Tanzania. Chicago: University of Chicago Press. 

McCarthy, Michael (1983) Dark Continent: Africa as Seen by Americans. Wesport, Conn.: Greenwood Press. 

Minear, L., Scott, C., and Weiss, T. (1996) The News Media, Civil War, and Humanitarian Action. Boulder: Lynne Rienner. 

Murison, Judith (1996) Fleeing the Jungle Bloodbath: The Method in the Madness, British Press Reporting of Rwanda. Edinburgh: Centre of African Studies, University of Edinburgh. 

Newbury, Catharine (1998) "Ethnicity and the politics of history in Rwanda". Africa Today, 45(1), 7-24. 

Newbury, Catharine (1988) The Cohesion of Oppression: Clientship and Ethnicity in Rwanda, 1860-1960. New York: Columbia University Press. 

Newbury, David (1995) "Rwanda: Genocide and after". Issue: A Journal of Opinion, 23(2), 4-7. 

Newbury, David (1997) "Irredentist Rwanda: Ethnic and territorial frontiers in Central Africa". Africa Today, 44(2), 211-222. 

Nightline (Aug. 9, 1994). Ted Koppel, ABC. 

"Operation Guardian Assistance" (Apr. 14, 1998). Available at: http://www.fas.org/man/dod-101/ops/guardian_assistance.htm. Accessed Aug. 21, 1998. 

Parks, Lisa (1998) Cultures in Orbit: Satellite Technologies, Global Media, and Local Practice. Unpublished doctoral dissertation, University of Wisconsin-Madison. 

Patterson, Chris (1997) "International television news agency coverage of 

Conflict". Journal of International Communication , 4(1) 50-66. 

Pottier, Johan (1996) "Relief and repatriation: Views by Rwandan refugees; lessons for humanitarian aid workers". African Affairs, 95, 403-429. 

Prunier, Gerard (1997) The Rwanda Crisis: History of Genocide. New York: 

Columbia University Press. 

Public Eye (1997) "Project description," Federation of American Scientists Homepage, March 6. Available at: http://www.fas.org/eye/project.html. Accessed: September 3, 1997. 

RefMon -- Refugee Monitoring Using High-Resolution Satellite Images (1996). Available at: einar/UN/refmon_proposal.html. Accessed [NEED DATE, ALSO CHECK FULL CITE] 

 Refugees International, "Another Round of Ethnic Cleansing," press release, Oct. 11, 1996. Available at: . Accessed July 12, 1998. (1996a) 

_____. "Crisis in Eastern Zaire," press release, Oct. 22, 1996. Available at: . Accessed July 12, 1998. (1996b) 

_____. "Emergency Action Needed Immediately To Save One Million in Eastern Zaire," press release. Available at: . Accessed July 12, 1998. (1996c) 

_____. "Hundreds of Thousands of Refugees Still at Risk in E. Zaire," Report #10. Available at: . Accessed July 12, 1998. (1996d) 

_____. "Presidential Leadership Needed to Save One Million Refugees," press release, DATE. Available at: . Accessed by July 12, 1998. (1996e) 

LISA -- I ADDED A, B,C TO ROSENBLATT CITES; WE CAN FIX IF NOT CORRECT 

Rosenblatt, Lionel (1996a) "The Media and the Refugee". In Robert Rothberg and Thomas Weiss (eds.) From Massacres to Genocide. Washington, D.C.: Brookings Institute, 136-146. 

Rosenblatt, Lionel. "Unless U.S. Galvanizes International Action, Central Africa Will Explode," Refugees International Report #8. Available at: . Accessed July 12, 1998. (1996b) 

Rosenblatt, Lionel. Statement to the House International Relations Committee Subcommittee on International Operations and Human Rights, Dec. 4, 1996. Available at: . Accessed July 12, 1998. (1996c) 

Scarry, Elaine (1985) The Body in Pain: The Making and Unmaking of the World. Oxford: Oxford University Press. 

Stix, Gary (1996) "Public eye: Spy satellite technology may assist government watchdogs". Scientific American. Available at: http://www.sciam.com/0896issue/0896scicit01.html. Accessed [NEED DATE] 

Thayer, Leah (1998) "I'm Not What You're Searching For": Frameworks, Perceptions, and the Construction of the Liberian Refugee in Cote d'Ivoire, 1989-1997. Unpublished Master's thesis, University of Wisconsin-Madison. 

United Nations High Commissioner for Refugees (1993) The State of the World's Refugees: The Challenge of Protection. Oxford: Oxford University Press. 

_____ (1995). The State of the World's Refugees: In Search of Solutions. Oxford: Oxford University Press. 

Uvin, Peter (1997) "Prejudice, crisis, and genocide in Rwanda". African Studies Review, 40(2) 91-115. 

Valentina's Nightmare (1997). Dir. Fergal Keane, Frontline, PBS. 

Vansina, Jan (1998) "The politics of history and the Crisis of the Great Lakes". Africa Today, 45(1), 37-44. 

Virilio, Paul (1996) Paysage d'evenements. Paris: Galiee. 

Wall, Melissa (1997). "The Rwanda crisis: An analysis of news magazine Coverage". Gazette, 59(2) 121-134. 

Weiner, Tim (1998) "Critics say U.S. ignored CIA warnings of genocide in Rwanda". New York Times, March 26, 1998, A10. 

[1] Estimates of the number of dead vary from 500,000 to 1 million. Part of the difficulty in determining the number of dead is that bodies were destroyed, hidden, buried, and burned, making exact counts impossible. Moreover, various estimates of the dead may reflect the interests of the political actor or organization in furthering a particular number. The same cautions apply to figures of refugees. For a short essay on problems of evidence, see Vansina, 1998. 

 [2] For some treatments of the history of conflict in the Great Lakes region -- Burundi, Zaire/Congo, Uganda, and in particular, Rwanda -- see, Lemarchand, 1970, 1994; C. Newbury, 1988, 1998; D. Newbury, 1998; Unvin, 1997; Vansina, 1998. See also, the special issues on Rwanda and Central Africa in Africa Today, 45(1), 1998 and Issue: A Journal of Opinion, 23(2), 1995. Prunier (1997) provides a general, almost journalistic account of events, and Destexhe (1995) draws on his personal experiences as the former director of Doctors Without Border. He includes a brief chronology of events through 1994 in Appendix 2. 

[3] Herein, we will refer to what is now the Democratic Republic of the Congo as Zaire/Congo. Zaire was renamed as the DRC on May 17, 1997, when Laurent Kabila's Alliance des Forces Democratiques pour la Liberation du Congo seized the capital, Kinshasa, and ended President Mobutu Sese Seko's regime. 

 [4] It should be noted that the Clinton administration, calling the genocide a "civil war," earlier refused to intervene with bilateral or international efforts. However, it did concede to involvement with humanitarian relief in an operation called "Support Hope" later in the year. 

 [5] Because of the instability of the region, refugees included Rwandans, both Hutu and Tutsi, as well as Zairian Tutsis, and Burundians. 

 [6] From the United Nations High Commissioner Report (1993: 11), the 1951 Convention Relating to the Status of Refugees, which is a legally binding treaty, defines refugee as: "a person who is outside of his or her former home country owing to a well-founded fear of persecution for reasons of race, religion, nationality, membership of a particular social group or political opinion Ö ." This definition has been variously expanded (for example to include in some instances gender-based persecutions). For an overview of definitions, see, Beigbeder, 1991; United Nations High Commissioner for Refugees, 1993; Loescher and Loescher, 1994. 

 [7] Many have commented on the power of nation-states to structure cultural, political, and historical ideologies about nation, nationality, and nationalism. For some of these works, see, Appadurai, 1996; Balakrishnan, 1996; Hutchinson, 1996; Anderson, 1991; Balibar and Wallerstein, 1991; Hobsbawm, 1991. 

[8] Tanzania gave refugees the option of repatriation or naturalization. Zaire/Congo, which was undergoing its own political crisis, demanded repatriation, often carried out by force.