¹ Imbusch, Peter, Misse, Michel and Carrión, Fernando, ‘Violence Research in Latin America and the Caribbean: A Literature Review’, International Journal of Conflict and Violence, 5: 1 (2011), pp. 87–154 Google Scholar; Arias, Enrique Desmond and Goldstein, Daniel (eds.), Violent Democracy in Latin America: Toward an Interdisciplinary Reconceptualization (Durham, NC: Duke University Press, 2010)CrossRefGoogle Scholar; Méndez, Juan, Pinheiro, Paulo Sérgio and O'Donnell, Guillermo (eds.), (Un)Rule of Law and the Underprivileged in Latin America (South Bend, IN: University of Notre Dame Press, 1999)Google Scholar; Nóbrega, José Maria Jr., Zaverucha, Jorge and Rocha, Enivaldo, ‘Mortes por agressão em Pernambuco e no Brasil: um óbice para a consolidação da democracia’, Revista de Sociologia e Política, 19: 40 (2011), pp. 43–58 CrossRefGoogle Scholar.

² Messner, Steven, Raffalovich, Lawrence and Shrock, Peter, ‘Reassessing the Cross-National Relationship between Income Inequality and Homicide Rates: Implications of Data Quality Control in the Measurement of Income Distribution’, Journal of Quantitative Criminology, 18: 4 (2002), pp. 377–95CrossRefGoogle Scholar.

³ Briceño-León, Roberto, ‘Tres fases de la violencia homicida en Venezuela’, Ciência e Saúde Coletiva, 17: 12 (2012), pp. 3233–42CrossRefGoogle Scholar; José Maria Nóbrega Júnior, ‘Os homicídios no Brasil, no Nordeste e em Pernambuco: dinâmica, relações de causalidade e políticas públicas’, PhD Dissertation (UFPE, 2010).

⁴ O'Donnell, Guillermo, ‘On the State, Democratization and Some Conceptual Problems: A Latin American View with Glances at Some Postcommunist Countries’, World Development, 21 (1993), pp. 1355–69CrossRefGoogle Scholar; ‘Polyarchies and the (Un)Rule of Law in Latin America: A Partial Conclusion’, in Méndez, O'Donnell and Pinheiro (eds.), (Un)Rule of Law, pp. 303–37.

⁵ Arias and Goldstein (eds.), Violent Democracy in Latin America; Auyero, Javier, Routine Politics and Violence in Argentina: The Gray Zone of State Power (Cambridge: Cambridge University Press, 2007)CrossRefGoogle Scholar; Auyero, Javier, de Lara, Augustin and Berti, Maria, ‘Violence and the State at the Urban Margins’, Journal of Contemporary Ethnography, 43 (2014), pp. 94–116 CrossRefGoogle Scholar; Cruz, José Miguel, ‘Criminal Violence and Democratization in Central America: The Survival of the Violent State’, Latin American Politics and Society, 53: 4 (2011), pp. 1–33 CrossRefGoogle Scholar.

⁶ Arias and Goldstein (eds.), Violent Democracy in Latin America; Caldeira, Teresa and Hoston, James, ‘Democracy and Violence in Brazil’, Comparative Studies in Society and History, 41: 4 (1999), pp. 691–729 Google Scholar; Alba Zaluar, ‘The Paradoxes of Democratization and Violence in Brazil’, paper presented at the Conferência Brasil e União Européia Ampliada, Universidade Federal do Rio de Janeiro (2004).

⁷ Arias, Enrique Desmond, ‘Understanding Criminal Networks, Political Order and Politics in Latin America’, in Clunan, Anne and Trinkunas, Harold (eds.), Ungoverned Spaces: Alternatives to State Authority in an Era of Softened Sovereignty (Stanford, CA: Stanford University Press, 2010), pp. 115–35Google Scholar.

⁸ Institutions here are considered as formal and informal rules and norms that guide individual and group behaviour, and ‘structure political, economic and social interaction … to create order and reduce uncertainty’: North, Douglass, ‘Institutions’, Journal of Economic Perspectives, 5: 1 (1991), pp. 97–112, here p. 97CrossRefGoogle Scholar.

⁹ It is acknowledged that there are important differences between lethal and non-lethal violence, and there is no wish to diminish these. Moreover, there is no assumption that lethal and non-lethal violence necessarily follow similar patterns or respond to similar policies. However, in discussing Pernambuco's experiences throughout this article, where the term ‘violence’ is present, it is used in reference to lethal violence or homicide.

¹⁰ Pernambuco was not the first state in Brazil to implement progressive public-security policies in order to address violence: see de Lima, Renato, Godinho, Letícia and de Paula, Liana, Os governos subnacionais na gestão da segurança cidadã: a experiência brasileira (Washington, DC: Inter-American Development Bank, 2014)Google Scholar. It has, however, arguably been the most successful.

¹¹ Waiselfisz, Julio, Mapa da violência 2012: os novos padrões da violência homicida no Brasil (São Paulo: Sangari, 2011)Google Scholar.

¹² On similarities in the north-east see Montero, Alfred, ‘A Reversal of Political Fortune: The Transitional Dynamics of Conservative Rule in the Brazilian Northeast’, Latin American Politics and Society, 54: 1 (2012), pp. 1–36 CrossRefGoogle Scholar.

¹³ On declining inequality see Ferreira, Francisco, Leite, Phillippe and Litchfield, Julie, ‘The Rise and Fall of Brazilian Inequality: 1981–2004’, Macroeconomic Dynamics, 12: 2 (2008), pp. 199–230 CrossRefGoogle Scholar. On redistributive policy see Barros, Ricardo, Foguel, Miguel and Ulyssea, Gabriel, Desigualdade de renda no Brasil: uma análise da queda recente (Rio de Janeiro: Instituto de Pesquisa Econômica Aplicada, 2007)Google Scholar.

¹⁴ Seawright, Jason and Gerring, John, ‘Case Selection Techniques in Case Study Research: A Menu of Qualitative and Quantitative Options’, Political Research Quarterly, 61: 2 (2008), pp. 294–308 CrossRefGoogle Scholar.

¹⁵ Though examining the ‘deviant’ Pernambucan case may sample on the dependent variable, characteristics of the dependent variable are often theoretically relevant, and in such cases can offer insights that can be later tested comparatively. Regarding challenges, see Geddes, Barbara, ‘How the Cases You Choose Affect the Answers You Get: Selection Bias in Comparative Politics’, Political Analysis, 2: 1 (1990), pp. 131–50CrossRefGoogle Scholar. On potential benefits, see Mahoney, James, ‘Qualitative Methodology and Comparative Politics’, Comparative Political Studies, 40: 2 (2007), pp. 122–44CrossRefGoogle Scholar. On the role of deviant cases in theory development, see George, Alexander and Bennett, Andrew, Case Studies and Theory Development in the Social Sciences (Cambridge, MA, and London: MIT Press, 2005)Google Scholar.

¹⁶ Snyder, Richard, ‘Scaling Down: The Subnational Comparative Method’, Studies in Comparative International Development, 36: 1 (2001), pp. 93–110 CrossRefGoogle Scholar.

¹⁷ Cornelius, Wayne, Eisenstadt, Todd and Hindley, Jane (eds.), Subnational Politics and Democratization in Mexico (Boulder, CO: Lynne Rienner, 1999)Google Scholar; Eaton, Kent, Politics beyond the Capital: The Design of Subnational Institutions in Latin America (Stanford, CA: Stanford University Press, 2004)Google Scholar; Herrmann, Julián Durazo, ‘Reflections on Regime Change and Democracy in Bahia, Brazil’, Latin American Research Review, 49: 3 (2014), pp. 23–44 CrossRefGoogle Scholar.

¹⁸ Alexander, Gerard, ‘Institutions, Path Dependence, and Democratic Consolidation’, Journal of Theoretical Politics, 13: 3 (2001), pp. 249–69CrossRefGoogle Scholar; Mahoney, James, ‘Path Dependence in Historical Sociology’, Theory and Society, 29: 4 (2000), pp. 507–48CrossRefGoogle Scholar. For Brazil, challenges to reform are likely given the autonomy states are granted under federalism, and particularly since changes in political processes and major policy shifts often begin locally rather than nationally. See Sugiyama, Natasha, ‘Theories of Policy Diffusion: Social Sector Reform in Brazil’, Comparative Political Studies, 41: 2 (2008), pp. 193–216 CrossRefGoogle Scholar.

¹⁹ Process-tracing methodology allows for examination of qualitative evidence to describe and analyse processes of social change and understand how sequences of events lead to particular outcomes. See Collier, David, ‘Understanding Process Tracing’, Political Science and Politics, 44: 4 (2011), pp. 823–30CrossRefGoogle Scholar; George and Bennett, Case Studies; Gerring, John, Case Study Research: Principles and Practices (Cambridge: Cambridge University Press, 2007)Google Scholar.

²⁰ Jeffrey T. Checkel, ‘Process Tracing’, in Klotz, A. and Prakash, D. (eds.), Qualitative Methods in International Relations (Basingstoke: Palgrave Macmillan, 2008), pp. 114–27CrossRefGoogle Scholar.

²¹ Interviews were conducted in Portuguese with a local assistant. Interviewees have not been named due to the political sensitivity of certain themes discussed and to protect respondents where necessary. Translations from interview transcripts are the author's own. These were supplemented by telephone interviews with relevant experts.

²² After stepping down as governor of Pernambuco in April 2014 to campaign for the presidency, Campos was killed in a plane crash in Aug. 2014 in Santos, São Paulo.

²³ I am grateful to an anonymous reviewer for emphasising this point.

²⁴ Hoelscher, Kristian and Nussio, Enzo, ‘Understanding Unlikely Successes in Urban Violence Reduction’, Urban Studies, 53: 11 (2016), pp. 2397–416CrossRefGoogle Scholar.

²⁵ Koonings, Kees and Kruijt, Dirk, Fractured Cities: Social Exclusion, Urban Violence and Contested Spaces in Latin America (London: Zed, 2007)Google Scholar; Moser, Caroline and McIlwaine, Cathy, Encounters with Violence in Latin America: Urban Poor Perceptions from Colombia and Guatemala (London: Routledge, 2004)Google Scholar.

²⁶ Often measured using homicide rates, some have also referred to this as lethal social violence – namely deaths due to violence between individuals or small groups without an explicit political motive. See Buvinic, Mayra, Morrison, Andrew and Shifter, Michael, Violence in Latin America and the Caribbean: A Framework for Action (Washington, DC: Inter-American Development Bank, 1999)Google Scholar; Fox, Sean and Hoelscher, Kristian, ‘Political Order, Development and Social Violence’, Journal of Peace Research, 49: 3 (2012), pp. 431–44CrossRefGoogle Scholar.

²⁷ Bates, Robert, Prosperity and Violence: The Political Economic of Development (New York: Norton, 2001)Google Scholar; Elias, Norbert, The Civilizing Process, Vol. 2: State Formation and Civilization (Oxford: Blackwell, 1982)Google Scholar; LaFree, Gary, Losing Legitimacy: Street Crime and the Decline of Social Institutions in America (Boulder, CO: Westview, 1998)Google Scholar; North, Douglass, Wallis, John and Weingast, Barry, Violence and Social Orders: A Conceptual Framework for Interpreting Recorded Human History (Cambridge: Cambridge University Press, 2009)CrossRefGoogle Scholar.

²⁸ Hegre, Håvard, Ellingsen, Tanya, Gates, Scott and Gleditsch, Nils Petter, ‘Toward a Democratic Civil Peace? Democracy, Political Change, and Civil War, 1816–1992’, American Political Science Review, 95: 1 (2001), pp. 33–48 Google Scholar; Urdal, Henrik and Hoelscher, Kristian, ‘Urban Youth Bulges and Social Disorder: An Empirical Study of Asia and Sub-Saharan African Cities’, International Interactions, 38: 4 (2012), pp. 512–28CrossRefGoogle Scholar; Fox and Hoelscher, ‘Political Order’.

²⁹ Nivette, Amy and Eisner, Manuel, ‘Do Legitimate Polities have Fewer Homicides? A Cross-National Analysis’, Homicide Studies, 17: 1 (2013), pp. 3–26 CrossRefGoogle Scholar.

³⁰ Hoelscher, Kristian, ‘Politics and Social Violence in Developing Democracies: Theory and Evidence from Brazil’, Political Geography, 44: 1 (2015), pp. 29–39 CrossRefGoogle Scholar.

³¹ Briceño-León, Roberto, ‘La comprensión de los homicidios en América Latina: ¿Pobreza o institucionalidad?’, Ciência e Saúde Coletiva, 17: 12 (2012), pp. 3159–70CrossRefGoogle Scholar.

³² Koonings, Kees and Kruijt, Dirk (eds.), Armed Actors: Organised Violence and State Failure in Latin America (London: Zed, 2004)Google Scholar; UNDP, Democracy in Latin America: Towards a Citizen's Democracy (New York: UNDP, 2005)Google Scholar, available at http://acme.highpoint.edu/~msetzler/LatAmPol/LAreadings/Report_Democracy_in_Latin_America_New.pdf (accessed 10 May 2017).

³³ O'Donnell, Guillermo, ‘Why the Rule of Law Matters’, Journal of Democracy, 15: 4 (2004), pp. 32–46 CrossRefGoogle Scholar.

³⁴ Kruijt, Dirk and Koonings, Kees, ‘Violence and Fear in Latin America’, in Koonings and Kruijt (eds.), Societies of Fear. The Legacy of Civil War, Violence, and Terror in Latin America (London: Zed Books, 1999), pp. 1–30 Google Scholar.

³⁵ Briceño-León, ‘Tres fases de la violencia homicida en Venezuela’.

³⁶ Davis, Diane, ‘Undermining the Rule of Law: Democratization and the Dark Side of Police Reform in Mexico’, Latin American Politics and Society, 48: 1 (2006), pp. 55–86 CrossRefGoogle Scholar.

³⁷ Shifter, Michael, Countering Criminal Violence in Central America (Washington, DC: Council on Foreign Relations, 2012)Google Scholar; Brinks, Daniel, ‘Informal Institutions and the Rule of Law: The Judicial Response to State Killings in Buenos Aires and São Paulo in the 1990s’, Comparative Politics, 36: 1 (2003), pp. 1–19 CrossRefGoogle Scholar.

³⁸ Arias, Desmond, Drugs and Democracy in Rio de Janeiro (Durham, NC: University of North Carolina Press, 2006)Google Scholar; ‘The Impacts of Differential Armed Dominance of Politics in Rio de Janeiro, Brazil’, Studies in Comparative International Development, 48: 3 (2013), pp. 263–84CrossRefGoogle Scholar; Javier Auyero, ‘Clandestine Connections: The Political and Relational Makings of Collective Violence’, in Arias and Goldstein (eds.), Violent Democracy, pp. 108–32.

³⁹ Moncada, Eduardo, ‘The Politics of Urban Violence: Challenges for Development in the Global South’, Studies in Comparative International Development, 48: 3 (2013), pp 217–39CrossRefGoogle Scholar.

⁴⁰ Armijo, Leslie, Faucher, Phillipe and Dembinska, Magdalena, ‘Compared to What?: Assessing Brazil's Political Institutions’, Comparative Political Studies, 39: 6 (2006), pp. 759–86CrossRefGoogle Scholar.

⁴¹ Holston, James, Insurgent Citizenship: Disjunctions of Democracy and Modernity in Brazil (Princeton, NJ: Princeton University Press, 2009)Google Scholar.

⁴² Hagopian, Frances, ‘“Democracy by Undemocratic Means”? Elites, Political Pacts, and Regime Transition in Brazil’, Comparative Political Studies, 23: 2 (1990), pp. 147–70CrossRefGoogle Scholar; Traditional Politics and Regime Change in Brazil (Cambridge: Cambridge University Press, 1996)Google Scholar.

⁴³ Cerqueira, Daniel, Lobão, Waldir and Carvalho, Alexandre, ‘O jogo dos sete mitos e a miséria da segurança pública no Brasil’, in da Cruz, Marcus and Batittuci, Eduardo (eds.), Homicídios no Brasil (Rio de Janeiro: Editoria FGV, 2007)Google Scholar.

⁴⁴ Leeds, Elizabeth, ‘Serving States and Serving Citizens: Halting Steps toward Police Reform in Brazil and Implications for Donor Intervention’, Policing and Society, 17: 1 (2007), pp. 21–37 CrossRefGoogle Scholar.

⁴⁵ Fiona Macaulay, ‘Problems of Police Oversight in Brazil’, Oxford Centre for Brazilian Studies Working Paper 33 (2002).

⁴⁶ Ahnen, Ronald, ‘The Politics of Police Violence in Democratic Brazil’, Latin American Politics and Society, 49: 1 (2007), pp. 141–64CrossRefGoogle Scholar.

⁴⁷ Bayley, David, Democratizing the Police Abroad: What to Do and How to Do It (Washington, DC: National Institute of Justice, 2001)Google Scholar.

⁴⁸ Bayley, David, Changing the Guard: Developing Democratic Police Abroad (Oxford: Oxford University Press, 2006)Google Scholar.

⁴⁹ Macaulay, Fiona, ‘Cycles of Police Reform in Latin America’, in Francis, David (ed.), Policing in Africa (New York: Palgrave Macmillan, 2012)Google Scholar; Bailey, John and Dammert, Lucia (eds.), Public Security and Police Reform in the Americas (Pittsburgh, PA: University of Pittsburgh Press, 2006)Google Scholar; Prado, Mariana, Trebilcock, Michael and Hartford, Patrick, ‘Police Reform in Violent Democracies in Latin America’, Hague Journal of the Rule of Law, 4: 2 (2012), pp. 252–85CrossRefGoogle Scholar; Uildriks, Niels, ‘Police Reform, Security and Human Rights in Latin America: An Introduction’, in Uildriks, Niels (ed.), Policing Insecurity (Lanham, MD: Lexington, 2009)Google Scholar; Ungar, Mark, Policing Democracy: Overcoming Obstacles to Citizen Security in Latin America (Baltimore, MD: Johns Hopkins University Press, 2011)Google Scholar.

⁵⁰ Eterno, John and Silverman, Eli, ‘New York City Police Department's CompStat: Dream or Nightmare’, International Journal of Police Science and Management, 8: 3 (2006), pp. 218–31CrossRefGoogle Scholar.

⁵¹ Eck, John and Maguire, Edward, ‘Have Changes in Policing Reduced Violent Crime? An Assessment of the Evidence’, in Blumstein, Alfred and Wallman, Joel (eds.), The Crime Drop in America (Cambridge: Cambridge University Press, 2000)Google Scholar; Walsh, William, ‘CompStat: An Analysis of an Emerging Police Managerial Paradigm’, Policing: An International Journal of Police Strategies and Management, 24: 3 (2001), pp. 347–62CrossRefGoogle Scholar.

⁵² Moncada, Eduardo, ‘Toward Democratic Policing in Colombia? Institutional Accountability through Lateral Reform’, Comparative Politics, 41: 4 (2009), pp. 431–49, here p. 433CrossRefGoogle Scholar. On horizontal and vertical accountability, see O'Donnell, Guillermo, Counterpoints: Selected Essays on Authoritarianism and Democratization (South Bend, IN: University of Notre Dame Press, 1999)Google Scholar.

⁵³ Risso, Melina, ‘Intentional Homicides in São Paulo City: A New Perspective’, Stability: International Journal of Security and Development, 3: 1 (2014)Google Scholar, http://doi.org/10.5334/sta.do (all subsequent websites cited in this article were last accessed 20 April 2017 unless otherwise noted).

⁵⁴ Goertzel, Ted and Kahn, Tulio, ‘The Great Sao Paulo Homicide Drop’, Homicide Studies, 13: 4 (2009), pp. 398–410 CrossRefGoogle Scholar; de Mello, João and Schneider, Alexandre, ‘Assessing São Paulo's Large Drop in Homicides: The Role of Demography and Policy Interventions’, in Di Tella, Rafael, Edwards, Sebastian and Schargrodsky, Ernesto (eds.), The Economics of Crime: Lessons for and from Latin America (Chicago, IL: University of Chicago Press, 2010)Google Scholar. This reduction was also partly due to strategies implemented by the prison-based criminal syndicate Primeiro Comando do Capital (First Command of the Capital, PCC), which eschewed the use of violence.

⁵⁵ This is despite numerous examples of democratic consolidation and support for security-sector reform and rule of law assistance. See Carothers, Thomas, Promoting the Rule of Law Abroad: In Search of Knowledge (Washington, DC: Carnegie Endowment for International Peace, 2006)Google Scholar.

⁵⁶ Skogan, Wesley, ‘Why Reforms Fail’, Policing and Society, 18: 1 (2008), pp. 23–34 CrossRefGoogle Scholar.

⁵⁷ Pereira, Anthony and Ungar, Mark, ‘The Persistence of the “Mano Dura”: Authoritarian Legacies and Policing in Brazil and the Southern Cone’, in Cesarini, Paola and Hite, Katherine (eds.), Authoritarian Legacies in Southern Europe and Latin America (South Bend, IN: University of Notre Dame Press, 2004)Google Scholar.

⁵⁸ Hadfield, Gillian and Weingast, Barry, ‘Microfoundations of the Rule of Law’, Annual Review of Political Science, 17: 1 (2014), pp. 21–42 CrossRefGoogle Scholar.

⁵⁹ Hadfield, Gillian and Weingast, Barry, ‘Law without the State: Legal Attributes and the Coordination of Decentralized Collective Punishment’, Journal of Law and Courts, 1: 1 (2013), pp. 3–34 CrossRefGoogle Scholar.

⁶⁰ Nathanson, Constance, ‘Social Movements as Catalysts for Policy Change: The Case of Smoking and Guns’, Journal of Health Politics, Policy and Law, 24: 3 (1999), pp. 421–88CrossRefGoogle ScholarPubMed.

⁶¹ Elias, The Civilizing Process.

⁶² This is important as state governors in Brazil are responsible for public security, and are therefore influential in shaping policy.

⁶³ I thank an anonymous reviewer for highlighting this point.

⁶⁴ Interviews, senior members of the municipal secretariat for urban security, March 2013; interviews, local NGO, March 2013. On the repressive tactics of Vasconcelos see also Alexandre de Freitas, ‘Fundamentos para uma sociologia crítica da formação humana: um estudo sobre as redes associacionistas da sociedade civil’, PhD Dissertation (UFPE, 2005).

⁶⁵ Interviews, senior member of the municipal secretariat for urban security, March 2013.

⁶⁶ Interview, former senior member of the Courts of Justice, Pernambuco, Feb. 2013.

⁶⁷ ‘There was no monitoring, no goals, no structures in place to protect crime scenes’: Interview, PC delegado (police investigator), Feb. 2013.

⁶⁸ Interviews, senior member of the PC, Feb. and March 2013. Such practices reinforced the political appointment of allies and the usurpation of the state apparatus for personal ends. See also Montero, Alfred and Samuels, David (eds.), Decentralization and Democracy in Latin America (South Bend, IN: University of Notre Dame Press, 2004)Google Scholar.

⁶⁹ This occurred both out of frustration that known criminals were evading punishment, and for officers’ own personal financial gain. Interviews, member and senior member of the PM, March 2013.

⁷⁰ Separate interviews, senior members of the PC and PM, March 2013.

⁷¹ Noto, Ana, Galduróz, José, Nappo, Solange and Carlini, E., Levantamento nacional sobre o uso de drogas entre crianças e adolescentes em situação de rua nas 27 capitais brasileiras – 2003 (São Paulo: Centro Brasileiro de Informações sobre Drogas Psicotrópicas, 2004)Google Scholar.

⁷² Interviews, senior member of the PM, March 2013; interviews, community leaders, Feb. and March 2013. A community leader noted: ‘If you compare here and a little [version of] Rio [de Janeiro], there crime is organised. Here it is just disorganised.’

⁷³ Author's translation from Júnior, José Maria Nóbrega, ‘Homicídios em Pernambuco: dinâmica e relações de causalidade’, Coleção Segurança com Cidadania, 1: 3 (2009), pp. 237–60, here p. 248Google Scholar.

⁷⁴ For further elaboration see Andréia Macêdo, ‘“Polícia, quando quer, faz!”: análise da estrutura de governança do “Pacto pela Vida” de Pernambuco’, Dissertation (Universidade de Brasília, 2012); and Ratton, José Luiz, Galvão, Clarissa and Fernandez, Michelle, ‘Pact for Life and the Reduction of Homicides in the State of Pernambuco’, Stability: International Journal of Security and Development, 3: 1 (2014), pp. 1–15 Google Scholar.

⁷⁵ CVLIs are PPV's main measure for evaluating impact in reducing lethal violence. CVLI data includes intentional homicides and deaths following assault, robbery or similar altercations (corresponding to first- and second-degree murder).

⁷⁶ Interview, former coordinator of NGO, March 2013.

⁷⁷ Interview, senior member of the Courts of Justice, March 2013; interview, senior colonel in PM, March 2013.

⁷⁸ While not mentioned in interviews, participatory budgeting processes in Recife during the 2000s may have increased citizens’ awareness of their rights and ability to challenge the status quo. I thank an anonymous reviewer for highlighting this. Also see Nuijten, Monique, Koster, Martijn and de Vries, Pieter, ‘Regimes of Spatial Ordering in Brazil: Neoliberalism, Leftist Populism and Modernist Aesthetics in Slum Upgrading in Recife’, Singapore Journal of Tropical Geography, 33: 2 (2012), pp. 157–70CrossRefGoogle Scholar.

⁷⁹ Interview, senior member of the PM, March 2013. See also Wolff, Michael Jerome, ‘Building Criminal Authority: A Comparative Analysis of Drug Gangs in Rio de Janeiro and Recife’, Latin American Politics and Society, 57: 2 (2015) pp. 21–40 CrossRefGoogle Scholar.

⁸⁰ Interview, NGO, March, 2013; interview, UFPE professor, Feb. 2013.

⁸¹ Interview, secretariat for urban violence, Feb. 2013; interview, journalist, March 2013; interview, NGO, March 2013. Particularly influential was the death of Dr Antônio Carlos Escobar, killed in one of the wealthiest parts of Recife in Dec. 2005 while intervening in an assault.

⁸² Many Latin American societies have responded to violence by supporting mano dura-style responses to violence, including other federal states in Brazil, particularly in the north-east.

⁸³ Interview, senior member of municipal secretariat for urban violence, March 2013; interview, journalist, March 2013.

⁸⁴ Ibid.

⁸⁵ Bogotá’s experience showed the importance of reforms that increased ‘public support for, and participation in, the reform process over time’: see Moncada, ‘Toward Democratic Policing’, p. 431.

⁸⁶ Hoelscher and Nussio, ‘Understanding Unlikely Successes’.

⁸⁷ Interviews, NGOs, March 2013. See Rique, Célia, Aguiar, Elaine, Lins, José and Barros, Leonardo, A criminalidade no Recife: um problema de amplitude nacional (Recife: Gabinete de Assessoria Jurídica às Organizações Populares, 2005)Google Scholar.

⁸⁸ See Assies, Willem, To Get Out of the Mud: Neighborhood Associativism in Recife 1964–1988 (Amsterdam: Centro de Estudios y Documentación Latinoamericanos, 1992)Google Scholar; on Pernambuco's historical civic activism, see: Mosher, Jeffrey, Political Struggle, Ideology, and State Building: Pernambuco and the Construction of Brazil, 1817–1850 (Lincoln, NE: University of Nebraska Press, 2008)Google Scholar; and De Carvalho, Marcus, Liberdade: rotinas e rupturas do escravismo no Recife, 1822–1850 (Recife: UFPE, 1998)Google Scholar.

⁸⁹ http://cremepe.org.br/2006/02/20/carreata-protesta-contra-violencia/.

⁹⁰ www.cartamaior.com.br/?/Editoria/Direitos-Humanos/37-mulheres-sao-assassinadas-em-31-dias-em-Pernambuco/5/ 9172.

⁹¹ www.alepe.pe.gov.br/2006/04/20/violencia-pernambuco-debate-seguranca-publica.

⁹² Interview, NGO leader, March 2013.

⁹³ Interview, secretariat for urban violence, Feb. 2013; interview, journalist, March 2013.

⁹⁴ Interview, NGO leader, March 2013.

⁹⁵ Campos came under considerable pressure to follow through on these campaign promises. See Michael Jerome Wolff, ‘Policing and the Logics of Violence: A Comparative Analysis of Public Security Reform in Brazil’, Policing and Society (2015), http://dx.doi.org/10.1080/10439463.2015.1093478.

⁹⁶ For example, in the north-eastern state of Alagoas, political will or capacity to respond to recent civil-society mobilisation in the face of violence was largely absent. See https://globalvoices.org/2012/06/01/brazil-alagoas-peace-most-violent-state/.

⁹⁷ It was arguably this dynamic that was present in some cases in the mid-2000s in Bahia, which now sees the highest homicide rates in the north-east: Paim, Jairnilson, Costa, Heloniza and Vilasbôas, Ana, ‘Política pública e controle da violência: um estudo de caso na cidade de Salvador, Bahia, Brasil’, Cadernos de Saude Publica, 25: 3 (2009), pp. 485–94CrossRefGoogle Scholar.

⁹⁸ Ratton, Galvão and Fernandez, ‘Pact for Life’.

⁹⁹ Interview, senior member of the SDS, Feb. 2013. ‘Spoilers’ were those who would lose out – materially or otherwise – through the introduction of new forms of accountability, including ‘old guard’ actors reliant on networks of patronage with legal and illegal actors and change-resistant bureaucrats opposed to a meritocratic bureaucracy.

¹⁰⁰ Interview, former journalist, Feb. 2013; interviews with senior academic, Feb. and March 2013. See also Wolff, ‘Policing and the Logics of Violence’.

¹⁰¹ See Ratton, Galvão and Fernandez, ‘Pact for Life’.

¹⁰² Macêdo, ‘Polícia, quando quer, faz!’

¹⁰³ Interview, senior member of the Courts of Justice, Pernambuco, March 2013.

¹⁰⁴ Interview, senior member of the SDS, Feb. 2013.

¹⁰⁵ Police were more closely vetted regarding involvement in illegal activity, for example, and promotions became meritocratic. Initial ‘resisters’, particularly within the police, either fell into line and acted in accordance with the new accountability ideals set out in PPV, or were demoted, ousted, or arrested. Interview, senior member of the SDS, Feb. 2013.

¹⁰⁶ Interview, senior member of the SDS, Feb. 2013. The author also attended several weekly meetings during fieldwork.

¹⁰⁷ This approach derived from hot-spot policing, a preventative methodology whereby extra resources and personnel are devoted to areas with higher crime: Braga, Anthony, Papachristos, Andrew and Hureau, David, ‘The Effects of Hot Spots Policing on Crime: An Updated Systematic Review and Meta-Analysis’, Justice Quarterly, 31: 4 (2014), pp. 633–63CrossRefGoogle Scholar.

¹⁰⁸ Important in this process was Campos’ leadership, who chaired PPV meetings once a month from inception until 2014.

¹⁰⁹ Interview, senior member of the SDS, Feb. 2013.

¹¹⁰ UN Human Rights Council, ‘Report of the Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions, Philip Alston: Addendum: Mission to Brazil’ (4–14 Nov. 2007), 14 May 2008, available at: www.refworld.org/docid/484d20cd2.html.

¹¹¹ A senior member of the PC remarked: ‘Police [were] also part of this context [of being investigated], and by doing this you show them that they are also vulnerable, that they can be arrested too. And we stick to it … The fact that [a police officer] can be arrested and made an example of, you show the others that going that way is not a good thing to do.’ Interview, March 2013.

¹¹² Statements that grupos had almost been eliminated came not only from high-ranking officers, but also from street-level police, academics and journalists, and many favela residents. Multiple interviews, Feb. and March 2013.

¹¹³ Interview, senior member of the SDS, Feb. 2013.

¹¹⁴ According to a senior judicial figure, March 2013: ‘Today the promotion of officers is made on merit … This meritocracy implemented by the state Government in the promotion of military and civilian police is a fundamental difference in Pernambuco.’

¹¹⁵ UN Human Rights Council, ‘Report of the Special Rapporteur’.

¹¹⁶ Targets and associated bonuses were tied to formal operational goals that focused on respect and protection of life and differed markedly from informal or extrajudicial ‘targets’ of killing criminosos that had been used in the past or elsewhere in Brazil.

¹¹⁷ During fieldwork several lower-ranking officers cited these bonuses as important in these regards.

¹¹⁸ Interview, senior member of the Courts of Justice, Pernambuco, March 2013.

¹¹⁹ Interview, senior delegado in the homicide investigation unit (Departamento de Homicídios e de Proteção à Pessoa – Homicide and Personal Protection Department, DHPP), March 2013.

¹²⁰ Ibid.

¹²¹ On the changes in police conduct, a community organiser remarked: ‘[Before] when the police cars came in, it was for sure there was going to be a shootout. They were not respectful. Today they come in on foot, walk straight in here … it has improved a lot.’ Interview, March 2013.

¹²² Interviews, community leaders in two favelas, Recife, March 2013.

¹²³ Interview, community leader, Recife, March 2013.

¹²⁴ Interviews, senior delegado in the DHPP and senior member of the PC, March 2013.

¹²⁵ Ibid.

¹²⁶ Since 2007 there has been a clear increase in identification and apprehension of suspects, cases brought to trial, and convictions. Prior to PPV only 3 per cent of homicides in Pernambuco were brought to formal trial. See UN Human Rights Council, ‘Report of the Special Rapporteur’. However, the penal system saw very little reform under PPV, and prison overcrowding is commonplace, despite improved case processing.

¹²⁷ Supporting the idea of strengthened deterrence effects, the Malhas da Lei (Mesh of the Law) initiative instigated by the Pernambuco SDS in 2010 focused on arresting ‘serious criminals’ or perpetrators of multiple homicides.

¹²⁸ Interview, senior member of the SDS, Feb. 2013.

¹²⁹ In numerous interviews in favelas across Recife, the general consensus was that communities were far safer places in the years following PPV, despite non-lethal criminality still being often problematic. As an example, Santo Amaro, considered to be the city's most dangerous favela, was transformed under PPV, with the government's official AIS-level statistics showing CVLIs declining from 178 per 100,000 in April 2007 to 44 per 100,000 in August 2013 – a trend supported by reporting in the local media. See, for instance, http://ww4.ufrpe.br/ruralnamidia_ver.php?idConteudo=6013 and http://jconline.ne10.uol.com.br/canal/cidades/noticia/2013/09/03/sds-registra-menor-indice-de-homicidios-nos-ultimos-dez-anos-96022.php.

¹³⁰ Interview, community leader in central Recife favela, March 2013. There is, however, still considerable distrust towards the police and the state among many residents, despite significant improvements. Interviews, community leaders and residents in several Recife favelas, March 2013. See also Wolff, ‘Policing and the Logics of Violence’.

¹³¹ Interview, community leader in central Recife favela, March 2013.

¹³² PPV was widely considered successful by both the state and citizens. During fieldwork, delegations from Uruguay, Paraíba and Bahia attended PPV meetings. Other states also implemented programmes modelled on PPV, notably PPV Bahia in 2012. This was less successful, however, probably due to resistance to reform and oversight amongst police, bureaucratic indifference, and lack of gubernatorial oversight. See www.insightcrime.org/news-analysis/what-makes-salvador-brazil-most-violent-city; and Paim et al., ‘Política pública e controle da violência’.

¹³³ The government of Pernambuco was recognised for ‘Improving the Delivery of Public Services’. See http://workspace.unpan.org/sites/Internet/Documents/Fact%20Sheet%202013%20UNPSA%20Programme%20Handbook%20_1_.pdf.

¹³⁴ Multiple interviewees within the police and government, for example, suggested that the expectations PPV had formed and the reductions in lethal violence it had fostered would make it near-impossible for future governors to drastically weaken the PPV model itself – regardless of future trends in lethal violence. A senior representative of the judiciary explained, with a flourish: ‘There are policies of government and there are policies of the state. Campos thinks holistically, plurally. He does not think to solve specific situations; he instead has a vision for a great future.’ Interview, March 2013.

¹³⁵ Elected in October 2012, Recife's mayor, Geraldo Júlio, also came from the PSB and was a key proponent in designing and implementing PPV. Potential benefits of vertical policy coordination are discussed in Hoelscher, ‘Politics and Social Violence’.

¹³⁶ This speaks to the importance of supporting violence reduction using both the formal aspects of the criminal justice system and other activities that positively impact on social, political and economic conditions. These can include activities related to urban planning and design, economic opportunity creation, youth-focused social programming, effective local governance, and creating social capital. Without this type of multi-sectoral approach, reforms may prove ineffective over the long term. See Hugo Acero Velásquez, ‘Os governos locais e a segurança cidadã’, UNDP consultation paper (2006), available at http://www.comunidadesegura.org.br/files/gestion_local.pdf; Kahn, Túlio and Zanetic, André, ‘O papel dos municípios na segurança pública’, Estudos Criminológicos, 4: 1 (2005), pp. 1–68 Google Scholar.

¹³⁷ Interviews, senior member of SDS, Feb. 2013; senior member of PC, March 2013. Reflecting early successes, CVLIs declined 4–5 per cent annually between 2007 and 2009 – prior to additional investment in policing and judicial capacity.

¹³⁸ Sentiments about popular mobilisation supporting political mobilisation were mentioned in several interviews.

¹³⁹ The north-east region has the greatest number of crack users in Brazil. See FIOCRUZ, ‘Estimativa do número de usuários de crack e/ou similares nas capitais do país’ (Rio de Janeiro, 2013), available at http://infograficos.estadao.com.br/especiais/crack/perfilusuarios.pdf.

¹⁴⁰ www.disquedenunciape.com.br/.

¹⁴¹ Confirmed also in interviews with members of the Departamento de Repressão ao Narcotráfico (Drugs Repression Department, DENARC), March 2013.

¹⁴² ‘Catching Up in a Hurry’, The Economist, 19 May 2011: www.economist.com/node/18712379.

¹⁴³ Waiselfisz, Mapa da violência 2012.

¹⁴⁴ This was mentioned in several interviews. See also Koster, Martijn, ‘Fear and Intimacy: Citizenship in a Recife Slum, Brazil’, Ethnos, 79: 2 (2014), pp. 215–37CrossRefGoogle Scholar. Conversely, the urban poor often have a more pragmatic view in dealing with insecurity. See Koster, ‘In Fear of Abandonment: Slum Life, Community Leaders and Politics in Recife, Brazil’, PhD Thesis, Wageningen University (2009).

¹⁴⁵ Ratton, Galvão and Fernandez, ‘Pact for Life’; and Ratton, José Luiz, ‘Apresentação seminário políticas locais de prevenção da violência’, in Ratton, José Luiz, Costa, Gino, Romero, Carlos and Soares, Luiz Eduardo (eds.), A segurança cidadã em debate (Recife: Provisual, 2012), pp. 10–23 Google Scholar (available at https:// www.academia.edu/4842389/Jos%C3%A9_A_Seguran%C3%A7a_Cidad%C3%A3_em_debate?auto=download).

¹⁴⁶ Implemented in 2003, Minas Gerais’ Fica Vivo (Stay Alive) programme was similar to PPV in combining progressive policing with social programming, yet with shallower reforms. See Alves, Maria and Arias, Enrique Desmond, ‘Understanding the Fica Vivo Programme: Two-tiered Community Policing in Belo Horizonte, Brazil’, Policing and Society, 22: 1 (2012), pp. 101–13CrossRefGoogle Scholar. Despite initial success, homicides and violent crime increased from 2010, highlighting challenges to long-term consolidation (http://www.seds.mg.gov.br/).

¹⁴⁷ See http://blogs.diariodepernambuco.com.br/segurancapublica/?p=9381; and http://jc.ne10.uol.com.br/blogs/rondajc/2016/01/15/pernambuco-tem-inicio-de-ano-mais-violento-da-historia-do-pacto-pela-vida/.

¹⁴⁸ http://jconline.ne10.uol.com.br/canal/politica/pernambuco/noticia/2016/09/24/pacto-pela-vida-morreu-diz-mentor-do-programa-254101.php (accessed 3 May 2017).

¹⁴⁹ Roxana Pessoa Cavalcanti, ‘“Over, Under and Through Walls”: The Dynamics of Public Security, Police–Community Relations and the Limits of Managerialism in Crime Control in Recife, Brazil’, PhD Thesis, King's College London (2017).

¹⁵⁰ See http://jconline.ne10.uol.com.br/canal/cidades/geral/noticia/2015/09/16/sensacao-de-inseguranca-ganha-as-ruas-199201.php.

¹⁵¹ Leeds, ‘Serving States and Serving Citizens’.

¹⁵² A concern for PPV is that if insecurity remains high, police tactics may become more hard-line, undermining institutional reforms and the emerging trust between police and communities.