Is there an inevitable relationship between cities and violence?
It certainly sometimes seems to be the case. We have records concerning urban violence that go back a long way, to the Sumerians and the Romans, and they often specifically associate the phenomenon with city living; the historian Livy for example wrote about the gangs associated with the urban political machine in ancient Rome. In fact, from the 19th century onwards, social science research explicitly began to argue that there was something about cities which led to violence. The idea was that cities were made up of a lot of people concentrated in a very small space, forced to interact with each other, and this caused frictions and led to violence, particularly evident in the form of urban crime. Cities were also widely seen as places where people could come together and overthrow regimes. The French Revolution, for example, started off as an urban revolution.
To a certain extent, this makes a lot of sense. If, for example, you take a paradigmatic form of urban violence, namely youth gangs, there are very few rural gangs, partly because a gang is a collective form and, obviously, if you have youth in different villages kilometres apart, it's quite difficult to get together. At the same time, however, when you look at different cities around the world, you find that some of them are violent and some of them are not. Ultimately, the key factor is less the fact that they're cities and concentrate people in a set space, but rather the way that they're organised. Although cities are definitely spaces that bring people together and force them to interact, the outcomes of this interaction are by no means necessarily given. It is the overarching political economy of cities that is important to consider in this respect: whether certain groups dominate, whether others are excluded, and whether there are mechanisms of inclusion.
So the motivation behind the violence is the decisive criterion?
Yes and no. On the one hand, there's been a lot of work in the recent years trying to define different types of violence according to their motivation – for example, distinguishing between political violence, economic violence, and social violence – in order to justify different types of policy approaches. I don't really agree with this kind of distinction. Although it can be useful to think about the existence of different categories of violence, these can rarely be clearly separated from each other when one faces the reality on the ground. Take crime, for example, which is frequently portrayed as a paradigmatic form of economic violence. Even if criminal practices can definitely be analyzed in purely economic terms, they are also clearly epiphenomenona of wider social forces. Many studies have shown that crime is linked to inequality, for instance, with more inequality leading to higher crime rates. Seen from this perspective, crime, although not necessarily an active political act, can be said to have a political dimension: it is a reaction against unequal social structures. This makes pinpointing the motivation criterion obviously very difficult.
So the motivation criterion is not important?
Once again, yes and no. Focusing on it in a narrow manner arguably diverts attention from certain fundamental issues. The dominant discourse on contemporary gangs in Central America, for example, contends that they are a form of economic or social violence. This ignores the huge levels of social exclusion, territorial exclusion, the lack of any employment opportunities, and perhaps most importantly of any significant policy efforts to actually try and generate employment, which are all fundamental to understanding the contemporary proliferation of gangs in the region. Indeed, in many ways you could say that this dominant discourse is very much something of a smokescreen, because labelling gangs as ‘economic’ or ‘social’ violence diverts attention away from the political issues that underlie their emergence. At the same time, you could also argue that at one level everything is political – not necessarily in the sense of big ‘P’ politics, that is to say of big men and women politicians and statecraft, but rather in the sense of small ‘p’ politics, the way things and society ‘fit’ together. The nature of the order of things the key question, in other words.
In this respect, armed conflicts, civil wars and major fighting were often taking place outside the cities and villagers were most affected. Even recently the conflict in Uganda merely affected the northern part and Kampala was hardly directly affected. How do you explain the shift of armed conflicts to urban areas?
As the anthropologist Eric Wolf famously pointed out, most of the great wars and revolutions of the 20th century were what he called ‘peasant wars’. During the past 50 years or so, a transition has clearly occurred, whereby these ‘peasant wars’ have become much more urban, partly as a result of demographic transition, but also because the global economy is much less agrarian, but increasingly urban production and manufacturing in bigger agglomerations. This has shifted the locus of conflict from the countryside to cities, to what my colleague Jo Beall at the University of Cape Town has termed ‘urban wars of the 21st century’.
So you are saying that a new pattern of urban warfare is developing? That there are less international armed conflicts and less clearly structured internal wars?
Definitely, yes. And it's not necessarily a pattern that is less violent than in the past. If you take Central America, for example, the levels of violent deaths afflicting the region are actually higher now than they were during the years of rather classical civil wars in the 70s and 80s. At the same time, urban violence is clearly much more complex than past forms of rural warfare. Old-style wars tended to involve clearly identifiable parties – those controlling the state and a belligerent group, for example – and generally revolved around taking over the state apparatus, this was the prize, so to speak. These days, it's much more complicated. The new urban wars of the 21st century involve a variety of actors who don't necessarily want to take over the state, but who might be defending some kind of resource, local territory, or may even just be trying to create spaces of order for themselves. Whether you want to call this warfare is another matter. In some of the work that I've been doing with colleagues at the London School of Economics, we often prefer to call it conflict, and we see such forms of violence as part of a continuum which goes from warfare to more prosaic forms of violence such as crime and delinquency.
Can we generally describe this new urban violence as collective or structured violence, i.e. a form of organized violence that at a certain level comes close to situations of armed conflict?
Yes and no. There are two ways of looking at this issue. On the one hand, yes, you can definitely have actively organized collective armed groups, say, an urban guerrilla group. On the other hand, you can also have groups, such as gangs in Central America, for example, which are not organized collectively as such, but can be said to represent something of a collective movement. It's estimated that there are anything between 100,000 and 500,000 gang members in that part of the world, which clearly constitutes a collective trend. When we see a small number of individuals behaving in particular ways, we can perhaps think that it's a coincidence. When dozens do it, then we might think that it's possibly a fad. But when, hundreds and thousands of people are doing it, there is definitely, at some level, some kind of collective movement going on. I realize that I'm probably departing from definitions of collective violence contained in international humanitarian law, human rights instruments or other branches of international law, but as an anthropologist, I'm very much bottom-up in my approach to things.
Let's take a concrete example: in your opinion, is the situation in Northern Mexico a war? The powerful drug mafia uses very violent means and the State reacts with military interventions? Thousand of people die every year
I'm not an expert on the situation in Mexico, but my sense of what's happening is that it's not so much a conflict between drug mafias and the Mexican state per se. Certainly, the drug traffickers are not trying to take over the Mexican state. When you look at the very high number of deaths, a lot of them, the majority result from drug dealers killing drug dealers – it is in other words largely an inter-drug dealer conflict. Obviously, bystanders and the local population are majorly affected, but the situation seems to me to be similar to the one that developed in Colombia during the 1980s, when drug cartels constituted a threat to the state by virtue of controlling certain areas and of having considerable fire power, but weren't necessarily trying to take over the state. That's very different to the conflict between the Colombian government and the FARC, where – at least originally, the situation having become more complicated over time with the latter's involvement in drug trafficking – you have a conflict between two societal models.
So how would you frame this different type of violent confrontation?
With my colleagues at the London School of Economics Crisis States Research Centre, we call these new forms of confrontations ‘civic’ conflicts, and distinguish them from ‘sovereign’ and ‘civil’ conflicts. Sovereign conflicts are conflicts between states, while civil conflicts are conflicts over the state between the state and one or more groups within a society. These can be seen as the ‘classic’ forms of armed violence. Civic conflicts, on the other hand, do not necessarily involve the state. They will involve different groups, some more organized than others. Historically, the trend is clearly for increasing civic conflict worldwide.
Communal riots in India, for example in Ahmadabad in 2002, would qualify as a form a civic conflict. The rioters didn't try to take over the state, although it could be argued that the riots were something of a response to particular state policies that fostered social exclusion. Certainly, research has shown that the triggers for these riots were state policies which led to deeper segregation within the city, a city which historically had been quite mixed. At the same time, the violence itself was quite spontaneous, very much mob-based rather than organized, even if certain groups actively sought to fan the flames of violence.
Crime can also be a form of civic conflict, especially when it can be linked to increasing forms of social exclusion, gated communities, the containment of certain populations, particularly the poor, into ghettos and peripheral areas of cities. The inhabitants of such areas can have difficulty getting out – sometimes even physically – or they become stigmatized and can't get jobs and so on, as the anthropologist Philippe Bourgois has described very powerfully in his work on East Harlem in New York, characterizing this kind of situation as a form of ‘urban apartheid’. This can lead to various forms of criminal and delinquent violence, including, as I have tried to show in my own work, gang violence in Central America.
Do you think that these civic conflicts often develop in post-war societies such as Nicaragua, for example?
Post-war circumstances are certainly a major factor to take into account, but they're not necessarily directly causal, often being more contextual in nature. In the case of the gang phenomenon in post-conflict Nicaragua, for example, it's often pointed out that the end of the war saw the demobilization of a lot of youth who had received military training – the age of conscription in Nicaragua was 16 – which at one level clearly contributed to the explosion of gangs, but not all of these youth joined gangs, of course, and other factors were much more important. In particular, regime change led to a breakdown of many state services, including policing, and crime levels spiralled, partly because the country was suffering a massive economic crisis. Gangs initially emerged as informal forms of local vigilante protection for neighbourhoods in a wider context of chronic insecurity.
Did this have anything to do with ideology, for example with the solidarity associated with Sandinismo?
It had less to do with ideology per se than with ideology as a symbolic reference point. In the Managua neighbourhood where I have been carrying out research since 1996, the youth who congregated in a gang around 1990 shared a very pro-Sandinista discourse. But this had less to do with the content of Sandinismo, but more their experiences coming from a traditionally Sandinista neighbourhood, as well as the fact that they had been demobilized from the Sandinista Popular Army. There are other neighbourhoods in Managua which were settled by Contra returnees post-1990, where the gang mystique is founded on the history of being Contra. To this extent, it's less the intrinsic nature of the ideology that is important, and more the fact that ideologies can become federating reference points.
How does the gang phenomenon relate to the youth phenomenon?
A lot of work has been done recently, particularly by the World Bank, about the risks associated with youth bulges. The basic argument is that if there's a large youthful – and male – population, then there's more risk of violence since most crime is perpetrated by young men. I'm not completely persuaded about this. Although gangs can definitely be associated with youth, insofar as there is an aspect associated with them that can be linked to adolescent development, it's critical to understand that not all youth joins a gang. Most studies seem to suggest that it's generally between 1 to 15% of a given local population that do so, although it's clear that you need a critical demographic mass of youth in order to have enough youth joining a gang.
You mentioned that gangs are often linked to adolescent development, and it's true that all of us have probably at one stage or another in our youth been part of a group that could be labelled a ‘gang’. How are we to differentiate this kind of experience from the Central American gangs of which you are talking? Is there an accepted definition of what is a ‘gang’?
Over 100 years of sociological, anthropological, and criminological research has been trying to answer this question and hasn't really come up with a satisfactory answer. Part of the problem is that the word ‘gang’ is used to describe all sorts of phenomena: from groups of adolescent youth who get together on street corners, and who fight each other irregularly, engage in petty vandalism, or even just hang out, to groups of youth who participate regularly in delinquent activities, to organized crime syndicates, to prison-based gangs, and even to political parties… A lot of research tries to create typologies, and categorize different types of ‘gangs’, but I actually think that this is not terribly useful, and can even lead to problems, insofar as too strict a definition can become quite limiting.
In my work I tend to adopt a rather loose definition, whereby a ‘gang’ involves a group of youth that is more or less recognized as an institutionalized collective unit by wider society, and where the core membership of the group is regularly involved in violence – the core of the group has to be relatively constant over a certain lapse of time, although the broader membership of the group will fluctuate over time – and most gang members tends to be under the age of 25, although this can vary tremendously considering that the notion of youth is a highly flexible socio-cultural category, and in certain societies you can be a ‘youth’ right into your 40s. At the same time, gangs are very volatile social forms, as the US gang researcher John Hagedorn has pointed out, today's youth gang can become tomorrow's drug cartels, which can become the day after tomorrow's ethnic militia. To this extent, what's important about gangs is less what they are, and more the broader social, political, and economic trends they fundamentally reflect.
Are gangs a social phenomenon related to the level of societal development?
This is a difficult question. Gangs are obviously more associated with impoverished than wealthy contexts, for example. However, one must be very careful with this kind of association, which is much more contextual than causal. Not everybody living in poverty joins a gang; in fact, the majority don't. In the neighbourhood in which I work in Nicaragua, gang members did not necessarily come from the poorest families within the neighbourhood; they came from a whole cross-section. Ultimately, individuals will encounter structural circumstances differently, depending on their individual life trajectories and possibilities. Sometimes individual agency and social structure will articulate in a way that leads individuals to make certain choices, but sometimes individuals in similar situations will take different directions.
Concretely, why do youth join gangs?
The reasons why youth join a gang vary tremendously, and there are few factors that emerge systematically. Some individuals join gangs searching for a sense of comradeship, looking for some alternative to family structures that might be dysfunctional. But this does not mean that youths from families that are not ‘broken’ don't join gangs, they clearly do. Some studies have shown that youth can join gangs because friends or siblings joined them, but others have highlighted much more individual factors – because they're attracted to various aspects of the gang life, for example the adrenaline of violence or taking drugs. Everybody has different reasons, so it's difficult to identify one factor which would explain things.
During the course of my research in Nicaragua, there's only one factor which emerged as systematically affecting gang membership: members of evangelical churches or evangelical families never joined gangs. One could speculate that this is because these churches provided totalizing ways of thinking and ways of living in a manner that is institutionally analogous to the gangs. Certainly, joining an evangelical church is an important way in which gang members actually end up leaving the gang.
Why do members leave gangs?
Most gang members realize that the gang life is not necessarily a long-term viable life choice. Beyond the obvious dangers associated with being a gang member, the intimate association of gangs with youth means that aging makes leaving the gang inevitable. Indeed, ‘maturing out’ is something that has been observed by gang researchers all over the world. Part of this is linked to the fact that although there are elements of the gang life that can be construed as constituting a ‘subculture’, it's one that is embedded within a broader societal culture to which gang members are not immune. For example, the two major reasons Nicaraguan gang members left the gang was either because their girlfriends became pregnant or because they got a job. The first was widely perceived as signalling a need to become ‘responsible’, and the second was of course an economic opportunity that could not be passed over in the wider context of a Nicaraguan economy affected by high levels of chronic unemployment. Both of these sentiments clearly responded to broader societal values rather than gang values. Other important ways in which gang members leave the gang is by migrating, or else – more rarely – by ‘graduating’ to more organized crime, such as drug dealing. And of course, many gang members die.
You mentioned that the element of violence is decisive regarding gangs. But violence can encompass many acts, for example from stealing to drug trafficking, from sexual violence to kidnapping, from assault to murder. What kinds of violence are Central American gangs involved in, and who does it affect?
Central American gangs are involved in a whole range of different forms of violence, although most studies suggest that the majority of gang violence in the region is actually quite low-level violence – muggings, petty delinquency, theft, etc. Serious forms of violence such as rape or murder are more occasional, although they are definitely committed by Central American gangs. The key question here, though, is less the kind of violence they are involved in but whether we should define them purely through this more episodic form of exceptional violence, or whether we should be looking at their more normative patterns of violent behaviour. It's also important to note that most gang-related injuries and fatalities are suffered by gang members themselves. Obviously, non-gang members are frequently affected too, but more incidentally, one could say, since a majority of gang violence is actually gang on gang violence.
The question of Central American gang violence that is further confused by the fact that there are two different types of gangs in the region, on the one hand the pandillas, and on the other the maras. The former are home-grown, local gangs, the origins of which can be traced back to the 1940s and 1950s, while the latter are at more recently transnational transplantation of US gang culture that has resulted from the mass deportation of illegal migrants since the beginning of the 1990s. Pandillas used to be widespread throughout Central America, but now only really exist in Nicaragua, as well as, to a much lesser extent, Costa Rica and Panama, as they have been supplanted by maras in El Salvador, Honduras, and Guatemala (for reasons related to their different migratory patterns, there are no maras in Nicaragua, Costa Rica, and Panama). The maras are overall clearly much more violent than the pandillas, partly because they are less well embedded in local social fabric, being transnational transpositions (even if they involve fewer and fewer deportees, with local youth now forming a majority of membership, albeit having adopted US gang customs). Pandilla violence, on the other hand, corresponded – at least initially – to a form of local vigilantism, geared around protecting local communities. This obviously makes a big difference, particularly for the inhabitants of these local communities.
Would you say that pandillas controlling and protecting their neighbourhoods is a good thing, in the sense that they are providing a service to local inhabitants?
To a certain extent, yes, you could argue this, although I think it's important not to romanticize this too much. For example, during the 1990s, Nicaraguan pandillas were underpinned by what could be termed a social logic. The gang that I studied engaged in semi-ritualized forms of gang warfare against other local gangs that through its ritualized nature provided a sense of predictability for local inhabitants within a wider context of chronic insecurity. The first battle of a gang war typically involved fighting with fists and stones, but each new battle involved an escalation of weaponry, first to sticks, then to knives and broken bottles, and eventually to mortars, guns, and AK-47s. Although the rate of escalation varied, its sequence never did – i.e. gangs did not begin their wars immediately with firearms. The fixed nature of gang warfare therefore constituted something of a mechanism for restraining violence, and provided local neighbourhood inhabitants with an ‘early warning system’. In addition, gang members always sought to protect local inhabitants in the context of gang conflicts, often at their own expense.
Although not the most efficient means of providing security services, since bystanders were often killed or injured, local inhabitants recognized that it was better than nothing, and there was generally support for the local gang in the neighbourhood at the time. Certainly, neighbourhood inhabitants never denounced gang members to the Police, and they'd regularly engage in joking banter with gang members on the street, as well as offering them water or buying them a Coke on hot days. This however changed completely when the whole logic of the gang transformed in the early 2000s as they became very much organized around drug dealing. Their logic now revolved around protecting the drugs trade rather than the local neighbourhood, and their violence turned against local inhabitants in order to create a climate of terror to make sure that nobody denounced them and that they could operate their drug dealing without impediments. This violence was also no longer ritualized, and was highly unpredictable, and local inhabitants clearly feared gang members, who would not hesitate to maim and sometimes even kill local inhabitants who had altercations with drugs consumers. There's very much a growing alienation in local communities vis-à-vis the gangs. In some ways, you could say that the logic of the pandillas in Nicaragua has been growing closer to that of the less locally embedded maras in El Salvador, Honduras, or Guatemala.
Would you say that the gangs in Central America have become more professional, then?
The gangs of Central America – both maras and pandillas – have clearly been professionalizing over the past decade or so. Perhaps the most obvious reflection of this professionalization is their increasing involvement in drug trafficking, which has meant that the gangs have been becoming smaller, meaner, and leaner. I've already mentioned how gang violence has become more brutal, and less protective of local neighbourhood inhabitants, for example, but this was also accompanied by a reduction in gang membership. In the case of the Managua neighbourhood where I carry out my research, the gang reduced from about 100 members in the 1990s to just under 20 in the 2000s because of the new focus on drugs trafficking, which is can only be profitable as an exclusive enterprise. The average age of gang members also increased. During the 1990s, gang members had ranged from 7 to 23 years old. In the 2000s, they ranged between 17 and 25 years old.
How have such changes affected the make-up of gangs, for example, have they developed more visible command structures?
The issue of command structure is complicated. In the 1990s, there wasn't really a command structure, as such. The gang was quite democratic, although you had some more influential individuals in the gang. You certainly didn't have a jefe, except in the context of gang conflicts, where what I suppose you could call a ‘war commander’ was appointed, because it's obviously impossible to engage in conflict in a democratic manner. The new drug-dealing gangs seem to have much more defined hierarchies, and there is often a narco at the apex of the pyramid, although his power tends to rest on being able to mobilize individuals to carry out acts of violence for him, and he is generally highly dependent on a small group of ‘lieutenants’. At the same time, drug dealing is quite decentralized, in some ways it is quite feudal in its logic, with a king at the top, and then local barons who pay tribute to the king, but otherwise conduct their affairs autonomously.
Is there some kind of code of conduct within the gangs?
Yes, definitely. Gangs aren't just anarchic social phenomena, they do have rules and prescribed patterns of behaviour, such as engaging in certain forms of violence or dressing in particular ways, for example. At the same time, these rules often derive from wider local culture. Take machismo, for example, which is very strong in Central America. Some aspects of the gang are clearly a heightened expression of machismo. Being a gang member can very be much about ‘being a man’ in the way that its understood locally, strutting around, exposing oneself to danger, showing that one is really strong. Partly for this reasons, there are very few female gang members in Central America. These kinds of codes of conduct have persisted over time, highlighting the way that these gangs are very much linked to wider culture.
Other codes of conduct have changed, however. For example, in the past the gang was often characterized by a strong sense of camaraderie and solidarity, whereby a member was part of a group, and could always count on other members to help and protect him. This has declined quite a bit, with relations much more ambiguous and also increasingly ‘professional’, in the sense of being linked to common participation in an economic enterprise (drug-dealing), but nothing else. One striking change over time has for instance been the transformation of what is called the traido, which is a kind of vendetta. Although this always occurred between individuals, normally between members of rival gangs, in the past conflicts related to a traido would end up involving all the gang members, while now it's seen just as an individual issue. It used to be ‘all for one, and one for all’, but now it's very much ‘each to their own’.
In warfare, especially in international armed conflict, codes of conduct are important to reduce the scope of violence. Do you think that something similar could be achieved with regards to gangs?
The idea of exploring the parallels between gang codes of honours and codes of conduct is definitely an interesting one. It's definitely one of the great achievements of international humanitarian law to have transformed modern warfare by modifying the perception that warfare is a zero sum game, to something that is seen as an activity where the two parties didn't necessarily have to lose everything in order for the other one to win. I'm not sure that that would work with regards to gangs at this point in time. It might have worked with Nicaraguan gangs in the 1990s, when they had a social logic, but I doubt that this would have much traction with contemporary drug dealing gangs. Although one scenario under which it might become possible with drug dealing gangs is if drugs were legalized. Illegality would no longer be the primary organizing issue, making a profit would be, and rules regulating competition could be introduced just as they have been for any other commercial activity. On the other hand, codes of conducts could possibly be developed in the context of gang versus state conflict, partly because of the generally asymmetric nature of this, in which case gangs might actually accept some basic rules. Whether the state would engage in this is another question, however, since in some ways, this would arguably be some form of legitimization of the gangs.
But aren't deals often struck between gangs and the authorities and/or politicians?
It depends where. In Indonesia, for example, under Suharto, gangs were for a long time the foot soldiers of the regime, and carried out a lot of the terrorizing and violent operations that sustained it on the ground. The Martin Scorsese film Gangs of New York also highlights the potential links between gangs and politicians, while the US anthropologist Desmond Enrique Arias has shown how there are numerous links between politicians and local gangs and drug traffickers in Rio de Janeiro, with the latter being mobilized to serve the financial and voting interests of the former. In other places, however, the complete opposite occurs, and non-cooperation with gangs is used as an excuse to carry out stringent repression that often spills over beyond the gangs and permits governments to repress or contain other groups, such as the poor, for example. This is something that is arguably the case in El Salvador, Honduras, and Guatemala, for example, where the introduction of the infamous Mano Dura approach seems to have very much been a means of trying to push the violence away from the city centre, away from the elite, and contain it in poor neighbourhoods – although not very successfully, one might add. At the same time, governments and politicians generally play a very tight game, sometimes they will cooperate with gangs, and sometimes they won't, depending on their interests at the time.
We've been discussing gangs as a paradigmatic form of Latin American violence, but you've just referred to Indonesian gangs, and one often reads about gangs in other parts of the world, for example in South Africa – could you perhaps tell us a little bit about gang dynamics in those parts of the world?
Let me begin by saying that although gangs are certainly one of the most visible forms of violence in much of Latin America, most prominently in Central America, they are by no means the only form of violence in contemporary Latin America, and this is very important to keep in mind. Domestic violence, for example, is a huge issue that can be related to the pervasive machismo in the region, and of course, as Gandhi famously put it, ‘the deadliest form of violence is poverty’. Having said that, I think that gangs can be considered ‘paradigmatic’ in the sense that they very clearly reflect some of the basic underlying processes associable with the new kinds of urban conflicts that have emerged in the region over the past few decades. Or to put it another way, quoting an early gang researcher, Frederick Thrasher, who studied gangs in 1920s Chicago, gangs are ‘life, often rough and untamed, yet rich in elemental social processes significant to the student of society and human nature’.
Perhaps not surprisingly in view of this, gangs are a global phenomenon, and can be found in most societies around the world. Although there exist quite a lot of in-depth studies in many different countries – including beyond Latin America, the USA, South Africa, Russia, France, Timor Leste, Nigeria, China, among others – we clearly need more research comparing their dynamics across contexts. Last year, I co-organized with Jennifer Hazen of the University of Texas-Austin a workshop which brought together researchers working on gangs in twelve different countries, and one of the cross-cutting issues that came out as important across the board was the relationship that gangs have to the state, for example. I've already mentioned the cooperation that existed in Indonesia, while in India gangs tend to be co-opted into the militant youth structures of political parties. Research presented on China suggested that the presence of the state right down to the family level plays a big role in cutting off spaces for gangs.
At the same time, one striking feature of comparative research is the similarities that exist between gangs across contexts. For example, I've done some comparative work with the Danish anthropologist Steffen Jensen, who works on gangs in South Africa. We looked at the dynamics of the gangs there and in Nicaragua, and although there were obviously major differences between the two, partly due to the different contexts and histories, there were also some surprising similarities. In both cases, gangs were undergoing processes of professionalization, and the role played by ideology – whether Sandinista or ANC – as reference point for mobilization and the construction of foundational myths and codes of conduct was very similar. Most importantly of all, the notion of exclusion was essential in both cases: South African cities were characterized by major spatial exclusion under Apartheid, and this has persisted and mirrors some of the new spatially excluding territorial re-organization of Managua, whereby poor neighbourhoods and shanty towns are increasingly disconnected from the rest of the city.
Do you think that a parallel can be drawn with ‘les émeutes des banlieues’ – the riots in the suburbs – in France?
You can certainly see a lot of common elements. The Parisian banlieues are very much isolated from the rest of the city. This was very well represented in the film La Haine, for example, both in the scene where the three young protagonists are in the centre of Paris and don't know what to do because they don't know the centre of Paris, as well as the fact that they get stuck in Paris due to the lack of trains going back out to the banlieues. If you look at Paris' urban development from a historical perspective, it's clear that there has been a gradual disconnection of the banlieues from the city centre over the past 50 years, with local train stations being closed down, and less and less bus routes operating. This is a ‘soft’ form of spatial exclusion, you could say, certainly by comparison to the walling of favelas, as has occurred recently in Rio de Janeiro, for example, but the basic dynamic is the same, and considering that gangs are fundamentally epiphenomena of broader structural circumstances, it's not surprising that we can see parallels between different contexts.
When there is spatial exclusion, high levels of unemployment, and very few opportunities for bettering one's situation, as is the case in both the banlieues in Paris and the shanty towns in Rio de Janeiro, it's not surprising that similar social forms will emerge. Now, the exact way in which they emerge, and the kinds of things they do, are dependent on other factors; situations are rarely identical. For example, Rio de Janeiro is a logical transit point for drugs, which Paris isn't – it is an endpoint. So that will also make a big difference, and partly explains why the gangs of the Parisian banlieues are not as dangerous and violent as those in the favelas of Rio de Janeiro. The availability of weapons is also very different in a place like Rio de Janeiro compared to Paris, as is the degree of state presence and control.
Is it that the state is no longer in control of the shanty towns in places such as Rio de Janeiro?
There's been a lot of talk about weak, fragile, failed, or crisis states over the past few years, and certainly one could argue that under such conditions, it would not be surprising that the reach of states in areas such as shantytowns has been considerably reduced. At the same time, I also think that in many cases states have also actively chosen to withdraw from such areas. The new global economy doesn't really need shanty towns anymore; it is less labour-intensive than in the past, and therefore doesn't need their populations as reserve armies of labour. This scenario has been very well outlined in a recent book called Planet of Slums by the US social commentator Mike Davis, where he basically suggests that slums now house a surplus population, and asks the question of what is to be done with it. The answer in most cases has been to exclude them, to keep them out of the lives of the urban elites, those living in gated communities, who are benefiting from the advantages of globalization and the new economy. The poor are contained in their shanty towns and left to fight amongst themselves for whatever informal scraps they can put together.
What's worrying is that this kind of process is clearly leading to levels of inequality increasing globally, and at some point or another, something is going to explode. I don't see how you can have the vast majority of the population living in poverty and not actually do anything about it without stirring social unrest. Certainly, when you consider contemporary Central America in terms of its social, economic and political conditions, it's striking that in many respects it's quite close to being where it was in the early 1970s, when the revolutionary struggles that led to over 20 years of armed conflict really took off. Few things epitomize this better than the current tax structure of Central America. Inequality in Nicaragua, for example, actually rises after taxation; since there are only about 9,000 individual taxpayers, this means that most taxes are indirect, and therefore being disproportionately born by the poor.
What is to be done in the face of such structural circumstances? Do you know of any policies that have been successfully implemented to reduce gang violence?
Let me begin with what we know doesn't work: repression. We definitely know that repressive measures are generally doomed to fail. Indeed, in Central America, they've actually increased violence. There are two main ways in which repression occurs. One is by specifically targeting individuals, such as gang members, which is what has happened in Central America. The other one is to target communities associated with violence, namely the poor, which is what has happened in Brazil in the past, for example, as well as Jamaica more recently, where numerous poor communities in Kingston literally came under siege. The problem with the first strategy, at least in Central America, is that it has led to gang members being increasingly treated as if they were ‘unlawful combatants’ – to use that particularly oxymoronic pseudo-legal expression – and has justified hugely disproportionate responses that often contravene international human rights law. The second strategy is almost a return to classical warfare – in a highly asymmetrical form, insofar as these local communities generally have nothing in the way of the state's firepower – except that it is directed internally, and as Abraham Lincoln once famously put it, ‘a house divided against itself cannot stand’.
The problem, however, is that repression clearly serves other purposes than simply trying to reduce violence. For example, although Mano Dura in Central America has been shown over and over again not to work, it persists, even if there is increasingly an alternative, more preventative discourse emerging, mainly because it serves the different purpose of demonizing gangs and diverting attention from societies which are very unequal and getting ever more unequal, largely due to a lack of will to change the situation by those in power. There is an absence of economic growth and job creation in Central America that is not just linked to global crisis but very much a function of the particular economic models that have been put in place during the past two decades, which are fundamentally exclusive.
So if we were to boil things down to one factor, would you say that inclusion is the solution?
Yes, I think that if you had to boil it down to one thing, inclusion is the solution. There's a variety of ways in which to try to include and integrate. It's not even just about creating jobs. There was a consciousness among gang members that I worked with in Managua of their exclusion, indeed of their spatial exclusion in particular, and that there were certain areas where they couldn't and wouldn't go. If they turned up, say in a posh mall, they wouldn't even be asked who they are and what are they are doing, they would just be kicked out. This obviously doesn't breed any kind of notion of collective living. On the other hand, take a city such as Buenos Aires, which is marked by significant socio-economic differentiation, and also significant socio-economic interaction. There are richer neighbourhoods and poorer neighbourhoods within the city, but you get a sense that people don't feel there are no-go areas, and people from rich areas do go to poor areas and vice-versa, although if one considers the whole metropolitan area of Buenos Aires, it can perhaps be said to resemble Paris, with its socially heterogonous centre, and excluded banlieues (known in Buenos Aires as partidos).
Concretely, though, what kind of integrative projects could be put in place to reduce gang violence?
There are very few projects that I know which have actually had any major success in diminishing gang violence. When you look at things historically, the most effective process has actually been economic boom; certainly, the history of US gangs suggest that they wax and wane with economic cycles. Not surprisingly, perhaps, in view of this, the most effective gang violence reduction programmes have been those which have created alternative opportunities for youth. But these have to be sustainable, which is not always an easy proposition.
One project that I have come across in Nicaragua, for example, would move gang members into the countryside, teach them how to be masons or carpenters over the course of three months, and then send them back to their neighbourhoods with a large loan in order to set up a business. The problem was that this was actually quite expensive, the skills being taught to gang members were very common ones, and few of the programme graduates managed to set up viable businesses. The programme had clearly not been thought out very well in terms of local conditions, and therefore had a huge failure rate overall.
Another project, this time in El Salvador, adopted a very different approach. Its starting point was that it was necessary to engage with the primary motivation for becoming a gang member – which was identified as the power and authority associated with being a gang member – and also to build on existing gang member skills. At that point in the project's inception, a lot of gang members in El Salvador were deportees from the US and spoke English, and so what the project did was train gang members to become English teachers, since as a teacher, you are of course placed in a relationship of power and authority vis-à-vis students. Unfortunately, there's obviously a limit to how many English teachers you can have, but this was probably one of the most original projects I've come across.
Finally, not all interventions need to be concerned with economic integration. One of the few forms of state intervention that occurred in the Managua neighbourhood where I have carried out most of my research during the past decade or so was the building by the municipal authorities of a basketball court. This occurred just before elections, and was clearly principally an instance of political demagoguery more than anything else, but it – unintentionally – had a significant effect on gang membership in the neighbourhood. It provided an alternative focus for a number of youths who would normally have hung out in the street, and otherwise gravitated organically towards the gang. Instead, they could now go and play basketball, thereby highlighting the benefits of institutionalized alternatives to hanging out in the street.
Finally, what role do you see humanitarian organizations potentially playing? Do you have any examples of a local or an international humanitarian organization really making a difference with regard to gang violence?
Gang violence, as I've said, is fundamentally an epiphenomenon; it is linked to wider structural problems, which are developmental rather than humanitarian problems. To this extent dealing with them might be construed as falling outside the mandate of an organization such as the Red Cross. Having said that, there are many local humanitarian organizations that do try to implement a range of initiatives to counter gang violence, including for example mediating and brokering truce agreements between gangs and getting gangs to lay down arms. None of these work for very long, however.
One area where I think humanitarian organizations such as the Red Cross may have a comparative advantage, however, is in bridging the lack of trust that often exists between the Police and gangs. This is critical, as the experience of one of the best known gang violence reduction NGOs in Central America, Homies Unidos, highlights well. This is an organization made up mainly of ex-gang members, who have given up the gang life, but still maintain links with active gang members, and run projects for them. Homies Unidos has huge problems obtaining any credibility from the El Salvadorean government, whom they furthermore distrust, partly due to a sustained campaign of harassment against them which culminated in the arrest of the Homies Unidos leader as a murder accomplice.
Irrespective of whether the latter accusation is true or not, both the Salvadorian authorities and Homies Unidos have displayed a complete unwillingness to actually sit down and talk to each other, or even accept that the ground rules for their interaction have to be determined together rather than unilaterally imposed. The Red Cross is obviously better positioned than most organizations to actually try and mediate here because of its neutrality, reputation, and because it will likely be listened to by both parties. Whether it has the political possibility to do so though is a different issue, and is arguably ultimately the fundamental one concerning gangs in a world which unfortunately sees them very much misguidedly as an inherently evil form of modern-day barbarism.
