Читать книгу The Dynamics of Violence in Central Africa - Rene Lemarchand - Страница 10

Chapter 2

The Road to Hell

If the fate of the African continent evokes hopelessness, nowhere is this sense of despair more evident than in former Belgian Africa. No other region has experienced a more deadly combination of external aggression, foreign-linked factionalism, interstate violence, factional strife, and ethnic rivalries. Nowhere else in Africa has genocide exacted a more horrendous price in human lives lost, economic and financial resources squandered, and developmental opportunities wasted. The scale of the disaster is in sharp contrast with the polite indifference of the international community in the face of this unprecedented human tragedy. What has been called Africa's first world war has yet to attract the world's attention.

The marginal ranking of Africa in the scale of international priorities is one obvious explanation for this generalized lack of interest in the Great Lakes crisis. Another is the sheer complexity of the forces involved. When one considers the multiplicity of political actors—domestic and foreign—the fluidity of factional alliances, the spillover of ethnic violence across boundaries, and the extreme fragmentation of political arenas, it is easy to see why the international community should have second thoughts about the wisdom of a concerted peace initiative. No other crisis on the continent seems more resistant to conflict resolution.

Adding to the confusion is the plethora of competing explanatory models that come to mind. How much credence should one give to Paul Collier's recent thesis that “it is the feasibility of predation which determines the risk of conflict”?¹ Is the crisis in the Great Lakes an extreme example of the “criminalization of the state”?² Or should one turn instead to Jeffrey Herbst's demographic argument and look for evidence of low population density, combined with the weakness of state boundaries, as an explanation for Kabila's inability to effectively broadcast the power of the Congo state?³ If Samuel Huntington's “clash of civilizations” model hardly applies, what of his contention that the “kin country syndrome” is the key to an understanding of regional instability?⁴ To these questions we shall return.

This chapter offers a different prism to view the roots of the crisis. The key concept around which much of this discussion revolves is that of exclusion. Political, economic, and social exclusion are seen as the principal dimensions that must be explored if we are to grasp the dynamics of domestic and inter-state violence in the Great Lakes. This is not meant to minimize the significance of external aggression. The capacity of Rwanda and Uganda to effectively project their military force into eastern Congo, albeit with mixed results for both, is unquestionably a major contributory factor to regional instability. External intervention, however, must be seen in the broader historical context of the forces that have shaped the tragic destinies of former Belgian Africa. Briefly stated, the central pattern that recurs time and again is one in which ethnic polarization paves the way for political exclusion, exclusion eventually leading to insurrection, insurrection to repression, and repression to massive flows of refugees and internally displaced persons, which in turn become the vectors of further instability. The involvement of external actors, as we shall see, is inseparable from the perceived threats posed by mobilized refugee diasporas to their countries of origin, as well as to specific communities within the host country.

Historical Backdrop

Let us begin with a brief reminder of basic historical facts.

RANKED SOCIETIES, EXCLUSION, AND INSURRECTION

In the context of ranked societies like Rwanda and Burundi, where a two-tier structure of ethnic domination tended to vest power and privilege in the hands of the Tutsi minority, political exclusion was the rule for roughly 80 per cent of the population, consisting essentially of Hutu peasants. In Rwanda, the Hutu revolution of 1959–62—powerfully assisted if not engineered by the Belgian authorities—brought to a close the era of Tutsi hegemony.⁵ While opening the way for the enthronement of the representatives of the Hutu, an estimated 200,000 Tutsi were forced into exile in neighbouring and other countries between 1959 and 1963—approximately 70,000 to Uganda, 25,000 to the Congo and 50,000 to Burundi.⁶

In Burundi, by contrast, where the “premise of inequality” was far less institutionalized and social relations more complex, ethnic polarization proceeded at a slower pace, allowing the Tutsi elites to consolidate their grip on the government and the army long before they faced the challenge of a servile insurrection. Every attempt made by Hutu leaders to overthrow the government—in 1965, 1969, and 1972—ended in dismal failure, each time resulting in extremely brutal repression, culminating in 1972 with the genocidal massacre of anywhere from 100,000 to 200,000 Hutu.⁷ Not until 1993, with the election of a Hutu to the presidency, Melchior Ndadaye, were the Hutu given to believe that they would soon control their political destinies, only to be robbed of this opportunity on October 21, when a radical faction within the all-Tutsi army killed the newly elected president, the speaker, and deputy-speaker of the National Assembly and overthrew the government. Six months later, after three and a half years of bitter civil war, opposing the predominantly Tutsi troops of the RPF against the FAR, Rwanda became the scene of one of the biggest genocides of the last century: between 600,000 and 800,000 people, mostly Tutsi, were sent to their graves by Hutu militias (interahamwe) and army men.⁸

THE BANYARWANDA OF EASTERN CONGO

Until then, the principal victims of political exclusion were the Tutsi of Rwanda and the Hutu of Burundi. Their closest analogs in eastern Congo were the “Banyarwanda,” a label that belies the diversity of their ethnic and regional origins.⁹ Included under that rubric are three distinctive communities: (a) Hutu and Tutsi who had settled in the Kivu region long before the advent of colonial rule, including a group of ethnic Tutsi indigenous to south Kivu (located in the Mulenge region) known as Banyamulenge; (b) descendants of migrant workers, mostly Hutu, brought in from Rwanda in the 1930s and 1940s under the auspices of the colonial state; (c) tens of thousands of Tutsi refugees who fled Rwanda in the wake of the 1959 Hutu revolution, and hence referred to as fifty-niners.

By 1981, following the promulgation of a retroactive nationality law, the Banyarwanda were for all intents and purposes denied citizenship because none could possibly meet the legal requirement of proof of ancestral residence before October 18, 1908, when the Congo Free State formally became a Belgian colony. By 1990, at the time of the RPF invasion of Rwanda, Banyarwanda resentment of Mobutu's exclusionary policies were matched by their growing sympathy for the cause of the RPF. Many did in fact join the ranks of the RPF and fought alongside their Ugandan kinsmen. By then both groups shared the deepest anxieties about their future in their respective countries of asylum. They would soon become critically important actors in the regional political equation.¹⁰

The devastating ripple effects of the Rwanda cataclysm were felt immediately in eastern Congo. The sudden influx of over a million Hutu refugees across the border, accompanied by the fleeing remnants of the FAR and interahamwe, brought a major environmental and human disaster to the region, while at the same time triggering a drastic reordering of ethnic loyalties. Almost overnight the Banyarwanda community split into warring factions, pitting Hutu against Tutsi.¹¹ Meanwhile, in the interstices of the Hutu-Tutsi tug-of-war, emerged a shadowy constellation of armed factions, the Mai-Mai. Drawn from ethnic groups indigenous to the region—Hunde, Nande, Nyanga, Bashi, and so forth—to this day, the Mai-Mai are notorious for the fickleness of their political loyalties, the fluidity of their political alignments, and their addiction to violence. Swiftly responding to changing circumstances, they first turned against Hutu elements, then against local Tutsi, and ultimately against the Rwandan invaders and their Congolese allies.

1996: THE TURNING OF THE TIDE

The destruction of the refugee camps by units of the Rwandan Patriotic Army (RPA), in October 1996, marks a turning point in the tortured history of the region. It signals the meteoric rise to power of Laurent-Desiré Kabila as the deus ex machina imposed by Museveni upon Kagame to lead the anti-Mobutist crusade under the banner of the AFDL. While the AFDL and its Rwandan allies fought their way to Kinshasa, forcing Mobutu to throw in the sponge in May 1997, the shooting up of the camps released a huge flow of refugees across the Congo, fleeing the RPA's search-and-destroy operations. The attack on the camps also marks the entry of new international actors in the Congolese arena, most notably Rwanda and Uganda. For a brief moment, the surge of popular enthusiasm caused by the overthrow of the Mobutist dictatorship seemed to submerge factional and ethnic divisions—but only for a while. With a substantial presence of Rwandans on the ground acting in military and administrative capacities, anti-Tutsi feelings rapidly spread among a broad spectrum of the Congolese population in North and South Kivu, in the Katanga, as well as in the capital city. Unable or unwilling to discriminate between Rwandan Tutsi, on the one hand, and Banyamulenge and fifty-niners on the other, for the self-styled “Congolais authentiques,” anyone with the looks of a Tutsi would be fair game when push came to shove in July 1998.

1998: THE TURNING OF THE TABLES

The next and most critical stage in the Great Lakes saga came in August 1998 when, sensing the liabilities involved in his dependency on Tutsi “advisors,” the new king of the Congo took the fateful step of turning against the kingmakers, thus paving the way for a replay of 1996. Yet the state of the play on the ground was now very different from the quasi-unanimous crusade of 1996. As 1998 drew to a close, no fewer than six African armies were involved, albeit to a greater or lesser extent, on the side of Kabila (Angola, Namibia, Zimbabwe, Chad, Congo Brazzaville, and the Sudan). Against this formidable coalition stood the fragile alliance of Rwanda and Uganda and their Congolese client faction, the Rassemblement Congolais pour la Démocratie (RCD), soon to break up into two rival groups, while a third rebel faction emerged in northern Congo, Jean-Pierre Bemba's MLC.

The 1998 crisis brought to light an immediate hardening of anti-Tutsi sentiment throughout the Congo, and particularly in North and South Kivu, where it was now the turn of the Congolese “autochtons” (i.e., non-Banyarwanda) to pay the price of exclusion. Denied all possibility of political participation, economically exploited by Rwandan interlopers, and trampled underfoot by foreign occupying forces, their most salient common characteristic is their visceral hatred of all Tutsi, whether of Rwandan or Congolese origins. Little wonder that today the Mai-Mai are increasingly training their gun sights on RPA units operating in the Kivu—and in the process, unleashing a terrible retribution upon civilian populations—as well as on the Banyamulenge, even though the latter fully qualify as “autochtons.” Evidently, their deep historic roots in South Kivu do not exonerate them of the suspicion of being in league with the Kagame government. The truth is that the Banyamulenge and the ethnic Tutsi, in general, are anything but united in their attitude toward Kigali. Many Banyamulenge resent the fact that they have been instrumentalized by Kagame, that they have become mere pawns in the regional poker game. Most of them, however, privately admit that Rwanda's military presence in eastern Congo is their sole protection against another genocidal carnage.

To sum up: exclusion does not just suddenly materialize out of the primeval fissures of the plural society; its roots are traceable to the rapid mobilization of ethnic identities unleashed by the democratization of societies built on the “premise of inequality” and to the profoundly discriminatory implications of public policies directed against specific ethnic communities. In all three states, however, refugee flows were the crucial factor behind the rapid polarization of ethnic feelings in the host countries. Everywhere, refugee-generating violence has produced violence-generating refugee flows.

Dimensions of Exclusion

In the context of this discussion, political exclusion means the denial of political rights to specific ethnic or ethnoregional communities, most notably the right to vote, organize political parties, freely contest elections, and thus become full participants in the political life of their country. Obvious cases are the Tutsi in post-revolutionary Rwanda and the Hutu in Burundi until the 1993 aborted transition to multiparty democracy (some might argue that relatively little changed since then), to which must be added the Banyarwanda of eastern Congo, after being disenfranchised by the 1981 nationality law, as well as the Tutsi refugee diaspora of Uganda, for whom naturalization was never envisaged. Admittedly, political exclusion is a relative concept, both in terms of the range of disabilities suffered by the excluded communities, and the context in which it occurs. It is easy to see why, for example, in the context of Mobutu's dictatorship, the withdrawal of citizenship rights from the Banyarwanda did not produce the same violent reaction as the refusal of the Burundi authority to recognize the victory of the Hutu at the polls in 1965. Again, it is one thing for a minority to be politically excluded and quite another for a group representing 80 percent of the population to be reduced to a silent majority, as is clearly the case today for the Hutu of Rwanda.

Economic exclusion, on the other hand, refers first and foremost to the denial of traditional rights to land. Given that land is the principal economic resource of peasant communities, denial of access to land use inevitably implies economic impoverishment or worse. Here again contextual factors are important. Although rising population densities and environmental degradation are everywhere a fundamental aspect of the land problem, nowhere is the problem more acute than where land has been redistributed to meet the needs of machine politics (as in pregenocide Rwanda), or reallocated to new claimants (as happened in North Kivu in the 1970s when tens of thousands of acres of land were bought off by Tutsi fifty-niners), or where rural insecurity becomes a pretext for massive population transfers in regroupment camps (as in Burundi and northern Rwanda).

Social exclusion goes hand in hand with the erosion of traditional social networks and the collapse of the safety nets that once supported the traditional social order of peasant communities. The result is a growing marginalization of rural youth. Deprived of the minimal economic security and coping mechanisms built into the customary social nets, yet denied the opportunity to make their mark in life through alternative channels, their life chances are almost nil.

To be sure, political exclusion does not always imply economic exclusion. If there is little doubt that the 1959 Hutu revolution in Rwanda received its impetus from the political exclusion of Western-educated Hutu elites, it is equally clear that economic exclusion had relatively little to do with the Hutu-Tutsi conflict. One might even argue that in some instances, withdrawal of political rights translates into rising levels of economic achievement for the excluded community, as shown by the large number of relatively well-to-do Tutsi entrepreneurs in pregenocide Rwanda. Nonetheless, processes of political, economic, and social exclusion are closely interconnected: just as refugee diasporas have exacerbated the problem of natural resource scarcities in the host countries, most conspicuously in eastern Congo and to a lesser extent in Uganda, the resultant shrinkage of land capable of cultivation, along with the dislocation of traditional social networks, must be seen as major contributory factors to the marginalization of youth and the rise of armed militias. The cumulative effect of these phenomena is nowhere more potentially disruptive than where specific ethnic communities bear the full brunt of economic and social exclusion.

Refugee flows provide the conceptual link among all three forms of exclusion. Not that refugees are always on the losing side economically, although in most cases they are. The more important point is that the side effects of large numbers of refugees moving into any given country of asylum translates into severe economic and social hardships for the host society. Rising commodity prices, the rapid depletion of environmental resources, and the frequency of petty crimes within and outside the camps, not to mention the systematic raiding of cattle, crops, and vehicles (as happened in eastern Congo in 1994), are all part of the catalogue of deprivations inflicted on the host communities. In such circumstances, refugees become an easy target for politicians eager to translate diffuse grievances into political capital. In different circumstances, however, they also can be mobilized by opposition groups to strengthen their hand against domestic foes, as indeed happened in Uganda in the 1980s and in Burundi in the 1960s. Refugee populations, in short, have served as a major political resource, either as foil or as a source of support.

The Politics of Mobilized Diasporas

Since 1959 the multiplicity of crises experienced by Rwanda and Burundi have generated four major refugee flows: (a) between 1959 and 1963 an estimated 150,000 Tutsis fled Rwanda in the wake of the Hutu revolution, the majority seeking asylum in Uganda, Burundi, and eastern Congo; (b) the second major exodus involved approximately 300,000 Hutu from Burundi fleeing the 1972 genocidal massacres of Hutu by the Tutsi-dominated army, most of them headed for Tanzania and Rwanda; (c) the next wave of Hutu refugees from Burundi, numbering perhaps as many as 400,000, of whom more than half ended up in Rwanda, followed the reciprocal massacres of Tutsi and Hutu, triggered by the assassination of President-elect Melchior Ndadaye on October 21, 1993, adding tens of thousands to the refugee camps in Tanzania, Rwanda, and South Kivu; (d) in 1994, the fourth and largest outpouring of refugees involved approximately two million Hutu from Rwanda fleeing the avenging arm of the FPR. Over a million settled in eastern Congo, the rest in Tanzania.

All of the above qualify as mobilized diasporas, in that they shared specific political objectives, were politically organized, and made a sustained effort to consolidate their grip on the refugee population. This is still the case for the Hutu diaspora from Burundi and what little is left of its counterpart from Rwanda. Ultimately, their overriding goal was to return to their homeland as citizens, by force if necessary. So far only the Tutsi refugees, under the banner of the FPR, after thirty-five years of exile were able to do so.

But if the saga of the Tutsi diaspora is a success story of sorts—but at what price!—its early history is a tale of consistent failure—political and military—causing enormous bloodshed inside Rwanda, a situation for which there are tragic recent parallels among the Hutu diasporas from Burundi and Rwanda.

Refugees are first and foremost an object of humanitarian concern; only at a later stage, after metamorphosing into a mobilized diaspora, do they emerge as a source of political concern for domestic, regional, and international actors. The obstacles in the way of effective political mobilization cover a wide gamut: the material and emotional costs of uprootedness, the geographical dispersal of the camps, the inadequacy of communication facilities, factional rivalries, and the constraints on political activities imposed by the host country are the usual handicaps faced by refugee diasporas. These disabilities vary enormously over time, however, and from one setting to another. The single most important conditioning factor, however, lies in the receptivity of the host country to the political goals and organizational efforts of refugee communities.

THE FIFTY-NINERS IN EASTERN CONGO: INYENZI AND MULELISTES

A brief comparative glance at the record of the first Tutsi diaspora, in the early sixties (the fifty-niners), with that of the second generation of refugee warriors in the 1990s, is instructive in this regard.¹² Even more revealing is the comparison with the Hutu diasporas.

Organizational strength, internal cohesion, leadership skills, and the ability to draw maximum tactical advantage from the domestic politics of the host country; these are the key ingredients that spell the difference between success and failure. On each count, the record of the Tutsi fifty-niners can only be described as dismal. Though formally affiliated with the monarchist Union Nationale Rwandaise (UNAR), the party virtually disintegrated after its leadership was forced into exile. While some Unaristes joined hands with the Muleliste rebellion in eastern Congo in 1964–65, a small group went to Communist China for military training; others, labeled inyenzi (“cockroaches”) by the new Rwanda government, opted for a “direct action” strategy and proceeded to launch armed raids from Burundi, the Congo, and Uganda, only to be repulsed—at great cost to themselves and Tutsi civilians inside Rwanda—by the Rwandan National Guard and their Belgian advisers.¹³Despite substantial support from a group of radical Tutsi politicians in Bujumbura (but not from the Crown), they never were able to translate this informal alliance into an effective military posture. In eastern Congo, their tactical alliance with the Banyamulenge of South Kivu proved short-lived; the Banyamulenge rapidly switched sides after the setbacks inflicted to the Mulelistes by the Armée Nationale Congolaise (ANC). Even more damaging to their ultimate goals was their international image as crypto-communists in league with Communist China.

THE SECOND-GENERATION TUTSI DIASPORA: UGANDA

The second generation of Tutsi exiles drew important lessons from their elders' inability to get their act together. None were more aware of the necessity to clean up their act than the Ugandan exiles who provided the spearhead of the military crusade that ultimately led to the capture of power in Kigali in July 1994. Though space limitations do not permit a full discussion of their troubled history, most observers would agree that the key to their success lies as much in their organizational skills as in their ability to make the most of the opportunities offered by the rise in 1981 of the anti-Obote guerrilla movement headed by Yoweri Museveni, the National Resistance Army (NRA).¹⁴

Already in the 1970s, the Rwandan Alliance for National Unity (RANU) provided a coherent organizational frame for mobilizing support within and outside Uganda: collecting funds, coordinating cultural activities, reaching out to the international community, and lobbying for the right to return to Rwanda. Between 1981 and 1986, when the NRA seized power in Kampala, a solid phalanx of second generation fifty-niners joined Museveni's movement, fought pitched battles in the Luwero triangle at a cost of 60,000 killed in action, and ultimately gained strategic access to Museveni's security apparatus when two of their officers, Fred Rwigema and Paul Kagame, respectively, rose to the positions of Deputy Minister of Defence and Deputy Chief of Military Intelligence.

Meanwhile a series of initiatives from Tutsi exiles in Uganda and the United States led to the birth, in 1987, of the RPF and the tacit endorsement by many of its leaders of the military option of a return by force. By the eve of October 1, 1990 and the attack on Rwanda, the RPF had grown into a powerful politicomilitary organization, combining political mobilization and military training with wide-ranging lobbying activities in the United States and Europe. By then, its recruitment net extended to Tutsi exile communities in Burundi, Kenya, Tanzania, and eastern Congo, infusing further strength into its ranks. Only after its capture of power on July 4, 1994, did the RPA develop into a formidable military machine, capable of effectively projecting its muscle into eastern Congo and beyond.

THE HUTU DIASPORAS

If the destinies of the RPF were served by exceptionally good fortune, the same cannot be said of the Burundi Hutu diasporas. Although the 1972 diaspora gave birth to the Parti de la Libération du Peuple Hutu (Palipehutu) in the Mishamo refugee camp in Tanzania in 1980, at no time was the party able to aggregate a range of political and military resources comparable to the RPF; its leadership never was able to match the organizational and strategic talent of a Rwigema or a Paul Kagame, let alone the latter's diplomatic skill in reaching out to external actors.

At no time was the party able to capitalize on anything like the extraordinary good luck of the FPR in Uganda in the early 1980s. Burundi exiles are notorious for their lack of internal cohesion.¹⁵ Their history is one of incessant splits, whether in Europe, in Rwanda, or in Tanzania. Their fissiparous characteristics became even more evident after the 1993 exodus and the emergence of several military wings of rival parties, the Front de Libération National (Frolina), the Forces pour la Défense de la Démocratie (FDD), and the division of the Palipehutu into three separate factions. Though some are said to draw benefits from the smuggling networks in Kigoma, and more recently from the shipment of arms from Zimbabwe and Kinshasa, their resource base is hopelessly inadequate for the task at hand: “All of the rebel groups (in Tanzania) complain of the lack of funding, arms and other resources necessary to carry out a sustained military campaign in Burundi.” ¹⁶Again, compared to the RPF's ability to draw international support (most notably from the U.S. Committee for Refugees in Washington) and visibility, the performance of the refugee factions on that score is less than impressive. “The main complaint of the rebels,” notes a recent International Crisis Group report, “is the lack of international support. As one rebel leader said: ‘ we don' t have anyone to support us the way the Banyamulenge are supported by Rwanda and Uganda.’ ” ¹⁷

The case of the 1994 refugees from Rwanda is unlike any other in terms of the magnitude of the human flow, the volume of weaponry transferred, the tightness of the political and military encadrement, the extensive support it received from the Mobutist state, its devastating impact on the natural environment, its catalytic effect on ethnic loyalties, the questions its raises about the political implications of humanitarian aid, and last but not least, the ultimate tragedy they suffered at the hands of the RPA. To review each of these dimensions would take us too far afield. Suffice to note that the seriousness of the threats posed to the new Rwandan state was without parallel in the history of mobilized diasporas. Exceptional circumstances called for exceptional measures. However, the destruction of the camps in October 1996 by the RPA was part of a wider underlying design, namely, not just to “secure” Rwanda's western border but to (a) to extend the search-and-destroy operations to the campsites in South Kivu and in so doing, deal a crippling blow to the refugee populations sheltering extremists (b) deny Uganda's armed opposition movements (notably Tabliq and the West Nile Liberation Front) access to safe havens in the Congo, and (c) pave the way for Kabila's “second coming.” On each count the Kagame strategy succeeded beyond all expectations, at least in the short run. From a wider perspective and with the benefit of hindsight, it is clear that the ultimate goal of the operation—making the Congo safe for Rwanda—has fallen somewhat short of the master planners' expectations.

THE TOOLS OF POLITICAL MOBILIZATION

As this discussion makes clear, contextual variables are of critical importance in explaining the success or failure of mobilized diasporas. Nonetheless, agency also matters. Context alone is not enough to explain the different tools and techniques that enter into processes of mobilization: ranging from coercion to ideological manipulation, from rumor mongering to arms smuggling, from practices and attitudes borrowed from the world of the invisible, and to the use and misuse of information designed to raise the political awareness of the rank and file. Not all of these are productive of success.

Coercive mobilization by some factions of the Hutu diaspora has often had the opposite effect of what was intended, causing tremendous disaffection among civilians and bitter rivalries among exile factions. What some factions view as legitimate means of ideological mobilization—such as the diffusion of historical narratives designed to demonize the Tutsi enemy, a favorite Palipehutiste technique—others tend to reject. Recourse to magic looms large in the arsenal of Congolese factions, notably among the Mai-Mai and the RCD, sometimes with disastrous consequences for the families and communities to whom, wrongly or rightly, magic powers have been attributed. Next to the availability of funding and weapons, information (or misinformation) is of critical importance. On that score, the performance of the Tutsi diaspora in Uganda ranks far above its Hutu counterparts.

Quite aside from its efficiency in collecting funds from exile Tutsi communities and gaining a privileged access to NRA equipment, compared to Hutu refugee movements, the Uganda exiles have been remarkably adept at mobilizing support through their skillful manipulation of information. This fact goes far in explaining its capacity to sway international public opinion long after the diaspora had become a nation.

Another Look at Theory

What new light do the theories mentioned earlier shed on the dynamics of ethnoregional conflict in the Great Lakes? The answer, in part, depends on how they “fit” into any particular aspect of the crisis.

Let us begin with Huntington. The whole drift of our argument, centered on the concept of exclusion, can be read as a refutation of the “clash of civilization” thesis; by the same token, his discussion of the “kin country syndrome” is of direct relevance to an understanding of the patterns of ethnic mobilization unleashed by refugee diasporas. As our previous discussion makes clear, where ethnic fault lines cut across national boundaries, conflict tends to spill over from one national arena to the next, transforming kin solidarities into a powerful vector of transnational violence. An action-reaction pattern sets in, whereby victims in one setting become instigators of violence in the other. Largely missing from Huntington's discussion, however, is a sustained attention to mobilization strategies, including the kinds of resources employed to mobilize support.

This is where Collier's paper offers some challenging insights. I refer specifically to his analysis of the role of diasporas and access to financial resources as crucial factors in explaining the risk of civil war. On the other hand, serious questions arise as to whether the financial viability of rebel factions, including refugee diasporas, is entirely reducible to the opportunities offered by commodity export economies. If this were the case, the whole of the continent would be tottering on the brink of insurrection. Not just any export commodity but gold and diamonds are the rebels' best friends.

Whether through gem trading or any other source of profits, financial viability matters. There is no denying the cardinal importance of the looting of gold and diamond resources in eastern Congo in the funding of the war effort by Kigali and Kampala and of the deadly rivalries over the loot in pitting Rwanda against Uganda in Kisangani. Nonetheless, “financial viability” only tells part of the story. Crucial as they are in explaining the failure or success of mobilized diasporas, contextual opportunities are not limited to financial viability; equally important is the political viability of rebel and refugee movements, most notably their ability to negotiate political and military support. This is true not only of the RCD factions today, but was certainly the case for the second generation Tutsi refugees in Uganda in the 1980s.

Where the Collier thesis seems most vulnerable is in the rejection of objective socioeconomic indicators as a source of civil violence: “Objective measures of social grievance, such as inequality, a lack of democracy, and ethnic and religious divisions, have no systematic effect on risk…because civil wars occur when rebel organizations are financially viable.” ¹⁸ Quite aside from the fact that the argument simply doesn' t hold up in the face of the overwhelming evidence to the contrary—a fact that Collier might conceivably explain away by relegating Rwanda, Burundi, and eastern Congo to deviant cases—one wonders why one set of independent variables (objective measures of social grievance) should exclude the other (financial viability).

Categorically dismissing rebellion as “protest motivated by genuine and extreme grievance,” Collier offers a striking analogy: “For a few moments suspend disbelief,” he writes, “and suppose that most rebel movements are pretty close to being large-scale variants of organized crime. The discourse would be exactly the same as if they were protest movements.” ¹⁹ Nowhere, however, does he consider the alternative proposition that the state might qualify as the criminal and the rebels as victims of state crimes. This is, of course, the central argument set forth by Bayart, Ellis, and Hibou in their recent work on the criminalization of the state.²⁰ This is not meant to deny the propensity of rebel and refugees, and refugees turned rebels, to engage in criminal activities, yet it is important to note that the phrase covers a wide spectrum of illegal activities and that such criminal activities often pale in comparison with those carried out by the state. Rwanda under Habyalimana, Zaire under Mobutu, and the Burundi armed forces under Buyoya all exhibit, to some degree or another, at one point or another, what can only be described as a criminal behavior of the worst kind, including political assassination, theft, and corruption on a grand scale. It is difficult to avoid the conclusion that the result has been to promote huge social and economic inequalities, along with corresponding “genuine and extreme grievances,” and thus pave the way for the exclusionary policies that lie at the heart of ethnic violence in the Great Lakes.

The Herbst thesis has the merit of looking at a range of variables seldom taken into account by political scientists: the combined effect on state failure of low population densities, weak and artificial boundaries, and the resultant inability of the state to control its hinterland; this, he adds, is in striking contrast with the historical record of European states, all of which have experienced “the brutality of interstate war” as a major ingredient of state consolidation.²¹ On each of these counts, however, the recent history of the Great Lakes offers massive counter-factual evidence. The region claims the highest population density in the continent; the precolonial boundaries of the interlacustrine kingdoms of Rwanda and Burundi were fairly well delineated, at least by comparison with the rest of Africa; control of these states over the hinterland was relatively well established; and the “brutality of interstate war” was a major feature of their precolonial histories, though by no means comparable to the devastation caused by the internal and interstate wars currently ravaging the region. What Herbst leaves out of the picture is the impact of colonial and postcolonial history. He leaves out what Crawford Young has so ably brought into view—the enduring disabilities arising from the impact of the colonial state on African societies.²² Predictably, Herbst makes no reference to the multifaceted crises of exclusion and social marginalization around which much of this discussion revolves and for which there are many parallels in the continent. Only by confusing optimism with fantasy and reality with illusion, can one take comfort in the view, implicit in the Herbst thesis, that the violent confrontations in former Belgian Africa will ultimately bring to the region the benefits of state consolidation along a bloodstained path similar to the one historically taken by European states.

Policy Implications

By postulating exclusion as a crucial dimension of the Great Lakes crisis, we do not mean to suggest that its conceptual opposite is the only solution to the region's woes. Inclusion is a theme that admits many variations. It can easily mask a policy of cooptation and serve as a substitute for a genuine sharing of power; carried to an extreme, with little or no attention paid to contextual realities, the result may be chronic instability, as happened in Burundi in 1995, following the so-called Convention of Government of 1994. The diffusion of ethnic violence across national boundaries, sustained by external forces, imposes severe limitations on the benefits of power sharing.

The case of Burundi is instructive in this respect. A key provision of the precarious peace deal worked out in Arusha (Tanzania) in July 2000, through Nelson Mandela's painstaking facilitating efforts, involves a broadly based three-year transitional government incorporating the representatives of fifteen parties, almost evenly distributed between predominantly Tutsi and Hutu parties. For the next eighteen months a Tutsi (Pierre Buyoya) will serve as president and a Hutu (Domitien Ndayizeye) as vice president; during the following eighteen-month period, the roles will be reversed. This power-sharing arrangement is bolstered by more fundamental concessions to the Hutu majority, such as a commitment to restructuring the all-Tutsi army on the basis of ethnic parity. Demands which until recently were non-negotiable have now been met, such as the presence of a 1,400-strong South African peace keeping force; others will be negotiated in months ahead, such as the restructuring of the army, the dismantling of regroupment camps, and the appointment of an international judicial commission to address issues of justice and impunity. Incentives to cooperate go beyond the allocation of portfolios to a wide array of coalition partners. To the extent that the Burundi state is becoming less hegemonic and more open to the demands of the Hutu majority, it has gone a long way towards promoting a climate of trust. Nor has the international community faltered in its efforts to reward cooperative behavior, as shown by the promise of a generous aid package from the European community ($440 million).

Despite efforts to widen the scope of Hutu-Tutsi cooperation, the Arusha framework faces an uncertain future. Inclusionary strategies so far have failed to convince the leaders of two extremist Hutu rebel groups, the FDD and the Palipehutu-FLN, to lay down their arms. That their obduracy is in part motivated by the financial and military support they have come to expect from external actors—whether from Zimbabwe or Congo-Kinshasa or from their ethnic kinsmen in the Congo or Tanzania—is reasonably clear. Equally plain is that Hutu extremism is bound to generate a response in kind from Tutsi hard-liners. Expectations of strategic assistance from external actors are not limited to any one group. They constitute major incentives for extremists at both ends of the ethnic spectrum not to cooperate in power-sharing arrangements.

If so, there are compelling reasons for encouraging the full implementation of the Lusaka accords (1998), especially as regards the withdrawal of foreign armed forces, the disarming of the so-called “negative forces” (the interahamwe militias and Mai-Mai factions), and the resumption of an inter-Congolese dialogue. The success of the Arusha accords is intimately linked to a global settlement in the Congo. Only if the rebel factions in Burundi are deprived of the support of secondary level participants (i.e., Zimbabwe or Congo-Kinshasa) will they join in the peace process or at least desist from violence. The same is true of the rebel RCD factions in the Congo, currently supported by Uganda and Rwanda.

Given the considerable economic stakes they have in the conflict, neither Rwanda nor Uganda are likely to envisage a withdrawal of their armies unless faced with vigorous pressure from the international community. Though providing convenient justification for their presence in eastern Congo, security concerns are of secondary importance for Presidents Kagame and Museveni; far more significant as a policy imperative are the enormous profits derived from the wholesale plunder of the Congo's mineral wealth. This is where a major reappraisal of strategic priorities is needed from donors, specifically from the World Bank, the United States, and Great Britain. By turning a blind eye to the “imperial” designs of Rwanda and Uganda in eastern Congo, while at the same time rewarding their economic performance, donors are in effect subsidizing their war effort. The time has come to recognize the fundamental contradiction involved in the pro forma support of the Lusaka accords and assistance policies designed to undermine their full implementation.

A more fundamental contradiction exists between the ethos of participatory politics and the exclusionary implications of the foreign-linked clientelism operating in much of the Great Lakes region. Reinforcing the neopatrimonial features of their domestic politics, Rwanda and Uganda have developed multiple linkages—economic, military, and political—with their respective client factions, but these linkages extend far beyond the boundaries of the Great Lakes. Corporate interests in the West and elsewhere in the world also have stakes in the rents generated through the illicit exploitation of the Congo's resources. The unpalatable truth is that the multiplicity of interests in support of the regional status quo far exceed the pressures of the United Nations for implementing the Lusaka accords. Nonetheless, no matter how daunting the obstacles ahead, Lusaka is the only roadmap for charting a new course towards peace. From all the evidence, this basic truth has yet to sink in among certain key members of the international community.