Life After Violence
Page 19
The differences between communes result entirely from the behavior – the aims, the ideologies, the power relations – of two small groups of local elites. The first group is composed of what one can broadly call the local administration and social elite. The communal administrator and various chefs de zone and colline, back then all nominated from above, are evidently in this group, but it also includes people like the directors of the local schools and hospitals, as well as some people not hired by the state: priests and monks, and the rare development professionals living locally. These educated and mostly administrative people act as the transmission belts between the center of the country, the government and the party (back then a single party), and the bases. They pass on the mots d’ordre, they control the population, they brief the higher levels on what is going on, who is malcontent and why. They truly are the crucial link between the center and the rest of the country. Whoever has the allegiance of these people controls the countryside (thus making local-level appointments a crucial matter not only of redistributing benefits to supporters, but also of ensuring political control over the territory).
Another group that influences local political dynamics is composed of the ressortissants – people who were born in the commune and have made it into senior positions in ministries, public enterprises, the UN or NGOs, and so on. Part of what these people do is simple development work. But they also engage in serious local political involvement and mobilization. Burundi’s decentralization law explicitly allocates a significant proportion of positions on the communal council to this group – an open invitation for political aspirants to maintain local networks of patronage and clientelism, to undermine local democracy and downward accountability, and to spread political ideologies and competition from the center to the periphery – an invitation that many of them gladly accept.
It is in the competition and alliances between these small groups of local elites that communal politics unfolds. They buy off local intermediaries, drive into the hills at night and organize ‘secret’ meetings, distribute machetes and beer if serious violence needs to be organized, etc. They know who thinks what, who has a conflict with a neighbor from a different ethnicity, who is desperate for money – all levers of potential violence, should the need arise. This is the micro-politics of violence. It also means that it is crucial to involve these people in conflict resolution and leadership workshops: they, much more than the population at large, are the crucial sparks to light the fire of violence. Explicitly targeting local ‘negative leaders’ – local elites who are known to be extremist and exclusionary or simply powerful – is absolutely crucial in this respect: development practitioners tend to avoid these ‘difficult’ people, but the real need is to work with them.
Insecurity The third factor that is commonly misunderstood or neglected in discussions about civil war in Burundi is violence and chaos. A lot of the violence, whether rural or urban, was committed in a climate of fear, chaos, and insecurity. This statement sounds tautological but what it means is this: in societies where the rule of law is close to non-existent and security forces are neither effective nor trusted, small groups of people willing to use violence can create enough chaos and fear to force everyone into making violent choices.
In other words, it is not necessary that all of society partake in extremist ideologies for extremists to force all of society into awful, often de facto ‘extremist,’ choices. Most people will simply flee, losing their meager possessions in the process. But many others will fight back to defend themselves and their families, will be tempted to opportunistically seek personal benefits, will become angry and strike back blindly, and so on – committing untold acts of so-called ethnic violence they would not have contemplated under normal circumstances. As an example, in our interviews the foremost cause ex-combatants gave us for why they joined rebellions, or the army, was fear and insecurity (Uvin 2007b). They had been attacked, they were afraid, school had been closed, they had fled and were without their parents. Many of these youths wanted to do other things with their lives but chaos and destruction – initially caused by a small group of people – turned their lives around. Many Burundians understand this: it is one of the reasons they do not necessarily want to discuss or punish all violent acts of the past, for they sense that things were done that people did not want to do, under extreme circumstances. Clearly, then, a law enforcement approach – the coercive imposition of order, as done by police forces, investigative units, the penal system – ought to be at the heart of conflict prevention, and it needs to work well before the chaos and violence become so widespread as to create a climate of fear. This is akin to the squeegee approach to law enforcement that was made famous by Rudy Giuliani when he was still chief of the New York police – the idea that, in order to bring down massive serious crime rates, you have to start with ending the many smaller crimes (things like jumping the metro, vandalism, intimidation), for they create a climate of fear and lawlessness that is conducive to serious crime. Note that these sorts of debates and activities are far beyond the comfort zone of the international development community.
Grievance So far we have treated the war in Burundi as a giant pogrom, an instance of mass violence – although one instrumentalized by elites, local and national, seeking their personal benefits. And part of the war in Burundi did display these characteristics: the pogroms against Tutsi after the murder of Ndadaye in many parts of the country, the ethnic cleansing of Bujumbura, and so on. But the civil war was also rooted in a rational political agenda, which was widely shared among many Hutu: the need to overthrow a regime and defeat an army that was widely seen as intent on maintaining exclusive control of the state and all the attendant benefits. One of the prime motivations for Burundian ex-combatants to have joined the rebels – the CNDD and the FNL – was a clear political agenda (discussed in more detail in Uvin 2007b). In my interviews, those who strongly identified with this agenda were more frequently (although by no means exclusively) older; they were also typically not the poorest or the most marginal.
These were people with political analyses and aims, who knew what they were doing. They voluntarily joined the fight,3 and they won their war. That is also why many of them voluntarily demobilized: the job was done, and it was time to return home. It is likely for the same reason that the integration of these ex-combatants – whether child soldiers or adults – went very well: almost all of them moved back to the localities and families they came from and report no problems (ibid.). The war may at times have degenerated into banditry and crime, but that was not its aim, and for many soldiers and their families the general sense of mission did not disappear (Samii 2007 has detailed data on this subject). Burundi’s civil war, then, was not based on the popular explanations of greed (Collier and Hoeffler 2001), natural resources, demography, or frustrated masculinity (Richards 2006).4
I draw two conclusions from this. First, each case is specific, and the general explanations that dominate international thinking about conflict in Africa must never be accepted at face value as universal explanations. At best they are general correlations, but they are surely not always true in the particular, even if the structural factors they are based on prevail in a particular country. The second, and more controversial, lesson I draw is, to state it as a slogan: structure is nothing, politics is everything. All the currently popular explanations for conflict in Africa focus on major structural factors – population density or age distribution; natural resource dependence; economic trends. But these factors explain nothing about why a conflict happens in a particular place at a particular time – and yet, as professionals of development and peace-building, that is what we need to know. Worse, they create the false feeling that we know what matters, which may lead to either false pessimism or false optimism. Hence, for a practice of conflict prevention, these theories are totally useless.
Masculinity and violence
The shock troops who commit most of this violence are typically composed of young men –
not grannies. This brings us to the dominant model in the social sciences today, which we find expressed in very different ways, using different methodologies and causal relations, in the works of Paul Collier (with Hoeffler 2001), Robert Kaplan (1994), Paul Richards (2006), Gary Barker (2005), or Henrik Urdal (2004) – all of which largely treat young men as an imminent danger (Sommers 2007: 2; 2006b: 6). This image is very widespread in Burundi as well, both among ordinary people and among policy-makers: I recall an interview with the top person in the Department of Youth in Bujumbura, who repeatedly justified the need for more resources for youth programs in terms of criminality. He proudly told me that his minister had been successful in getting more funds for a youth employment plan by arguing to his fellow ministers that ‘if you don’t fund this, they will come into your houses to steal your possessions and rape your wives.’
This book has confirmed that many of the factors that scholars describe – joblessness, humiliation, incapacity to marry – do exist in Burundi. Where our analysis differs is in our understanding of the reaction of young men and the societies they live in toward these trends. When young men face great difficulty in achieving normative manhood, they do what most of us do when confronted with major challenges in our lives – they try harder than ever, they seek to innovate, they try to move and find opportunities elsewhere, they turn to God for strength, they hang out with friends and complain – but they do not necessarily become murderers. Burundian masculinity centers on responsibility – taking care of wife and children, as well as of parents. It is true that this definition devalues female contributions to the household, relegating them to the private and the invisible. But it is equally true that this masculinity is not automatically or even primarily violent.
The power of masculinity to explain civil war in Africa has been exaggerated in the scholarly literature so far. The overwhelming majority of young men in Burundi, as elsewhere, faced with the same poverty and thwarted masculinity, have chosen many life paths other than violence, even during the awful years of war (UNDP 2006b: 27ff.). Indeed, less than 3 percent of Burundian young men joined an armed movement during the war (see too Sommers 2007: 3; Barker 2005: 157, 181). Much of this literature on masculinity generalizes far beyond what is acceptable, as well as being overly ‘miserabilizing.’
The relevance of the ‘frustrated masculinity as a driver of violence’ explanation is even more limited in terms of actual programming. Its primary policy implication is the notion that a focus on young men – through education and jobs – is a tool for conflict prevention. While this is certainly relevant – there is nothing young men and women want more than education and jobs – it is not a solid or actionable basis for conflict programming.
There exists no social science that can tell us precisely the identity of the young men who will take up arms, or engage in criminal and violent and destabilizing behavior. As said earlier, it takes only small numbers (and the political entrepreneurs who typically organize and equip them) to create a climate of fear and anarchy that is propitious to the spread of violence. Among the shock troops of violence, the young, the school leavers, the urban and peri-urban, those with disrupted or non-existent families (Brett and Specht 2004: 3), and ex-combatants are probably over-represented, but that still leaves hundreds of thousands of people to ‘target’ for conflict prevention activities. At the same time, the resources available are far from sufficient. Even if one managed to lift a hundred, a thousand, or even ten thousand of these young men out of poverty – an amazing success few projects ever achieve – there will be many, many more who remain equally easy to mobilize or recruit, and many more who have received jobs but who would never have been recruited in the first place. In short, development assistance cannot prevent civil war by providing potential perpetrators of political violence with jobs. Creating jobs for youth is absolutely crucial for development, but it is not a useful conflict prevention tool.
If one wishes to use development funds in such a way as to reduce social dynamics leading to violence, other ideas are more promising. First, there is the already mentioned law enforcement approach, provided this can be done in a way that is impartial and legitimate – a tall order in any country. This was probably impossible in pre-war Burundi, given the composition and the record of the security apparatus. Now, with the integration of the army and police, there are many more possibilities, and this should be a priority for donors and the government.5
Second is improved local accountability: the spread of central-level conflict to the rest of the country takes place through the transmission belt of local elites of all stripes. Any process that increases the availability of information at the local level; that strengthens the habit and capacity of ordinary people to use the legal mechanisms available to control those who govern them; that empowers a broad range of informal and community leaders to act as intermediaries and alternatives to local elites – all these local governance mechanisms are conflict resolution tools in a country where people want to go beyond ethnicity and fear, but have precious few opportunities to learn how to make that happen.
Third: mechanisms and social processes by which central leaders are restrained from using violence as a key tool of doing politics by other means. Burundi has made great progress in this field: the emergence of a vibrant free press, as well as a much deeper awareness by ordinary citizens of the way they are being manipulated. But more can be done, including much more forceful diplomatic action by the international community when political leaders blatantly and consciously destabilize domestic political situations.
The fourth point is related to the first: improvements in economic and political fairness are crucial – the rule-of-law agenda, a tough nut to crack. Even at high levels of poverty, if ordinary citizens feel they have a fair chance of succeeding, they will maintain a stake in the system. Burundian youth are capitalists: they deeply believe in education, hard work, personal initiative, and individual responsibility. The general corruption and social exclusion (including through unequal access to higher education) offend them, make them cynical and angry, and make violence easier to justify or accept (Ndikumana 2005: 16). That is why the fight against corruption is a crucial element of a long-term conflict prevention strategy. From a conflict prevention perspective, more important and more doable than creating jobs, then, is to create a climate in which young people can believe that their hard work will pay off – no development, but also no peace, without institutional change.
Gender
Burundians’ lives are profoundly changing under the tectonic pressures of continuous impoverishment and insecurity, reducing to rubble all that they held dear and thought would last for ever. Gender roles are no exception. Traditional expectations of men and women continue to make up the core of Burundian identity, even among young people. But they are extremely hard to achieve. Failure to achieve normative masculinity does leave young men unhappy and frustrated. And women’s lives, both the joys and the sufferings, unfold in gendered ways as well.
But these same economic and political pressures are simultaneously unleashing different, opposing forces. These are dynamics wherein girls and women are encouraged to study as long as they can, where female dynamism and mutual respect between spouses are increasingly sought, where traditional marriage expectations are relaxed. This conforms to Barker and Ricardo (2006) when they write that ‘Various studies and research […] confirm that many young men simultaneously hold traditional and rigid views about gender alongside newer ideas about women’s equality.’
There is an excessive miserability in much of the masculinity literature. It resembles initial gender – read feminist – scholarship of a few decades ago. When scholars first focused on the situation of women in development a few decades ago, disaggregating women’s experience from men’s and trying to understand the way in which the social construction of gender impacts on social change, the result was overwhelmingly negative. It laid bare – for good reason – the plight of women, the gendered way
s in which their experiences were not taken into account and how they often failed to benefit from so-called development. This early work was soon reproached on the ground that it neglected female agency and creativity and ongoing changes in society. A variation on this theme seems to be prevalent now that the gender focus has started to include masculinity. The new literature is extremely miserabilistic as well – men as suffering from economic crisis and social and political disempowerment. The difference is that these men, presumably, all take it out on others (or themselves). And, again, while much of what is said is true, this initial approach also neglects agency and change. The overwhelming majority of men do not turn violent. Faced with stunning constraints, they seek different ways to survive, to innovate, to find respect, and often in so doing some of them slowly begin reinterpreting gender roles as well – not without resistance, for sure, but appreciably none the less. In its relentless focus on violent behavior and its almost automatic association between young men and violence, the literature has – mostly unintentionally – created a picture that does injustice to the dignity of young men.
I argued that much of the social science scholarship is excessively miserabilistic (which would explain the difference between my results and those in the general scholarship), but it must also be acknowledged that Burundi differs from the other countries where much of this work was undertaken (primarily Sierra Leone and Liberia). Burundi seems more able to adapt to young men’s inability to achieve normative manhood than other societies, and as a result young men’s frustration and marginalization may be less severe than elsewhere. There are some good reasons why this might be so. Burundi is a much more non-hierarchical and amorphous society than most others in Africa and elsewhere, with no strong village chiefs who have the power to enforce rules, no initiation rites to maintain purity. Life has always been more individualistic and centered on the nuclear family than elsewhere, as the traditional dispersed mode of habitat graphically illustrates.6 Second, Burundian society has always valued flexibility, the capacity to bend, but never break, to adapt to changing circumstances. And third, the civil war in Burundi actually had a major ideological aspect to it for most participants, who were defending the future of their people, and were widely seen as doing so by their communities. Perhaps these factors explain why young men in Burundi do not seem to conform as much to the ‘young men equals frustration equals aggression’ model that has come to dominate scholarship.