Community Policing in Latin America : Lessons from Mexico City

Community policing programmes are widely perceived and promoted as an important solution for the pressing problems of insecurity in contemporary Latin American cities, and for improving citizen-police relationships. By drawing on the results of empirical fieldwork conducted in Mexico City, the article presents a critical analysis of the local community policing effort. The article demonstrates that this policing effort is overly determined by a local context, characterized by clientelism, political factionalism and police corruption, which therefore renders its contribution to a sustainable improvement of local accountability and police legitimacy unlikely. Against this background the article calls for more empirical studies on this topic and a greater sensitivity for the embeddedness of policing programmes within a wider political context.

In Latin America, the local democratization processes embedded in state downsizings, the state's withdrawal from economic and social welfare programmes, an increase of informal economic activities, and the dissolution of corporatist party and policy structures were all accompanied by a substantial increase in crime, violence (Koonings andKruijt 1999, 2004;Davis and Pereira 2000) and an '(un)rule of law' throughout the region (Nolte 2002, Méndez et al. 1999, O'Donnell 1999).Related studies indicate that these processes have converted Latin America into one of the most violent regions in the world (Portes andHoffman 2003, WHO 2002).Given the concentration of Latin America's population in urban areas, this 'new violence' has a predominantly urban face (IBRD 2008, Koonings and Kruijt 2007, Rotker 2002).Faced with this situation, Latin American urban policy makers increasingly look abroad for solutions to these pressing security problems.One important outcome of this 'looking abroad' has been the import of policing strategies and concepts, most of all from Anglo-Saxon countries.The most prominent respective policing imports have been zero tolerance and/or 'broken windows' policing efforts, which contributed to the hegemony of mano dura (strong hand) or tolerancia cero security agendas in many countries of the region (Glebbeek forthcoming, Müller 2009a, Corva 2008, Davis 2007, Dammert and Malone 2006).Whereas these concepts have been widely criticised by academics and civil society organizations for their repressive nature and the human rights violations, which frequently accompanied their implementation, another imported Anglo-Saxon policing model has received a more favourable response: community policing.
In fact, since the 1990s, an increasing number of Latin American countries have opted for a variety of community policing efforts.These efforts focus on enhancing the level of cooperation between local residents and the police at the neighbourhood level.Such projects have been implemented in Argentina, Brazil, Chile, Colombia, Costa Rica, Ecuador, El Salvador, Guatemala, Mexico and Peru (Feth 2008, Dammert and Malone 2006, Föhrig et al. 2006, Sozzo 2005, Dammert 2004, Eilbaum 2004, Frühling 2004).Although there is still considerable academic debate as to how to define community policing, the following citation captures the dominant expectations associated with this concept: Officers and neighbourhood residents approach each other as co-equal partners in the design and implementation of programmes designed to address local problems.Such 'co-production' practices can plausibly lead to greater effectiveness and greater legitimacy: effectiveness because the community would be more fully involved; legitimacy because the police would be open to democratic oversight (Herbert 2001, 448).
Therefore, community policing programmes not only promise an improvement of the urban security environment.In addition, they seek to make the governance of security more democratic, participatory and accountable, increase the local populations' confidence in the police, and improve police sensitivity to citizens' security concerns.
The increasing attractiveness of community policing for Latin American urban policy makers, I suggest, can be related to three interconnected dimensions.First, the hegemony of Anglo-Saxon policing discourses and practices seems to have become the global hallmark of 'a new crime control establishment that draws upon the new criminologies of everyday life to guide its actions and mould its techniques' (Garland 2001, 17).Second, and closely related, the emergence of an international development discourse -with obvious neoliberal undertones -calls for police reform measures and promotes the principles of ownership, participation and accountability (Buvini et al. 2005, 11-13; for an important critique, see Stanley 2006).Finally, in the newly democratized political environment of Latin America, policy makers and police reformers are struggling more or less successfully -and with more or less political commitment -with the heritage of authoritarian rule and 'political policing', which have seriously undermined confidence in the local police institutions and their credibility in addition to the problems of frequent and endemic police abuse and the participation of police officers in criminal activities and large scale corruption.
These dimensions have mutually reinforced each other and contributed to the emergence of a political consensus among many academics, local NGO activists and policy makers regarding the usefulness of community policing strategies as an 'international best practice', whose implementation throughout Latin America is not only perceived as a promising and sustainable security effort, but also as an important contribution to rebuilding trust in the local police forces.
What these different community policing promoters frequently ignore is the fact that even with regard to Anglo-Saxon countries that first developed community policing strategies, the alleged 'success' and concrete achievements of community policing programmes are still an unresolved matter of dispute (Herbert 2006, Stanley 2006, Waddington 1999).Furthermore, the Latin American community policing community also tends to ignore critical evaluations regarding the implementation of community policing programmes in other regions of the so-called 'developing world' (see for example Lau 2004, Ruteere andPommerolle 2003).Summarizing much of these criticisms, Mike Brogden and Preeti Nijhar, in their detailed analysis of community policing efforts in 'transitional and developing countries', not only indicate that community policing is situated in the long and problematic tradition of 'exporting used goods from the Western supermarkets of policing and other legal institutions to so-called developing and transitional countries'.They further suggest that it is 'at best, unproven practice.At worst, it is simply a practice that reinforces existing schisms and inequalities' (Brogden and Nijhar 2005, 229, 161).
Most of these critical comments, it seems, have been ignored by the current literature supporting and promoting the import of community policing to Latin America.In addition, much of this literature is characterized by a problem-solving orientation whose principal concern is the promotion of concrete action (e.g. the implementation of community policing programmes), and not in-depth analysis of the community policing concept itself (Feth 2008). 2 This, of course, does not imply that these studies are not aware of possible problems and shortcomings of community policing efforts.However, these problems are most of all perceived as technical problems that can be resolved through more efficient and sustainable implementation procedures (see for example Frühling 2004, 28-36).This primarily technical and problem-solving orientation, I suggest, contributes to an important shortcoming of most of the literature advocating large-scale implementation of community policing programmes in Latin America.This shortcoming resides in the absence of a deeper contextualization of local community policing efforts and in the lack of sensitivity to the impact of established political, social and policing structures on such programmes, as well as to the political appeal of such programmes for local policy makers -beyond immediate security concerns.This paper wants to address this shortcoming by offering a critical analysis of a community policing project in Mexico City, the so-called Policía de Barrio programme.

Community policing in Mexico City: the context
Mexico City has recently been labelled as 'a relative newcomer to the community policing philosophy' (LaRose 2006, 286).This view has to be corrected.The Moral Renovation Campaign of Mexican president Miguel de la Madrid (1982)(1983)(1984)(1985)(1986)(1987)(1988) already included a project of community-oriented neighbourhood policing (Policía de Barrio) in the 1980s.This project, it was hoped, would create new confidence in the police forces by promoting the image of the local cops on the beat as 'everybody's friend' (SEGOB/Presidencia 1988, 69-70).The existing documents and literature do not permit an appraisal of the project's success or failure.However, it seems that the project was more and more disregarded and finally abandoned.The restoration of this policing effort had to wait until the beginning of the new millennium, when a community policing revival, closely related to the local democratization process, occurred in Mexico City.To understand this relationship we first have to take a closer look at the local democratization process itself.This, in turn, requires an understanding of the pre-democratic characteristics of Mexico City as a starting point.Without going into detail, the following citation captures the most important features of the pre-democratic political structure, dominated for nearly 70 years by the Partido Revolucionario Institucional (Institutional Revolutionary Party, PRI): From 1928 to 1997, the Federal District had no independent rights and became a centralized, managed entity under federal power.The mayor, or regente, was hand-picked by the president and the 16 delegaciones or precincts had delegados hand-picked by the mayor, without a city council to offer the necessary supervision or framework of accountability.The mayors ranged from political heavyweights to close friends of the president, and prior to 1988 not a single one of them was a Mexico City native.The political structure was organized around issues of political and social control, caving in only when necessary and managing dissent in the same fashion as the PRI ruled the country as a whole (Castillo 2008, 184).This structure of governance was dramatically altered in 1997, when, due to constitutional changes, the local population could directly elect the mayor of Mexico City for the first time since 1929.In this election, the majority of the residents voted for Cuauhtémoc Cárdenas from the Partido de la Revolución Democrática (Party of the Democratic Revolution, PRD).When Cárdenas was elected mayor of Mexico City, he was expected to improve the local security situation that the majority of the local politicians and Mexico City residents perceived as having been in a state of constant deterioration since the mid 1990s (Pansters and Castillo Berthier 2007, 41;Davis 2006, 65;César Kala 2000, 222-223).This development was widely interpreted as evidence of the failure of the security policies of the preceding PRI administrations and the deficiencies of their authoritarian, abusive and highly corrupt police apparatus. 3 Probably the most important factor for this political success was the PRD's discursive commitment to democratization, which placed an important emphasis on citizen participation (Davis and Alvarado 2004, 136).Cárdenas had already promised a more democratic government and a participative governance model during his electoral campaign that would differ significantly from the decades of authoritarian politics under the PRI.This focus was clearly expressed in his campaign motto 'Juntos!Gobernaremos la ciudad' (Together!We will govern the city).A cornerstone of this effort was the introduction of the principle of effective citizen participation in local politics.According to this official democratic commitment, the local population would also receive more opportunities to participate in local security matters.This participation, in turn, was expected to contribute to a closer police-citizen relationship, which would then result in a more successful fight against criminality.
In this regard, the introduction of a community policing programme by the Cárdenas administration, which found its expression in the creation of the policía comunitaria (community police), served as the ideal strategy to combine these objectives (Davis 2003, 20).Following official discourse, the police units of the Secretaría de Seguridad Pública del Distrito Federal (Secretary of Public Security for the Federal District, SSPDF) assigned to this programme were to patrol the neighbourhoods most affected by the current wave of crime.Through their presence in these areas, it was expected that they would contribute to the establishment of a closer and better relationship between the local population and the police, thereby rebuilding trust and confidence in the police apparatus and enhancing the capacity of the latter to fight against criminality.'To accomplish this, the SSP sec-retary selected the "highest-crime areas" and gave officers a three-month training course to teach them to coexist with the members of the community.Selected patrol cars and police guard stations were painted with the words "Community Police", and that was all that was done' (Arroyo Juárez 2007, 430, emphasis added).
Notwithstanding these ambitious objectives, the programme soon fell by the wayside in the following years.However, due to the still unresolved security (and police) problems, as well as the related citizen mistrust in the local police forces, this programme was reinstituted and renamed a few years later under the PRD government of Andrés Manuel López Obrador (2000)(2001)(2002)(2003)(2004)(2005).The programme, formally established in 2003, was now called Policía de Barrio.In his inaugural speech, the then-police chief (and current mayor of Mexico City) Marcelo Ebrard stated the centrality of re-establishing a police-community relation based on mutual confidence and trust as an essential element of the local policing strategies.The creation of the Policía de Barrio was to serve this purpose.
We have said, and that was also the order of the mayor, that the most ambitious objective of every police officer is support from the community in their actions.In many circumstances, the police and the community distance themselves from each other.One loses confidence, contact and communication.[…] [T]he principal instrument for combating crime is this confidence, which is the support of the community for the actions of the police, and, on the other hand, the support of the police for the community.We call this programme, with this ambitious goal, Policía de Barrio, because this is how the police have operated in Mexico City for a long time, and this is what has been lost for some decades for different reasons.We have lost this contact, and the result has been increasing crime indices; a sense of isolation within the police when they do their work; the perception or sense from the community that their legitimate demands are ignored.
[…] Therefore, we need to overcome this distance.This is the most important effort that we can promote as a security strategy.We can have the best equipment, but if we do not construct this bridge and this mutual support, it will be very difficult to achieve positive results.In this respect, we take up the tradition of our city of having a neighbourhood police, and we update it.This is what the international experiences, be it of Japan, Chile, or Spain with whom we have been working, tell us.This is what the oral traditions tell us.We did not undertake a study of this topic in Mexico City, but we recall that in the 1950s and throughout most of the 1970s, the police had a very high level of support form the community (SSPDF 2003).
As explicitly acknowledged in this speech, the re-establishment of this programme was not accompanied by an evaluation of its predecessor or the structural conditions of Mexico City and its police apparatus.Instead, as the passage quoted from the then-police chief demonstrates, the decision to implement this programme was based on what was heard from the experiences of other countries, as well as on an evocation of an idyllic, golden-age-like picture of the 1950s and 1960s, 4 when the population of Mexico City was said to have had confidence in their police forces (Arroyo Júarez 2007, 430).Notwithstanding this lack of scientific and professional evaluation, during the following years, the Policía de Barrio programme was implemented throughout the city.

The Policía de Barrio in theory
The main objective of the Policía de Barrio programme, according to official statements, can be identified as the (re)establishment of confidence in the local police forces by bringing them into closer contact with the local population and by making them more accountable to the local residents through permitting the latter a voice in the evaluation and planning of police work and strategies.These steps are perceived as essential components of a more efficient strategy in fighting local criminality.An official police document offers the following description of the Policía de Barrio programme: The goal of the Policía de Barrio model is to generate profound changes with respect to police conduct, and in this respect it is one of the biggest challenges.It is based on approximation, communication and confidence between the police and the citizenry in order to serve [the citizenry's] demands and to organize solutions.The goal is the sustainable consolidation of the proximity police model in most areas of the city.The Policía de Barrio officer must be known by the inhabitants whom he serves and with whom he needs to achieve constant communication, confidence, proximity, and, most of all, openness [apertura] to listening to their demands, complaints and suggestions, in order to guarantee that these contribute in the most efficient manner to decisions and to the most pertinent solutions.The central axis is the contact between the police and neighbours -with a high degree of mutual integration -which seeks to overcome the communicational void which only benefited delinquents (SSPDF 2006, 225).
In 2006, the Policía de Barrio operated with 1,836 police officers, drawn from the Preventive Police (about 6 per cent of its active members), which were assigned to 169 patrol areas (characterized by high criminal indices), in 15 of the 16 boroughs of the Federal District (SSPDF 2006, 225-226).The basic police work of the officers assigned to this programme is rather simple.According to information provided by members of the local administration of justice and local NGOs, it consists of a pair of beat cops assigned to a certain patrol area, whose presence and patrol turns have to be confirmed by selected residents (Código Águila).In addition, members of the Policía de Barrio participate in the meetings of the local comités vecinales (neighbourhood committees).This participation, in turn, was to contribute to the evaluation and reorganization of local police strategies according to the security needs of the communities.However, although the Policía de Barrio is assigned to a certain area in a respective Mexico City borough, it is important to stress that neither the precinct mayors nor the local programmes for crime prevention and citizen participation have direct control over the Policía de Barrio agents, who remain under the central authority of the SSPDF.
The neighbourhood committees were created as a central element in the citizen participation strategies of the government of Cuauhtémoc Cárdenas on the basis of the Ley de Participación Ciudadana del Distrito Federal (Law of Citizen Participation in the Federal District, LPC) passed in 1999.On 4 July 1999, the electorate of Mexico City voted to elect the representatives of the 1,352 neighbourhood committees.However, only about 10 per cent of the 6.1 million voters participated in the elections (Davis and Alvarado 2004, 139, 150).The neighbourhood committees are collegial bodies that consist of one coordinator and between six and fourteen committee members, their number depending on the number of registered voters in the neighbourhood.The committees are expected to articulate and represent the interest of the neighbourhood vis-à-vis the Federal District authorities.Furthermore, they are expected to inform the neighbours about government programmes and initiatives that concern the neighbourhood.This implies that they, at least partially, substitute the functions of intermediary institutions -like political parties -at the local level (Harbers 2007, 45).
The role of the neighbourhood committees with regard to questions of public security were established in Article 80 of the LPC.This article, although in rather vague and imprecise terms, states that the neighbourhood committees are responsible for the 'verification of public security programmes'.In 2004, the law was modified, the title of the neighbourhood committees was renamed as Comités Ciudadanos (Citizen Committees), and their role in citizen participation was redefined.It is Article 88 of the new law, which defines the role of the committees as being to 'emit opinion concerning the public security and administration of justice programmes of the territorial coordinations'.In addition, Article 92 states that the internal organization of the committees has to include a Coordinación de Seguridad Ciudadana y Prevención del Delito (Coordination for Citizen Security and Crime Prevention).
These legal changes have been explained in interviews with administrative personnel at the borough level and with NGO members as a political manoeuvre by the local government to come to terms with the criticism concerning the fact that, though by law the members of the neighbourhood committees should be re-elected every three years, there have been no re-elections since.All of my interview partners on this topic continued to use the term 'neighbourhood committees' and stated that the legal reform had not substantially improved the public security related faculties of the committees that serve as the primary basis of their contact with the Policía de Barrio programme.

Community policing in practice: clientelism, mistrust and politics
As community policing programmes emphasize the crucial role of citizen participation and accountability, and because in the case of the Policía de Barrio the neighbourhood committees are the single most important aspect in this regard, it makes sense to take the neighbourhood committees as a starting point for addressing the question of how the programme works in practice.As mentioned above, the concrete legal function of the neighbourhood committees with regard to their participation in local public security provision is vaguely defined.This implies that in the absence of a clearly defined set of legal functions, their respective activities, as well as the resulting outcomes, depend on the concrete local context as well as the practices of their members.Taking into account the fact that the democratization of the Mexico City politics was not accompanied by the dissolution of political clientelism and paternalism that characterized Mexico's capital city politics under the decades of PRI one-party rule (Hilgers 2008;Bartra 2007, 64-69;Durand Ponte 2007;Schütze 2005), it should be no surprise that the activities of the neighbour-hood committees are also haunted by clientelistic practices (Zermeño et al. 2002, 245-251;Rodríguez Luna 2007, 244-245).The lasting impact of these relations can be easily identified in the realm of policing and the activities of the Policía de Barrio.Members of neighbourhood committees and also representatives from the local administration of justice interviewed for this study frequently stressed that in many cases the members of the neighbourhood committees tended to 'privatize' the Policía de Barrio officers assigned to their neighbourhood and use them for private purposes.For example, Antonio, 5 an ex-director of a local crime prevention programme, stated with respect to his experience with the local neighbourhood committees: In reality, there was a lot of corruption going on in the subcommittees [of public security].Members of the subcommittees had the Policía de Barrio agents practically under their private authority.They converted those police officers into private vigilantes, keeping an eye on their houses, cars, shops etcetera.All of this was highly arbitrary (interview April 2008).
Such behaviour is generally tolerated by the local authorities, as it represents a crucial feature of the practices of negotiations or gestiones that still stands at the heart of capital city politics in Mexico City: in exchange for the delivery of services to local residents, the latter offer political support to people capable of providing these services.However, this relationship is frequently not a direct one, but mediated through local brokers or intermediaries.Their power is based on political capital derived from their capacity to mobilize people.This in turn enables them to function as informal spokesmen of local-level collective interests and to represent these interests to state authorities.This capacity gives such brokers a privileged access to state personnel and resources that they can then appropriate to a certain degree for their private and political purposes (Müller 2009b, chapter 8;Hilgers 2008, 137).The following passage from the interview with Antonio offers us an inside view of these practices and their seemingly inevitable nature within the context of Mexico City politics: Yes, this system obviously produces a vice of certain privileged relationships, but this goes on everywhere.There are certain personalities, certain groups, which manage to establish a direct relationship with some public official and, well, this opens the door to a more expedient path.For example, we have a good relationship with a Manuel, a local member of congress.When the people asked for more light on one street, well what did we do?We asked the people to send a request to the person responsible for the sector [jefe del sector], the local chief of public security.But this is nothing more than a formality.We picked up the phone, called the person responsible for the sector, whom we know and who is familiar with our relationship with Manuel, and, well: 'Listen chief, these people need these things.When can we meet to speak about this with you personally?' And, well, isn't it great that things can be taken care of so easily?What we do, and this is completely within the same logic, is take advantage of a privileged contact in order to solve a concrete problem.
In such a context, the political and legal nature of the neighbourhood committees converts them into ideal sites for the concentration of these practices of negotiation and into privileged spaces for the arbitrary appropriation and distribution of public (security) resources for private and political purposes.Against this background, it is not surprising that when Pamela, a local member of a neighbourhood committee in the middle-class neighbourhood of Coyoacán, with political ambitions, was asked what the police do in order to improve the local security situation, she replied: 'Well, of course they'll send me officers, because I have the phone number of the local police chief [jefe del sector].I call him, they come' (interview September 2007).
Such forms of privileged access to and the private appropriation of public security resources, however, are not exclusively the domain of neighbourhood committees in middle-class districts.For example, Rodrigo, a shopkeeper and former member of a neighbourhood committee in the marginalized Mexico City borough of Iztapalapa, explained that during the time he was active in the neighbourhood committee, he organized other shopkeepers in his area to collect money to be passed on to the local Policía de Barrio officers to keep an eye on their commerce zone and have 'a little bit more surveillance out here' (interview November 2007).
Under such conditions, it is evident that the evaluation of the Policía de Barrio work by the neighbourhood committees, as well as the resulting police-community relations, may frequently be based more on the neighbourhood committee integrants' personal interests than on the security concerns of the entire community.In general, local members of the administration of justice and NGO activists remained sceptical as to if, when taking into account their political history and embeddedness in broader political structures of clientelism and local political bossism, the neighbourhood committees could be expected to function as authentic representative bodies of the respective neighbourhoods and as efficient accountability institutions for the Policía de Barrio.This absence of real accountability mechanisms perpetuates the institutional problems of the local police forces, such as informality or involvement in criminal and corrupt practices (on these topics see Müller 2009b, Davis 2006, Martínez de Murguía 1999), thereby making the expected positive impact on police-community relations, as well as the positive contribution to the local security situation, questionable.
However, it is important to stress that the informal privatization of the Policía de Barrio and the resulting fragmentation of their security provision is not limited to the activities of the neighbourhood committees.Rather, Policía de Barrio officers can be 'appropriated' by a broad variety of actors, ranging from shopkeepers to ordinary citizens.Although, the neighbourhood committees undeniably have a privileged access to the Policía de Barrio agents, this privilege does not imply exclusivity.In a local context where the majority of the police officers opted for this employment out of a general interest in 'making money' (Artega Botello and López Rivera 2000), 'buying' private protection from the police is a rather common phenomenon (Pansters and Castillo Berthier 2007, 45;Anozie et al. 2004, 4), and the Policía de Barrio seems to be no exception in this regard.Therefore, the observation made by Marcos, a member of a local NGO, who explained that in prosperous zones of the Federal District parts of the local population would 'donate' money to the Policía de Barrio officers, who, motivated by these financial incentives, in turn showed a larger and more regular presence in these areas than in more marginalized zones, suggests that money-driven market logics also (nega-tively) affect the quality of the local community policing effort and contribute to a high degree of spatial selectivity and geographical fragmentation (interview March 2006).
The informal appropriation of police officers is not the only problem related to the activities of the Policía de Barrio.Other features include arbitrary and even criminal behaviour, frequently contributing to widespread negative perceptions of the Policía de Barrio programme.For example, Rafa, a shopkeeper in the middleclass neighbourhood of Colonia del Valle explained that although the Policía de Barrio officers assigned to the zone have a clearly defined patrol area and schedule, in practice, the agents show up whenever they want.Although in theory their patrols have to be individually confirmed after completion by selected local residents, in practice, he (in his function as such a controller) has been frequently 'asked' by the local Policía de Barrio agents to confirm all of their patrols for the entire day at one time.After they received the signature, the agents would not return that day in order to continue with their patrols.These facts, according to his opinion, seriously undermine not only the efficiency but also the credibility of the programme (interview September 2006).
A further element that seems to discredit the local implementations of the Policía de Barrio programme was mentioned in an analysis of neighbourhood reactions towards insecurity problems by the NGO Democracia y Derechos Humanos.The authors refer to an incidence in the marginalized Mexico City borough of Iztapalapa, in which the residents mentioned that the local Policía de Barrio agents are involved in extortion of local residents and delinquents (Alvarado et al. n.d.,29).In this regard, a spokesman of a local business organization in the middleclass borough of Coyoacán declared that after having been informed by local authorities about their plans to assign Policía de Barrio units to patrol his residential zone, he as well as other members of the organization vehemently (and successfully) rejected this proposal based on fears that the permanent presence of and observation by Policía de Barrio agents would permit the latter to gather sensitive information concerning the daily routines of the residents and therefore, taking into account the bad reputation of the local police forces and their frequent connections to organized crime, would represent a serious security risk for them and their families (interview December 2007).However, as the following interview passages demonstrate, members from the lower social classes and marginalized segments of the local population have equally negative views of the Policía de Barrio -although for different reasons.For example, Rebecca, a sex worker working in downtown Mexico City, stated: Preventive Police, Policía de Barrio, Judicial Police, that doesn't make a big difference.Customers are a problem, most of all, when they refuse to pay, but the authorities represent another risk for us.They try to extort us.When you refuse to pay, they take you to the delegación, where they can keep you for 24 hours without any possibility of making a phone call.They tell you that you have been arrested for committing a crime.To achieve a good relationship with the police, you have to give them the money they want and everything else they ask you to.There are a lot of colleagues that had to enter into a sexual relationship with a policeman (interview October 2006).
And Héctor, a resident from the marginalized Mexico City neighbourhood of Iztapalapa, mentioned that due to the behaviour of Policía de Barrio officers in his neighbourhood, local residents perceive their presence as an aggression.
[There] is a programme that they call Policía de Barrio, where you have police officers who walk around and keep an eye on the neighbourhood, but in reality they are not very efficient, because when the people here see eight police officers walking around, they perceive this as an aggression.It's not that the Policía de Barrio officers show up and talk colloquial with, let's say an old women, as friends [cuates].No, when they come, they are very indifferent to what's going on here.So with this attitude they will never be able to integrate with the community.
You just said that the Policía de Barrio operates here with groups of eight people?
Well, in other neighbourhoods I was told that there are only two police officers on patrol.Here it seems that this is somehow different?Yes!And they don't show up every day.Sometimes they come once in a week, sometimes every two weeks.Policía de Barrio... Yes...Here are some police posts [módulos de vigilancia] that belong to the citizen participation programme.In these posts there's sometimes a Policía de Barrio officer, but this guy locks himself up in his post.So, he is safe, but in the street before him, there's no security.There's no security because he is inside his post and doesn't make his patrols.When they finally go on patrol, well then always in a group of many, and the people here have an aggressive view about this (interview September 2007).
However, it seems that such perceptions of mistrust, mentioned by Héctor, are not one-sided, but mutual.As Carlos, a local policeman from the same borough, explained, due to the prevailing lack of confidence in the police by the local residents in his patrol area, he would prefer not to make his patrols on foot, as required by the Policía de Barrio programme, in order to avoid problems with the residents.Instead, he uses a police car for collecting the signatures for the Código Águila (interview March 2008).
Against the background of the aforementioned observations, it is of little surprise that Fernando, an instructor of the National Institute for the Study of Criminal Science (Instituto Nacional de Ciencias Penales), a public agency that offers training for public servants involved in the prosecution and administration of justice, gave the following description of nearly six years of community policing in Mexico City: The only thing the SSPDF has done was to put more police on the streets.More police presence on the streets of certain neighbourhoods.This has nothing to do with approaching the citizens.There is no direct contact with the citizens.All they do is give the policemen a paper with which they go to the store and ask the clerk for a signature, they go the pharmacy and do the same, they go the beauty salon and ask the owner for a signature.This is what the Policía de Barrio is all about for them (interview December 2009).This outcome, in addition to the abovementioned problems of clientelism and police corruption, which according to my interview partners are well known to the respective authorities, raises the question why this programme continues to be presented and actively promoted by the local government and the SSPDF as a key strategy for addressing the security demands of the local citizens.A possible and convincing explanation of this paradox was given by Ramón, a local crime prevention functionary.He explained that the creation of the Policía de Barrio programme should not be interpreted as a serious commitment to citizen participation in security issues.Rather, it represents the symbolic intent to demonstrate a political will to improve the local security situation in a democratic way and to express sensitivity to the respective demands of the local civil society by avoiding the transformation of the traditional structure of the local Preventive Police.This permits a kind of illusory decentralization by maintaining central political control over the local police apparatus at a time when local politics are marked by growing inner-party rivalries between the different factions of the PRD (interview July 2007).In a similar vein, Antonio, the ex-crime prevention functionary introduced above, stated: In the end, there is no interest in converting public security into something more closely related to the community.There is an interest, a necessity, and a decision for control.They [the government of the Federal District] are convinced that this is something necessary, and because of this, they are not interested in participation or even at the very least in interlocution with the community.The police in Mexico are very corrupt, and because of this, I think that there is no will from the government of the Federal District to resolve these problems.This is why they don't want better local control and supervision.With a centralized police structure, the control of the corrupt superiors over their agents in the Federal District is much more efficient (interview April 2008).
In this respect, NGO activists and members from the local administration of justice interviewed for this study indicated that the Policía de Barrio projects reflect a more general development in the realm of local policing, a development which can be described as 'symbolic policing'.In Mexico City, local authorities increasingly tend to address security problems and critical issues in the local law enforcement agencies without a serious commitment to structural reforms.They are instead concerned with publicly visible and displayable measures such as the acquisition of new, more fashionable uniforms, new equipment, or the creation of new police units.The increasing attractiveness of such forms of symbolic policing, as the interviews indicated, is closely related to the local democratization process, the related pluralization and competition among local and national political actors, as well as the resulting politicization of security issues in Mexico City.Leaving aside political conflicts between the local city government (PRD) and the federal government (first PRI, later the Partido Acción Nacional, the National Action Party, PAN) this politicization is also reflected inside the governing architecture of Mexico City and is most evident at the level of the boroughs.Here, internal factional-ism within the dominant party (PRD) frequently determines not only the supply and security equipment assigned to specific boroughs, it also impedes a further decentralization of policing faculties and the creation of impartial accountability structures at this administrative level, because the government of Mexico City prefers a centralized police structure that can more easily facilitate the maintenance of a certain degree of control over the local public security apparatus.However, in order to demonstrate their commitment to democracy and citizen concerns in a political conjuncture where security problems rank among the most important political issues in local opinion polls, local politicians tend to create seemingly 'innovative' and 'progressive' police units and programmes (on the historical roots of this pattern of 'symbolic policing', see Davis forthcoming).These programmes, however, due to their embeddedness in the prevailing political context and the absence of serious political will for a structural police reform effort, are in many cases a failure by design.Therefore, such efforts like the Policía de Barrio, 'may serve for show as much as for anything else' (Davis and Alvarado 2004, 149).

Conclusion
The active participation of citizens in the provision of public security through the implementation of community policing programmes has become an important urban policing strategy in contemporary Latin America, actively promoted by local politicians, NGOs, national and international think tanks and academics.This support is in large part due to the assumed potential of this 'international best practice' to contribute to a more efficient, democratic and accountable policing, and to improve citizen-police relations.
This article presented a critical analysis of a Mexico City community policing programme, the so-called Policía de Barrio project.By confronting the officially stated goals and objectives of this community policing programme, with the community policing practices on 'the ground', this article demonstrated that the experiences in Mexico City significantly differ from the widespread positive expectations regarding the democratic potential of community policing programmes and their contribution to improving citizen-police relations.First, the Mexico City case study demonstrated that the local community policing efforts are overly determined by established structures of clientelism and police corruption that permit and facilitate the private appropriation/distribution of community policing resources for private purposes.Second, the case study also showed that the programme's contribution to an improvement of citizen-police relations is doubtful.Far from finding a harmonious cooperation between police and citizens, contributing to a mutual beneficial co-production of security, we encountered descriptions of illegal, unduly abusive conduct from Policía de Barrio officers, as well as strong sentiments of mutual mistrust between community policing officers and local residents.Finally, the findings of this article stress the centrality and negative impact of politics.The abovementioned observations suggest that the Policía de Barrio programme should most of all be interpreted as an effort of 'symbolic policing' aimed at the public and as a discursive display of an 'authentic' commitment to citizen participation and to democratization, without a real political will to confront the structural problems of the local police forces or to enhance police accountability.
Therefore, the findings of this article support much of the scepticism regarding the alleged democratic potential of community policing efforts and their global export that is frequently ignored by Latin American community policing promoters.Against this background, the present paper should be read as a call for empirical studies on community policing efforts in Latin America that are more sensitive to the embeddedness of such policing imports in wider socio-political relations and local 'cultures of control'.In fact, in Latin America, as well as in other regions of the so-called 'developing world', 'most of the analyses relevant to policing are driven by today's policy requirements, or are rooted in an uncritical liberalism; nuanced theoretical analyses are missing, as are ethnographic studies' (Hills 2009, 212).In order to overcome the resulting shortcomings, future research should try to address these topics by moving beyond the 'excessive formalism' (O'Donnell 2006, 287) which haunts most of the related studies and their frequently apolitical and de-contextualizing treatment of security and policing issues in contemporary Latin American cities.  [11][12][13][14]2009.I would like to thank Ruth Stanley and the two anonymous reviewers for their helpful comments and Frank Müller for his invaluable research assistance.2. That Latin American policy makers, academics and NGO activists are searching for solutions to the pressing local security problems is understandable and important.However, this search in many cases, leads to an uncritical affirmation of the presumed efficiency of Anglo-Saxon policing models, representing what Wacquant called the 'new security doxa', which manifests itself 'in the form of a concatenation of "scholarly myths" ', that is, according to Pierre Bourdieu, a web of statements that intermingle 'two principles of coherence: a proclaimed scientific appearance, which asserts itself by proliferating outward signs of scientificity, and a hidden coherence, mythic in its principle' (Wacquant 2009, 247).3. On these topics, see Davis (forthcoming, 2006), Piccato (2001) and Mártinez de Murguía (1999).
For a formal description of the Mexico City police forces, see Reames (2007).
Müller is lecturer in politics at the Institute for Latin American Studies, Freie Universität Berlin.His research focuses on (in)security, policing and state-society relations in Latin America.Recent publications include: 'Wenn 'Null Toleranz' und 'Zerbrochene Fensterscheiben' auf Reisen gehen: Globalisierung und die Restrukturierung des historischen Zentrums in Mexiko Stadt', in: Kriminologisches Journal, 41/2 (2009), pp.82-99 and 'The Struggle over Human Rights in Mexico', in: Hoffmann-Holland, Klaus (ed.) (2009) Ethics and Human Rights in a Globalized World.Tübingen: Mohr Siebeck, pp.43-62.<muellerm@zedat.fu-berlin.de>Notes 1. Research for this article was conducted between 2006 and 2009 within the project 'Public Security as Governance?Policing in Transitional and Developing Countries', of the Research Centre (SFB) 700: 'Governance in Areas of Limited Statehood -New Modes of Governance?' funded by the German Research Foundation (DFG) and located at the Freie Universität Berlin.An earlier version of this paper was presented at the XXVIII International Congress of the Latin American Studies Association (LASA), Rio de Janeiro, June