Introduction
The documents I am looking at are written on the pages of a ring bound school notebook. At first glance they look hastily scribbled. The words are heavily slanted to the right and there are many abbreviations. Many are misspelt, reflecting a phonetics of the spoken word. But on closer examination, one sees a depth to the organisation and significance of the words on each page, tabulated into categories as they are. Taken together they relate a kind of modest bureaucracy that defies the relative impermanency and amateurism of misspelt and handwritten type. Each page is in a standardised format, copied and completed by hand. On each page there are at least 11 fields. And there are more than 100 pages in this or a similar format, each a distinct record of its own. The fields require a standard set of categorical and revelatory answers: name, member number, nickname, neighbourhood, last three prisons, red telephone number, date of baptism, place of baptism, godfathers, punishment, last three responsibilities, those involved in the punishment.Footnote 1
These pages are records of punishment of members, affiliates and others that have been carried out and meticulously documented by São Paulo's organised crime group, the Primeiro Comando da Capital (First Command of the Capital, PCC). They relate a centralised system of governance and authority with systematic and gradated levels of punishment carried out through a routine process requiring the involvement of multiple members and reflecting a complex division of labour. At their core, though, records like these relate something distinct – a structure of decision-making and authority over the process of lethal and non-lethal punishment. Yet that structure of punishment does not so much reflect a desire to kill anywhere near as much as it privileges reasonable, balanced and measured consequences for actions deemed wrong. As I will detail, the vast majority of these records relate incidents where individuals have been subject to comparatively minor punishments. This is precisely why these records are necessary: they are numerous; they catalogue otherwise nominal offences and, in doing so, they allow for repeat offenders to be judged vis-à-vis their past offences and patterns of recidivism. Nearly all of the records that I have suggest that the normal kind of punishment is non-violent and relatively temperate, including temporary suspension, prazo, and permanent expulsion, exclusão. The following entry is illustrative in this regard.Footnote 2
Name: Romero Santos
Member number: none
Nickname: Bushman
Last three prisons:
Neighbourhood: Vila Pequena
Red telephone number: 011 9999999
Date of baptism: 25/09/2011
Place of baptism: on the street in Vila Pequena
Godfathers: Silvio Big Nose, Crooked Face, Little Mike
Punishment:
Last three positions: Street leader of Vila Pequena
Involved in punishment: Disciplinarian of Vila Pequena, Regional leader, Liverpool,Interior leader, Mike, Regional Disciplinarian, Clement, Bookkeeper, Marco.
After each of the filled fields is a succinct concluding note, written slightly off to one side: ‘Thrown out on 06/10/12 by Rogerio’. This document is one of a number of records of expulsion carried out by the PCC and stands in contrast to a single record of an execution by the organisation.
In this article I will argue how bodies come to be, and, more importantly, do not, is not an outcome of public administration, policy or ‘the state’ as is widely presumed. Homicide statistics, their oscillations and effects are, rather, a product of a particular configuration of governance that tends to obscure groups like the PCC within larger systems of perceived authority. Widespread acceptance and assumptions about homicide numbers by both policy and academic users further hides the influences prominent in the production of bodies and numbers, to great consequence for security, democracy and social science knowledge.
Homicide statistics are largely considered the most reliable of crime metrics. ‘One body, one tally’ as the idea goes. Believed difficult to ‘fudge’, either by police, political actors or coroners, these statistics have become the measuring stick for comparisons across continents, states, cities and neighbourhoods, serving as the near-universal standard for the state of security in time and space. For policy-makers, these statistics, or the likely prospect of them, are the basis for both foreign and domestic action. For criminologists, they are the hard currency of scientific inference. They serve as the foundation for subsequent analyses that test myriad demographic, physical and social variables, from toxic substance(s) consumed, to urban space, ethnicity, race and family structure, as well as hypotheses about policy effectiveness and crime ‘booms’.Footnote 3 , Footnote 4 , Footnote 5 The findings of these analyses underscore key theoretical assertions with regard to vital sociological and political theories about states, social ‘disorganisation’, subcultures, delinquency and policing.Footnote 6 They are also the impetus for vast and sweeping police reform efforts.Footnote 7 Homicide statistics, in other words, are an epistemological buttress for a great deal of our contemporary beliefs about safety, security, the need for policy intervention and, particularly in the Global South, the vitality of nascent democratic institutions.
But if the emphasis has largely been on the statistic as an outcome, there has been comparatively little rigorous study of the organisation of homicide, even in the Global North.Footnote 8 Increasingly, though in a limited way through the availability of data on motives, scholars have considered the relevance of the predictable patterning of homicides.Footnote 9 But this is not just a question of social networks and conflict between social groups, like gangs.Footnote 10 It can also be a question of a distinct kind of social control over entire neighbourhoods and communities, particularly in recently urbanised cities in the Global South.Footnote 11
Based on three years of ethnographic fieldwork alongside homicide and other detectives in the city of São Paulo, I argue that such an emphasis is urgent. Via sustained observation of homicide investigations, interactions between detectives, members of organised crime and residents of communities controlled by these same groups, I detail how a focus on the process of homicides, what happens before police encounter and count the bodies, illuminates a pressing concern for the study of crime and violence, intimately related to the regulation of life and death, that is, of why and how (and how many) dead bodies come to be.
At the heart of this concern is a challenge increasingly evident in urban regions in the Global South, the emergence of a new kind of organised crime group. These groups, such as the PCC of São Paulo have shown the desire not just to control illicit markets and supply chains of drugs and guns. They also, and perhaps primarily, seek to regulate the amount and type of crime and violence, including but not limited to homicide, in a given time and space. Threats to security, like gang war retribution for a murder are tightly governed. Here, honour as a motive for murder is irreconcilable with the mores of centralised order. This seems far from the individualist American ‘Code of the Street’.Footnote 12 Emergent from contexts of urban violence in periods of post-dictatorship and Cold War governance, these groups, evident in places as far afield as Jamaica and Jakarta, are increasingly prominent regulators of security and death in historically violent parts of cities.Footnote 13
But all of this has a key backdrop. An urban region with 20 million residents, São Paulo is celebrated for a remarkable trend in its homicide rate.Footnote 14 Since 2000, the number has dropped by upwards of 85 per cent according to the statistics of the Secretary for Public Security.Footnote 15 Not only that, in 2011, the United Nations Office on Drugs and Crime highlighted São Paulo in its global report on homicide as an example of violence reduction, without any evidence as to how such a reduction occurred.Footnote 16
I will detail here how the homicide rate of the city obscures the organised practice of homicide within it. There are two systems of criminal ‘law’ in this city. But they are not inseparable. Homicide detectives demonstrate that police and organised crime often work with overlapping notions of who should live and who may die in São Paulo. The formal justice system itself is deeply intertwined with the alternative legal structures of organised crime, such that the two do not necessarily operate in parallel. I argue that they are nested systems that together are responsible, in good and bad, and for better or worse, for the number of homicides in São Paulo.
Methodology
This article is built upon three epistemic foundations. First, it draws upon a multiyear ethnographic study of homicide and other detectives of the Civil Police in the city of São Paulo. This research examined both the social world of these detectives, but also the patterns of homicide and police killings known as resistências seguida de morte carried out in the city. I undertook this research between 2009 and 2012. During this time I accompanied detectives in the stations, delegacias, of the city and in two specialised precincts, hate crimes, crimes de intolerância, and homicide. This entailed watching police practices and conducting informal interviews with police both on the job and off, as well as accompanying them as they visited different crime scenes, took breaks in local coffee shops and did all other manner of routine police related tasks. Second, I analyse the set of PCC documents given to me by individuals within the public security system. These documents demand a specialised knowledge of language, context and politics to be made legible. I draw upon my in-depth knowledge of the criminal justice system gained through time following the Civil Police as a means to make sense of the words, structure and ideological position of the PCC. Third, this analysis hinges on research that I carried out in a PCC community in 2006 and again in 2011. In these periods I interviewed residents of a neighbourhood on the East Side of São Paulo. These residents had witnessed first-hand the transition from the uncontrolled neighbourhood violence of the late 1990s through to the PCC's consolidation of authority and the establishment of a dramatically more reliable system and sense of security for community members. Their street-level experience with the PCC's system of justice, including the establishment and use of tribunals, ties together my use of these three multi-scalar and epistemically diverse sources of data, each providing a complementary perspective on the questions I deal with here. In bringing these three epistemes together I seek to unite otherwise diverse and multi-scalar perspectives through a storyline following how bodies come (or do not come) to be.
However, there is a caveat. This article is part of a larger study and set of publications on the regulation of life and death in Brazilian cities.Footnote 17 In particular, it is an expansion and an elaboration of an argument I make in passing in a recently published book.Footnote 18 There I detail the everyday relations between the PCC and the public security system more broadly, and especially as it regards the investigation of homicides and police shootings of citizens, known until recently as resistências seguidas de morte, by detectives of the Civil Police in the city of São Paulo. In this article I admittedly skim past some of the larger issues I discuss elsewhere. I do so both for brevity and lack of repetition. At hand is a small piece of a larger story about the mundane practices of shared sovereignty and its oscillations, life amidst the PCC system and the efforts of street-level police to navigate a complex existence alongside that same system, both in their homes on the margins of the city and on the job.
The Conditions of Life and Death
Control over the conditions of life and death is a principal tenet of many theories of sovereign power, whether read through Schmitt's ‘exception’, Mbembe's ‘necropolitics’ or Weber's ‘legitimate use of force in a given territory’.Footnote 19 In the contemporary world, the sovereign power vested in states is reflected in the ability to define and regulate when, how, where and why people can be killed, either directly by those empowered by the state, or indirectly, as constituencies are abandoned and deemed expendable.Footnote 20 This is true not just of a police officer shooting a citizen or a state administering a deadly cocktail to those on death row. It is also a much more diffuse practice of who is left to die in a given society.Footnote 21
As an extension, homicides are a barometer of the sovereign ability to define the spatial, ethnic, racial, gender or other constituencies that can die more often, without decisive objection. That homicides in much of the Global North are often ‘disorganised’, with motives reflecting individual or small group action, passion, robbery, sexual assault and gang disputes, is no surprise. Though some homicides are patterned, as with the gang turf wars evident in cities like Chicago, we tell ourselves that homicides rarely reveal an alternative and sustained attempt to define, via violence, who may live and who may die.Footnote 22 With some prominent exceptions like Northern Ireland and Southern Italy, homicides do not seem to evoke a premeditated, organised, uncontested and sustained effort to order social relations, other than that of the state. According to dominant theoretical understandings, to do so would be to contest sovereign power and to incite a decisive and defining response on the part of the state.
In many places, then, homicide statistics seem insured against default by sovereign power. They are a largely inviolable reflection of the provision of public security as a redistributed service, as orchestrated by states, watched over keenly by police and gate kept by political handlers via systems like Compstat. This is particularly true of regions of the world such as the United States and the United Kingdom that are the focus of much of the criminological research, where the regulation of homicides remains the exclusive purview of states, with police (and homicide detectives, in particular) as the street-level emissaries of that regulatory order.Footnote 23
It has become increasingly apparent, though, that some states do not maintain the same semblance of control over the conditions and definitions of life and death.Footnote 24 Following the Cold War, many national security states that had been propped up by explicit or clandestine support from the United States were forced to devolve power in the face of widespread democratic reform.Footnote 25 The subject of widespread suspicion and scorn, tightly centralised systems of security, policing and counter-subversive inquisitions were dissolved or pared back to their bones. Not only that, public security in many of these places had always been a deeply exclusionary practice, especially on the margins of society, in cities and for certain races and classes.Footnote 26
With public security a mirage, and compounded by neoliberal pressures for smaller state footprints, many states and the primary cities within them have crumbled into violence in recent decades. In places as dissimilar as Johannesburg and Jakarta, nascent gangs, organised crime, extermination groups, para-state militias and police fight among themselves, particularly in parts of cities with historically weak state-society relations, evident in the urbanisation of ‘slums’.Footnote 27 Emergent from civil-war-like conditions of violence, many of these groups developed strong self-protection rationales seeking to shelter their members (and, sometimes ‘constituents’ as well) from the perceived harm of other groups, including police. Not just that, in the absence of state provided security in this new era of ‘democratic’ governance, many Latin American communities, from Argentina to Mexico, have since banded together and resorted to their own localised means of ordering violence, relying on tools like community patrols, lynching, social cleansings and extrajudicial killings of perceived sources of disorder and violence.Footnote 28
There are strong historical precedents. Bottom-up ‘legality’ and order maintenance (or making) has deep historical roots in Latin America, and, indeed, in the South more widely.Footnote 29 Research has shown that community-based ‘alternative dispute mechanisms’ are especially strong in urban spaces that have a contentious relationship with the state and its law in practice and on the books. Often on the receiving end of social control and legal outcomes, communities like ‘Pasargada’, an informal settlement detailed by Santos, rely heavily on local associations to resolve conflicts.Footnote 30 These local associations of ‘the oppressed’ provide parallel legal systems that attempt to resolve and find ‘justice’ for the kinds of disputes deemed irrelevant or unimportant by the dominant legal system and its actors.Footnote 31 In doing so, they also mimic some of the legal and bureaucratic formalities of the state's system. The kinds of problems that residents' associations deal with in places like Rio de Janeiro are those that occur in spite of the state's letter and practice of the law. These are problems like the transfer or sale of squatted land, the resolution of disputes between businesses that do not formally exist, the movement of boundaries of property and ‘ownership’, and the sharing of utilities, water, electricity, that, in different periods, were otherwise considered ‘stolen’ from the public infrastructure. In Rio de Janeiro, for example, many have described the central role that residents' associations play in resolving everyday community disputes, and how those associations have a close but conflicted relationship with the larger formal legal system.Footnote 32
It has since become much more complicated. Some such groups have grown both deep and wide in terms of social control, becoming expansive in scale and social relations as they exert authority over illicit economies, urban space and, more importantly, the use and the limitation of violence in a given territory. In doing so, they also transcend urban space, indelibly linking that sphere of marginality with another in the prison system. The categorical difference between the resolution of everyday disputes that do not cause physical or emotional harm, and those that, in any system, would be considered ‘criminal’ is central to this shift. Santos argued that residents' associations like that in Pasargada shied away from resolving problems of a criminal nature, neither attempting to find solutions nor referring these cases to police. As Santos states, given that ‘favelas’ of the time were empirically and in larger perception home to workers and ‘criminals’ or other disdained groups, sex workers, child labourers and street children, for example, residents' associations did not want to be seen as either placating violence or criminalising their own community. This left a gap in the regulation of ‘crime’ and violence across the two existing spheres of legal pluralism that would eventually be filled.
That 1970s Rio de Janeiro and parts of cities in similar structural conditions did not have prominent local institutions or mechanisms for regulating and resolving these questions set the stage for something larger. This became a deeply consequential problem, allowing, perhaps fostering, the future growth of groups interested in using and regulating violence. In Rio, that void was eventually filled by the Comando Vermelho (Red Command, CV) that emerged from the prison system as both a criminal and ‘legal’ entity.Footnote 33 The CV, and later groups like the Amigos dos Amigos (Friends of Friends, ADA), established deep roots, and indeed gained much local legitimacy, for their management of violence and other sources of criminality within the community.Footnote 34 This was in spite of how seemingly arbitrary their own violence could be.Footnote 35 The rise of these groups as regulatory actors occurred in an inseparable but complex (and complementary) relationship with residents' associations, meant that there were arguably three interlinked ‘legal’ systems, that of the state, that of non-violent disputes and another for violence.Footnote 36
The Primeiro Comando da Capital of São Paulo
In the early 1990s the prison system of São Paulo state gave rise to a prison organisation known as the Primeiro Comando da Capital (PCC). The PCC emerged from a set of violent circumstances that collided in one prison in the city, known as Carandiru. Prison conditions at the time allowed for rampant sexual abuse, an epidemic of disease and uncontrolled violence in a prison population that vastly exceeded its capacity. Varella, a physician working in the unit at the time, has referred to these conditions as ‘hellish’.Footnote 37 In the midst of a riot in 1991, a SWAT-like police unit known at the ROTA entered the prison but promptly killed over 110 prisoners. Recent testimony during a trial for some of these police revealed that the shootings were savage, with evidence leaving little doubt that most of the prisoners died while cowering in their cells.
For survivors, these conditions consolidated a shared notion of persecution and identity. Within two years, a codified form of government emerged that was destined at reducing violence.Footnote 38 In 1993, the PCC drafted a formal statute with a rhetoric of ‘peace, justice and liberty’, and vowed never again to accept violence, either from the state or among the prison population.Footnote 39 In the span of months, the organisation's moral order of rules backed up by the threat and use of violence was being enforced in the majority of the state's prisons as the de facto system of governance. Today, as one public prosecutor told me, the PCC controls 135 out of 152 prisons, the other 17 are institutions for the mentally infirm, sexual offenders, foreigners, women, wayward police and two small rival groups. Some of these, like sexual offenders, rivals and wayward police must be separated from the PCC population. Under the PCC's governance São Paulo's prison system, with its 195,695 inmates as of December 2012, has been transformed in moral terms, even as it has grown almost exponentially, by greater than 400 per cent since 1994.Footnote 40 Sexual violence has been abolished. No crack cocaine consumption is allowed, but other drug use is, and prisoners continue to maintain a system of self support in which all provisions, from hygiene products to bedding as well as clothes and food, are guaranteed by the organisation, paid for by the dues of members on the outside.
More importantly, though, the PCC has also become prominent in what were the most historically violent parts of the city. In an undeadly symbiotic turn, districts like Jardim Angela, Cidade Tiradentes, Brasilandia and Campo Limpo, as well as ‘São Gotardo’ where I did research in 2006 and 2011, saw unprecedented change. From 2000 to 2004, these communities, and many similar, saw their homicide rates drop virtually overnight, by up to 90 per cent in some cases.Footnote 41
In its urban form, the PCC calls this new system of governance a ‘Peace among criminals', which it derives from an ‘ethic of crime’.Footnote 42 As I examine elsewhere, this ‘ethic’ is surprisingly monopolistic, top-down and centralised. Maps of the city reflect the spatial coordinates of this dynamic, highlighting two distinct homicide patterns. Central areas and wealthy neighbourhoods have higher concentrations of police, more or less reliable provision of public services and historically strong state-society relations. Most distant districts urbanised informally, with most houses having been self-built, bit-by-bit, by their owners. Historically, basic infrastructure, public services such as hospitals and transportation, and codified urban planning only came to these districts following civil society mobilisation and public demands from residents.Footnote 43 In terms of security though, these areas have typically been the subject of particularly violent, reactionary and sometimes extrajudicial policing. We can see that these two also reflect distinct patterns in homicide rates. Through the 1990s, districts on the urban periphery inflated the city's homicide rate, while those nearer the city centre had rates that sometimes came close to those of northern countries.
These urban periphery neighbourhoods are precisely the kinds of districts where the PCC has found a home. Just as the PCC took control over the terms of violence inside the prison system, it did the same outside, enforcing where, how, upon whom and why certain people can, and cannot, be robbed, beaten or killed. Occasionally details of this system reach the media, but often as material easily labelled as salacious and exaggerated. This is because what is reported suggests something of a scale that seems unbelievable. The PCC uses a system of tribunals as a means to adjudicate crimes that are not allowed to occur according to the rules, ranging from minor infractions such as non-payment of membership dues to homicide without the authorisation of the organisation. They are set up in ways that imitate the formal system, complete with defendants, accused, ‘legal’ representation, witnesses for both parties, and jurors, involving over 50 people.
The tribunals follow a process, leading from whether or not a given incident occurred, whether to punish, what a just punishment is given the circumstance, and what the consequences will be for those involved, on both sides. They are coordinated by the PCC's local leadership, and are overseen via mobile phone by members in the prison system who hold a certain position, known as a disciplina, a ‘disciplinarian’. After all, one thing nearly all PCC members have in common is a personal encounter with the injustice of the formal justice system. Justice, they seem to suggest, has not failed because of the bureaucratic configuration of the formal legal system, but because of its practices.
Descriptions of different tribunals are usually eerily similar. The spatial locations are also somewhat unsurprising. They occur all around the city of São Paulo, and in the larger cities of the interior of the state. Commonly, though, reports indicate that they take place in poor neighbourhoods and spaces of informal urbanisation where social ties between neighbours are more prominent. To date there are no known cases of tribunals occurring in wealthier parts of the city.
An analysis of tribunals demonstrates what the PCC's bureaucratic documents may not. Namely, that the process of meting out ‘justice’ is a complex undertaking involving procedure, time and diligence. It also does not illustrate that this process requires communicating with members in distant prisons who hold positions akin to judges. Nor does it show that though at least ten PCC members were involved in deciding the circumstances of Santos, the introductory case, there may have been many other community members also present or who participated in one way or another. Most notably, though, that document does not show how or why the PCC justifies killing someone, and why, and how much more often, they do not.
During a return trip to the city in 2014, I stayed overnight at a hotel in Guarulhos. On my arrival at the hotel, I picked up a local newspaper. On the cover was a sub-headline about police breaking up a tribunal do crime, a ‘crime tribunal’, in which five people were being tried by the PCC for allegedly raping a local woman. But if the first 50 words were on the newspaper's cover, the rest was buried at the back. The reason for this is that such stories have become commonplace. Tens if not hundreds of examples of such cases have been reported, and are available on the Internet as YouTube videos, newspaper excerpts and blog posts. Though buried near the back of the newspaper this case seemed an emblematic example:
The intelligence service of the Military Police has arrested 17 people suspected of involvement in a ‘crime tribunal’, an informal court organised by members of a criminal faction to judge people accused of serious crimes in the community. According to the police, nearly 40 people were present, though most managed to get away when police arrived with squad cars. The incident occurred on Thursday afternoon, near Liberty Hill in Guarulhos.
Police suspect that up to 25 members of the Primeiro Comando da Capital (PCC) criminal group participated, the same group that organised the largest group of prison riots in the country in 2001. Information provided by the police states that five people were being tried for their involvement in the rape of a local girl in the region.
The 16 men and one women detained were taken to the 7th precinct, where they have been subjected to criminal record checks. One of them, according to the police, has already been identified as a member of the PCC. Police fear, though, that most will be released because of an absence of evidence since none was armed and none is willing to provide information about the incident.Footnote 44
This incident is much the same as the tribunals described in detail by Brazilian scholars, including Gabriel FeltranFootnote 45 and of those described to me by interlocutors within the public security system. One prosecutor explained to me that they once had followed one such tribunal live via wiretap. It lasted more than 24 hours in total, with breaks for meals, coffee and a night's sleep in between. Having authority only to tap the one telephone line, they knew only the physical location of the disciplinarian, in a prison west of the city, but could not figure out where the tribunal was taking place. They listened until the conclusion, when the disciplinarian ruled that the ‘defendant’ was sentenced to a beating from the neck down for having stolen from a woman in PCC territory.
That the PCC is behind many punishments and homicides should surprise no one. Take this brief description of a homicide from a newspaper report in 2011:
A 27-year-old man was found dead the night before last in Campinhas. The body of labourer Ronaldo Silva was located in a barren field on Miguel Santos Road, in Parque Sol. He was found by his family at 7 pm.
When police arrived, family members explained that he had been missing since Saturday afternoon. The incident was handled by detectives from the investigations unit.
The family does not believe that Silva was killed to level a score, though he was an occasional drug user. The victim was a resident of Jardim das Flores. Police will continue to investigate the case.
A different interpretation of this same case exists, and it is in the hands of those responsible for deciding whether Ronaldo Silva should have been killed. The outcome of their decision is filed as one of many PCC records of punishment, dated the same month as the newspaper report. It follows:
Nama [sic]: Ronaldo da Silva
Membership number: 999.999
Nickname: Copperhead
Neighbourhood: Campinhas
Place of baptism: CDP Pinheiros
Godfathers: MC David, Little John, Néné
Date of baptism: 28/10/2010
Date of exit: 04/12/2010
Position: Moneyman of Campinhas
Date of death:
Motive: diverted funds, hand in the pot
Amount of debt: Reais$ 1,100 in crack cocaine, March 2011
If tribunals often result in severe punishment, they are also somewhat exceptional in terms of how often they must occur. Other seized PCC documents show the complexity and scale of the PCC's organisation and system of punishment to a much larger degree. These include everything from membership rolls, Excel spreadsheets of marijuana, crack and cocaine sales by area code by week, the rules of the organisation's ‘gun library’, and an outline of a new fund to help members get ‘back on their feet’ following release from prison with a US$ 2,500 equivalent loan and access to a gun. These documents outline much more clearly the structure of punishment for members and include page after page of smaller-scale punishment records. These documents shed a great deal of light on how routine low-level tribunals occur for small offences among members that entail much less severe punishment.
How and why the PCC decides to punish is evident in some of the brief descriptions of cases contained in their paperwork, which illustrate a variety of small-scale infractions and responses. Most such punishments are for non-payment of drug debts. On the first non-payment, members are suspended for 15 days and may not obtain more drugs. On the second non-payment, or if the debt is not paid, that suspension is extended to 90 days. If the debt remains unpaid, the individual is expelled from the organisation. Alternatively, the organisation sometimes downgrades members to companheiro status, in which they cannot obtain or sell drugs, and risk marginalisation if they enter the prison system. Examples of these gradations of punishment are common in the documents:
He was removed from the organisation because of a debt on 14/09/2010, is in the neighbourhood and has been on the line with our disciplinarian brothers in the regional leadership, MC Luke and Pedrinho, and said that he was having difficulties and was looking at ways of paying the debt.
He was suspended because of this debt and from there onward has never been seen, and has disappeared from the neighbourhood.
He was removed on 10/06/2010 and is in an unknown prison for expired private debt.
He had disappeared for ten months and then was in the can. He is now in jail in Saracoba, and was removed for turning his back on the family (abandonment). He has now been downgraded to affiliate because of this debt from selling cocaine.
He took two boxes of crack cocaine in 9/11 and has not been seen since 30/11. He is out of sync and has not sought the finance people of the region to settle his debts.
He was removed on 18/09/2010 and is in the neighbourhood. He got on the line with the regional disciplinarians and said that he had paid his debt with Iron. He was responsible for finances of region while Iron was offline in prison.
He was removed because of his debts on 11/12/2010, has disappeared from the neighbourhood and the brothers from finance in the region have not been able to find him to communicate the message.
He was removed. Brother Blackey sent a message that he is related to him and says he is in prison, in solitary at P5 General.
He was removed on 14/09/2010 and is imprisoned in CDP Magnum.
His debt is not dated as it is from old times and the area at the time did not have control of the spreadsheets like we have today.
Instances like this go on and on, with each reflecting tight control over access to resources and a format of punishment when individuals do not conform. But they also show a certain morality in their leniency such as when an individual has made an effort to show that he was trying to settle a debt or has reached out favourably to the debt collectors.
The PCC In Police Practice
For homicide detectives and other police, the sway of the PCC is unavoidable, both on the job and off. The organisation's system of governance is uncompromisingly clear in seized documents, homicide investigations, witness, victim, and suspect testimonies, or the absence of them, as well as the everyday lived experiences of the street-level police who come from poor communities themselves and live alongside (and sometimes under) the rules of the organisation. Police strategies for coexistence with the PCC range from looking the other way, to disguising the fact that they are police officers, leaving gun, badge and uniform at the station, assuming other public personalities, to coexisting in a tedious and largely unspoken negotiation of identities.
A major part of the problem for the police is now the scale and complexity of this other system. When bodies come to be, an alternative architecture of violence and authority is often implied in the case, whether by the characteristics of the victim, the location, the pattern of occurrence or the near absolute impossibility of getting someone local to say something, anything, about the case in question. What has become routine, as a result, is a kind of resignation to the larger problem at hand. This was routinely evident in my field notes:
You could smell it when we drove past. Samuel, a police detective, stops the Blazer. There is a truck there from the coroner's office that will take the body after we're done. We get out and chat with the other detectives from the other Blazer that has accompanied us all day. There are also some Military Police there who have been guarding the scene. It isn't clear who they would be guarding it from though. The last few houses were a couple of kilometres back. There is no traffic on this dirt road. So little, in fact, that anyone who drives by would probably be noticeable.
Walking down the road a little bit, we come to a clearing in the trees. Descending from the side of the road is wet embankment. It is strewn with bits of garbage, metal refuse, mud and plastic. At the bottom is a burnt out but relatively new model car resting in a small creek bed. Somewhere down there is a decomposing body. No one can see it. Someone strings up a rope to help get down. Of the ten police from the homicide division, plus myself, only two of us decide to go down. Someone passes me some rubber gloves. I grab the rope and step down the wet slope.
We were basically on top of them when we realised it. Not one but two burnt bodies, covered in beetles and decomposing. Up close you could make out much more. Covering them, particularly around their heads and upper bodies, were what looked like wires, rusted and frayed as they were. ‘A man and a woman’, the police officer beside me said. (Pointing) ‘The torso on this one is much smaller’. It didn't seem so obvious to me. The bodies kind of blended into the mud, their burnt dark earthy tones made similar over time. Somehow, the smell wasn't as bad this close to the actual bodies. But in the absence of the smell, the sight of the two damp, partly burnt and decomposing bodies was a shocking enough replacement.
The car was down the hill a bit further. It was empty. No seats, no engine, no tail-lights nor a windscreen. The police up above surmised about what went over the hill first, the bodies or the car. Without a motor, the car would have had no way to get to this remote place without a rig towing it. If the motor and everything else were removed after the deed, it would have been a heck of a job to get it up the hill. If the car had come after the bodies, it would have run right over them, likely spreading the remains all over the place. It must have been there before, they figured, since people dump stuff like this all the time. They concluded that the car, though burnt and not identifiable apart from brand and model, was unrelated to the bodies. Not that any one of them even went near the car to look for a serial number on the dash or anywhere else.
For the police, the bodies and the scene gave few immediate clues. Plus, it stank. There was an air of resignation and impatience, like the bodies had been there for so long that nothing obvious would come of this. The detectives in the regional squad would pick this up and be better suited to deal with it once the autopsy report comes back in two months' time.
But from up closer, you could see a lot. The wires were virtually all parallel and uniform, curving slightly and running under and over both of the bodies. It was tyre belting. The rubber had burnt away entirely, leaving only hoops of now-rusting tyre belting behind. Those tyres would have burnt hot for hours – if not days – with thick black smoke. The presence of tyres was more than a curiosity. These two people perished in what is known as micro-ondas, the ‘microwave’. They died after someone forced two or three tyres over their shoulders, doused them with some kind of accelerant and set them on fire. They would have been burnt alive while standing. This is a form of particularly horrific and infamous killing that is used both as a way to create a horrible amount of pain for those suffering it, but also to emotionally shock anyone witnessing or picturing it in their mind's eye. Less well known but not unheard of for the PCC in São Paulo, the ‘micro-ondas’ is much more commonly used by drug trafficking groups in Rio de Janeiro. It seemed to me, in this case, and despite my obvious naivetés, to be a big loose end.
The police up above, who by some strange effect were getting a much stronger smell, started prodding us to go. Someone tossed down two body bags. One of the men from the coroner's office came down. I grabbed the rope and scrambled up. Someone reached down to help me take the final step up. I took off the gloves. ‘They're falling apart’ moaned one of the coroner's workers as he tried to pick up part of the body. They laid out one bag. With the body language of revulsion, he grasped one of the bodies by the spine, picked it up and tossed it on the bag. The arm, head and the all the rest from the first body came flying in soon after.
The regularity of the PCC's control over homicide has meant that their system of governance is often taken for granted or assumed by homicide detectives themselves. In fact, as has been argued more fully elsewhere, the PCC is behind what is the ‘normal homicide’ in São Paulo. This was routinely evident when I rode around to homicide scenes with detectives. This case, like many others, reflects a world in which police see justice in homicides. Police tend to believe that those who are killed have failed not just the formal legal system, but also the laws of the PCC. In moments like these, where the circumstances of violent death are obvious but deemed unimportant, the moral outlooks of the police and the PCC align.
But if a case tells us that the PCC can inflate the homicide statistics, how do we know if the PCC is also limiting the number of bodies? Sometimes the influence of the PCC is acutely obvious, walking directly into the homicide division, as I saw at other times during fieldwork. One such case occurred early in 2012, after a call came in to the homicide division. The body of a young woman whom I will call Gabriella had been found dead on the north side of the city. She was visibly pregnant and was found naked in the bushes at the edge a road at the very beginning of the urban sprawl. She was found with an ID card, and the initial report came with some description about her last known whereabouts. Residents had reported that she was addicted to crack cocaine. Her boyfriend had also been recently arrested.
Detectives discounted many possibilities: Her boyfriend was out for a holiday break and had arranged for her to be killed. A sexual predator was at loose in the community, as evidenced by a handful of reported rapes. Considering the nature of the violence and the difficult imagery that accompanied a death like this, there was some concern that it might get a great deal of media attention. All the stops were pulled out. The chief detectives ordered examinations to be done on the body, from cavity examinations for semen, to subcutaneous examinations for skin and blood, and tests for blood, drugs and alcohol.Footnote 46 The other detectives were using all of the databases, motor vehicle, criminal record, national identity, among others, for more details about the husband, his known accomplices, and for more about this potential sexual predator. At the end of a long day of fieldwork myself, I decided to leave, making a mental note to ask them about the case next time I saw them.
Days later I saw the same detective again and asked what had come of the case. He explained: arriving at the scene they had tried desperately to solve the crime, canvassing the area, talking to residents, seeking out known sex offenders and digging into recent sexual crimes in the area. Their efforts had been fruitless and none of their possible theories, suspects or potential witnesses had panned out. With few leads and even fewer witnesses the case was passed on to secondary investigators, which meant the chances of solving it were slim.
And yet out of the dearth of answers came a conclusion. Some days later, the supposed offender walked right into the station and gave himself up. The same detective told me the story: the man claimed that Gabriella's death was not a rape, but, rather, that she had died when a consensual sex-for-crack exchange went bad. They had gone to the bushes to consummate the ‘deal’. After the sex, she had collapsed and became irresponsive. Scared about what to do, he fled, leaving her to die naked in the scrub by the side of the road. He had not been around when the detectives came looking for answers, but he had heard all about it.
This was not his problem. People in the community were talking about him and what he had done. There was hearsay that he had raped Gabriella, and it spread throughout the community quickly. A formal PCC tribunal was organised and took place the man explained. No one in the community disputed that the man giving himself up was involved in Gabriella's death. That much was clear. He himself admitted that there was intercourse. Her death was not the problem. Rather, the detective intoned, the case hinged on one crux: for the PCC the most important piece of information, also the hardest to discern, was whether the sex had been consensual or not.
No one could say for sure what the problem was. For the PCC there was no clear evidence that he had been violent, taking advantage of her on a dark street. There was of course no examination of the body, which had been taken away to the medical examiner. Nor was there evidence that she had consented because she wanted crack, despite her habit and reputation as a user. After a great deal of deliberation but no unanimity about who was at fault, they came to a decision about how to settle the problem.
The local PCC affiliates decided to give him a choice: either, walk in to the police station, take responsibility for Gabriella's death and go to prison, or be killed. According to the man, if he could have proved that the incident was not a rape, but instead that the whole mess was just about a girl, who happened to be pregnant, who was willing to pay for crack with sex, then they would have let him go. Failing to do so, he took the only real option open to him.
The irony, of course, is that this man would only leave the PCC's moral space for days, if not hours. After writing up the report, giving his statement and completing other formalities in the police station, he would go straight into a prison system controlled almost in its entirety by the PCC's moral regime. There he would be governed by the same system, now much more obviously embedded within the state, which forced him to surrender to it in the first place. In many ways, it seemed like the PCC had decided that the suspect deserved jail time, not capital punishment, and used the detectives at the police station to move him into what is essentially their own regime. Read this way, police detectives seem like little more than bureaucratic cogs in the PCC machinery.
More nuanced still, is the fact that this man was likely judged, at least in part, from inside the formal system. Tribunals occur when PCC members responsible for punishment connect from inside prison via mobile phone with local PCC members at tribunals to judge a specific case. On the one hand, this moral regime is contingent on the relative safety and protective confines of those in the prison system. On the other, it hinges on those outside of the prison system being subordinated to a leadership distant from the location of the ‘crime’.
The outcome for this man and the case was entirely a result of PCC governance. And yet, he became a vital thing for a formal system so auspicious in its inability to ascertain one single meaningful piece of information or evidence about the case. There was not a single person willing to go on the record for the police. And yet they got credit for solving the case. This man was now a representation of the epitome of good police work in contemporary times: a statistic of a closed murder case. More valuable even was the near certainty that the man would be found guilty, despite the fact that there was no evidence and it was only because he himself admitted to doing the crime. But perhaps most importantly, the man was one fewer dead person in the city.
The existence of PCC moralities inside the formal justice system also ensures that those believed to be on the wrong side of the PCC's law will rarely, if ever, go unpunished. This was put on public display in 2012, following a particularly heinous crime that shocked the city. During a home robbery in a poor community on the west side of the city, five men beat up a Bolivian family for not giving them enough money. In the tumult a young boy named Brayan started to cry, begging the men not to hurt his mother. In short order one of the men shot Brayan point blank, killing him beside his mother. Taking what money had been given to them already the five men then fled. For weeks the case captured public attention. The Bolivian ambassador spoke out and Bolivians and Brazilians held street rallies calling for justice and an end to impunity. Television personalities hosted long diatribes railing against the failure of public security in the city. The sense of disgust was so thick that even police themselves were wrapped up in it.
Sometime after the event I spoke to some of the homicide detectives. They told me that there was intense pressure from upstairs to resolve the case, rather than the detectives themselves feeling any less desire than their bosses to make sure the five men got their due. Their worry was not so much that the five men would get away with murder. Mostly everyone that I spoke to understood that the resolution to this case would come. Of that there was little doubt. But everyone knew it would come after a foot race. At hand was who would find the men first, the PCC or police. A standing order was issued to homicide detectives to notify their superiors of any bodies found in or around the district where the robbery took place. But who would get credit for finding the five men? Credit for the resolution would go either to police or the PCC, or, perhaps, both. Either way, many detectives said, the five men would get what was coming to them. The first two were found less than ten days later, but neither could be identified without fingerprint analysis, which would take a month. Both had been shot in the head multiple times, their bodies found separately but near each other on the outskirts of a neighbourhood not far from the international airport. As police awaited word on the identities, they found two of the other men alive. Broadcast live, and incessantly afterwards, both men were mobbed and nearly lynched by the public during their arrival at the precinct. They were arrested, charged and eventually forwarded to the pre-trial detention centre, an institution that was known to everyone in the public security system to be controlled by the PCC.
Within hours both were dead. According to police, both had been forced to drink a lethal cocktail of cocaine, creatine, water and Viagra, colloquially known in the prison system as ‘Gatorade’. One of the city's two major newspapers reported that when the two bodies were removed from the jail the prison staff were told they did not need to look for the other two any longer. Both had been dealt with. A week later the fingerprint results for the first two came back positive.
Multiple newspaper reports related that the fifth, a minor, was being held in a juvenile detention centre. He would apparently survive. In the aftermath of the four dead bodies ultimately paying the price for the death of Brayan, hundreds of media broadcasts celebrated the resolution as just and merited. There seemed exceptionally little debate about the ‘good’ of the resolution.
The race between police and the PCC to find the five men seemed a tie. Even so, the message of the PCC came across much more clearly. Newspapers posted ‘infographics’ of the timelines of the killings, linking them directly to the PCC.Footnote 47 One of São Paulo's two large-circulation daily newspapers published an editorial that concluded:
Up until now, the Homicide Department has not attributed the death of the four suspects in Brayan's death to the PCC. ‘We don't have details, but no hypothesis will be dismissed’, said detective Itagiba Franco. What is ironic is that of the four dead criminals, one was serving time in a penitentiary, and, after being allowed to go on temporary leave to spend Mother's Day in freedom, he never came back. He was a high-risk criminal that was given partial parole [regime da progressão] – a benefit that even the PCC found to be absurd in this case.Footnote 48
Much more noteworthy and clear was a message from the PCC leadership that leaked out of the prison system. Dated three days after Brayan's death, the message made it unequivocally clear how the organisation would treat killings of children. The message read:
The leadership [sintonia final] informs everyone that from this day on, no violence against children will be tolerated. Anyone who commits or is involved in such violence will be found and eliminated. We are, above all, human beings. We have family and we know our feelings for those around us. We cannot respect someone who commits this type of crime, which is totally contrary to the basic principles of any honourable criminal. The ends of these actions do not justify the means. We leave this in the hands of each regional leadership. Via this note, it is decided that there will be zero tolerance for committing this crime. It is worth remembering. The comando is the voice of crime, and we will not allow it. Thanks to all. Signed, Sintonia Final, the leadership.Footnote 49
From a distant prison 612 kilometres away, the leadership of the organisation mobilised its members in a massive man hunt for the young men involved. In a labyrinthine urban region of 20 million people, they found them in short order. More savvy, flexible and smarter that the formal justice system, and certainly (but not necessarily) more violent, the message was unambiguous: if you do not follow the rules and morality of this system, formally affiliated or not, you will not escape punishment.
Conclusion
Andreas and Greenhill have shown some of the problematic ways that statistics are produced, politicised and appropriated in the public sphere and in scholarly publication.Footnote 50 In the scholarship on violence in Latin America there is already a great deal of tentativeness towards the use of homicide and other crime statistics because of a fear of misrepresentation of statistics by public authorities, as well as a lack of trust in the validity and categorical fluidity of such data. I have sought to show here that there are even larger concerns at hand when it comes to the use of statistics. These concerns relate directly to modes of governance in contemporary times. In cities like São Paulo, there is little reason to believe that sovereign authority insures homicide statistics. The assumption that states control the street-level terms of policy when it comes to questions of life and death does not hold, empirically. Rather, the empirical practice of homicide regulation indicates deep influences on the part of an organised crime group that is ordering the conditions of violence in the most historically violent parts of the city. Who dies, how and why is definitively patterned, reflecting an alternative social and ‘legal’ order. But this order does not stand in opposition to the formal legal order. Rather, they are connected and, in many ways, nearly impossible to disaggregate, both in moral terms and in the everyday workings of the criminal justice system. As it turns out, ‘violence reduction’ is a PCC policy.
This practice of homicide regulation raises questions about the homicide decline in São Paulo. These concerns are reflected throughout Latin America. Medellín, Colombia saw a marked decline from 381 per 100,000 to less than 20 per cent of that from 2002 to 2007.Footnote 51 Now once again on a steep rise, the influence of a paramilitary kingpin's micro-monopoly over previously violent neighbourhoods during the decline is well known.Footnote 52 El Salvador is currently experiencing a reversal of a dramatic decline in its homicide rate, the product, some say, of a truce and its implosion between the two mara gangs.Footnote 53 That these kinds of groups, organised crime, paramilitaries, gangs, are both protagonists of violence and potential regulators of it is a fact not well considered in the scholarship, particularly when it comes to common assumptions about how to measure ‘security’, the state's capacity to govern it or how to reform police organisations.
Statistics are produced, but they are also productive. When Hannah Arendt wrote about the bureaucratic manipulation of reality by ‘statistical treatments’, she cautioned against the world they could create. The ability to make and control statistics as a legitimate source and foundation for knowledge is premised upon a more or less unquestioned assumption of state-enshrined power. This allows for the construction of an alternative reality towards a given end, in Arendt's case of a nationalist and genocidal Nazi regime. Today, state centric assumptions born out of histories of a particular northern state formation foster a more or less untested consensus that the state is the only political authority capable of producing a certain category of statistic. This category relates to the primary function of the state itself: the regulation and legitimate monopolisation of violence. Yet under conditions where empirically existing and conceptual states are believed indistinct, we bestow upon today's states a double gift: the legitimacy to produce statistics, and the ability to use them to create what they need or desire. As we are evermore submersed in an era of big data-as-solution, none of the deeply problematic underpinnings of how (even ‘small’) data is used, and by whom, have been adequately addressed. What this means for the longevity of local regulatory practices of life and death is very much in question.
It is clear that many states do not provide adequate security as a public service at the subnational and urban scale. There is little reason to make sweeping comparisons or prescriptions based on the experiences and histories of state formation from northern locations, where the configuration of political power could scarcely be more dissimilar. If the nature of homicide, violent death and empirical structures of governance in São Paulo can tell us anything, it is that we need a research agenda that is both methodologically sensitive to the everyday circumstances of the region, but also, that is willing to look to other regions of the South where similar questions abide. It is by looking at those lived experiences of security and democracy, especially in cities, that may provide us with new ways of thinking about the practices of governance in contemporary times.
The dissonance between what we presume homicide statistics imply and what they may really mean for lost lives has wide-ranging consequences from policy, to social science knowledge, and the everyday functioning of fledgling democracies. Sometimes the rationale behind volatile homicide rates is debated publicly, with news organisations reporting extensively on the role of organised crime and politicians taking public positions, to one effect or another. Other times the influence of organised criminal groups is shut out of public discussion, with media groups shying away, communities reticent to speak out for fear of retribution and elected officials loath to do anything but try to sweep the existence of these groups under the rug. The problem, ultimately, is when the existence of these groups can be swept under the rug, and when doing so, while benefiting from the numbers, is in the interest of those in charge. Much more importantly, though, this story should raise many larger concerns about the state of security and safety in Latin America, and perhaps much of the world. Doing so does not mean doing away with statistics altogether, of course. Rather, it demands a particular sensitivity to the assumptions of such statistics, and how the use of this kind of data, given its assumptions and production, can be adequately separated from dominant understanding of what security should be – and how we must get there in the current historical moment defined by democracy, rights and universalist ethics.
