Brazil's Pacifying Police Units (Unidade de Policia Pacificadora - UPP) are not an unprecedented security model. Rio de Janeiro state has seen previous experiments in community policing, like one unit known as the Special Areas Policing Group, or the GPAE, established in 2000. Led by a police commander who had a degree in sociology and who handed out his cell phone number to favela residents, the GPAE had many of the same responsibilities currently handled by the UPPs. This included basic community policing duties like daily street patrols, manning a crime tip hotline and talking to residents about crime concerns. At the time, such an approach felt revolutionary in Rio de Janeiro, which has some of the highest police brutality rates in the world.

The GPAE faced its share of challenges. Unit officers still accepted bribes from drug traffickers, even though corruption was less widespread than in other police bodies. In some ways, the very mission of the GPAE contained contradictions. Authorities presented the community policing project as an alternative to the "shoot-to-kill" policy favored by the security forces in Rio de Janeiro. On the other hand, the superficial training given to officers ensured that the GPAE lacked the institutional support needed to be effective on a mass scale. The GPAE was treated as a short-term solution to crime and violence, even though the core of its mission -- changing the way police interacted with poor communities -- is a long-term mission.

The same questions currently face the GPAE's successor, the UPPs. Created in 2009, Rio de Janeiro's security secretary ambitiously described the police units as "a new tool to end conflicts," signaling the end of the "reign of the gun." The UPPs were intended to give many favela residents their first experience of a permanent police force stationed in their neighborhoods. Accompanying the UPP's arrival in many neighborhoods were cable, electricity and sanitation companies, intended to help the government's stated mission of re-establishing "control in territories long lost to militias and factions."

Now numbering 17 battalions in about 13 neighborhoods, the UPPs have been accompanied by dramatic drops in homicide and theft rates. This is similar to what happened when the GPAE was first deployed to Rio de Janeiro's suburbs. But crime rates haven't necesarily gone down because the security forces were successful in breaking up the city's criminal gangs. In the cases of both the UPP and the GPAE, pacifying forces only entered the favelas after the area had been "cleared" by the special forces unit known as the BOPE. The elite squad is basically responsible for purging the favelas of drug gangs, who then shift their operations into other neighborhoods.

Much of the city's violence is caused by drug gangs fighting each other over territory, or police engaging in shoot-outs with suspected criminals or innocent civilians. The BOPE's aggressive entry into the targeted favela, followed by the UPP, forces drug traffickers to assume a much lower profile. As a result, armed confrontations between rival gangs have become much more rare. Police use of lethal force has also dropped in the neighborhoods where UPP units are based.

So far, these are among the UPP's top accomplishments. Another encouraging sign is that the program has been well received by residents, and some neighborhoods without UPP units -- including Rocinha, Rio's biggest favela -- have started lobbying for one.

But the program is beginning to show many of the same cracks that afflicted the previous GPAE program. UPP officers reportedly lack resources like decent headquarters, bullet-proof vests and weapons. One UPP unit commander was recently fired for accepting bribes. Even the governor of Rio, who made the UPPs a central message of his 2010 re-election campaign, admitted that the program was showing "weaknesses."

The investigation and punishment of corrupt elements in the UPP could be read as a sign of strength. The same goes for the government's admission that the program is in need of re-evaluation. But the corruption allegations also raises the question of how much the UPP has learned from the failures of the GPAE. High salaries are supposed to deter UPP officers from accepting bribes, but it seems some of the old institutional patterns are lingering on.

The fear is that the UPP will turn out to be just another community policing experiment, limited in terms of scope and benefits. Critics point to the fact that the first neighborhoods which saw UPP units are relatively affluent favelas in the south, while many western neighborhoods remain battlefields. Many of the areas with UPPs are those that must urgently need to be cleaned up for the World Cup and Olympics, lying near sports grounds or highways.

The fact that the UPP project has been criticized for strategically selecting which favelas are high-priority areas -- and it makes sense in economic, political and strategic terms that these should be the southern neighborhoods -- is a symptom of the high expectations generated by the program. In a 2009 U.S. State Department cable published by WikiLeaks, the state security secretary is quoted as saying that only 10 to 12 favelas out of 1,000 are critical for the UPP program. It is possible to argue that this approach neglects security in other violence-racked neighborhoods in the city's western zone. But given the UPP's problem of limited resources, the government's choice to be selective makes sense.

If the UPP's goal is defined as promoting safer environments in select poor communities, the project has already succeeded. The question is whether the UPP will achieve the more ambitious objectives set out by the government. This includes changing the fundamental relationship between the police forces and favelas. It means creating a more enlightened police force which has the necessary institutional support to discourage corruption. It means considering the successes and failures of prior community policing initiatives like the GPAE, and learning from them. If the UPPs can survive changes in governance, and the Olympics and World Cup fever, Rio de Janeiro may yet be able to boast a policing model that the rest of Latin America would do well to imitate.