By gathering a powerful means of physical coercion, that is, firearms, and by enriching themselves, gangs end up becoming a political force and setting up a system of local power.
Introduction
Hundreds of drug-trafficking gangs have governed favelas in Rio de Janeiro for more than four decades. While most scholarship focuses on the emergence of prison-based factions during Brazil’s military dictatorship (1964–85) as the starting point of these arrangements, the real history of Rio’s gangs begins long before, on the streets and in the alleyways of these neighborhoods. This chapter traces the origins of Complexo da Maré’s first embryonic gangs. In doing so, I seek to answer a series of interrelated questions: Where do gangs come from? How and why did they begin to govern? And what did those first forms of governance consist of?
These are important questions because we continue to lack a systematic understanding of gang formation and evolution in Rio de Janeiro. While today’s drug-trafficking gangs look little like the small groups of young men that began to organize themselves more than a half century ago, several of these organizations can be traced directly from these initial groups to their modern incarnations. Most studies of gangs, however, deal little with their origins. This is for good reason. Gang formation is mostly lost to history. The lives and activities of their members are almost never written down or recorded, only existing in the memories of those involved or the residents living in these neighborhoods at the time. In this chapter, I combine dozens of oral histories conducted with former gang members and longtime favela residents, reports from local newspapers, and documents from community archives to reconstruct the origins of Complexo da Maré’s gangs. In doing so, I make several interrelated arguments concerning gang emergence and governance.
First, gangs formed in Complexo da Maré’s favelas at very different times and only after previous governance arrangements had broken down.Footnote 1 Increasing migration over the course of the 1960s and 1970s led to worsening conditions in Maré. Without sanitation, electricity, or basic infrastructure, Maré’s favelas became downright squalid. A variety of other non-state actors that had previously maintained order and provided some services to residents could no longer effectively govern this burgeoning population. Rio’s favelas were also increasingly marginalized during Brazil’s military dictatorship (1964–85) as the abusive and violent practices of the police became more widespread. These combined factors led to spiking levels of crime and violence and, eventually, the emergence of several gangs in the early 1970s in Nova Holanda, where the conditions deteriorated most quickly. In Maré’s other favelas, gangs would only emerge more than a decade later as existing governance arrangements more effectively controlled violence and crime and, thus, delayed the formation of gangs. Eventually, however, they too would be incapable of preventing gangs from forming.
I also seek to explain how and why these gangs began to govern. I argue that competition between rivals, police enforcement, and governance were inextricably intertwined from the very beginning. Quickly after their emergence, several nascent gangs began to compete over increasingly valuable drug-selling turf. The more successful gangs destroyed or incorporated their rivals through conquest or alliance and extended their turf to these areas. The expansion to governance activities was, in some sense, inevitable. Residents and all local political and economic actors became subject to gang authority as they demanded obedience amid increasingly competitive dynamics with their rivals. Gangs also sought the protection and cooperation of residents as police began to focus their attention on these organizations. In Nova Holanda, multiple fledgling gangs violently competed for dominance over the course of the 1970s and early 1980s. After several years of gang warfare, one gang would eventually win out and, in the process, develop the capacity to engage in more significant governance activities. For the other two case studies investigated in this chapter, a single gang would emerge from Parque União and another from Morro do Timbau. Unlike Nova Holanda, there would be no such violent competition for territory, at least initially. These gangs would only begin to exert territorial control and govern like their Nova Holanda counterpart after they integrated into the prison-based drug-trafficking factions in the 1990s.
The final point I will make concerns the role of residents. Favelas were not, as has often been argued, places of rampant criminality or anarchic environments from their beginnings. On the contrary, favela communities often developed highly cohesive and stable governance mechanisms even amid contexts of precarity. Before gangs emerged and began to govern, a mix of other favela actors provided governance services that resident populations desired: a stable form of order, crime fighting, dispute resolution, and some access to welfare. In many cases, it was the AM president, local strongman, or prominent businessman, through their connections to the police or military that took on this role. Although the emergence of gangs would not be a positive development within these communities, gangs did take over this pivotal function within the community. They imposed an essential, if perhaps violent, form of order while variously providing certain benefits that favela residents sorely needed. In return, residents offered gangs their tacit or explicit support by not denouncing them to their rivals or the police. This exchange formed the foundation of a social contract that still exists to this day.Footnote 2
This chapter is organized as follows. The below section provides a brief overview of the early history of Rio’s favelas and the predecessors of gangs in the city. The rest of the chapter focuses more intensively on Complexo da Maré and the cases that form the empirical backbone of this book. I first trace the pre-gang history of Maré’s favelas, focusing on the settlement patterns and the various actors that provided governance prior to the emergence of gangs in the area. I then offer a detailed account of the deteriorating living conditions, increasingly abusive and repressive behavior of the public security apparatus, and the breakdown in preexisting governance arrangements that eventually led to the emergence of the gangs. I take considerable care to outline the diversity of these pre-gang governance arrangements that would impact when and where gangs formed and eventually governed. I conclude with a short description of the emergence of the prison-based factions in the 1970s, their eventual spread from the prisons into Rio’s favelas, and the integration of Maré’s gangs into these larger networks.
A Short History of Crime and Governance in Rio de Janeiro’s Favelas
According to a popular myth, Rio’s first favela was established in 1897 by veterans of the Canudos War who pitched their tents on a hillside in the center of Rio de Janeiro to await payment for their service from the federal government (Valladares Reference Valladares2008, 28–30). The ex-soldiers, many of whom were former slaves, called the hill favella, after Monte Favella, a small mountain in a remote region of Bahia where they had fought – and many died – during the war.Footnote 3 The real birth of this community, however, was several years earlier, in January 1893, following the destruction of Cabeça do Porco (Pig’s Head), a massive cortiço (tenement) in the center of the city. Cabeça do Porco, which housed roughly 4,000 urban poor, was one of hundreds of tenements that provided unsanitary and cramped living quarters for many of Rio’s newly freed slaves and recent migrants from Brazil’s hinterlands and parts foreign.Footnote 4 After demolition by municipal workers, former residents gathered what few belongings they managed to save and began constructing ramshackle huts on a nearby hillside (Chalhoub Reference Chalhoub1996, 15–20). The ex-soldiers would join them on that hill several years later.
Tenement removals were common around the turn of the twentieth century as Rio began to model itself after major European cities by carrying out massive public works and urban reforms to realize this self-image.Footnote 5 After being suddenly and, in many cases, violently dispossessed of their housing, residents were forced to search for alternative accommodations. The steep hillsides in the center of the city, difficult terrain though they were, had available land on which they could build homes and these fledgling communities grew from a smattering of huts into permanent settlements with hundreds of domiciles and thousands of residents. Just as the tenements were viewed as public health risks and dens of vice and iniquity before, favelas would take up this mantle. They faced an increasingly threatening and invasive government. Many politicians, urban planners, and numerous government agencies collaborated in favela eradication and removal campaigns. It was often Rio’s Military Police that was responsible for evicting residents and sometimes burning down these settlements. These policies were largely driven by pervasive prejudice and racism toward the inhabitants of these communities, assisted by sensationalized media accounts.Footnote 6
The reality was that favelas were neither as violent nor as full of criminals as was commonly believed though some types of criminality existed in these communities. Favelas of the period probably contained some low-level drug dealers, neighborhood capoeira (a Brazilian form of martial art combining music, acrobatics, and dance) groups called maltas that may have engaged in some violence, as well as possibly some cangaçeiros (bandits and thieves) that had migrated from the Northeast of Brazil where banditry was common. Misse has argued that favelas of the era had malandros (scoundrels), valentes (warriors), and malfeitores (wrongdoers), three categories of low-level criminals that engaged in theft and robbery, smoked marijuana, and were considered, to varying degrees, dangerous and antisocial (1999, 251–62). Notwithstanding such different understandings of what constituted a criminal, more organized forms of criminality were little present in Rio’s favelas in the early part of the century.
By the 1920s, only a couple dozen favelas dotted the hillsides in and around the city center but by 1940 there were at least 116 such informal settlements in the quickly expanding city (O’Hare and Barke Reference O’Hare and Barke2002, 228). In this second phase of favela development, a host of other actors would participate in land invasions and squatting, not just the indigent freedmen and women who could not afford formal property or lodging. Wealthy land speculators, politicians, lawyers, the middle class, traditional rural poor, foreign immigrants, and farmers seeking land to raise crops and livestock, among many others all vied to get a piece of the highly valuable real estate market.Footnote 7 Many of the favelas from this era would look little like their predecessors. The lands these assorted groups occupied were no longer solely on steep, almost uninhabitable hillsides; they were defunct estates, unused church or state property, unoccupied land, and even swamps that residents would fill in over time. Many of these areas were far from the city itself but within the vast municipal territory.Footnote 8
Complexo da Maré would emerge in this second phase of favela development. At the end of the 1930s, several miles north of Rio’s bustling city center, an immigrant couple from the neighboring state of Minas Gerais built a small hut on a hill surrounded by Guanabara Bay’s massive tidal plain.Footnote 9 The area was almost completely unsettled except for a colony of local fisherman who used the area to tie their boats at night (Santo, Gonçalves, and Silva Reference Santo, Marinho Gonçalves, Sousa Silva, Andréia Martins de Oliveira and Eliana2013, 21; Santos Reference Santos1983, 44). Slowly, other immigrants followed and by the middle of the 1940s, a small community had taken shape on Morro do Timbau, Maré’s first favela.Footnote 10
Several years later, in the late 1940s, Avenida Brasil, a major highway connecting the center of Rio to surrounding suburbs would spur migration to the area. Other industries would quickly follow, including a petroleum refinery. Many of the male residents of Morro do Timbau found steady but grueling work in these factories. Two other favelas quickly materialized: Baixa do Sapateiro (1947) and Parque Maré (1950). These communities would be located on the only other two surrounding areas of land that were permanently dry. When these areas were settled, other migrants began to build palafitas (shacks on stilts) above the tidal plain (see Figure 4.1).

Figure 4.1 Aerial photograph of Morro do Timbau and Baixa do Sapateiro, 1960s
Parque Rubens Vaz (1951) and Parque União (1954), Maré’s next two communities, followed a slightly different pattern of settlement in which one resident dictated much of the organization and migration to the area.Footnote 11 Margarino Torres, a lawyer and member of the Communist Party, came to Parque Rubens Vaz in 1958 to protect the community from removal by the police and quickly became the undisputed community leader (Vaz Reference Vaz1994, 158). He would eventually organize the occupation of an adjacent area, where a local industrial firm, Tekno, had already filled in some of the low-lying swampland. This area would later become known as Parque União.
As the percentage of the municipal population living in favelas steadily increased, so too did the possibility for these communities to influence public policy. The most important politician of this era, Getúlio Vargas, sought greater support from the burgeoning lower classes and favela residents. During his tenure as dictator (1930–45) and then President (1951–54), Vargas helped found the Fundação Leão XIII and the Municipal Guard, two government agencies that transformed the way the government dealt with favelas. Public policy shifted from being exclusively focused on removing favelas to improving their infrastructure and providing services while simultaneously attempting to control the social movements emerging from within them. President Juscelino Kubitschek (1956–61) would also promote more development-oriented policies, even sponsoring a law that placed a two-year ban on all favela evictions while supporting vocational training programs for industrial workers, which largely benefitted favela residents. In addition, the Catholic Church, amid transformations of its own in the Second Vatican Council (1962–65), began promoting democratic principles and services for the poor, inserting itself into favela communities with vigor.Footnote 12 As these progressive changes were occurring, favela communities including those in Maré began to form AMs to represent their interests to government agencies and political parties. Across the city, these informal democratic institutions elected local leaders, formed coalitions, petitioned the municipal government, and carried out many essential functions for residents. In some favelas, AMs managed to bring significant public services and infrastructure to the community.
And yet, the outcome of this period for favela residents was ambivalent. Forced removal remained an important municipal policy when it came to favelas. The 1938 Building Code was the first official document to recognize the existence of favelas though it simultaneously declared them illegal and prohibited their creation or expansion (Perlman Reference Perlman2010, 26–27). In addition, residents of favelas faced increasing violence from the police in the middle decades of the twentieth century. Extortion, detention, mass arrest, illegal invasions, and even dispossession became standard police procedures and, by the 1960s, some precincts were even engaging in torture and murder of suspected criminals (Fischer Reference Fischer2008, 204–8). In many favelas, residents seldom cooperated with police during investigations and refused to make statements or offer testimony of any kind. For these reasons, “many poor communities began to effectively remove themselves from police jurisdiction for all but the most serious of crimes” (p. 206).
Favela communities used whatever means were at their disposal to prevent removal and provide themselves with services. Through local community leaders, they created complex client-patron networks with various state and non-state actors: the Catholic Church, Fundação Leão XIII, political parties, and the Ministry of Justice, just to name a few. For more local matters, such as crimes and disputes within the favela itself, residents sometimes sought out the police. Knowing that police would likely detain and possibly abuse those accused of crimes even without evidence, residents began to use the police as a coercive tool to resolve disputes and exact revenge (Fischer Reference Fischer2008, 207; Kant de Lima Reference Kant de Lima1986). In other cases, residents sought out respected community leaders who could be called upon to arbitrate disputes. Finally, residents could sometimes turn to emerging organized crime networks, represented by the jogo do bicho (animal game, akin to the numbers racket in the US), drug dealers, or other bandits whom residents “respected,” either because their verdicts were just or because they had the means to enforce their decisions.Footnote 13 The cumulative effect of this piecemeal rule of law was that traditional forms of justice and citizenship would become even more difficult to find in Rio’s favelas.
Complexo da Maré’s evolution closely maps onto these larger citywide dynamics. In the early 1960s, two other communities were inaugurated in Maré that were temporary housing projects built by the municipal government for residents who had been removed from other favelas in the city. One of these housing projects, Centro de Habitação Provisório de Nova Holanda (Provisional Housing Center of Nova Holanda), was constructed in the low-lying swampland between Parque Maré and Parque Rubens Vaz in 1962 (see Figure 4.2). Conjoined single and two-story wooden homes were built with straight wide streets and a more “rational” checkerboard design (see Figures 4.2 and 4.3). What was intended to be a “temporary community,” turned into a permanent settlement, and with only minimal investment, Nova Holanda quickly came to differ little from the surrounding favelas.

Figure 4.2 Aerial photograph of Complexo da Maré, 1979

Figure 4.3 Map of Complexo da Maré, 1974
In 1964, a military coup overthrew the democratically elected President, João Goulart, and instituted a twenty-one-year dictatorship during which favela policy would be characterized even more by removals and repression. The influence of the AMs waned as petitioning authoritarian institutions became more difficult. The military regime created new government agencies with the purpose of eradicating the favelas from the urban landscape. From 1968 to 1975, the city of Rio removed seventy favelas and more than 100,000 residents (Perlman Reference Perlman1976, 258–60). Most of these evictions were from highly desirable locations in and around the famous beaches, wealthy neighborhoods, and the city center where property values had already begun to skyrocket. When and if AMs attempted to intercept such forced removal, they were dealt with harshly by public security and military forces (Gay Reference Gay1994, 20).
For residents of Maré, life progressively became more difficult during the dictatorship. Continued migration to the area led to a swelling population and with very little investment in public infrastructure, the living conditions became unequivocally terrible. By 1980, some 68,000 residents were living in Maré’s nine favelas, increasing to nearly 80,000 by 1987 (Ribeiro da Silva Reference Silva2006, 194) (see Figure 4.3).Footnote 14 Roughly a quarter of those residents lived in palafitas, which generally consisted of one room with no running water, electricity, or sanitation. The palafitas were connected by a series of precariously constructed planks (see Figure 4.4), which led to many accidents, sometimes proving fatal when small children fell into the water or mud below (Vieira Reference Vieira1998, 72). In interviews with longtime residents, they recalled how unsanitary and dangerous it was to live in Maré at that time. According to one resident, “The situation was really bad. The community kept growing. Lots of people from the Northeast [region of Brazil] kept coming and there were no services. The standard of living was really terrible.”Footnote 15 One of the most difficult aspects of local life was that there was no source of fresh water in the community. Residents had to carry buckets or roll barrels across the busy Avenida Brasil highway. “I saw many people die getting hit by trucks trying to cross over,” João told me.Footnote 16

Figure 4.4 Photograph of palafitas in Complexo da Maré, 1969
In addition, the public security apparatus’ tendency for violence and abuse only increased during the dictatorship. Of Maré’s nine existing favelas during the 1970s, four had a Destacamento de Policiamento Ostensivo (Ostensive Police Detachment, hereon DPO), or a Posto de Policiamento Comunitário (Community Police Post, hereon PPC), often just a small building with an office and a cell, located within the community (Ribeiro da Silva Reference Silva2006, 202). By this time, police had been carrying out periodic violence and removals of residents for several decades. They had even managed to expel and destroy a handful of embryonic communities from the area (pp. 190–92). Moreover, the use of torture by the police was widespread, especially following Institutional Act number 5, issued by the military regime in 1968, which suspended some civil and political rights including habeas corpus.
In some of Maré’s favelas, the relationship between residents and police deteriorated to the point that police seldom left their post and residents refused to enter. In interviews with longtime Nova Holanda residents, they described the horrifying brutality with which the police carried out their job. Artur recalled living across the street from one of the police posts. He explained how police would round up the extremely poor who were accused of vadiagem (vagrancy). “Sometimes they would arrest up to ten people at a time and lead them in a line by rope back to the post. My house was across the street and we could hear them scream when they tortured them. All these tactics were approved by the dictatorship,” he said.Footnote 17
In other areas of Maré, however, the relationship with police had not deteriorated so precipitously.Footnote 18 In Praia de Ramos, for example, where another of the era’s PPCs was located, resident reactions were not so uniformly negative. Few recalled episodes of torture and generally considered relations with the police decent to the point where residents would still enter the post to make complaints and fill out reports.Footnote 19 In Morro do Timbau, where the AM worked closely with the military, and Parque União, where the AM maintained connections to the police, residents continued to report crimes and ask for dispute resolution from public authorities until the 1980s.Footnote 20
During this middle period of the twentieth century, with the exponential growth of Rio’s favelas and the dramatic change of the urban landscape, drug traffickers began to emerge as major players in favelas and urban politics. Misse (Reference Misse1999) traces the beginnings of o movimento (the movement), or the initial forms of drug trafficking, to two of the city’s oldest favelas in the 1940s: Providência (formerly Morro da Favela) and Mangueira. By the 1950s, several other favelas were known to have drug trafficking and by the 1960s it was in most favelas. Despite the single term, drug trafficking did not have a uniform expression on the ground. Drug traffickers had no specific organizational structure and the individuals and groups involved were often not interconnected (p. 319). In most cases, it was only a local boca de fumo. Even before the retail drug market arrived, however, the role of the Dono was already well-established through the presence of the jogo do bicho or through a local group of bandits and thieves who organized themselves as a unit (pp. 313–15). The relations that Donos maintained with favela communities varied incredibly but were founded on his – they were exclusively male – ability to control the abuse of residents, defend the community from bandits or thugs, and the mediation of disputes or help in times of need that allowed him to maintain his dominion (p. 343).
At the beginning of Brazil’s military dictatorship, gangs did not exist in Maré, but by its end, they would be present in every one of the Complexo’s favelas. In the sections to follow, I detail the emergence and eventual ascendance of three gang organizations from different favelas in Maré: Nova Holanda, Parque União, and Morro do Timbau. Across these cases, several factors helped incentivize the formation of gangs and facilitate their expanding influence and power. First, these were difficult times in Maré. Migration to the area continued and the population swelled but with little investment in infrastructure, living conditions deteriorated. The burgeoning drug trade, armed robbery, and theft became important sources of income for families who had a difficult time making ends meet. In addition, increasing numbers of young men were being excluded from the formal labor market as they became the focus of police enforcement. Arrest, imprisonment, and powerful social stigma often prevented them from acquiring decent and stable employment, thus expanding the number of men willing to engage in crime for a living. Finally, increasingly repressive and indiscriminate policing tactics and the marginalization of previous forms of governance, a former source of stability and order in these communities, led to growing distrust of state institutions. By the 1980s, residents seldom reported crimes to the police. The informal governance arrangements that existed previously were incapable of dealing with rising levels of crime and violence.Footnote 21 What each of the following cases makes clear, however, is that gangs emerged and began to govern at very different times across Maré that corresponded to these local conditions. This temporal variation provides initial evidence as to the dynamics influencing how and why gangs govern.
The Emergence of Criminalized Governance in Complexo da Maré
Nova Holanda
In January 1962, the municipal government of Rio de Janeiro began relocating several thousand residents that had been removed from a handful of favelas (Praia do Pinto, Morro da Formiga, Esqueleto, Macedo Sobrinho, and Morro de Querosene) in more desirable areas of the city to the temporary housing project known as Nova Holanda.Footnote 22 Many of those eventually relocated were residents who could not afford the cost of monthly payments for properties in more permanent housing projects, such as Cidade de Deus, Cidade Alta, or Vila Kennedy (Ribeiro da Silva Reference Silva2006, 99). For some, Nova Holanda represented an improvement in living conditions, at least initially, with mostly stable multiroom houses that had electricity, sanitation, and running water. For many others, the experience was an incredible trauma, having been violently removed from their former homes, loaded onto trucks with what few belongings they could gather, and unceremoniously left in unfamiliar territory. Fabiene described how her family arrived to Nova Holanda:
My grandfather was a bricklayer so their house in Macedo Sobrinho [in Humaitá] was one of the few brick houses on the hill and it was a very big, comfortable house. There was a whole support network of family and friends close by and my grandmother ran a birosca [bar and convenience store] out of the house so she could work and stay home to take care of their six children and her brother who had a disability. But my grandfather died and when a social worker came to assess my grandmother’s economic level, they decided that she was one of the poorest because she was a widow with six children. So instead of Vila Kennedy or Cidade de Deus, they threw her in a wooden shack with a dirt floor in Nova Holanda. She didn’t have any support structure, no place to work, and six children, the youngest of which was six months old. It was a very, very difficult history. …My grandmother never spoke of it because it was too painful.Footnote 23
Fabiene’s family story is supported by the historical record. In fact, many of Nova Holanda’s wooden homes were not finished when residents arrived – the floors were still dirt and there were no faucets or electrical outlets (Vaz Reference Vaz1994, 129–33). In some cases, like those of Fabiene’s grandmother, these living conditions were far worse than the favelas from which they had come. Moreover, after only a couple of years, the sanitation, water, and electrical systems ceased to function in most of the houses and most of the wooden structures would eventually be at risk of collapsing.
Nova Holanda was initially placed under the supervision of Fundação Leão XIII, an organization born of a partnership between the Catholic Church and the municipal government. Although Leão XIII represented a shift to less repressive public policy toward favela inhabitants, its larger motive was to suppress popular unrest, prevent Communist influences, and control much of community life with the threat of expulsion (Valladares Reference Valladares2008).Footnote 24 Leão XIII did not allow residents to make any improvements to their homes even as the living conditions deteriorated (Vaz Reference Vaz1994, 125–27). Leão XIII did, however, provide some governance services to residents, at least for a while. They mediated disputes, punished and prevented crime through their connections to the police, and offered education and hygiene classes to “civilize and urbanize” a population that the state thought was unprepared for permanent apartments (Varella, Bertazzo, and Jacques Reference Varella, Bertazzo and Jacques2002, 40). In addition, through their connection to Leão XIII, police began to extort local residents by making them pay to move into the community in the first place (Vaz Reference Vaz1994, 128). Many of Nova Holanda’s original inhabitants almost immediately turned away from Leão XIII because of its corrupt behavior and intransigence to local realities. Moreover, the memories of their violent expulsions from their former homes often led to powerful grievances. Thus, when problems arose between residents or someone needed help within the favela, most residents sought out informal leaders and not Leão XIII.Footnote 25
The drug trade likely arrived to Maré with the inauguration of Nova Holanda. The temporary housing project received thousands of residents from much older favelas where “the movement” was already active. At first, it was just a few residents, mostly women, in fact, that grew marijuana in their houses and sold it to their neighbors to help make some extra income to support their family.Footnote 26 For many residents from the Northeast of Brazil, this was a familiar practice. Nova Holanda residents from this period also remember some bandidos (bandits) who lived in the neighborhood but there were no organizations or groups yet operating. According to Vicente, “One guy or another would have a small pistol that he would use to rob people outside of the favela but crime was very underground at the time.”Footnote 27 Other longtime residents told me that if there was ever violence in the community during this period it involved knives and not guns.Footnote 28 Beto remembered that there were a couple of older men with pistols in the 1960s but they didn’t display them and almost never used them. “It was all hidden,” he said.Footnote 29
Longtime residents could not recall very much violence in Nova Holanda before the 1970s and no one could remember the presence of gangs until at least a decade after the neighborhood’s creation.Footnote 30 By the late 1970s, however, six different gang-like groups had emerged and began laying claim to small pieces of turf within Nova Holanda. Each operated within just a couple of blocks or streets and their members were generally all from the same favelas from which they had been removed. According to Vaz:
The immense ruptures created by the loss of their former homes (as well as friends and relatives) traumatized the removed who, upon arriving to [Nova Holanda], were divided into groups according to the favela of origin. These became demarcated territories in which, for example, a resident of Praia do Pinto could not walk in the ‘territory’ occupied by residents from Esqueleto (1994, 136).
It is difficult to say with certainty because we have no systematic documentation of these initial groups, their membership, or the exact activities in which they were engaging, but it is likely that the gangs initially formed in response to local prejudices and fears.Footnote 31 Young men turned to forming groups to prevent physical and verbal abuse by others.Footnote 32 After forming for mutual protection, these groups then began to use their organization to make money by developing the drug trade or engaging in daring robberies and assaults outside of the favela.Footnote 33 By all accounts, the gangs coexisted relatively peacefully until the end of the 1970s when small conflicts and physical confrontations gradually grew to include gunfights and revenge killings.Footnote 34
By the early 1980s, intergang violence began to seriously disrupt community life. The Irmãos Metralha (Beagle Boys), named after the Disney cartoon characters, were the most powerful gang in the area. They were led by the youngest of five teenage brothers, Wanderlei, referred to as Lelei. Through direct conflict with some of the other local gangs, the Irmãos Metralha had won control of roughly half of Nova Holanda. The other half was contested by a handful of less powerful gangs: Candinho’s gang, Vavá Rosalvo’s gang, Tiãozinho’s gang, Juarez’s gang, and the 8th Street gang.Footnote 35 Candinho’s gang would eventually become the most powerful of these and initiate an all-out war for control of Nova Holanda with the Irmãos Metralha. Although some of the other gangs persisted, they were quickly destroyed or incorporated by Candinho’s gang during the conflict.
At this time, neither of the two gangs yet had high-powered rifles. Most of the members had pistols of varying calibers and one or two of the members possessed a submachine gun.Footnote 36 Despite their lack of high-powered weaponry, because of their relatively equal size and strength – roughly twenty members, ranging in age from late teens to early twenties – these two gangs remained in a constant state of violent conflict for several years (Silva Reference Silva2012, 106–9). One gang would routinely invade their rivals’ territory in the early morning and kill as many members of their rival as they could find.Footnote 37 Despite the presence of a DPO in Nova Holanda and police efforts to find and apprehend gang members, the police lacked the capacity (and perhaps desire) to effectively pursue and capture the gang members or put an end to the violence.
During the two-year gang war, the Irmãos Metralha and Candinho’s gang began to intensively control space by patrolling the border with their rival and preventing residents from crossing over. Each gang also began to engage in extreme forms of punishment for anyone suspected of cooperating with their rival (Silva Reference Silva2012, 106–9). There were few limits to what either of the gangs would do to try to defeat their rival: They tortured their victims by cutting off limbs, mutilating their bodies, and leaving them in the street. No one was beyond suspicion. In 1983 alone, more than fifty people were murdered in Nova Holanda.Footnote 38 The violence became so bad that residents stopped crossing over the border between the two gangs’ territories and began referring to the two sides of the community as Nova Holanda I and II.Footnote 39
Overall, favela residents had little recourse to stop these battles for outright supremacy. The history of Maré does, however, contain one prominent example of community collective action to try to bring an end to gang turf wars. Amid the conflict between the Irmãos Metralha and Candinho’s gangs, several local social movements and community institutions, including members of the Catholic Church, a women’s group, community health workers, the local samba school, and members of Maré’s União das Associações de Moradores da Maré (an association of various AMs in Maré) decided to try to put an end to the violence. They organized several protests, hung a massive banner spanning the border between the gangs that read “Vamos Acreditar na Paz” (Let’s Believe in Peace), and met with the two gangs’ leaders, individually, demanding an end to the war (Silva Reference Silva2012, 106–9). Lelei and Candinho eventually agreed to a truce.Footnote 40 In a historic scene, the two fully armed gangs met at a predetermined time on the border between their territories. In front of a large crowd, gang members threw their weapons on the ground, thus bringing an end to their yearslong conflict.Footnote 41 Although each gang retained control of their respective turf and continued to engage in a variety of illegal activities, their war and all its associated violence came to an end. This, however, would be the last time residents managed to collectively demand an end to gang violence in Maré. The territorial disputes and intergang conflict that would arrive to Maré in the 1990s would be of an entirely different magnitude.
Despite the truce, the rivalry between the two sides continued. At the end of 1983, the Irmãos Metralha side of the community, which contained many of the palafitas, was destroyed by the municipal authority and all of the residents were removed to housing projects in other areas of Maré (see Chapter 7).Footnote 42 Without a reliable base of operations, several of the Irmãos Metralha were quickly imprisoned and their leader, Lelei, was eventually killed in 1985 (Silva Reference Silva2012, 103).Footnote 43 Candinho’s gang immediately took over the Irmãos Metralha territory. Around this time, Candinho was arrested and his second in command, a man known as Jorge Negão (Big Black Jorge), became the gang’s new leader. For the next decade, Jorge Negão would be the most powerful gang leader in Maré. He developed an intensive and personalistic form of governance in Nova Holanda, in which he became the local arbitrator of disputes, punisher of criminals, and distributor of benefits.
Shortly after becoming Dono, Jorge Negão distributed pamphlets to residents explaining that any violence within his territory would no longer be permitted.Footnote 44 He showed residents that he was true to his word by harshly punishing those that disobeyed. Over the years, Jorge Negão would become renowned for his adherence to this rule.Footnote 45 He and his gang often turned to brutal forms of punishment, including execution, for crimes committed within the community, such as stealing or rape.Footnote 46 While such spectacular forms of punishment would seem to indicate a more disordered and competitive environment, residents of the era welcomed the harsh form of justice. Many longtime residents described Jorge Negão’s tenure as idyllic and peaceful.Footnote 47 In an interview from that era, one resident reportedly said, “Here in the favela no one steals. If you steal, you are expelled or killed. These guys [the gang] give us protection. We get more protection from them than we do from the police.”Footnote 48 Another resident described Jorge Negão’s regime in the following way: “There are people who think they can come out of nowhere and stab each other… and there are lots of feuds between families. If it weren’t for the gang, there would be a lot of death here.”Footnote 49
Especially following several years in which crime was mostly uncontrolled within the community and levels of violence reached incredible levels, Jorge Negão’s strict form of order was welcomed by many. Homicides fell from an astonishing fifteen per month in 1984 to less than one per month in 1986.Footnote 50 Jorge Negão was especially popular among shop owners as he implemented a security patrol in which gang members would ride bicycles around the community to ensure there were no crimes occurring.Footnote 51 Other than the brutal punishments – directed mostly at young men caught stealing or engaging in other prohibited activities within his territory or gang members accused of vacilação (disloyalty) – Jorge Negão and his gang refrained from using threats and violence more indiscriminately. Residents were able to move freely and the gang did not monitor their borders like during their war with the Irmãos Metralha.
The principal reason why Jorge Negão could maintain a less coercive posture was that he and the other existing Donos in Maré refrained from violent competition. After the truce between the Irmãos Metralha and Candinho’s gang, no one could remember any intergang violence occurring until the arrival of the factions in the 1990s.Footnote 52 There were no invasions or other gang confrontations for the better part of a decade.Footnote 53 According to gang members of this era, the Donos from Nova Holanda, Morro do Timbau, and Parque União respected each other and allegedly even attended baile funk parties in one another’s turfs, something that would become unimaginable after the arrival of the factions.Footnote 54
In addition, with the arrival of cocaine to Maré in the 1980s and the increasing profits of Jorge Negão’s gang, many residents began to seek the gang leader out to help pay for medicine, food, or other expenses. On the one hand, as one of the wealthier members of the community, and given the degree of poverty among some residents, it was seen as his duty to help those in need. This was probably the case for other more economically stable members of the community as well but there was a quid pro quo nature to these exchanges that likely did not exist for other more licit enterprises. Jorge Negão was a wanted man, and the police regularly scoured the area for him.Footnote 55 The Dono would have suffered greatly – he eventually did – should he have been captured. Both he and Nova Holanda’s residents knew this.
Jorge Negão was a famously popular and well-liked Dono. He often bought snacks for kids, medicine for the elderly, and occasionally paid for holes in the street to be filled.Footnote 56 His gang regularly stole trucks full of chickens or robbed rice, beans, and other staples from grocery stores that they then distributed to residents in what Osvaldo referred to as “a Robin Hood thing. …If you asked my mother straight up, she would still say Jorge Negão was the best gang leader ever.”Footnote 57 The Nova Holanda gang also organized weekly baile funks, where DJs would play an emerging, underground style of music that combined explicit lyrics with a heavy bass track.Footnote 58 Jorge Negão and his gang also threw holiday parties on Easter, Mother’s, Father’s, and Children’s Day, and for the Festa Junina (June Festival), a Brazilian holiday celebrating the harvest and rural life. Jorge Negão even funded drag shows called Noite das Estrelas (Night of the Stars), in which trans and queer members of the community would perform to large audiences (Noite das Estrelas 2021). Jorge Negão also often set up a television in the street so that everyone could watch the very popular novellas (soap operas) of the era.Footnote 59
Decades later, men and women who grew up during this period recounted how they had idolized Jorge Negão. Some of the men I interviewed admitted to pretending to be him when they playacted with their friends, shooting at police or other bandits with his characteristic submachine gun.Footnote 60 When Jorge Negão was finally killed by a Federal Police agent in 1992, hundreds of residents put candles in their windows and along the streets to commemorate his death.Footnote 61
Jorge Negão was not the first Dono in Rio to engage in the provision of benefits – such practices were one of the primary ways that Donos throughout the city could popularize their authority and maintain the support of the community. Like other Donos of the era, Jorge Negão’s relationship with residents of his territory was always a proprietary and personalistic one. This was Jorge Negão’s gang and his community. He was the one on the streets doing the punishing, arbitrating disputes, and providing benefits. His incredible popularity likely delayed the integration of the Nova Holanda gang into the expanding prison-based factions by several years (see Chapter 5).Footnote 62 Although the role of the Dono would continue to be an important element in future criminalized governance regimes, no longer would the Dono be so visible or at the forefront of governance. Such highly personalistic rule would mostly come to an end. Instead, governance activities would be implemented by an expanding organization.
Parque União
The gang that would eventually emerge and govern Parque União followed a very different trajectory. Unlike in Nova Holanda, Parque União had little to no crime during its initial stage of development. Residents could not remember any thieves, robbers, or drug dealers living within the quickly growing favela for the first two decades of the community’s existence, until the end of the 1970s. Tight control of the community, initially by Margarino Torres, a Communist lawyer, then by other community leaders, likely prevented crimes from occurring and gangs from emerging. In addition to organizing the settlement of the area and prohibiting the entrance of “bad elements,” Torres charged all residents a tax, prohibited prostitution and gambling, and burned down houses that were built without his authorization (Vaz Reference Vaz1994, 167). Through his legal expertise and connections, Torres also helped consolidate the area as a permanent settlement by preventing police from entering the community without search warrants (Ribeiro da Silva Reference Silva2006, 88–89).
In 1961, just a few short years after the initial occupation, Torres abandoned Parque União and Antônio Azevedo took over the administration of the area. Azevedo would continue many of Torres’ practices and was instrumental in the organization of an AM (Vaz Reference Vaz1994, 167). Parque União continued to rapidly grow during the 1960s as the settlement’s wooden shacks were gradually replaced with brick-and-mortar houses. In the coming decades, many residents managed to build multiple levels to their homes and Parque União would become the most populous neighborhood in Maré.
In the 1970s, a PPC was installed along Avenida Brasil, on the edge of Parque União. Residents remembered that, initially, police in the PPC worked with the AM to resolve problems and fight low-level crime within the community.Footnote 63 One longtime resident recalled that the police would sometimes arrest or detain “vagabundos” (bums), take them to the PPC, and force them to clean the bathroom or just check to make sure their papers were in order.Footnote 64 No one I interviewed from Parque União, however, could recall the police engaging in the kinds of torture and abuse that I commonly heard from Nova Holanda residents.
There are several possible explanations for this disparity. First, the PPC in Parque União was located on the edge of the favela and not in the very heart of the community like in Nova Holanda; torture could have been occurring and residents were just unaware. An alternative explanation is that the police in the two posts behaved differently. According to Bernardo, for instance, the police in Nova Holanda felt like they needed to deal with criminals in a more violent and coercive manner because they were failing to control violence and criminality.Footnote 65 Moreover, the population in Nova Holanda was inhabited overwhelmingly by descendants of slaves who had lived in Rio’s older favelas for generations whereas Parque União was mostly settled by mixed race migrants from the impoverished Northeast region of Brazil.Footnote 66 The different practices of the police could have also been driven by their perceptions of and prejudices toward these two communities. Finally, it may have been that the levels of abuse and torture were similar in the two communities but that due to the histories of removal and their disdain for the police, residents of Nova Holanda remembered this violence as excessive and unacceptable. Whatever the explanation, residents recalled that the relationship with the police in Nova Holanda had always been more precarious than in Parque União. Nonetheless, over time, the relationship with the police in Parque União would similarly deteriorate and, as drug traffickers and groups of robbers became more common, the behavior of the police became as aggressive and abusive as other areas of Maré.
By the late 1970s, several individuals and small groups began to engage in crime and violence in Parque União. Vinícius, who moved to the area as a child in 1963, recalled that Lulu, a local musician who played in a Música Popular Brasileira (Brazilian Popular Music) or MPB band, and who would later become the leader of the gang, sold marijuana on the side during the 1970s.Footnote 67 His drug dealing was clandestine and he did not carry any weapons at first. In addition, the jogo do bicho was also active in the community and there were several groups of young men who were known as robbers and thieves.Footnote 68 Importantly, none of these small groups were part of larger organizations and they lacked the capacity to monopolize criminal activities and violence within the community.
Lulu would eventually be arrested in the mid-1970s and sent to prison where he made connections with elements of what would later become CV. Upon his return to Parque União several years later and with access to CV’s trafficking networks, Lulu started selling cocaine and slowly expanding his gang. Residents began to refer to him as the Dono.Footnote 69 Despite the use of this moniker, Lulu still wielded relatively little authority within Parque União. Unlike Nova Holanda, where multiple gangs emerged and began to compete for control of the drug trade, forcing them to become cohesive and militarized units, there was no similar process in Parque União. Residents said that Lulu controlled the drug trade with only a pistol and that although he had a bunch of friends who were also involved in drug dealing, “they weren’t really an organization.”Footnote 70 Without a rival threat, Lulu and his gang were not incentivized to militarize nor to create a more cohesive organization such as Nova Holanda’s gangs.
For the duration of the 1970s, Lulu threw some parties and organized the occasional soccer tournament but did not provide much in terms of resources to help people in the community. He did little to control violence within the community and mostly did not mediate disputes between residents. As an example of the relative lack of power of Lulu and his fledgling gang, residents retained the right to have and use firearms until the 1990s. Evaristo, for instance, remembered that his grandfather, a local bar owner, used to have a gun behind the bar just in case there would be any problems with customers.Footnote 71 He recalled that his grandfather would shoot his gun in the air when customers got into fights to calm a situation down.
There are two interesting things to note in Evaristo’s recollection. First, it is almost unheard of for residents not involved with the gang to have or use firearms. Once a gang takes control of a community, they will never allow residents to have or use weapons. The fact that Evaristo’s grandfather continued to have a firearm signifies that the local gang still had not monopolized violence within their turf. Second, the fact that there were fights and problems within the community serious enough that a firearm was necessary also means that the gang remained relatively weak. Serious fights and destruction of property other than if sanctioned by the gang itself is almost nonexistent in favelas with a powerful gang because everyone knows that severe forms of punishment await anyone engaging in such behavior. Moreover, Evaristo recalled that even in the early 1980s, police would still frequent his grandfather’s bar during the day and then the gang members could be seen at the same bar in the evenings. In the coming decades, this would become an impossibility. Police would no longer enter favela territories so casually nor spend much time in them.
At the beginning of the 1980s, the AM in Parque União still maintained significant authority within the community. One president, Custódio Belardino, was particularly active, resolving disputes between residents, offering a variety of services through the AM, and successfully demanding a series of infrastructure projects from the municipal and state governments.Footnote 72 In addition, the Catholic Church as well as several Evangelical Churches already maintained a strong presence within Parque União. Residents also sought out these organizations to resolve disputes and gain access to services.Footnote 73
In the early 1980s, crime and violence began to increase in Parque União. The archive of O Globo newspaper contains dozens of articles detailing the activities of criminals, mostly thieves, robbers, and kidnappers, who used Parque União as a home base, yet another example of the weakness of Lulu’s embryonic gang.Footnote 74 A stronger, more cohesive gang would not have allowed such unaffiliated individuals to use or live in their turf for fear of attracting attention to their own activities.Footnote 75 Such forms of violent crime, naturally, led to increased police attention, which eventually forced Lulu to start imposing a more responsive form of order.Footnote 76 According to Gustavo, after the police started conducting more operations, Lulu made sure that people did not steal within the neighborhood and expanded his governance activities by resolving a lot of the residents’ problems.Footnote 77 In this way, it appears that the gang’s first forays into governance were less driven by competition with rivals and a need to assert their dominance, as in Nova Holanda, and more incentivized by a need to avoid enforcement.
Over the next few years, Lulu treated the community very well, avoided antagonizing the police, and prevented crime and violence within the neighborhood. He was eventually killed in 1986 by a member of his own gang, Djalma Negão, who then tried to take over leadership of the gang.Footnote 78 Lulu’s brother, Joãozinho, however, ended up winning the support of the rest of the gang, forcing Djalma to flee. One month later, Djalma returned with a gang of his own, and attempted take the community back by force, the only time anyone could remember that a gang tried to invade Parque União. He was killed by police in the ensuing shootout between the two gangs.Footnote 79 According to Vinícius, Joãozinho was a very social and benevolent Dono but he also died at the hands of police a short time later.Footnote 80 The leadership of the gang then passed to Leandro, Joãozinho’s nephew, who was only a teenager at the time. According to Vinícius, “Leandro was a cruel and wicked kid.”Footnote 81 He disrespected residents, refused to mediate disputes, failed to combat crime, and provided few services. This was one of the worst periods for residents in Parque União because they were still unable to go to the police for fear of retribution but Leandro had little sense of what he should provide to the community in return.Footnote 82 Continuing enforcement efforts should have incentivized Leandro to provide responsive benefits to the community as Lulu and Joãozinho had done before but he failed to understand this fundamental relationship and what he needed from the community (perhaps due to his age). Overall, this meant that Leandro never gained the authority or fame of some of his counterparts in other areas of Maré (longtime residents of Parque União seldom even remembered his name) and perhaps explains why he would also eventually be murdered by a member of his own gang.
Despite Leandro’s deviation from expectations in terms of his treatment of the community, the case of Parque União, like Nova Holanda, supports the theory developed in Chapter 3. One of the unique qualities of Parque União during this period was its relative calm. By the late 1980s, most other communities in Maré had already experienced violent turf wars for control of the drug trade. Parque União was an exception. The primary reason for this lack of competition was that a large drainage canal separated Parque União from the rest of Maré. To enter the community, residents had to pass by the PPC along Avenida Brasil, likely preventing other gangs from invading. Conversely, the canal also prevented the Parque União gang from attempting to expand their own territory. This lack of rival competition meant that the gang did not need to expand the organization by increasing its manpower and weaponizing to the same degree as in Nova Holanda. Without the incentive to monitor and control their territory more intensively, the gang did not monopolize violence until almost a decade after Jorge Negão and his gang in Nova Holanda. And yet, due to increasing pressure from the police, the Parque União gang and its leaders (other than Leandro) began to resolve disputes and provide a more responsive form of order within the community. In these ways, the Parque União gang demonstrates the close connection between enforcement and the provision of benefits.
Morro do Timbau
The gang that would eventually emerge from Morro do Timbau followed yet another path of development. From the very beginning of Maré’s first settlement, the military controlled migration to the area and dominated resident life with the threat of expulsion. Residents eventually formed a close relationship with the military that helped to prevent gangs from forming until much later than in either of the other two cases detailed earlier. A single gang would eventually emerge on the Morro, more than a decade after Nova Holanda gangs had begun to violently compete with one another and after Lulu’s drug-trafficking operation had already formed. As a result, despite the Morro predating the other favelas in Maré by several decades, the Morro do Timbau gang’s governance activities began much later than the other two cases.
Shortly after the founding of Morro do Timbau, an adjacent area was selected for the Brazilian Military’s 1st Regiment of Combat Vehicles (1º Batalhão de Carros de Combate) and a base was built in 1947. An army Sergeant named Adauto quickly appeared on the Morro, telling residents that the entire area was the property of the military, and that they had to pay a monthly fee to avoid expulsion (Santos Reference Santos1983, 44). Because they were squatting on land they did not own, the residents had no legal recourse. According to military personnel, this was unsanctioned behavior, and the senior officials had no knowledge that Adauto was engaging in this activity (p. 44). Nonetheless, Adauto allowed the residents to remain on the Morro and even helped the community expand. This began a symbiotic if unequal relationship between the community and the military that remained in place for most of the next four decades.
For the first several decades, control of life by the military was near absolute. Adauto required residents to sign a contract that stipulated that the lands on which they lived belonged to the Armed Forces and that they would behave and organize their homes as if they were in the military (Ribeiro da Silva Reference Silva2006, 205). Anyone who wanted to move to the area was forced to pay a fee and meet several standards of respectability and financial security.Footnote 83 In the 1950s, the military implemented even stricter rules under a new sergeant’s (Julio) tenure by prohibiting any new constructions or renovations and limiting entry and exit to the favela, while continuing the protection scheme (Ribeiro da Silva Reference Silva2006, 77). When Sergeant Julio found that new houses had been built without his authorization or that renovations to existing homes had been made, he destroyed them immediately with no regard for families or possessions. Military personnel even carried out curfew checks at 10 pm to ensure residents were in their homes (Santos Reference Santos1983, 7–8). If there were interpersonal conflicts or problems within the community, Sergeant Julio or one of the Commanders at the base would resolve the dispute.Footnote 84 Finally, they even built a barbed wire fence around the Morro with a guard post where residents were forced to present documentation to enter and exit and which prevented them from having visitors outside of the specified hours (Santos Reference Santos1983, 8).
On one hand, the military’s imposition of this extremely strict form of order had several benefits. First, the Morro became the least densely populated favela in Maré because many would-be migrants to the area were turned away and others expelled. Also, by controlling entry and exit and by imposing curfews, much crime and violence was likely prevented. Those with the need or desire to engage in illegal activities were wary of doing so through the credible threat of expulsion. In this regard, the Morro differed considerably from the neighboring favelas of Baixa do Sapateiro and Parque Maré, which were not supervised by the military and lacked a significant form of organization. Settlement of these areas was totally unplanned. Criminality emerged much sooner in these areas and they were considered by residents to be, “a very dangerous place… full of bandits” (Vaz Reference Vaz1994, 9). Though this description is likely hyperbolic, the presence of thieves and bandits in other areas of Maré significantly predates their emergence in Morro do Timbau.
Such absolute control by the military, however, made life difficult for residents. Although the community was highly organized, residents were unable to improve their housing situation, lacked the resources to improve the infrastructure of the community, and the military often dealt with them in a rather violent and brutal fashion.Footnote 85 With the prospect of eviction and removal constantly hanging over them, residents lived in a precarious and untenable situation. They eventually began to push back. First, residents became increasingly distrustful of the military’s claim to their land. Groups of residents disseminated lists with hundreds of signatures to various authorities demanding property rights, even inviting several political party members and local politicians to visit the Morro (Vaz Reference Vaz1994, 11). In return for promised support during elections, these politicians became highly critical of the military’s treatment of residents.Footnote 86 The response of Sergeant Julio and the base Commanders was initially retaliatory but, by the 1960s, the residents and the military came to an agreement. The military could maintain their security regime but residents were allowed greater freedom of movement and, with the authorization of the Commander, the ability to renovate and improve their homes (Santos Reference Santos1983, 13). This led to several public works projects, the first of which was the installation of three huge water cisterns followed by the inauguration of a Light Commission in 1967 that provided electricity to many of the residents (pp. 14–15). Finally, residents acquired guarantees that any existing homes would not be demolished or evicted (Vaz Reference Vaz1994, 10–12).
In the late 1960s, Agamemnon, a young student activist, became president of the AM, founded in 1954, which began a period of infrastructure development and an even closer partnership with the military. Agamemnon worked to improve the water networks, extend the electrical grid, pave streets, and build retaining walls and sewage drains (Santos Reference Santos1983, 20). All of this was done under the watchful eye of Sergeant Cruz, who began his tenure of control in 1970. He made sure that all infrastructure improvements did not impede military access to any part of the community and conformed to the security specifications of the base. Sergeant Cruz even lived on the Morro, in housing built for military personnel, which further solidified the close relationship between the Morro and the military (p. 19).
If any problems within the community arose – such as disagreements between community members – “Agamemnon gave the verdicts and the Sergeant legitimized them” (p. 20). Occasionally, soldiers from the base would get drunk on açucana (sugar cane liquor) and abuse or threaten residents. In these cases, Agamemnon went to Sergeant Cruz or the base Commander to complain and the soldiers were disciplined.Footnote 87 Many residents also used to go to Agamemnon to resolve their problems; couples, neighbors, virtually anyone who needed something to be done went to Agamemnon.
For the entirety of the 1960s and 1970s, no one could remember any thieves, bandits, or drug traffickers being present in Morro do Timbau. Beginning in the mid-1980s, however, important changes were taking place. First, the military began to extract itself from the daily governance of the community. After years of investigation, the property claims of the military were found to be spurious and, after residents gained property rights, the governance arrangement between Sergeant Cruz and Agamemnon collapsed (Santos Reference Santos1983, 147). As this relationship broke down, a group of residents challenged Agamemnon’s authority, arguing that only he and a small group of friends were profiting from his “despotic” and “violent” methods (Santos Reference Santos1983, 23). In 1984, another candidate won the AM elections and the close relationship between the military and residents of Morro do Timbau finally came to an end.Footnote 88
Without the threat of eviction and punishment by the military, some young men began to engage in criminal activities. In the mid- to late-1980s, several high-profile bank robberies were carried out by residents of the Morro.Footnote 89 Eventually, a man nicknamed Kito emerged as the leader of this group of bandits.Footnote 90 He was initially more interested in robbing banks and armored cars than developing the drug trade but eventually took over the drug business from a man known as Rene do Pô (Powder Rene) in a “golpe sem sangue” (bloodless coup).Footnote 91 Although he did not dominate the area like Jorge Negão in Nova Holanda, he was able to slowly expand his organization, eventually taking over the drug trade in neighboring Baixa do Sapateiro from a man nicknamed Paulo Brasil.Footnote 92 Residents of the area began to refer to Kito as the Dono.
For several years, Kito’s group of bandits engaged in no violent competition with other gangs and, as a result, they remained uncoercive.Footnote 93 They were little present on the streets and did not monitor the borders of their turf. If and when Kito dealt with crime in the area, residents remembered him preferring to expel those involved instead of employing violence.Footnote 94 Kito and his men provided few other benefits to the community initially. The attention of the police to the emerging drug trade on the Morro and Kito’s fledgling gang seems to have remained far below what Jorge Negão and his Nova Holanda gang experienced. This difference in enforcement is perhaps because Kito’s gang were less violent and only engaged in theft and robbery outside of the favela itself.Footnote 95 In addition, unlike Nova Holanda, many residents still reported crimes and dealt with disputes by going to the PPC stationed in Baixa do Sapateiro.Footnote 96 Like Parque União, the relationship with the police was not nearly so bad on the Morro as it was in Nova Holanda. It is curious, then, that the police did not receive more complaints about the gang and its behavior though perhaps so long as Kito and his men did not engage in violence and crime within the community itself, residents did not mind their presence. It was only after Kito and his gang integrated into the TC faction, that they became a significant focus of police enforcement efforts. Thereafter, the gang would quickly become more responsive to resident demands to resolve disputes, provide economic stimulation, and organize parties. This integration also coincided with higher levels of rival threat and the increasing use of coercion to control these communities (see Chapter 7).
The Birth and Expansion of Rio de Janeiro’s Prison-Based Factions
At this point, it is necessary to address the other side of the organizations that would come to govern Rio’s favelas. While small and fledgling street gangs were slowly developing in Maré and hundreds of other favelas across the city, another type of gang was emerging from within Rio de Janeiro’s prisons. The marriage of these two organizational forms (prison and street gangs) would eventually change the nature of violence, crime, and politics in the city. In this section, I provide a brief history of the birth and expansion of CV and Rio’s other prison-based factions.
The first modern gang faction, CV, was born in the 1970s during Brazil’s military dictatorship (1964–85). In 1969, after a series of bank robberies by leftist revolutionaries, the Brazilian congress passed Decree No. 898 or the Lei de Segurança Nacional (National Security Law) through which the government could try political dissidents in military courts. Article 27 of that law stipulated that any attempt to rob a bank would lead to a sentence of 10–24 years or longer if anyone was killed (Gay Reference Gay2015, 81–84). This meant that not only were revolutionaries tried in military courts but also many less ideologically inclined, “common” bank robbers and bandits. All these men were then incarcerated together in the Penal Institute of Cândido Mendes, the famed Caldeirão do Diabo (Devil’s Cauldron) on Ilha Grande (Big Island), a former leper colony.
In Cândido Mendes, the revolutionaries and common prisoners were housed together in Cell Block B. The relationship between the common criminals and the revolutionaries was fraught from the beginning (Amorim Reference Amorim1993). This was not the meeting of two groups of equals with a common cause of ending the state’s brutal, racist, and repressive behavior. Rather, it seems that the political prisoners always sought to distinguish themselves from their fellow cell-block mates. They saw these mostly dark-skinned men from favelas as real criminals and incapable of understanding their ideological movement. From the beginning, there were conflicts between the two groups and the political prisoners resorted to using violence when they thought one of their cellmates was out of line (Amorim Reference Amorim1993, 77; Misse Reference Misse1999, 357). The political prisoners eventually demanded the prison administration separate the two groups and a brick wall was hastily constructed (Amorim Reference Amorim1993, 66; Misse Reference Misse1999, 357).
There is still significant controversy concerning the role of the revolutionaries in the eventual creation of CV.Footnote 97 According to Amorim (Reference Amorim1993), the presence of the revolutionaries was instrumental in CV’s creation while Misse (Reference Misse1999) argues that the connection between the political and common prisoners was weak and that the prisoners would have likely organized on their own. Either way, the political prisoners created a highly structured organization within Cândido Mendes.Footnote 98 Each member had a role (secretary, manager, internal duty, and political obligation) and the group prevented drug use, sexual abuse, gambling, or fighting amongst themselves (Amorim Reference Amorim1993, 66, 77). The revolutionaries were not, however, the only inmate organization at the time. In fact, four other prison gangs that included anywhere from a dozen to twenty men already existed within each of the other cell blocks. These organizations, referred to as Falanges (Phalanxes) by the media, were called Sul (South), Coreia (Korea), Jacaré (Alligator), and the Neutros (Neutrals). Each had their own unwritten rules, norms, and means of identification but their key characteristic was the abuse and predation of weaker members within the cells (pp. 54–56).
Sometime in 1974, a group of common prisoners in Cell Block B eventually decided that they would form their own organization to end the predation, abuse, and terror that characterized much of prisoner life. In the beginning, their focus was on protecting prisoners from abuse but, as the group grew, it would eventually make demands on the prison administration. Initially referred to as the Falange LSN (LSN Phalanx) after the National Security Law, the Falange Vermelha (Red Phalanx) for their alleged connection with leftist movements, or just O Coletivo (The Collective), they demanded an end to beatings by the guards, freedom to circulate in the common areas (cells were always locked before this), better treatment for visitors, including an end to invasive searches and the ability for visitors to spend the night on the island (Amorim Reference Amorim1993, 83–84).Footnote 99 They also founded a soccer team, created the Clube Cultural e Recreativo dos Internos (Cultural and Recreational Club of Inmates), which funded different activities for prisoners, and built a library (Amorim Reference Amorim1993, 97–98; Misse Reference Misse1999, 364). They even created a slogan for their movement, “Paz, Justiça e Liberdade!” (Peace, Justice, and Freedom!).
By 1979, the political prisoners had already been transferred off the island as part of the Lei de Anistia (Amnesty Law) and Falange Vermelha had incorporated most of the prisoners in Cell Block B. The group that would later become CV gained at least some respect among prisoners of the other areas for their policy of ending abuse and violence between prisoners but increasing tensions with the Falange Jacaré eventually led CV to issue an ultimatum: submit to their authority or be killed (Amorim Reference Amorim1993, 99). Jacaré members refused to disband. In the early morning hours of September 17, 1979, dozens of CV members invaded Cell Block C where the leaders of Jacaré had sequestered themselves in a cell with thirty of their most trusted members. CV members, armed with knives, sharpened spoons, pieces of wood, and pipes, broke open the cell and confronted the Jacarés, eventually killing all the major leaders of the group and injuring many others. The rest of the Jacaré members decided to give themselves up or were later killed as all the other prison gangs immediately submitted to CV’s authority (pp. 101–2).
Many of the original CV leaders were quickly transferred to other prisons where they spread the organization further. Although it remained a tightly knit group inside Cândido Mendes, this expansion would irrevocably change the organization. CV would never regain the hierarchical structure and cohesive leadership it had initially. In the first phase of expansion from the prisons (1979–86), the original leaders escaped or were released and returned to the favelas from which they came. At first, this group organized kidnappings and bank robberies to fund the organization and support its members and their families. They also formed strategic alliances with local gangs by loaning guns and personnel to those who sought to take over new territory (Misse Reference Misse1999, 315, 321). To become part of CV, local gangs had to agree to an alliance against enemies as well as promise to respect, support, and protect residents and friends of the organization. CV’s initial concern with normative behavior like preventing abuse of prisoners and respect for women, children, and the elderly in favela communities – values that some gangs had already developed independently – quickly gave way to a focus on profit as they realized the enormous moneymaking potential of their organization (Dowdney Reference Dowdney2003, 29; Leeds Reference Leeds1996, 54; Misse Reference Misse1999, 367).
During the second period of expansion (1986–94), CV spread rapidly to most of the favelas in the city. CV gangs engaged in increasingly violent competition with preexisting local groups for control of the drug trade. Many of the original leaders of CV were killed and a new leadership emerged. These men were younger, more ambitious, and, overall, had less respect for favela communities (Misse Reference Misse1999, 315). Gang behavior in Rio’s favelas became more coercive in the second phase of CV’s evolution, especially because CV gangs faced increasing challenges to their dominance by a rival faction.
The origins and evolution of TC are still disputed. Misse (Reference Misse1999, 319, 322) points to another group of prisoners that organized within a different prison (Frei Caneca Penitentiary) in the 1970s with the name Falange Jacaré, which later became known as TC. Amorim (Reference Amorim1993, 55) argues that Falange Jacaré from Cândido Mendes was destroyed but that two of the other Falanges (Sul and Coréia) would later unify in a different prison to form TC. According to the autobiography of William da Silva Lima (Reference Lima1991, 63), one of the founders of CV, several members of Jacaré would also be transferred to Ary Franco Prison after the massacre, where they started TC. According to O Globo reports from the 1980s, TC either broke from Falange Jacaré or Falange Vermelha inside Cândido Mendes but the two remained entirely separate organizations that controlled different prisons during the 1980s.Footnote 100
Perhaps the best evidence regarding the origins of TC comes from the Jornal do Brasil, which documented the rise of a rival to CV inside Cândido Mendes in the early 1980s.Footnote 101 The group was originally just called Terceira Galeria (Third Gallery), a reference to the part of the prison where they were housed, and was first formed in response to the increasingly brutal and violent treatment by CV after the original leaders had been transferred, escaped, or killed. By 1983, Terceira Galeria was also allegedly present in two other prisons in Rio (Lemos de Brito and Helio Gomes).Footnote 102 In that same year, 104 prisoners from the Terceira Galeria in Cândido Mendes were transferred to Ary Franco Prison, where CV and Falange Jacaré were also present. As late as 1991, O Globo reported that the three factions (CV, Terceira Galeria now referred to as TC, and Jacaré) were still fighting for control of Ary Franco Prison, where TC and Jacaré would allegedly form an alliance against CV.Footnote 103 After this, Falange Jacaré seems to have disappeared entirely.Footnote 104 Perhaps it dissolved or was fully incorporated into TC. Either way, from the early 1990s on, TC would be the sole competitor to CV for control of the prisons and favelas throughout the city.
Then, in 1994, the CV faction split, the result of ongoing disputes between the older and newer generations of the organization and a series of events that occurred in Complexo do Alemão, a group of a dozen favelas sprinkled over several hills in the Zona Norte of the city, just a short distance from Maré. In July of that year, Orlando Jogador, the Dono of Morro do Alemão (German Hill), and one of the most respected CV gang leaders in the city, was murdered by Ernaldo Pinto de Medeiros, nicknamed Uê, the Dono of Morro do Adeus (Goodbye Hill), a nearby favela also within the Complexo do Alemão. Uê’s motivations were likely multiple. He was allegedly an incredibly ambitious young man and wanted control of Complexo do Alemão for himself but also desired revenge on Orlando Jogador for shooting and paralyzing his brother several years earlier. Whatever his motivations, Uê and his men pretended that he had been kidnapped by the police and asked Jogador to help pay the ransom. Jogador agreed and quickly gathered R$60,000 but when he met Uê’s men to hand over the money, they immediately killed him and his entire entourage. The killing of a CV Dono without the authorization of the leadership of the faction was explicitly against the rules of the organization and the younger members of CV, many of whom idolized Jogador, demanded vengeance.Footnote 105 Some of the old guard of CV maintained close connections to Uê, however, and refused to punish or expel him from the faction. The disagreement fractured the organization. The younger leaders christened themselves Comando Vermelho Jovem (Young Red Command, hereon CVJ). For the next several years, the CV and CVJ engaged in a series of assassinations and battles for control of favelas in the city.Footnote 106
Then, in 1998, Uê left CV to create his own faction, Amigos dos Amigos (Friends of Friends, hereon ADA), with Celso of Vila Vintém. ADA quickly formed a strategic alliance with TC in 1999 to better compete with the more dominant CV factions, which precipitated the reintegration of CV.Footnote 107 It would take several years to fully reincorporate the more than 120 separate gangs that comprised CV and CVJ but by the end of 2002, CVJ no longer existed.Footnote 108 In fact, the most powerful leaders of CV from this point on (especially Fernandinho Beira-Mar and Marcinho VP) would be former members of the breakaway CVJ faction.Footnote 109 Regardless of the name, the changing of the guard within CV was finally complete. Meanwhile, the alliance between ADA and TC ended on September 11, 2002, when Uê and several other leaders of ADA and TC were brutally murdered in Bangu I, Rio’s maximum security prison, by Beira-Mar and the new leaders of CV. The massacre would precipitate the creation of a new faction, Terceiro Comando Puro (Third Pure Command) or TCP, as I further detail in Chapter 7. From this point on, every gang throughout the city would at least be nominally connected to one of the three remaining factions: CV, TCP, or ADA (see Figure 4.5).

Figure 4.5 History of Rio de Janeiro’s prison-based factions
The Integration of Maré’s Gangs into the Factions
In the early 1990s, each of Maré’s gangs integrated into one of the two existing factions. The Nova Holanda and Parque União gangs became CV and the Morro do Timbau gang allied with TC. Integration would irrevocably change these organizations and their relationship to Maré’s communities. First, instead of just neighborhood gangs with mostly local ambitions, they would increasingly seek to expand their organizations and the control of the drug trade. They quickly began to violently compete for control of Maré. Second, when Maré’s gangs integrated, they each had a few dozen members, mostly young men in their late teens and early twenties. Their structure was simple, dominated by the Dono (see Figure 4.6) who was assisted by a braço direito (right-hand man). The roles of the vapor and avião or aviãozinho had already been clearly established but the rest of organization did not have a lot of structure.Footnote 110 Most members would take on various roles, including security, participate in robberies outside the favela, or assist with some other aspect of the drug trade.Footnote 111 Integration meant increasing access to cocaine through the faction’s international suppliers. The number of bocas de fumo in each gang’s territory multiplied and became stationary.Footnote 112 The gang’s structure would also become more elaborate as the roles diversified. Overall, the size of Maré’s gangs would increase from just a couple dozen members to three to four times that by the end of the decade.

Figure 4.6 Organizational structure of Maré’s gangs prior to faction integration
The integration into the prison factions would change the size, capacity, and resources available to Maré’s gangs and shape their governance in myriad ways. Most importantly, the factions would encourage the expansionary behavior of local gangs by providing weapons and men to conquer rival territories, thus incentivizing the more coercive aspects of these organizations. However, the factions would simultaneously control and manage most conflict within the faction, thus creating stable alliances and networks of mutual support. In addition, when there was the need or desire to remove a gang leader, it could only be done with the explicit approval of the faction’s leaders in prison.Footnote 113 In this way, the factions would also help stabilize gang organizations throughout the city.
Despite the significance of Rio’s prison-based factions in the forms of criminalized governance implemented across the city, the favela-based gangs would largely retain their local orientation and autonomy. As mentioned in previous chapters, Rio’s factions are better understood as loose and horizontal networks of neighborhood-level gangs than vertically integrated and hierarchical organizations. All decisions regarding governance and the treatment of residents are determined by the local gang and are not dictated by the faction from inside the prisons. Thus, there remains significant variation in governance across even similarly affiliated gangs due largely, I argue, to the very different security environments in which these organizations find themselves.
Conclusion
This chapter has traced the emergence of the three gangs that would come to dominate Complexo da Maré. In doing so, it has made a series of contributions to our understanding of criminalized governance. First, the three cases demonstrate how gangs emerged and began to govern at very different times that depended heavily on local context. In the case of Nova Holanda, gangs emerged earlier as the breakdown in previous governance arrangements occurred quickly after the construction of the community. As violence and criminality increased, multiple gangs emerged and quickly began to compete for control of turf, which eventually precipitated more intensive and coercive forms of governance. The Nova Holanda gang that eventually emerged victorious against its rivals would monopolize violence and provide significant benefits to residents early on.
The other two gangs that emerged, in Parque União and Morro do Timbau, would only begin to govern in a similar manner more than a decade later as they remained weak and, at least initially, faced little competition for territorial control. Competition between gangs is fundamental to the move to govern as gangs will seek to control their territory more intensively while also developing organizations capable of providing more benefits. In this way, coercion and benefits are intimately intertwined. That said, the two weaker gangs that emerged in Parque União and Morro do Timbau did provide some limited benefits, especially the control of crime and dispute resolution, even before they began to compete with rivals as they needed the support of residents to avoid police enforcement. Overall, these dual threats have shaped governance from the very beginning of each of these gangs’ organizational lives.
The second contribution of this chapter is that resident demands for goods and services are essential to understanding how gangs began to engage in governance. Residents of favelas have always demanded governance: crime fighting, dispute resolution, and some forms of welfare. Before gangs, other favela actors provided these services to varying degrees. In many cases, it was the AM president that maintained contacts with the police (or military) that provided some form of order within these communities. As favelas became further marginalized during the military dictatorship and existing connections with public security institutions weakened, gangs would eventually subsume these governance tasks. Many residents also sought out gang leaders to deal with a variety of local issues: crime, violence, order, and even welfare. By doing so, gang leaders legitimized their authority among residents and were offered the community’s silence and even support to evade enforcement. Thus, the relationship between gangs and communities in these early days of criminalized governance should not be understood as a purely coercive one but as more symbiotic and reciprocal as well.
In Chapters 5, 6, and 7, I focus on how each of these gangs and their governance activities evolved over the subsequent two decades, from their integration into the factions until the Brazilian military occupied Maré in April 2014 (see Figure 4.7). Throughout each, I use multiple types of data, including eighteen months of participant observation, 206 in-depth interviews with current and former gang members, NGO workers, AM representatives, and longtime residents in each of the gang territories, more than 400 newspaper and community-based journalistic accounts, and microlevel quantitative data in the form of anonymous denunciations. Together, these cases offer an unprecedented view inside these communities and organizations, showing how and why criminalized governance has evolved over time.

Figure 4.7 Maré gang case studies






