Epidemic levels of organized and interpersonal violence are present in parts of Latin America and Central and Southern Africa. Despite rapid economic growth and political and social transformations in both regions, many states and cities exhibit among the world’s highest rates of real and perceived victimization (Szabó de Carvalho, Garzón and Muggah 2013). Personal and household insecurity is statistically correlated with reduced living standards. It is also associated with reversals in poverty reduction and service delivery. Predictably, Latin American and South African citizens are registering a declining confidence in public institutions and an increased fear of leaving their homes.

For decades public and private entities in Latin America and South Africa have invested in, designed and implemented a wide variety of responses to promote national security and public order. The majority of these interventions were pursued by military and policing institutions, a legacy of authoritarian regimes, and with elite consent. Security was thus constructed in Weberian terms, as the duty of the state, with law and order mediated exclusively by the armed forces, law enforcement agencies and their auxiliaries. The results of these efforts were dramatic, including the introduction of draconian legislation, stiffer penalties, soaring homicide rates, and overflowing prisons.

Since the late 1990s, a different approach to promote public safety has emerged across Latin America, the Caribbean (UNDP 2013) and even some cities in South Africa. The concept of ‘citizen security’ prioritizes responsible statehood and proactive citizenship (Muggah and Szabo 2014). This change was a direct reaction to national security paradigms that had dominated during the dictatorship and apartheid eras. As a set of discourses and practices, citizen security emphasizes preventive approaches, risk reduction, and enhancing protective factors in crime-affected areas. It promotes community-oriented models, citizen participation, and data-driven interventions. Brazil, Colombia, Mexico and South Africa are all veritable laboratories of experimentation, with both successes and failures.

However, citizen security is a concept that is more easily described than defined. In functional terms, it consists of a wide array of primary and secondary violence prevention, community and proximity policing, rehabilitation of at-risk youth and innovative justice mechanisms. Major international donor agencies such as the Inter-American Development Bank (IADB) and the World Bank have started supporting city-, state- and federal-level partners using the lexicon of citizen security. And they are not alone. A recent assessment by the Igarapé Institute counted no less than 1,350 citizen security programs and projects across Latin America and the Caribbean since 1998 (Muggah and Aguirre 2013; Muggah 2014). Another assessment of similar types of activities in South Africa yielded almost 60, though the number is likely much higher (Muggah and Wenmann 2011). In spite of their rapid spread and the growing appetite for citizen security among governments, bilateral and multilateral agencies, foundations and research institutions, there is very little evidence of whether or not they improve the real and perceived safety and security of citizens.

The Citizen Security Dialogues was launched in 2014 precisely to advance debate on what works, and what does not. With support from partners such as the International Development Research Centre (IDRC), the African Police Civilian Oversight Forum (APCOF), the Foundation for Ideas for Peace (FIP), the Institute for Security and Democracy (INSYDE) and the Inter-American Development Bank (IADB), the Dialogues are designed to foster a conversation between leading experts in public security and their public and private sector counterparts. Between 2014 and 2015 Dialogues in Rio de Janeiro, Mexico, Bogotá and Cape Town will provoke debate domestically, but also within and between Latin America and Southern Africa. The intention is to encourage South-South interaction and allow partners to identify, evaluate and explore replication potential of effective citizen security interventions. Findings will be captured in special editions of Stability: International Journal of Security & Development.

Citizen Security in Brazil

With roughly 50,000 murders each year (OAS 2012), Brazil is one of the most violent countries on the planet. Lethal violence is also on the rise: in 1996 the national homicide rate was 24.8 per 100,000 inhabitants and by 2011 it had risen to 27.1 per 100,000. Brazil also features some of the most violent cities in the world, with urban centers across the north-east regularly featuring in international rankings (Waiselfiz 2013). Currently, approximately one in every ten victims of lethal violence in the world resides in Brazil (Muggah and Mack 2014). There are a number of factors contributing to the country’s spiraling violence, not least drug-related crime, repressive policing, high rates of impunity and structural inequality.

Despite overall increases in lethal violence in Brazil, the spatial and temporal dynamics of violence are neither linear nor homogenous (UNODC 2013). For example, the homicide rate in the south-east of the country has decreased over the past decade due in part to massive declines in São Paulo and Rio de Janeiro (Muggah 2013). Meanwhile, murder rates have sky-rocketed in the north and north-east of the country – by as high as 75 per cent – owing to insecurity in Pará and Amazonas (Waiselfiz 2013). Likewise, lethal violence has also increased steadily in the south and central regions, Paraná, Santa Catarina and Goiás. Remarkably, the 17 states featuring the lowest homicide rates in 2001 have all seen increases: in Alagoas the homicide rate shot-up from 29.3 per 100,000 in 2001 to 72.2 in 2011 (Waiselfiz 2013). The vast majority of this violence is committed with firearms and is concentrated among low-income male youth.

While there does not appear to be a coherent ‘national’ strategy to prevent homicidal violence, there are many subnational efforts to reverse these trends. Indeed, efforts have sprung up in major cities – including Belo Horizonte, Canoas, Recife, Rio de Janeiro and São Paulo – across the country. A growing cadre of scholars, often working with state public security secretaries and non-governmental groups – has started to assess the outcomes and impacts of these efforts. However, many of these efforts have been pursued in isolation of one another, hardly surprising in a country the size of Brazil. And yet there are remarkable insights and lessons to be learned from greater collaboration. In this special edition of Stability, the focus is on learning from citizen security innovations from the country’s leading specialists.

It is appropriate that the first article of the special edition focuses on a major violence prevention initiative pursued in Pernambuco state. As José Luiz Ratton, Clarissa Galvão and Michelle Fernandez show, this is the only north-eastern state to have experienced a reduction in the homicide rate – by over 26 per cent between 2006 and 2011. In their view, the Pacto Pela Vida (Pact for Life) program – an intervention designed to reduce criminal violence – was singularly responsible for reductions in homicides by some 40 per cent between January 2007 and June 2013. A key ingredient of success was the direct engagement of the state governor – highlighting the critical functions of political leadership on an otherwise delicate subject.

The special edition then turns to innovative data-driven efforts to reduce criminal violence in the capital of Minas Gerais state. Cláudio Beato and Andrea Silveira start their article by reviewing previous evaluations of crime prevention policies adopted by the state government. They detect a greater emphasis in the evaluation literature on studies addressing ‘outcomes’ rather than ‘processes.’ Very few assessments consider the means by which policies and programs were established and implemented. More fundamentally, the authors also note that in the haste to adopt multi-dimensional strategies to fight crime, there has been comparatively less attention to their monitoring and evaluation. There is a serious weakness, then, in measuring the overall effectiveness of citizen security effortswhich could hamper their spread in the state, and the country.

Much has been made in media circles of the ‘São Paulo miracle’ which resulted in plunging violence rates over the past decade. Melina Risso considers some of the more recent activities undertaken in São Paulo state and highlights the ways in which they may have contributed to violence reduction. Key lessons relate to the importance of restructuring the military police, investing in information systems to facilitate smarter operations, and expanding the capabilities of the civil police homicide department and their ability to protect witnesses. Risso also underlines the positive dividends of arms control and local violence prevention initiatives. And yet many challenges remain, not least the continued involvement of police in extra-judicial killing.

In the final article of the special edition, Robson Rodrigues, the former commander of the so-called Pacification Police Units (UPP) in Rio de Janeiro, provides some reflections on the past five years of experimentation. The initiative is credited with reductions of more than 65 per cent in homicidal violence since its inception in 2009. Rodrigues highlights the many risks, opportunities and achievements of the pacification process. He detects a critical transition from an erstwhile reactive model of repressive policing focused on interdicting drugs to one that is more proactive and emphasizing violence reduction. And yet the UPP project also has many challenges. Indeed, it is still evolving and is not a substitute for a wider project of social, economic and political investment. At a minimum, a clearer doctrine is required, improved metrics, and also more systematized means of collecting data.

There are many insights that can be drawn from the Brazilian experience that apply in other Latin American and even Southern African settings. The highly urbanized nature of citizen insecurity, the exacerbating effects of the war on drugs, and the sharp divisions between haves- and have-nots are all factors that resonate far beyond the country’s borders. Of course there are also many variables that are unique to Brazil, and the temptation to generalize should not be taken too far. Even so, by considering the lessons emerging from innovative citizen security practices, this Special Edition invites readers to critically reflect on what works and what does not. It highlights the central place of verifiable data and analysis, as well as the need to adopt comprehensive approaches moving well beyond policing.