Skip to main content

Dear Governor Sérgio Cabral,

Since we last met in December 2009 to discuss the findings of our report “Lethal Force,” Human Rights Watch has followed your administration’s efforts to improve policing in Rio de Janeiro with great interest. On several recent visits to Rio state, we have conducted in-depth interviews with state officials, prosecutors, police officers, public security experts, civil society organizations, and favela residents. I am writing now to share our observations based on these visits and to recommend crucial steps we believe you should take to improve public security and curb ongoing police abuse in your state.

Human Rights Watch is encouraged that official statistics point to a decline in killings by police in Rio state over the past two years (from 855 cases in 2010 to 524 in 2011), as well as in the overall number of intentional homicides (from 4767 cases in 2010 to 4286 cases in 2011). While some public security experts and academics have questioned the reliability of these statistics, most agree that there has been a significant decrease in both categories.[1]

We have also been encouraged by your government’s pursuit of two innovative policies aimed at improving public security in the state: the System of Goals and Results Tracking and the Pacifying Police Units (UPPs).  While it is too early to determine the long-term impact of these initiatives, we believe that both have the potential to promote more effective policing and reduce police abuses in Rio.

At the same time, however, we are concerned that the number of police killings in Rio—despite the recent decrease—remains much too high (averaging almost two deaths per day since 2010).  And we are concerned that there appears to have been little progress in addressing the critical problems—documented in “Lethal Force”—that tend to perpetuate unlawful violence by police. These problems include police cover-ups, deficient investigations, and impunity for officers implicated in human rights crimes.

Incentives for Better Policing

A very promising public security initiative you have pursued is the System of Goals and Results Tracking (Sistema de Metas e Acompanhamento de Resultados). As you know, this program, launched in 2009, has two key components. First, public security officials track homicide and other crime rates in various policing districts, meet with police commanders to identify challenges and design more effective law enforcement strategies. Second, officials set crime reduction targets for each policing district and award bonuses of R$3000 to all officers in districts that meet their targets for each semester. Districts that make the most progress are awarded prizes of up to R$9000. A total of R$40.5 million were distributed among approximately 9,000 police officers in April 2012 alone. 

We were especially pleased when, in January 2011, your administration expanded the tracking and incentive program to cover killings by police. This variable is now included with other forms of intentional homicide in a single policing target, entitled “violent lethality.” Because all intentional homicides—whether committed by drug gangs, petty criminals or fellow officers—now affect the same policing target, police officers have a greater incentive to prevent them.

It is important that police do not misconstrue competent job performance and respect for human rights as requiring additional pay. However, this system of goals and rewards can create a powerful set of incentives for police officers to work together to improve public security. In particular, the program may improve collaboration between military and civilian police, which in the past have failed to coordinate their operations adequately and share critical information. 

Pacifying Police Units (UPPs)

Another policy that can potentially help to reduce both common crime and police abuse is the creation of Pacifying Police Units (UPPs), police units in low-income communities meant to re-establish territorial control and the rule of law, as well as ease availability and access to public services. UPPs appear to have helped reduce violence in some communities, according to favela residents and public security experts we interviewed. 

Formidable challenges lie ahead, however. The most obvious is how to sustain the current UPPs and expand the UPP model throughout the capital city and state. There are more than 1000 low-income communities in Rio state, according to official figures[2], yet only 23 UPPs have so far been established in approximately 100 communities, most of which are located in the wealthiest areas of Rio or near the Maracanã Stadium. Meanwhile hundreds of communities in other parts of the state remain under the rule of criminal gangs or armed militias, and there are growing concerns that violence may be migrating from Rio’s “pacified” favelas to other cities such as Niterói.

We understand that one obstacle to expanding the UPP project more rapidly is the lack of qualified police officers, though your administration has taken steps to enable more candidates to pass the certification exams. Another is apparent dissatisfaction of many UPP police officers, who would prefer to work in traditional battalions, according to research done by the Centro de Estudos de Segurança e Cidadania (CESEC) of the Cândido Mendes University.[3] This dissatisfaction could undermine the job performance of these officers and the long-term viability of the UPPs.

There have also been allegations of excessive use of force by some UPP policemen and possible corruption among UPP officers and criminal gangs. In September, three officers from the Santa Teresa UPP were detained for allegedly turning a blind eye to drug trafficking activities in exchange for thousands of reais per month.  More recently, a former São Carlos UPP commander was arrested on February 16 on the basis of similar corruption allegations. We commend the investigation of police officers implicated in these corruption schemes. The long-term viability of the UPP program hinges on the steadfast commitment and transparency of its officers. 

Lack of Progress on Accountability

Despite these two promising initiatives, there has been limited progress in addressing one of Rio de Janeiro’s most critical challenges in the area of public security: ongoing obstacles to accountability for police abuses.

“Lethal Force” documented how legitimate efforts to curb violent crime in Rio state were undercut by police who misreported executions as “resistance killings” and investigators who routinely failed to conduct proper inquiries into these cases, violating basic tenets of homicide investigation. Police investigators often did not seek out and interview non-police eyewitnesses, adequately question police officers involved, or conduct basic forensic tests such as crime scene analysis. Together, these failures resulted in criminal inquiries that did not clarify events or provide sufficient evidence for prosecutions in cases of wrongdoing.

The misreporting and inadequate investigation of police killings remains a very serious problem today, according to local prosecutors and public security experts. The problem was evident last year in the widely-publicized “Case of Juan,” involving the 11-year-old Juan de Moraes, who disappeared on June 20 after an incident in the Danon favela in which three other persons were shot by military police, one fatally. The police reported the incident as a “shootout” with “armed assailants”, yet for a week, no serious investigation was undertaken to determine what had taken place. It was only after the case received extensive media attention that civil police investigators analyzed the scene of the shooting and sought out eyewitness testimony. Police investigators found Moraes’ DNA at the crime scene and other forensic evidence indicating that there had been no shoot-out. When they discovered Moraes’s body after 10 days by a river in the Baixada Fluminense, it was initially identified as a girl. DNA tests ultimately established that the body was in fact of Moraes.

Unfortunately, the “Case of Juan” was hardly an isolated incident. That same month, a street candy vendor, Diego Beliene, was shot to death by military police in the Salgueiro favela. The police reported the death as a “resistance killing,” claiming that Beliene was wounded during a shootout in the street. However, civil police investigators found forensic evidence and eyewitness testimony indicating that a police officer shot Beliene after he climbed over a wall into a property that had been occupied by the police. Police subsequently held Beliene in their custody for more than half an hour as he bled to death, refusing pleas by his family members to allow them inside to assist him. Police eventually took him to a nearby hospital in a plastic body-bag. He was dead upon arrival.

In a third case from June 2011, André Ferreira was shot in the back by a military police officer in the Pavão-Pavãozinho favela. The officer reported the incident as a “resistance killing” and claimed to have shot Ferreira in self-defense after hearing a gunshot. Yet witnesses and forensic evidence gathered by civil police investigators indicate that only one shot was fired.  Moreover, Ferreira’s autopsy report showed that he was struck by a bullet with a downward trajectory, contrary to the shooting police officer’s account that he had fired uphill at the victim.  

Other cases of misreporting by police include the shooting of two boys in the Pica-Pau favela in April 2011. Military police officers reported that they found the wounded boys (ages 15 and 16) with the help of locals, yet the boys later told civil police investigators that they had been trapped in a dead-end street and shot by one of the officers. After examining the case, civil police investigators concluded that the military police officers’ account of events was “impossible” given their close proximity to the boys at the time of the shooting.

In addition to misreporting events, another practice highlighted by the “Case of Juan”—and documented in “Lethal Force”—is the failure to preserve the scene of a potential crime by, among other things, removing deceased victims before a proper investigation can be carried out.  An example of this practice is the shooting death of Valdira de Souza Godinho in the Salgueiro favela in May 2010. Military police officers reported that they came upon her body after a shootout with drug traffickers. However, multiple witnesses told investigators that there had been no shootout and that police had shot Godinho. And the firefighter who ordered the removal of Godinho’s body admitted that she knew Godinho was already dead and failed to follow protocol requiring that when police encounter deceased individuals, they must preserve the scene of the incident and await the arrival of forensic experts.

Among the most disturbing cases we documented was the shooting of six people at a barber shop in the Nova Holanda favela on June 11, 2010. Military police reported that they had come across three victims after a shootout. However, forensic evidence collected by civil police investigators indicated that all the shots during the “shootout” had been fired by the police officers themselves. Prosecutors concluded that the victims were unarmed and that the officers had given false testimony. In addition, witnesses apparently saw police escort one man in handcuffs to a police vehicle. The man was walking at the time, according to the witnesses, yet it appears that he arrived at a hospital with a bullet wound in the head and was reported (incorrectly) as “dead” by police officers. Medics ultimately managed to save his life.   

In all the cases mentioned here, misconduct by the military police was exposed thanks to the work of civil police investigators. Unfortunately, however, according to prosecutors with whom we spoke, civil police investigations of cases involving police violence often suffer from serious shortcomings that make it difficult for prosecutors to determine criminal liability.  

The misreporting and investigatory failures are major factors contributing to widespread impunity in police killing cases, according to state prosecutors. While the precise scale of impunity is difficult to determine, the official data that is available lends support to the view that impunity in police abuse cases is the norm. For instance, from 1999 to March 31, 2012, the Rio Police Ombudsman’s Office recorded over 11,300 complaints against police officers concerning criminal conduct. These complaints generated only 43 criminal charges by state prosecutors and a paltry four convictions.

Moreover, in the majority of cases that Human Rights Watch reviewed in 2009 containing credible evidence of police extrajudicial execution, no police officers have been held accountable. For example, no one has been brought to justice in connection with the Complexo do Alemão police killings of 19 people on June 27, 2007, despite extensive evidence that multiple extrajudicial executions occurred, crime scene evidence was deliberately destroyed, and investigators negligently failed to request obvious forensics analyses.

We were encouraged when, in response to the “Case of Juan,” Civil Police Chief Martha Rocha acknowledged on July 7, 2011,that “it is time to learn from our mistakes” and issued instructions regarding the mandatory steps to be taken before registering cases as “resistance” killings.[4] These include immediately isolating and analyzing the crime scene, testing the guns used by police during the operation that led to the shooting, collecting testimony from eyewitnesses, and conducting separate interviews with all police officers involved in incident.[5] While it is important to note that Chief Rocha’s instructions reflect practices that are already mandated by Rio state law, we are hopeful that—with adequate monitoring and enforcement—they will lead to consistent compliance. 

Moving Forward

The most effective way to curb police abuse—and to ensure the full success of the important initiatives that your administration has undertaken—is to make certain that police officers who break the law are brought to justice.

Toward that end, a central recommendation we put forward in “Lethal Force” is that state prosecutors should conduct a thorough review of resistance killing cases from recent years, especially in areas with disproportionately high levels of police violence, to determine where there may have been unlawful use of force and other forms of misconduct and to ensure that the officers responsible are appropriately sanctioned. A second key recommendation is to create a permanent specialized prosecution unit within the state Attorney General’s Office dedicated to handling police killing cases going forward. Human Rights Watch discussed these proposals at length with prosecutors, including top officials in the Attorney General’s Office, who fully supported our ideas.

While implementation of these two recommendations would be the responsibility of the Attorney General’s Office, we believe that it is crucially important that you, as governor, use your position of leadership and authority to promote and actively support such efforts.

Your office has already taken an important step in this regard by entering an agreement with the attorney general to grant prosecutors unhindered access to the civil and military police databases. In addition to this agreement, there are several other steps we would urge you to pursue:

  • Establish a special unit of civil police investigators to assist prosecutors charged with reviewing “resistance” cases;
  • Ensure that Chief Rocha’s instructions on proper crime scene protocol in police killing cases are strictly enforced, and that officers who fail to do so are appropriately sanctioned; and
  • Require police officers to notify prosecutors of “resistance” killings immediately after they take place.

We believe that these recommendations are not only consistent with your other public security initiatives, but are also critically important for ensuring their success. It is worth noting that several of these measures have already been pursued in one part of your state – São Gonçalo – with promising results. 

Beginning in 2008, a state prosecutor, Paulo Roberto de Mello Cunha, worked with local police commanders to improve policing practices and accountability. Cunha made repeated presentations to police officers in which he made clear that he would actively prosecute cases of unlawful killings by police. He followed through on that commitment by filing at least 46 criminal charges against 90 officers. This collaboration culminated in a written agreement in 2009 in which local military police commanders and civil police precinct chiefs committed themselves to ensuring that “resistance” cases would be handled properly: military police would preserve all evidence at shooting sites and immediately notify the civil police, which would then carry out crime scene analysis. Police killings in São Gonçalo subsequently fell by almost 70%. And the intentional homicide rate did not go up, dispelling fears that increased accountability for police officers would lead to higher crime.[6]

Unfortunately, the brutal murder of Judge Patricia Acioli in São Gonçalo last August—in retaliation for her commitment to applying the law to police officers charged with homicide by prosecutor Cunha—has badly undermined this sense of progress. Yet her tragic death has also made crystal clear the urgent need for Rio to do more to curb abusive practices by those police officers who consider themselves to be above the law.

In closing, I would like to reiterate that Human Rights Watch is encouraged by your initiatives to improve the performance of Rio police forces. However, in the absence of a concerted and comprehensive effort to ensure full accountability for police abuses, we fear that the impact of those initiatives may be limited, unlawful police killings will continue, and Rio de Janeiro’s legitimate efforts to curb criminal violence will suffer.

Thank you for your attention to this important matter.

Jose Miguel Vivanco

CC: Cláudio Soares Lopes, Attorney General
CC: Eduardo da Costa Paes,  Mayor
CC: Erir Ribeiro Costa Silva, Commander of the Military Police
CC: José Mariano Beltrame, Secretary of Public Security 
CC: Leonardo de Souza Chaves, Deputy Human Rights Prosecutor
CC: Marcelo Freixo, President of the Human Rights and Citizenship Commission of the Rio de Janeiro Legislative Assembly
CC: Marta Rocha, Chief of the Civil Police
CC: Antônio Claret, Secretary of Social Assistance and Human Rights


[1]See for example, Daniel Cerqueira, “Mortes Violentas Não Esclarecidas e Impunidade no Rio de Janeiro,” October 2011.

[2]According to a census by the Brazilian Institute of Geography and Statistics (IBGE), there were 1,332 “subnormal settlements” (aglomerados subnormais) in the state of Rio de Janeiro in 2010, including “favelas, “invas, grotas, “baixadas, “comunidades, “vilas, “ressacas, “mocambos, and “palafitas”, among others.  

[3]Center for the Study of Security and Citizenship(CESEC), “Police Pacifying Units: What police officers think,”May 2011.

[4]“Após erros em caso Juan, policia vai mudar investigação de confrontos,” O Globo, July 9, 2011.

[5]Portaria PCERJ No.553, dated July 7, 2011, “Establishes the basic guidelines for police authorities to follow in investigations of ‘resistance killings’ and makes other determinations”.

[6]According to statistics of the Public Security Institute (ISP) of the Public Security Secretariat of Rio de Janeiro, there were 53 and 62 cases of resistance killings in 2007 and 2008, respectively, and 20 and 21 in 2009 and 2010. Intentional homicides numbered 440 in 2007 and 404 in 2010. 

Your tax deductible gift can help stop human rights violations and save lives around the world.

Region / Country