The 2019 Rio de Janeiro Armed Groups Map was an unprecedented pilot project carried out by 5 organisations with the objective of demonstrating the territorial reach of drug factions and militias in the state of Rio de Janeiro and at the same time providing journalists and researchers with a tool to better analyse and understand the living conditions of the Rio’s citizens and the impacts of public security choices. The project was inspired by the certainty that not It is a project that believes in the importance of quality and free information to transform the reality of citizens.
Territorial control is one of the historical and distinctive characteristics of the dynamics of armed groups in Rio de Janeiro since the 1970s, and an unavoidable variable not only for public security, but also important for other urban public policies, from transportation and housing to education and health. Over the years, groups have emerged, fragmented and grown. But despite it’s failure, the state’s response over the past 40 years has remained the same: confronting these groups through direct confrontation operated by the police, aligned with the international policy called War on Drugs. As a consequence, beyond its natural beauties, Rio has become known for violence and stray bullets. Here, control over space is disputed by guns, both by organized crime and by the state itself. This situation exposes the population at any time to the crossfire, and affects the supply of any and all public and private services, especially in the peripheries where armed groups are mostly concentrated. The most tragic effect of this scenario can be understand by the fact that a 6-year-old boy from a favela in Rio has a better chance of being shot than obtaining a college degree. For this reason, it is surprising that the mapping of armed territorial domain has not been carried out to date or that, when done, it has not become of public knowledge and access. The lack of a reliable historical map of territorial control of these groups – and the disputes between them – not only hinders the elaboration and implementation of public policies in Rio, but also hinders the economic development led by private initiative, since any investment made without this information incurs a high level of blind risk. The project lauched in last october adresses exactly this missing information, and envisions to use the prototype to build.
The project was carried out by 5 organizations with recognized experience in the area of public security (Fogo Cruzado, Pista News, Disque Denúncia, Geni/UFF and Nev/USP). The Disque Denúncia database was chosen as the primary source to classify the presence and control of armed groups over certain areas, due to its richness of details and territorial and temporal coverage. For the 2019 prototype, 37.883 anonymous denunciations were analyzed, which mentioned militias or drug trafficking, using natural processing language techniques. Of these, 10.206 were considered valid after a process of recognition of the groups mentioned and the type of activity mentioned. The valid denunciations were georeferenced, plotted on the map and processed from polygon maps, following statistical/mathematical criteria, to identify whether or not each polygon was controlled by the groups studied. The maps of polygons were constructed from the expertise of Pista News in mapping, slums and housing estates (informal geographic units, smaller than the official neighborhoods recognized by the city halls). The process resulted in a map that exposes the classification of each polygon identified according to the dominant group (ADA, CV, TCP or Militia) or as an area in dispute, according to data from 2019.
The same analysis was also done to evaluate the distribution of groups by neighborhoods (official classification provided by the Rio de Janeiro Public Prosecutor’s Office). In 2019, 92% of the neighborhoods in the city of Rio, where almost 98% of its population lives, were partially or wholly under the control or in dispute for armed groups. The group with the greatest prominence were the militias, paramilitary groups formed by state security agents, who controlled partially or wholly 25.5% of the neighborhoods in the capital of Rio de Janeiro, where 33% of its population lives.
What was the hardest part of this project?
The construction of the 2019 map of armed groups presented 3 major challenges. The first of them was to read and process thousands of anonymous denunciations (which deal with various types of crime), choosing only those that in fact present convincing indications of territorial control by a certain armed group. To solve this challenge, researchers with PhDs in areas related to public security defined objective criteria for determining territorial control and, based on the reading of a sample of the material, built dictionaries for subsequent automatic classification of text by machine, using natural language processing techniques. At the same time, a team of data and statistical scientists, sought to georeference the complaints of the Whistleblower. This task proved to be a challenge not only because of the amount of records, but also because of the low quality of the address information available due to spelling mistakes and standardization not corrected in the original database, but also because of the high level of urban informality of Brazilian cities, which makes it difficult to locate the addresses accurately. The solutions found were organized in an R package to facilitate new georeferencing rounds. Finally, the last challenge was the construction of maps that reflect the reality of urban informality in Rio. The official neighborhood map, made available by the MPRJ, is interesting for analysis of the coverage of militias (that generally operate in entire neighborhoods), but it is not granular enough to understand the reach of drug factions (usually restricted to favelas or housing estates). The solution was to build 2 complementary analyses. The partnership of Pista News, an organization that maps the control of armed groups in Rio’s favelas and housing complexes in real time (but not with historical perspicacity) in a collaborative manner, was essential for the construction.
What can others learn from this project?
The project is a tool for journalism in 2 main directions. Directly, by making the map available for free, becoming a tool for journalistic analysis to reveal trends, from corruption relations between some of these groups and local police units, to relations between politicians, through the direct impact of these groups’ disputes on the access of population to services.
As the map’s shapefile is open, the journalist can cross-check this information with any other data (health, education, sanitation) that have geolocation and produce diagrams that until then were not produced by the government. Such investigations can reveal the impact of the control of armed groups on the daily lives of the local population in various instances.
The project also encourages journalists to look for non-governmental sources of information to report reality and not to get stuck in the political narrative of the facts. In Rio de Janeiro, most newspapers and TV programs do not name factions and militias so that, according to some journalists, they do not value these groups. But that decision proved innocuous. With the map, we show that the action of these groups has grown and that it is important to give names to everyone, ponting who rules where, what is their progress and representativeness.
In other words, the Map of the Armed Groups of Rio de Janeiro is a journalistic opportunity to question the government in a multiciplinary, didactic and transparent way – and claim for better policies.