THE WORLD BAN K Urban Poor Perceptions of Violence and Exclusion in Colombia 20747 July 2000 CAROLINE MOSER CATHY McILWAINE FOREWORD BY ANDRES SOLIMANO *1~~~~~~~~~~~~~ CONFLICT PREVENTION AND POST-CONFLICT RECONSTRUCTION Urban Poor Perceptions of Violence and Exclusion in Colombia Caroline Moser Cathy Mcllwaine Foreword by Andres Solimano Latin America and Caribbean Region, Environmentally and Socially Sustainable Development Sector Management Unit The World Bank Washington, D.C. i 2000 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433 All rights reserved Manufactured in the United States of America First printing July 2000 1234 0403020100 The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organiza- tions, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data included in this publication and accepts no respon- sibility for any consequence of their use. The material in this publication is copyrighted. The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. Permission to photocopy items for internal or personal use, for the internal or personal use of specific clients, or for educational classroom use is granted by the World Bank, provided that the appropriate fee is paid directly to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA.; telephone 978-750-8400, fax 978-750-4470. Please contact the Copyright Clearance Center before photocopying items. For permission to reprint individual articles or chapters, please fax a request with complete information to the Republication Department, Copyright Clearance Center, fax 978-750-4470. All other queries on rights and licenses should be addressed to the Office of the Publisher, World Bank, at the address above or faxed to 202-522-2422. Cover photo: Teen gang members "converted" to militias with 7.65 pistols, Medellin, Colombia 1991. Timothy Ross/Imageworks. Caroline Moser is a head specialist, Social Development, Latin America and Carribean Region at the World Bank. Cathy Mcllwaine is Lecturer at the Department of Geography, Queen Mary and Westfield College, London. She was on a one year sabbatical to the World Bank to carry out this research. Andres Solimano is Country Director for Colombia, Ecuador, and Venezuela, The World Bank. ISBN 0-8213-4731-4 Library of Congress Cataloging-in-Publication Data Moser, Caroline O.N. Urban poor perceptions of violence and exclusion in Colombia / Caroline Moser, Cathy Mcllwaine. p. cm. - (Conflict prevention and post-conflict reconstruction) Includes bibliographical references. ISBN 0-8213-4731-4 1. Violence - Colombia - Public opinion. 2. Urban Poor - Colombia - Attitudes. 3. Public opinion - Colombia. 4. Violence - Colombia. 5. Marginality, Social - Colombia. I. Mcllwaine, Cathy, 1965- II. Title. III. Series. HN310.Z9 V5526 2000 303.6'09861-dc2l 00-04632 Contents Foreword ..................................... v Acknowledgments ............................. xi Executive Summary ............................. 1 1 Introduction ................................... 7 2 Summary Findings: Perceptions of General Problems and Violence ......................... 15 3 Social Violence in the Family and Household ....... 31 4 Economic Violence and Drug Consumption ......... 43 5 Unemployment, Exclusion, and Economic Violence ... .57 6 Perverse Social Capital and Economic and Political Violence ............................. 71 7 Youth, Exclusion, and Violence ................... 93 8 Community Perceptions of Solutions to Violence ....99 Notes ...................................... 113 Annexes .................................... 115 Bibliography ................................ 121 iii Foreword Societies in Crisis, Globalization and Violence Andres Solimano Societies in crisis: Concept and scope A society in crisis presents both a challenge and an opportunity. This is certainly the case in the context of Colombia, now in the fifth decade of its bitter civil war. The crisis in which Colombia finds itself today rep- resents an integral challenge to the economy, the institutions, and the values of its society. At the same time, it provides an opportunity for changing structures that no longer work. Although this may vary according to different interests in society, the needs of the most exclud- ed and poorest are often marginalised. Yet, as this study shows, they have very clear perceptions of those structures that they consider important to change The elusive concept of crisis that underpins this study has long been an important concern. The German philosopher Jurgen Habermas for instance, provides a useful taxonomy of systemic crisis comprising a fourfold classification of (a) economic crisis, (b) rationality crisis, (c) legitimation crisis, and (iv) motivation crisis (Habermas 1972). An eco- nomic crisis, termed a "realization crisis" by Habermas, may be a reces- sion, an economic depression, or an inflationary phenomena; a rational- ity crisis, in Habermas's words, is a breakdown of the "rational admin- istrative" practices necessary to maintain the economy in due course. This can be interpreted as the inability of the government in a crisis sit- uation, to properly manage and regulate the economic system. A legiti- mation crisis, in turn, is characterized by a breakdown in the level of public support, credibility, and trust on existing institutions. A motiva- tion crisis is a crisis in the realm of values, traditions, and norms in soci- ety. Both economic and rationality crises belong to the economic sphere, legitimation crises are political in essence and motivational crises v VI URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION belong to the sociocultural realm. The boundaries may shift, in turn, depending on the specific nature of a crisis in a country during a given historical period. Thus, a societal crisis is a comprehensive phenomena occurring simul- taneously at an economic, political, socio-cultural (value-system) level. It is a systemic, rather than a partial or local, phenomena in which institu- tions become dysfunctional in terms of their ability to process internal societal conflicts, as both the formal and informal rules that mediate social interaction collapse. Such a breakdown often has far-reaching, mostly negative, effects. At the economic level, the investment climate deteriorates in an environment without well-defined rules (e.g., property rights may not be enforced), with ensuing adverse effects in terms of eco- nomic growth and employment creation. Countries in crisis rarely expe- rience economic growth and development is postponed. At the social level conflict erodes trust and social cohesion-the social capital so impor- tant in development processes. Another manifestation of crisis is the emergence of violence, a phe- nomena linked to a breakdown in rules of behavior (see below). High levels of violence and insecurity not only deteriorate the investment cli- mate, but have economic as well as human, social, and political costs which exacerbate the crisis. In a crisis situation, public objectives may start to be replaced by private interests. Accountability and monitoring of the "agent" (e.g. governments or public officials) by "principals" (par- liament, judiciary system, the people) weaken substantially. Often the result is corruption, a phenomenon that has received substantial atten- tion in the 1990s, particularly in the development community. Corruption has a demoralizing effect on the public (and on honest offi- cials in the public sector institutions), eroding social creditability in institutions. In Habermas's framework, corruption, and, to an extent, violence, can be understood from several angles. At one level, corrup- tion is a manifestation of a "rationality crisis". In this sense common administrative practices of sound governments weaken to such an extent that conditions prevail for the acquisition of public assets for pri- vate benefit. In turn, when corruption reaches significant proportions a legitimation crisis may develop, eroding trust and confidence in gov- ernment and other public institutions. Finally, if corruption became an ingrained social practice, affecting the norms and values of society, this then contains the typical features of a motivational crisis. Thus Colombia is experiencing a dirty war, economic recession and high lev- els of corruption, all of which are affected by and in turn affect the cur- rent crisis. FOREWORD VII Crisis and Violence: Global and national causes Since the theme of this book is violence, it may be useful to put the issue into historical perspective. Marx, as well as other 19th century political theorists highlighted. the fact that the most important political changes in history were surrounded by violence, sometimes acute and dramatic. Such was the case of the French, Russian and American revolutions- to name a few important examples of radical social change. In a more con- temporary context, the post-cold war period of the 1990s has also been characterized by violence and internal armed conflict. Examples of this include the following: the disintegration of the Soviet Union, followed by violence in former soviet states such as the Chechnya war; the dissolution of the former Yugoslavia, leading to the more recent armed conflicts in Bosnia and Kosovo. The degree of violence and armed conflict in Africa in the 1990s, such as in Rwanda and Somalia to name but two of the con- flicts affecting the region, have frightened the international community. In Latin America, violence and armed conflict have also been present, such as in Chiapas, Mexico. The long lasting Colombian conflict is prob- ably the most intractable, armed conflict in South-America, given the complex interplay of guerrilla, narcotraffics, paramilitary, and the army. On the positive side, the 1990s in Central America witnessed the signing of peace agreements, and the end of civil wars in El Salvador, Nicaragua and Guatemala. These coincided with the end of the cold war. Yet these countries continue to experience high levels of violence in terms of street violence, social violence and violent crime which pervade all sectors of their societies, thus hampering their post-conflict recovery. Looking ahead in the early 21st century it seems that political and eth- nic violence will, unfortunately, be prevalent phenomena in several regions of the world. The historian Eric Hobsbawn (1999) has recently pointed out that a decade after the dissolution of the Soviet block, the process of state-building in the former soviet republics and former Yugoslavia is still an unfolding process. Needless to say, those processes are far from being peaceful and orderly events. The Andean region of Latin America also turned very volatile in the late 20th century, a trend that is bound to pervade into the early 21st century. Beside war-torn Colombia, traditionally peaceful Ecuador shows the first signs of unrest that could lead to political violence-as a consequence of the tensions accu- mulating from deep economic crisis, increased regional division, political fragmentation, and the eruption of an active and powerful indigenous movement with a radical economic and political agenda. In addition, Peru is experiencing complex reactions to its current presidential elec- VIII URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION tions, and Venezuela is engaged in an attempt at internal political and institutional change the final outcomes of which still remain uncertain. Therefore, the end of the cold war and the onset of globalization in the 1990s, contrary to initial expectations, have not been followed by wide- spread growth and social and political cohesion. Rather the last decade has seen an increase in inequality, exclusion, and violence around the globe. However, the nature of violence has changed when compared to that of the cold-war years. Wars between countries have become less important (almost disappearing), while internal armed conflicts within countries have become a more important source of violence. In addition, internal conflict in countries such as Colombia, have changed their nature from a conflict with a relatively important ideological component to a conflict tied to (and financed by) the large economic rents generated by the drug-industry and kidnapping. It would be interesting to know more about the effects of a globalized (probably illegal) arms market that pro- vides the weapons that sustain such conflicts. Furthermore, as this study shows, Colombia is affected by many different types of violence at dif- ferent levels that rise from problems not likely to be resolved by any peace process. Summing-up, several factors can be highlighted as 'new' determinants of violence in the post-cold war, globalized world of the late 20th and early 21st century. These include the reconfiguration of new national states after the collapse of the soviet block, the eruption of underlying ethnic conflicts in several countries, the emergence of an apparently important global arms market, the competition for increasing shares of the large economic rents generated by narco-trafficking, internal social conflict generated by deep economic crisis and the loss of legitimation (Habermas's 'legitimation' crisis) of existing political regimes because of corruption and poor management. In the Latin American context, increasing inequality and social exclusion are leading to higher levels of tension and frustration throughout the region, often resulting in increased levels of violence as countries such as Guatemala, El Salvador, Nicaragua, and Peru fail to move away from the cultures of violence they lived in the 1980s. These will be some of the main challenges facing Colombia in a post-conflict context, a country which has lived through a simmering civil war for half a century, in which unemployment, unequal income distribution, and social exclusion are some of the main determi- nants of violence, a situation only worsened by its recent unanticipated economic crisis. FOREWORD Ix Reforms to rebuild societies in crisis and reduce violence We have seen that crises are a multidimensional phenomena with sever- al manifestations: sluggish economic growth, dysfunctional institutions, corruption, and violence. Interestingly enough, societal crises often accompany and/or trigger important historical transformations. The most recent examples of important political and economic events at a global scale are the end of the cold war in the late 1980s and the process of economic globalization in the 1990s. Overcoming societal crisis require comprehensive economic, institutional, and political reform. Measured in a historical-time scale, economic reforms can proceed at a faster pace than institutional reforms. In turn, institutional reforms that alter formal rules can take place more rapidly than changes oriented to modify informal rules of human behavior, based on cultural norms, val- ues, traditions, and other historical factors. These considerations are valid when addressing the problem of violence. Its determinants are multiple, and linked to such factors as the occurrence of legitimation and motiva- tional crises, broad historical transformations, and economic crisis. Therefore, a comprehensive strategy to reduce violence must include broad economic and institutional reforms, specific interventions to address factors that propagate violence, and, ultimately, a change in val- ues and cultural norms that certainly underlay the phenomena of vio- lence. In addressing such issues the perceptions of poor communities themselves are essential. Through the participatory research methodolo- gy employed in this study the inter-linkages between different types of problems and their relation to different types of violence can be explored from the voices of the poor themselves. In this way the findings from this study complement more widely known economic and statistical data on violence, to show the complex reality of violence and exclusion in poor urban communities. This study follows on from a sector study on vio- lence in Colombia also published in this series (Solimano 2000, World Bank 1999). It provides the next stages of the critical work necessary to better understand the forces involved in violence reduction and the tran- sition to peace in Colombian society in the coming years. Acknowledgments This study is based on research conducted during January and February of 1999, using a participatory urban appraisal methodology. The study is part of a larger initiative within the Environmentally and Socially Sustainable Development Departrment, Latin America and Caribbean Region, World Bank-the Urban Peace Program-directed by Caroline Moser (Lead Specialist Social Development). The Swedish International Development Authority (SIDA) has pro- vided funding for this program, which also includes a similar study on Guatemala. For her vision in supporting this very new initiative, particu- lar acknowledgment is owed to Eivor Halkjaer. In Colombia Donny Meertens, of Universidad Nacional de Colombia, was the adviser for the project, and Daniel Selener of the Instituto Internacional de Reconstrucci6n Rural in Quito was the participatory urban appraisal trainer. The study was carried out by the authors, in col- laboration with four teams of researchers from the Universidad Nacional, CEMILLA (Centro Microempresarial del Llano), Fundaci6n Mujer y Futuro, and a group of consultants. They included the following mem- bers: Universidad Nacional Fundaci6n Mujer y Futuro Juliin Arturo Isabel Ortiz Haidi Hernandez C6rdoba Christiane Lelievre Lya Yeneth Fuentes Carlos Carlos Arnulfo Mendoza Maria Eugenia Vasquez Jose Luis Mufioz Cathy Mcllwaine CEMILLA Independent consultants Santiago Parra Roman Ana Daza Claudia Zulima Jim6nez Angelica Acosta Margarita Molina M. Juan Pablo Fayad Rafael Roman Francisco Hurtado G. Alba Rocio Rosas G. William Rodriguez Titus Moser xi XII URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION In the World Bank's Colombia Country Management Unit in Washington, D.C., Andres Solimano (Director), Jonathan Parker (Sector Leader), and Connie Luff (Country Officer) provided important commit- ment and advice to the project. In Bogota at the World Bank Mission, Felipe Saez (Resident Representative), Jairo Arboleda (Civil Society and Social Development Representative), and Maria Teresa de Henao (Sr. Operations Assistant) all provided invaluable assistance and support. The authors are particularly grateful to the entire mission staff for their interest and help to the project throughout the research process. World Bank peer reviewers for this study were Norman Hicks (LCSPR), Markus Kostner (SDVPC), Joel Reyes (LCSHE), Lavinia Gasperini (LCSHE), and Anthony Bebbington (SDV). As a member of the Urban Peace Program Team, Carolina Ladino, with assistance from Fiona Clark, made important contributions to the data analysis of this study. Thanks are also due to Barbara Karni, Daphne Levitas, and Carole-Sue Castronuovo of Communications Development for the editing and pro- duction of this document. Cathy Mcllwaine would like to thank the Department of Geography, Queen Mary and Westfield College, University of London, for granting her a year's leave of absence to pursue this research. Above all, the authors thank the many people in the nine communities in Colombia who participated in the research. They not only welcomed us into their lives, shared their time and perceptions, but in some cases took risks in order to contribute to the study. For safety reasons, they must remain anonymous. Executive Summary Despite sustained improvements in its social and economic indicators over the past several decades and its rich stock of natural and human resources, Colombia remains plagued by violence. The very high level of violence reflects a variety of factors, including the country's simmering 50-year-old civil war, the increase in armed conflict, the rise in urban and rural crime, and drug cartel-linked violence. As the government struggles to reach peace agreements with guerrilla and paramilitary groups, political violence and armed conflict have been the primary focuses of political analysts and civil society groups alike. The perceptions of violence by people living in poor communities have received much less attention. This report addresses this issue by provid- ing the results of a participatory study of violence conducted in low- income urban communities in Colombia. Objectives of the Study The study documents how people living in poor urban communities in Colombia perceive violence. Specifically, it identifies the categories of violence affecting poor communities, the costs of different types of vio- lence, the effect of violence on social capital, and the causes and effects of social exclusion. To describe the relationships that produce and sustain this cycle of vio- lence and to begin to identify interventions to break it, the study devel- ops a violence-capital-exclusion nexus-an analytical framework that links different types of violence both to society's capital and to the exclu- sion of its poor population. To incorporate the rarely heard voices of the poor, the study uses the participatory urban appraisal methodology, which emphasizes local knowledge and enables locals to make their own analysis of the problems they face and identify their own solutions. Fieldwork was undertaken in nine predominantly low-income com- munities located in seven cities or towns that are representative of Colombia's urban areas. These communities, identified by pseudonyms, included three barrios in Bogota (Embudo, 14 de Febrero, and Jeric6); two barrios in cities long connected with the drug cartel (P6rtico, Medellin, 1 2 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION and El Arca, Cali); two barrios in cities or towns with large numbers of displaced people (Amanecer, Bucaramanga, and Rosario, Gir6n); and two barrios in frontier towns located in areas rich in natural resources (Cachicamo, Yopal, and Colombia Chiquita, Aguazul). Types of Violence Violence-related problems emerged as the single most important type of problem facing the urban poor. Within this category, drug use was iden- tified as a major issue in many communities. Lack of physical capital was the second most important type of problem, with unemployment the most frequently cited specific problem. Lack of social capital was identi- fied as a problem more often than lack of human capital. Lack of natural capital was cited as a problem only in recently established settlements. Focus groups in the nine communities listed an average of 25 different types of violence, with one community distinguishing 60 different types of violence. The various types of violence were grouped into three inter- related categories: political, economic, and social. Economic violence was cited most often (54 percent of all types of violence), followed by social violence (32 percent) and political violence (14 percent). Perceptions of violence varied across cities and demographic groups. Intrafamily violence emerged as especially important in Bogota, gang violence as very important in Cali and Medellin, and political violence as important in frontier towns and towns with large numbers of dis- placed people. Elderly people were most concerned with insecurity and drugs. Adult women focused on violence against children, whereas adult men were most concerned with political and youth violence. Young people were especially troubled by the drug problem. Young men were also con- cerned with gang and militia violence, and young women were worried about rape outside the home. Children's perceptions, elicited from draw- ings, revealed the association between fear and guns. Costs, Causes, and Consequences of Violence Different types of violence are interrelated in a highly complex and dynamic manner. Social violence within households and families, for example, may lead young people to take drugs and join gangs, which may in turn lead to economic violence-including robbery and killing- or political violence associated with guerilla or paramilitary groups. Understanding each type of violence is thus critical to understanding the nature of the problems affecting people living in poor communities. EXECUTIVE SUMMARY 3 Families, Households, and Social Violence Intrafamily violence is a daily occurrence in Colombia, and it is closely linked to other types of violence. Study participants identified some 20 different types of violence perpetrated within the home, including incest, sexual abuse, and murder. People in the communities associated the level of intrafamily violence with various factors, including changes in the economy and the rate of unemployment, alcohol (and to a lesser extent drug) use, and machismo among men and submissiveness among women. Intrafamily violence was perceived as undermining how households functioned intemally in terms of constructing norms, values, and trust. It also eroded social capital networks between households and reduced the human capital endowments of children and young people. Violence within the home was perceived as leading to violence outside the home. Violence was perceived as permeating the spectrum of social relations within poor urban communities, with the critical nexus being households and families. With trust in the home severely eroded by violence, chil- dren spend long periods of time in the street with their friends. Young men often join gangs or military groups associated with political vio- lence. Young women engage in sexual relations at an early age, often becoming pregnant. Drug Consumption and Economic Violence Drug consumption, particularly among young men, was perceived as the leading cause of economic violence in most of the communities studied. Children were reported to begin consuming marijuana at age 8, moving on to petrol and glue by age 12. Teenagers began using bazuco (a type of cocaine) at 14, later moving on to perico (another type of cocaine), the most expensive of the drugs used. The most frequently cited causes of drug consumption were intrafam- ily violence and conflict, peer pressure, and parental example. Other causes included the lack of organized recreational opportunities, espe- cially sports and leisure facilities, and unemployment, which made it dif- ficult for young people to fill their days. Drug consumption eroded the human capital of young people, who often dropped out of school after getting involved with drugs. School dropouts were rarely able to secure employment, leading them to engage in illegal activities, such as drug dealing and robbery. Drug consumption also increased fear in communities, with drug addicts perceived as the perpetrators of assaults and robbery. Many people responded to the per- ception of danger by remaining indoors in the evenings. The result was an erosion of community-level social capital. 4 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Unemployment, Exclusion, and Economic Violence Economic violence was also found to be tied to the level of unemploy- ment, which was very high in the communities studied as a result of both the nationwide recession and the large number of people fleeing political violence in the countryside. Employment prospects were also perceived as being reduced by the stigma of coming from the barrio, which many people outside the community associate with criminal activity. High levels of unemployment were closely linked to various types of violence. In some cases violent crime was a last-resort survival response to unemployment. In other cases it reflected desperation and frustration created by lack of economic opportunities. Many people reported that using drugs was a way of dealing with lack of work and that robbing was a way of paying for drugs. Another factor tied to economic violence (and violence in general) was exclusion of young people, which many young people dealt with by using drugs or becoming involved in violent or criminal activities. A major cause of exclusion was high levels of intergenerational conflict, often the result of intrafamily violence. Community Level Social Institutions, Perverse Social Capital, and Political Violence Violence not only affects individuals and households, but communities themselves. Study participants identified 371 social institutions across the nine research communities. These institutions included both institutions that benefited the community (that is, created positive social capital) and institutions that benefited their members while hurting the community as a whole (that is, created perverse social capital). Women's and childcare groups and state-run social service delivery organizations (primarily schools and health centers) were trusted by most members of the community. Institutions connected with the perpe- tration or prevention of violence, such as gangs or state security and jus- tice institutions, were the least trusted. Perverse organizations were the most prevalent membership organi- zations, with 16 types of illegal groups functioning in the nine communi- ties. Guerrilla and paramilitary groups perpetrating political violence were universally feared in areas where their presence was dominant. Underlying this fear was the lack of trust among community members, who were afraid to talk openly about the problem. The fear of reprisals eroded solidarity, replacing it with the belief that people must look after themselves to survive. EXECUTIVE SUMMARY 5 Community Perceptions of Solutions to the Problem of Violence Study participants identified four types of strategies for dealing with violence: avoidance, confrontation, conciliation, and other strategies. Most people responded to violence by keeping silent about it out of powerlessness or fear of retribution. Many changed their mobility pat- terns, avoiding taking certain routes or simply staying home in the evening. People in the communities recognized that the continuum of violence requires that a variety of solutions be implemented simultaneously. Almost half of the interventions proposed involved creating social capi- tal. Within this category, the promotion of family values and dialogue between families and communities was the most frequently mentioned proposal. Other proposed interventions included improving education, establishing more drug and alcohol rehabilitation centers, improving employment opportunities, and integrating young people into society. Many community members also endorsed increasing the resources of community organizations and obtaining external assistance to establish new organizations. Some endorsed social cleansing and harsh police actions to crack down on violence. Public Policy Recommendations Local communities identified three national-level binding constraints that policymakers need to address. These include: - The pervasive nature of political violence. Negotiation of peace with the guerrillas, paramilitary organizations, and other groups is an important precondition for the success of other violence reduction interventions. • The serious problem of displaced people, which affects the daily lives of people in all communities. • The lack of employment, which leads to drug use, crime, and violence. To address these problems, people in the communities suggested the following interventions: * Create job opportunities in the formal, informal, and self-employment sectors. * Attack the problem of drug use. * Reduce society's tolerance for intrahousehold violence. 6 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION * Rebuild trust in the police and the judicial system. * Strengthen the capacity of community-based membership organiza- tions, particularly those run by women. * Target interventions at young people. Chapter One Introduction Colombia has long been plagued by violence. Despite being one of the most enduring democracies in South America and sustained improve- ments in its social and economic indicators over the past several decades, the country has one of the highest homicide rates in the world. The very high level of violence reflects the country's 50-year-old simmering civil war, the rise in armed conflict, urban and rural crime, and the presence of drug cartels. Until the late 1980s, economic development continued unabated in Colombia. Armed conflict affected primarily those involved in the polit- ical conflict and marginal populations in remote rural areas. Urban crime and violence affected mainly low-income barrio dwellers. In the past decade the scale and intensity of violence has changed, with violence now dominating the daily lives of most Colombians. Remote guerrilla conflict has turned into a countrywide war that involves other actors, such as paramilitary groups and drug cartels. The causes of vio- lence have also changed. These include external events such as the col- lapse of the Cold War, which has affected funding sources for guerrilla activity. It also includes internal changes relating to economic liberaliza- tion, which have had implications for the demand for labor (and levels of unemployment), and the growing levels of inequality, which are associ- ated with areas of the country in which coal and oil developments have occurred. Since 1982 successive governments have tried to find both mil- itary and political solutions for reaching peace with different guerrilla groups. State agencies, nongovernmental organizations (NGOs), and pri- vate foundations have spent vast resources attempting to reduce levels of urban crime, while military and police forces have endeavored to break the drug cartels. The World Bank's 1997 Colombian Country Assistance Strategy (World Bank 1997), which adopted a participatory approach that includ- ed consultations with civil society, identified violence as the country's key development constraint. That constraint is now affecting macro- and microeconomic growth and productivity in Colombia, as well as reduc- ing the government's capacity to alleviate the poverty, inequality, and 7 8 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION exclusion that affect the majority of its residents in both urban and rural areas. The Country Assistance Strategy recommended a comprehensive intersector policy, with violence reduction-and its counterpart, peace and development-identified as one of six key areas of strategic impor- tance in which the World Bank could assist Colombia in its development process. In 1999 the Bank completed a sector study on violence based on back- ground papers commissioned from the foremost experts in their fields, most of them Colombians (World Bank 1999). The study included three critical components: a conceptual framework, which identified a continu- um of violence, including political, economic, and social violence; an assessment of the costs of violence in terms of its erosion of the country's capital and associated assets (especially its social capital); and a brief framework for a National Strategy for Peace and Development, comprising a national-level peace program, sector-level initiatives to integrate violence reduction into priority sectors, and municipal-level social capital projects. Objectives and Research Framework of the Study In studying violence, most journalists and research violent6logos (violence experts, a uniquely Colombian discipline) focus on political violence and armed conflict. The perceptions of violence by poor communities have received much less attention. To address this issue, this report provides the results of a recently completed study conducted in poor urban communities in Colombia. The study was undertaken as part of the third stage in the Bank's contribution to developing opera- tional interventions for peace and development. The objective of the study is to document violence in Colombia as per- ceived by poor urban communities in terms of the following four ques- tions: * What categories of violence affect poor communities? Is the political vio- lence that dominates the newspapers and preoccupies politicians the only important source of concern for poor people, or are other types of violence also important? Building on the work of violence experts in Colombia and the World Bank Colombia Sector Study (World Bank 1998), the study distinguishes among political, economic, and social violence, identifying each in terms of a particular type of power that consciously or unconsciously uses violence to gain or maintain itself. * What are the costs of different types of violence? What is the financial or psychological cost of violence to poor communities, households, and individuals in terms of the erosion of physical, human, natural, and social capital and their associated assets? INTRODUCTION 9 * Does violence erode orfoster the creation of social capital? Does "social cap- ital for some imply social exclusion for others" (Harriss and De Renzio 1997; p. 926)? In fact social capital may lead to negative outcomes, and social capital itself may be created by activities that do not serve the public good. In examining this issue the study distinguishes between productive, or positive, social capital and unproductive, or perverse, social capital (Rubio 1997). * Is violence the cause or consequence of exclusion? The complex relation- ship between violence and poverty has been widely debated. Social exclusion (the process through which individuals or groups are excluded from full participation in the society in which they live) may be a more useful concept than poverty for understanding violence, because it involves a more dynamic and multidimensional conceptu- alization of deprivation. The study thus seeks to identify the causal linkages between violence and exclusion. In order to identify the relationships that produce and sustain violence in poor urban communities in Colombia and to begin to identify inter- ventions to break this cycle, the study develops a violence-capital-exclu- sion nexus (figure 1.1). This analytical framework links different types of violence to both society's capital and the exclusion of its poor population. Figure 1.1. The Violence - Capital - Exclusion Nexus Violence * Political * Economic * Social Capital * Physical * Human Exclusion * Natural * Social 10 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION The Participatory Methodology and Its Implications for Policy Recommendations Like poverty, violence can be measured in different ways (see Baulch 1996 and Moser 1998). Both phenomena can be measured objectively using large, random sample household surveys that use measures of income or consumption as proxies for the variable being measured (Ravaillon 1992). Both can also be understood subjectively using partici- patory assessments that collect data on multiple indicators that emerge out of the complex and diverse local realities in which the poor live (Chambers 1992, 1995). The same is true of violence. Extensive statistical and political analyses of Colombia violence exist (World Bank 1999). To complement those findings with the rarely heard voices of the poor, this study uses a participatory urban appraisal methodology. This approach emphasizes local knowledge and enables local people to make their own appraisals, analyses, and plans.' Its itera- tive approach to research is suitable for the investigation of the complex causal relationships that affect violence (Moser and Mcllwaine 1999). The reliability of the findings is increased through triangulation-the use of a variety of techniques and sources to investigate the same issues and ver- ify results. Qualitative research such as this study, which relies on in- depth investigation of a small number of communities, also uses purpo- sive rather than random sampling. This means selecting communities that are considered representative of the issue under investigation and conducting a participatory urban appraisal with sufficient groups to be representative of each community. (See annex A for a summary of par- ticipatory urban appraisal techniques.) Participatory urban appraisal involves an extensive number of differ- ent tools (see annex A). The most important ones used in the current study are listings that provided the basis for the quantitative data analy- sis, causal impact diagrams that analyze the causes and effects of partic- ular issues, and the institutional mapping diagrams that allow the iden- tification of social institutions perceived as important within communi- ties. All participatory urban appraisal tools are implemented in focus groups facilitated by two researchers and comprising between 2 and 20 people (occasionally they are conducted with one person). The primary aim of participatory urban appraisal is to allow the peo- ple to express their own ideas and perceptions. Therefore, in the focus groups people are encouraged to design the diagrams and provide the associated text themselves. This process often is referred to as "handing over the stick" (in the urban Colombian context it usually involves pens and pencils). The rationale behind this methodology is the transfer of power from the researcher to the researched. Consequently, all the dia- INTRODUCTION 11 grams reproduced in this document were drawn by people in the com- munities themselves and use their language.2 The study describes community perceptions of the causes, costs, and consequences of violence and identifies local perceptions of potential solutions to the problems described. Whether perceptual data can legiti- mately be used to influence or define violence reduction policies or strategies is currently an issue of debate. The Bank's recently completed report "Global Synthesis: Consultations with the Poor" (Narayan and others 1999) and its endorsement by President James Wolfensohn in his 1999 annual meeting address has certainly given a measure of legitimacy to this approach in international agencies. During the past decade several innovative interventions have been proposed to reduce violence (table 1.1). All of the approaches reflect dif- ferent solutions, although in general there has been a shift away from the control of violence toward violence prevention and most recently to rebuilding social capital. By presenting bottom-up solutions to violence, this study aims to con- tribute to the search for sustainable solutions. The solutions recommend- ed are those that local communities themselves perceive as appropriate. The approach adopted here, therefore, is one of a number of approaches that can guide policymakers concerned with reducing violence. Problems Associated with Using a Participatory Urban Appraisal to Study Violence Determining perceptions of violence in communities deeply affected by violence is difficult for several reasons. First, the law of silence makes many people reluctant to discuss violence directly or indirectly. This unwillingness to speak out was most evident in communities affected by guerrilla and paramilitary activity that had experienced killings or threats to community members. To deal with the problem, the researchers made appointments to talk at safe times, when people were sure that the paramilitaries or guerillas were not present, and conducted focus groups in back rooms of houses rather than in the street. Second, intrafamily violence is a highly sensitive issue. Young people were often more willing to discuss the issue than older people, and women were more likely to raise the issue than men. The problem of alco- hol abuse, a major cause of domestic violence, was often used as a con- duit to discuss violence in the home. To ensure the safety of the people who participated in the study and to prevent retribution, the researchers changed the names of all study participants and communities.3 To ensure the researchers' safety and help negotiate with gatekeepers, the research teams included people with guaranteed access to the communities. Table 1.1. Policy Approaches to Violence Intervention Type of violence Policy/planning Approach Objective addressed intervention Limitations Criminal Violence deterrence and Economic Top-downstrengthening of Limited applicability to justice control through increased judicial, penal, and police political and social violence; arrest and conviction rates systems and associated success highly dependent and harsher punishment institutions on enforcement > Public health Violence prevention Economic, Top-down surveillance; Almost exclusive focus on e through the reduction of social risk factor identification; individual; often imposed ° .0 individual risk factors resultant behavior top down; highly sensitive modification; scaling up to quality of surveillance of successful interventions data; limitations in indicators Conflict Nonviolent resolution of Political, Negotiations to ensure con- Often long term in its impact; transformation conflict through negotiated social flict reduction between dif- often faces challenges in H terms between conflicting ferent social actors, often using bringing parties to the table z parties using third-party mediation. and in mediating conflict u) May be top down or bottom up. 0 Human rights Legal enforcement of human Political, Top-down legal enforcement Legalistic framework often dif- c rights and documentation of social reinforced by bottom-up ficult to enforce in a context of abuses by states and other popular participation and lawlessness, corruption, and z social actors NGO lobbying impunity; documentation of r) abuse sometimes dangerous > Social capital Creation of social capital to Political, Bottom-up participatory ap- Less well-articulated than reduce violence in both economic, praisal of violence; institutional other approaches; fewer m informal and formal social social mapping to address problems; indicators developed r) institutions, such as families, community participation in community organizations, violence reduction measures uz and the judiciary 0 INTRODUCTION 13 Description and Categorization of Communities Studied Fieldwork was undertaken in nine predominantly low-income settle- ments or communities, located in seven cities and towns that are broad- ly representative of Colombia's urban areas (table 1.2). The communities reflect coverage of different geographical areas of the country as well as different types of violence. The nine communities studied can be categorized into four main urban area types: the capital, large metropolitan areas with a long ties to the drug cartel, medium-size cities or small towns with large numbers of displaced people, and frontier towns located in areas rich in natural resources. In the capital, fieldwork was conducted in three communities: Embudo, Jeric6, and 14 de Febrero. Embudo, a colonial settlement, is located in the central area of Santa Fe. Jeric6 and 14 de Febrero, both established in the 1970s and 1980s, are located in the southern part of Bogota. Jeric6 was founded by three families who invaded a large hacienda and then illegally sold off subplots. The community of 14 de Febrero was established following land invasions by the Central Nacional Provivienda, a left-wing party linked to the Colombian Communist Party. Two communities, P6rtico and El Arca, were selected in cities with ties to the drug cartel. P6rtico is located in Medellin, Columbia's second largest city. El Arca is located in Cali, the third largest city in Colombia. Both cities continue to be dominated by the drug cartel, despite the arrest and later assassination of Pablo Escobar of the Medellin cartel in 1993 and the capture of the Rodriguez Orejuela brothers and other leaders of the main Cali cartel in 1995. Communities in medium-size cities or small towns included Amanecer and Rosario, both located in Bucaramanga. Amanecer is locat- ed within the city limits of the Santander departmental capital. Rosario is located within the jurisdiction of the town of Gir6n. Metropolitan Bucaramanga has a large number of displaced people, many of them from the Magdalena Medio region. In Rosario, a relatively new, uncon- solidated settlement with high levels of squatting and cheap rental hous- ing, displaced people represent more than half of the population. Two frontier town communities were included in the study. Cachicamo is located in Yopal, the capital of the Department of Casanare. Colombia Chiquita is located in Aguazul. Both towns are located in the oil-producing region of the Llanos, which has been affected by the booms and busts associated with oil development. During the mid-1980s the oil boom brought significant economic activity to both communities, as international oil companies constructed facilities. Once the facilities were completed, the economies of both communities collapsed, as demand for labor fell dramatically. Table 1.2. Descriptions of Urban Communities Studied 14 de Colombia Characteristic Embudo Febrero Jerico P6rtico El Arca Amanecer Rosario Cachicamo Chiquita Location Bogota Bogota Bogota Medellin, Cali, Valle Bucaramanga, Gir6n, Yopal, Aguazul, Antioquia de Cauca Santander Santander Casanare Casanare Type of city Large Large Large Large Large Medium-size Small town Frontier Frontier >O metropolitan metropolitan metropolitan metropolitan metropolitan city with with town in town in > area/capital area/capital area/capital area with area with displaced displaced area rich area rich Z city city city drug cartel drug cartel populations populations in natural in natural I 0 history history resources resources 0 Socio- Poor Poor Poor Poor Poor Poor Poor Poor and Poor economic middle- status income t Intracity Inner-city Peri-urban Peri-urban Peri-urban Peri-urban Peri-urban Peri-urban Central Peri-urban O location urban 0u 0 Date of t establishment 1596 1971 1980 1972 1980 1980 1991 1935 1994 r° Form of Colonial Invasion of Invasion of Purchase Invasion of Purchase Invasion of First Invasion m establishment residential private land private of lots on private land of lots on unused settlement of private Z area by Central farmland private by M-19 private public land registered farmland m Nacional and pur- farmland guerillas farmland in munici- > Provivienda chase of lots pality c 01 01 S Chapter Two Summary Findings: Perceptions of General Problems and Violence Rather than ask respondents specifically about violence, the participato- ry urban appraisal first focused on people's perceptions of the main prob- lems affecting them and their community. Thus it did not assume that violence would necessarily be an important issue in people's daily lives. Perceptions of Problems in Poor Urban Communities Violence was the single most frequently cited problem facing the urban poor (table 2.1). Respondents identified 14 types of violence-related problems, with drugs leading the list. Drug consumption and its asso- ciated problems represented 21 percent of all violence problems (annex B, table 1). Important differences exist in perceptions of violence across commu- nities. In P6rtico, Medellin, the Peace and Conciliation Process of the Table 2.1. Frequency Listings of Types of Problems Identified in Nine Urban Communities Percentage of Type of problem total problems cited Violence-related 43 Lack of physical capital 28 Lack of social capital 14 Lack of human capital 12 Lack of natural capital 3 Total 100 Source: 159 focus group listings of general problems. 15 16 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Metropolitan Area of Aburra, initiated in 1994, reduced the perception of violence. In contrast, in Colombia Chiquita, Aguazul, where 14 people have been assassinated since 1998, the relatively low levels of reported perceptions of violence probably reflect conmmunity fear of retaliation. The perception of drugs as a community problem reflected both the existence of drugs and differences in levels of tolerance of drug con- sumption. In Rosario, Gir6n, where tolerance of drug consumption is low, one-third of all violence problems were identified as drug related. In contrast, in P6rtico, Medellin, where drug consumption was widespread but tolerance levels were very high, only 5 percent problems identified were related to drug consumption. Ranking of Perceived Problems in Poor Communities Focus groups ranked problems according to their importance (box 2.1). These findings reinforced the frequency findings and revealed the inter- relationship among different types of problems. In three communities (Embudo and 14 de Febrero, Bogota; and Rosario, Gir6n), drug-related problems were perceived as most important. In another three communi- ties (P6rtico, Medellin; Cachicamo, Yopal; and Colombia Chiquita, Aguazul), unemployment was the most significant problem. In El Arca, Cali, and Jeric6, Bogota, the most important problem cited was insecuri- ty, a catchall term widely used. In Amanecer, Bucaramanga, the leading problem cited was robbery. Problems related to violence were thus iden- tified as the most serious problem facing three of the nine communities. In communities identifying unemployment as the most significant prob- lem, it was invariably linked with violence and insecurity. Lack of Physical Capital Lack of physical capital was the second most frequently cited general problem (see table 2.1). Within this category, unemployment problems were mentioned most frequently (annex C, table 1). Aggravated by the recent decline in national economic performance, the lack of employment opportunities was severely undermining the productive capacity of many communities. In El Arca, Cali, the decline in the drug cartel-linked construction industry following the recent capture of key drug traffickers increased unemployment to about 80 percent. In Colombia Chiquita, Aguazul, about 90 percent of the community were reportedly unem- ployed. The high rate of unemployment was associated with the comple- tion of the exploration phase of the region's oil boom and the resulting decline in demand for construction workers. Recent legislation prohibit- ing the sale of goods from stalls on the streets was also perceived to have increased unemployment (see chapter 5). SUMMARY FINDINGS 17 Box 2.1. Ranking of Problems in P6rtico, Medellin, Prepared by Mixed Group of Six Adults The focus group listed 11 community problems and then prioritized them using an onion diagram. The most serious problems were placed in the cen- ter ring of the diagram, with those of lesser importance placed in the outer rings. The most serious problems identified were unemployment and lack of training; the least important problems were lack of parent-child education and lack of information about health services. Problems related to violence were placed in the third ring, although the group indicated that these prob- lems were interrelated with unemployment and lack of unity. Unemployment Lack of information on health services Lack of training Lack of parent- child education g (~~~S~> ; 8 \ ~~Lack of unity Lack of tolerance Lack of Los Muchachos gang communication Vices/drugs Insecurity in the face of violence Note: "Vices" refers primarily to drug use. It can also refer to alcohol and cig- arette use. Throughout the document "mixed group" refers to a group mixed by sex. Source: Five women and one man, ages 20-45 years. Even when job opportunities existed, people felt shunned by employ- ers, who stereotyped urban poor communities as seedbeds of guerrilla activity, delinquency, prostitution, and drug consumption. Residents with academic qualifications found the stigma of living in a poor urban area difficult to overcome. 18 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Lack of adequate public service provision, including water, electricity, sanitation, and garbage collection, was also a major preoccupation. While most communities had access to water and electricity, the cost and qual- ity of service were problematic, with a number of households pirating electricity. Housing concerns often related to issues such as overcrowd- ing and high rents charged by landlords. Lack of Social Capital Overall lack of social capital was identified as a slightly greater problem than lack of human capital (see table 2.1). Specific problems included lack of unity within the community; lack of trust in social institutions, partic- ularly the police; and a sense of exclusion or discrimination by the barrio as a whole or by particular demographic groups. Young men, for exam- ple, often mentioned that the rest of the community dismissed their views ("no one takes us into account"). Middle-age and elderly people tended to complain about an increasing sense of isolation in their com- munities and a lack of communication. In figure 2.1, for instance, social Figure 2.1. Flow Diagram of Problems Affecting 14 de Febrero, Bogota, Identified by a Mixed Group of Six Adults (aged 25-64) Druag adcts t ' <~~~~~~~~~~~~~~~~~' Lkofunity 14 de Febrero in the barrio KYoath gangs- 2/\ Lack of secrity Note: The larger the circle, the more important the problem. SUMMARY FINDINGS 19 capital-identified as "lack of unity"-was perceived as the most impor- tant problem in 14 de Febrero, Bogota. Lack of Human Capital Concerns over lack of human capital emerged frequently in all commu- nities. Lack of sufficient primary-school places was seen as a particularly serious problem. Because of the current economic crisis, some middle class people are no longer able to afford to send their children to private school, something that was once the norm among the middle class. This increased pressure on the state system means that children from the poorest families often have fewer opportunities. In addition, many peo- ple complained about the lack of money to pay matriculation fees. In the case of health care services, complaints related to lack of services, increas- ing costs, and distances to health posts. Lack of Natural Capital The lack of natural capital was cited frequently only in the recently estab- lished settlements: Rosario, Gir6n and Colombia Chiquita, Aguazul. In both communities, the most commonly cited problem was that residents felt threatened by the adjacent river's lack of flood barriers. Perceptions of Problems by Demographic Group Perceptions of problems differed across different age and gender groups. Elderly women, for example, cited declining state public services provi- sion, especially health care, as well as the decline in social capital, recall- ing with nostalgia the days when everyone knew one another and worked together for the good of the community. Adult women focused on community problems rather than individual concerns. Community mothers (childcare providers employed by the Colombian Family Welfare Institute) frequently cited mistreatment of children, lack of adequate recreation for young people, and lack of com- munity unity. Teenage girls viewed community problems more from an individual perspective and were more likely than other groups to cite problems such as rape and harassment by men in the street. Teenage girls also cited intrafamily violence regularly. Girls under the age of 12 mentioned intrafamily violence, rape, and fear in the streets as well as problems in school, especially teacher violence and harassment from boys. In contrast, elderly and adult men focused on employment and taxa- tion. Less directly affected by violence in the community than other groups, they blamed community violence on young people, often describing young men as "full of vices" and young women as "street girls" (ninas callajeras). 20 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Young men cited exclusion, unemployment, lack of educational and recreational opportunities, police harassment, and drugs as major issues. Teenage boys cited family discord, drugs, and violence at school. Children and young people were more likely than other groups to raise taboo subjects, including rape, intrafamily violence, and guerilla activity. Significant differences emerged across socioeconomic lines. Although all the participatory urban appraisals were conducted in low-income communities, a number of the older settlements comprised sections with people of higher socioeconomic status. In Cachicamo, Yopal, for example, several focus groups were conducted with professionals and business- people. These groups tended to emphasize the effects of corruption, pay- ment of bribes, and political violence on their businesses. Perceptions of Violence in Poor Communities Focus groups in the nine communities listed an average of 25 different types of violence, with one community (Embudo, Bogota) distinguishing 60 different types of violence. Types of violence ranged from intrafamily violence to fights between rival gangs and local militias to robbery and assassination (table 2.2). Types of Violence Listings of violence were grouped together under three rubrics: political, economic, and social violence (World Bank 1999; table 2.3). Social vio- lence was disaggregated into violence within the home, violence outside the home, and either (to categorize rape when it was not clear where it occurred). The types of violence represent a continuum of categories that overlap rather than being mutually exclusive. Drugs, for example, are categorized as economic violence, because drug consumers were per- ceived mainly as a problem linked to robbery to feed their habit. For drug consumers themselves, taking drugs was linked primarily with seeking an identity, conforming to youth culture, and succumbing to peer pres- sure, therefore making it a social violence issue. Together economic and social violence were mentioned more frequently than political violence in all nine communities. Economic violence was most dominant, representing over half of all types of violence. Social violence accounted for about one-third of all violence, with political violence constituting 14 percent (tables 2.3 and 2.4). Types of violence varied across communities. In Embudo, Bogota, almost 60 percent of violence was identified as social, despite the fact that drugs (a form of economic violence) dominated the community. In Jeric6, Bogota, intrafamily violence was mentioned more often than in any other SUMMARY FINDINGS 21 Table 2.2. Types of Violence and Insecurity in El Arca, Cali, Identified by Mixed Group of Six Adults Sphere of influence Violence Insecurity Individual Physical and verbal aggression Fear Household Fights between husbands Bad neighbors and wives Drug addicts Fights between grown-ups Leaving children alone Killing wife and burying in the house her in the yard Deficient housing construction Interpersonal Disharmony Hypocrisy Armed fights among friends Lack of confidence or knowledge to talk to friends Neighbor Armed robbery Lack of telephone to call Lack of conscience in case of emergency Bringing strangers to the barrio Dangerous friends Lack of unity Community Guerilla confrontations Naming people from outside with civilian victims the barrio as representatives Army accusations of civilians Unknown neighbors being guerilla members Politicians city Youth gangs Policies of Pastrana Satanic sects Government coup Source: Three women and three men, ages 24-40. community; in Amanecer, Bucaramanga, and Rosario, Gir6n, economic violence represented 70-80 percent of all references to violence. In communities with a history of drug cartels, economic violence pre- dominated, although political violence was also important. El Arca, Cali, for example, was established after members of the now demobilized guerilla group the Movimiento 19 de Abril (M-19) invaded the area. In both Cali and Medellin, levels of violence have escalated as guerilla groups (mainly the National Liberation Army, or ELN), militias, and informal protection forces and gangs have fought with one another with- in communities (see chapter 6). Political violence was an important source of violence in Cachicamo, Yopal, and Colombia Chiquita, Aguazul. This violence was tied to the Table 2.3. Types of Economic, Social, and Political Violence as Percentage of Total Violence in Each Community 24 de Colombia Type of Embudo, Febrero, Jeric6, Amantecer, Rosario, El Arca, P6rtico, Cachicamo, Chiquita, violence Bogotd Bogota Bogota Bucaramanga Gir6n Cali Medelltn Yopal Aguazul Total Economic 31 57 54 71 83 60 51 52 28 54 Drugs 18 22 12 15 33 12 11 9 0 15 > Insecurity 4 13 14 15 9 16 11 18 15 13 Robbery 9 17 15 28 15 25 7 16 8 15 o Loitering 0 3 3 9 25 0 2 2 3 5 Gangs' 0 2 10 3 0 7 18 2 2 5 Prostitution 0 0 0 1 1 0 2 5 0 1 - Social 58 40 42 27 16 23 40 18 25 32 0 Inside the home 11 10 16 6 11 13 5 7 8 10 Intrafamily violence 11 10 16 6 11 13 5 7 8 10 Outside the home 32 28 22 15 1 10 33 11 10 18 Fights 23 20 10 8 1 4 13 4 5 10 r11 Deaths 8 2 2 3 0 0 20 5 0 4 n Alcoholism 0 6 4 3 0 5 0 2 2 2 > z Encapuchados 0 0 0 0 0 1 0 0 0 1 (hooded men) x Otherb 1 0 6 1 0 0 0 0 3 1 r 0 z Table 2.3. cont. 14 de Colombia Type of Embudo, Febrero, Jeric6, Amanecer, Rosario, El Arca, P6rtico, Cachicamo Chiquita, violence Bogota Bogota Bogota Bucaramanga Gir6n Cali Medellin Yopal Aguazul Total Outside or inside the home 15 2 4 6 4 0 2 0 7 4 Rape 15 2 4 6 4 0 2 0 7 4 Political 11 3 4 2 1 17 9 30 47 14 Police abuses 11 3 2 1 0 8 7 0 2 4 Guerrilla activity 20 0 2 0 0 1 0 2 0 1 War 0 0 0 0 0 1 0 26 0 3 Paramilitary activity 0 0 0 0 0 0 0 2 22 2 Assassination 0 0 0 0 0 1 0 0 18 2 Activity by private security forces 0 0 0 0 0 7 0 0 0 1 Extortion 0 0 0 0 0 0 2 0 0 0 Threats 0 0 0 1 0 0 0 0 0 0 Violence by displaced people 0 0 0 0 0 0 0 0 5 1 Total 100 100 100 100 100 100 100 100 100 100 a. Gangs were classified as an economic form of violence because of the close nexus with theft. b. Includes machismo, discrimination, and mistreatment of children in the street. Source: 104 focus group listings of types of violence. I'3 24 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Table 2.4. Types of Violence Identified in Nine Urban Communities Percentage of types Type of violence of violence identified Economic 54 Drugs 14 Insecurity 13 Robbery 15 Other 12 Social 32 Inside the home 10 Outside the home 18 Inside/outside the home 4 Political 14 Police abuse 4 War 3 Paramilitary activities 2 Assassination 2 Other 3 Total 100 large guerilla and paramilitary presence associated with the oil boom. Initially, these organizations competed for territory and power; with the collapse of the oil sector they turned to economic extortion and corrup- tion, terrorizing the local population. In Colombia Chiquita, residents were assassinated by paramilitary groups. In Cachicamo the main con- cern was "war" in general. Ranking Types of Violence Focus groups identified the most serious types of violence in their com- munities. In the seven communities where violence was prioritized, drug-related violence was most important in Embudo and 14 de Febrero, Bogota, and in Rosario, Gir6n. In Jeric6, Bogota, intrafamily violence was considered most significant (box 2.2), while in P6rtico, Medellin, killing was viewed as the most serious problem. In El Arca, Call, mistreatment of children and robbery were identified as particularly severe. Perceptions of violence also varied by gender and age. Elderly women and men tended to be concerned mainly with insecurity and drugs. Elderly men discussed how young women were involved in violence and drug use, something that represented a new phenomenon for them. SUMMARY FINDINGS 25 Box 2.2. Types of Violence in Jeric6, Bogota, Ranked by a Group of Four 11-Year-Old Girls A group of 11-year-old school girls listed the main types of violence affect- ing their community and ranked each (4 = most serious problem, 3 = next most serious, 2 = third most serious problem). They unanimously identified drug-related violence as the most serious problem, with marijuana, bazuco (a form of cocaine), cocaine, and Boxer (a type of glue) the most common drugs used. Three of the four girls had experienced pressure from their friends to take drugs, with one girl noting that a girlfriend said sniffing Boxer was "like being on the moon." Trying drugs was a sign of bravery; refusal meant being called pollo (chicken) or bobo (fool). The girls were also concemed about rape. They felt most threatened by men they knew (friends, boyfriends, and male relatives). Listing of type of violence Score Prioritization Rape 13 Drugs Robbery 4 Rape Fights among gangs 4 Robbery Verbal aggression 2 Fights among gangs Drugs 16 Verbal aggression Family arguments - Adult women (especially among the community mothers) focused on violence against children and insecurity. In contrast, adult men discussed political violence, as well as violence among young people, especially drug-related violence and delinquency. Young people were especially concerned with the drug problem. Young men also mentioned gang and militia violence, with most being aware of the illegal and informal organizations operating in their barrios. Young women also discussed these themes, but they tended to be more concerned about rape outside the home. Children's perceptions were elicited not only from group discussions but also from drawings (figure 2.2). Of a total of 244 drawings created in response to the question "What are you afraid of?" almost half (115) dealt with violence with two-thirds of the drawings associating fear with guns. In 80 percent of the drawings, men were depicted as perpetrators of vio- lence. Most of the violence depicted took place outside the home, usual- ly in the streets (82 percent of cases). Several children also depicted 26 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Figure 2.2. Drawing of "What Are You Afraid Of" by 13-year-Old Girl in El Arca, Cali Note: Peleas fights; violaciones rapes; borrachos drunks; maltrato = mistreatmnent; pandillas = gangs. intrafamily violence, although many were reluctant to do so. In 14 de Febrero, for example, two young boys (ages 7 and 8) refused to draw what they were afraid of. Eventually they admitted that they were afraid of their fathers, who beat them regularly. Drawings of both types of vio- lence invariably showed victims, usually women or children, in tears. Drawings of drugs usually depicted men inhaling cocaine or bazuco or smoking marijuana on street corners in the evenings. Perceptions of the S'patial Nature of Violence T'he incidence of violence varied across types of communities (table 2.5). Drug-related violence, insecurity, and robbery were important in all areas. Intrafamily violence was especially important in Bogota. In the SUMMARY FINDINGS 27 Table 2.5. Variations in Types of Violence Urban category/community Predominant types of violence Large metropolitan area/ Economic and social (drug-related capital city violence, insecurity and robbery, Embudo, Bogota intrafamily violence) 14 de Febrero, Bogota Jeric6, Bogota Large metropolitan area with Economic, social, and political drug cartel history (drug-related violence, insecurity El Arca, Cali and robbery, gang violence) P6rtico, Medellin Intermediate cities and small Economic and social (drug-related towns with high levels of violence, insecurity and robbery, displaced populations loitering) Amanecer, Bucaramanga Rosario, Gir6n Frontier towns in natural resource Economic and political (insecurity rich areas and robbery, war, paramilitary Cachicamo, Yopal violence, assassinations) Colombia Chiquita, Aguazul large metropolitan areas with histories of drug cartel activity, gang vio- lence was especially marked. The effects of political violence in the coun- tryside were experienced in the intermediate towns and cities through displaced populations. Loitering among young people was also per- ceived as an important problem. In frontier towns rich in natural resources, political violence-including war, paramilitary violence, and assassinations-was most marked. Political violence by guerilla and paramilitary organizations dominat- ed the area around Aguazul, (shown in figure 2.3 as the city) and the out- lying barrio of Colombia Chiquita itself. The community was surrounded by guerilla and paramilitary groups, both in the highland area (known as the Piedemonte) and the lowlands (known as the Llanos). People living in peri-urban communities located on the outskirts of large urban areas often felt ostracized by the city itself due to the "area stigma" associated with perceptions of high concentrations of violence (see chapter 5). In Bucaramanga the city center is located on a plateau, while the community of Amanecer is situated in a lowland area around the city. A 19-year-old woman from Amanecer who had also lived in the city center noted that Amanecer was primarily associated with crime and 28 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Figure 2.3. Dangerous Areas in Casanare Around Colombia Chiquita, Aguazul, Drawn by a Mixed Group of Eight Adults PRG P PG BRRI G AeOLO PG CHIQUT G G = guerilla P = paramiliaries "people who make you afraid." She assumed that it was better to live in the city center. At the community level, the dangerous areas most commonly cited included street corners, basketball courts, parks, and river banks-all places where drug addicts, sellers, or gangs congregated. In most com- munities, girls and women feared river banks because of the danger of rape in these secluded locations (see chapter 4). In El Arca, Cali, residents felt unsafe in most of the barrio (figure 2.4). While some areas were dangerous only at night, most were also danger- ous during the day. Most places, such as the basketball court, were dan- gerous for both men and women. Places where drugs were sold and gangs met were perceived as unsafe only for women. Ironically, the police SUMMARY FINDINGS 29 Figure 2.4. Unsafe Locations in El Arca, Cali, Drawn by a Mixed Group of Four Young People Ravine River bank Pr and wome station DglAM& W MC W (:Danger at night during >Xwrt .~~~~~~~~~~~~~~~~~~~~tain Figure 2.5. Causal Impact Diagram of Interrelationships Among Intrafamily Violence, Insecurity, and Social Capital in Amanecer, Bucaramanga, Analyzed by Three Young Men from a Youth Center Social mistrust _ Lack of unity Intrafamily violence Insecurity fl + < ~~~~~Fear Lack of social institutions Robbey Delinquency| 30 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Figure 2.6. Causal Impact Diagram of Interrelationships Among Intrafamily Violence, Violence Between Neighbors, and Violence over Land in Colombia Chiquita, Aguazul, Prepared by a Mixed Group of Nine Adults Interrelationships among Different Type Mistreatment d ain gagt setween spouses and from pangs tu ld t ecnoi violence outid the home, inldn obr,atcsn . Ts uLia o ote ve o i sionofsocialcapita a m s ack of y f ins~~titductions.I Coomi Chqia Agaupltcal volec ledst fiendshi etldmg/) / 4 Lac~~~~~~~~~~~~~~k of s r 1 ) V ~~~~~~~mone bh olence between ( r 2.neighbors6 ioec vr an_ ;NChldefen efih ~~~~~~~~~~~~~~~~~~s s of \ , the popu~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~latio~n station was also perceived as unsafe, reflecting negative attitudes toward the police (see chapter 6). Interrelationships among Different Types of Violence Different types of violence are interrelated in a highly complex dynamic mnanner. In Amanecer, Bucaramanga, intrafamily violence (the primary cause of insecurity) leads family members, especially young people, to take drugs and join gangs (figure 2.5). Drugs and gangs in turn lead to economic violence outside the home, including robbery, attacks, and killing. This ultimately leads to other negative outcomes, including ero- sion of social capital (social mistrust, lack of unity, fear, and lack of social institutions). In Colombia Chiquita, Aguazul, political violence leads to both social violence between neighbors and intrahousehold violence (figure 2.6). Chapter Three Social Violence in the Family and Household Throughout the communities studied, young and old alike agreed that violence begins in the home. A young boy in 14 De Febrero, Bogota, noted, "Violence begins in the home. It is one of the most important fac- tors in the [lack of] harmony of the community. It brings about lack of respect in everyone." All communities identified intrafamily violence as a daily occurrence, and all perceived it to be linked to violence in the community. Because of the "law of silence," however, the frequency with which intrahousehold violence was reported was low (10 percent for the sample as a whole).4 The topic emerged much more forcefully in focus group, discussions with children, adolescents, and adult women (adult men rarely raised the issue). In El Arca, Cali, a group of community mothers estimated that at least one in three children in the community was mis- treated at home. In Cachicamo, Yopal, teachers estimated that intrafami- ly violence affected 70 percent of the community. The Nature and Scope of Intrahousehold or Intrafamily Violence Urban residents identified some 20 different types of intrahousehold vio- lence (box 3.1). Three types of intrafamily violence emerged as most important: violence between spouses, violence against children, and sex- ual abuse of children. Violence between Spouses Violence between spouses was widespread, with husbands more likely than wives to be the perpetrators. Violence, ranging from verbal and physical aggression to murder, occurs on a daily basis in most communi- ties. As a 25-year-old woman from Amanecer, Bucaramanga, noted, "Here it is an everyday thing that husbands beat their wives. Some have been wounded with knives and bottles." 31 32 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Box 3.1. Types of Intrafamily Violence Identified by Urban Residents in Nine Communities * Mistreatment of children * Physical violence between spouses * Verbal violence between spouses * Husband killing wife and burying her in the yard * Hitting children hard and making them bleed * Aggression against children by parents * Children hitting parents * Mistreatment of the elderly * Fighting * Violence of husbands towards mothers-in-law * Rape of girls by fathers * Rape of girls by renters * Rape of girls by stepfathers * Rape of girls by fathers under the influence of bazuco * Children having sexual relations with their mothers when they take bazuco * Mothers raping sons under the influence of drugs * Fights between siblings * Abandonment of children * Physical and emotional trauma Source: 159 focus group listings of general problems and 104 listings of violence. Violence against Children Violence against children includes beating, usually by fathers but also by mothers. Miguel, a 10-year-old boy from 14 de Febrero, Bogota, acknowl- edged that he was afraid to go home because his father beat him with wooden sticks and a leather cattle whip, particularly when inebriated. Although his mother did not hit him, she did nothing to stop the father's abuse. "I think my mother loves me, but not my father," said the boy, capturing a frequently repeated theme. Sexual Abuse of Children Sexual abuse of children, including incest within the home, was cited fre- quently in focus groups. The most severe abuse was identified in. SOCIAL VIOLENCE IN THE FAMILY AND HOUSEHOLD 33 Box 3.2. Perpetrators of Rape in 14 de Febrero, Bogota, Identified by a Group of Six Community Mothers A group of community mothers in 14 de Febrero, Bogota, identified rape by tenants as the most important problem affecting the children they looked after. According to them, tenants were more likely than relatives, stepfa- thers, or others to rape children. Listing Ranking by votes Family (father, uncle, relative) 18 Step-father 12 Renter 21 Delinquent 8 Embudo, Bogota, where it was associated with high levels of drug con- sumption. Fathers and mothers were identified as raping daughters and sons when under the influence of bazuco. Rape of girls by male relatives was mentioned in most communities. A community mother in 14 de Febrero, Bogota, noted, "One hears a lot about the rights of children, but there is one rape a day of children." Sexual abuse of children by tenants or lodgers was especially acute in communities with high levels of renting (box 3.2). Thus while housing can be used as a productive asset with which to generate income, the effect on children can be harmful. Factors Affecting the Level of Intrafamily Violence Intrafamily violence is considered a permanent feature of life in the com- munities studied. As a local leader from 14 de Febrero, Bogota, noted, "Mistreatment within the family has existed since Adam and Eve, and still we have this illness." The level of intrafamily violence appears to be affected by various factors, including changes in the economy and the rate of unemployment, alcohol (and to a lesser extent drug) use, machis- mo among men and submissiveness among women, and the intergenera- tional transfer of abuse within the home. Changes in the Economy and the Rate of Unemployment Residents of all nine communities perceived that intrafamily violence had intensified since the mid-1980s, partly as a result of a decline in the economy and an increase in unemployment. In 14 de Febrero, Bogota, for 34 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION example, intrafamily violence has risen since 1985, when unemployment became a problem (figure 3.1). In El Arca, Cali, where male unemploy- ment has increased since 1991, men have spent more time at home and taken out their frustration by abusing their children. Yet in Cachicamo, Yopal, the boom in employment opportunities that began in 1985 also increased intrahousehold violence. People migrated to Yopal to work for the oil companies, often earning good salaries. The combination of money, drugs, prostitutes, and fighting resulted in increases in intra- household violence. Alcohol and Drug Consumption Alcohol-and to a lesser extent drug use-was identified as a critically important determinant of intrafamily violence. Celebrations and festivals at which large amounts of alcohol are consumed often resulted in vio- lence. A focus group in Amanecer, Bucaramanga, identified the worst periods of the year as Mother's Day in May, the festival of the "Day of Love and Friendship" in September, and Christmas-all occasions of family gatherings. Violence within the home also tended to increase on weekends, when alcohol consumption was greatest. Machismo among Men and Submissiveness among Women Gender-based violence was caused both by machismo among men and submissiveness among women-referring to the manner in which women put up with continual violence or abuse from husbands. Three community mothers from El Arca, Cali, linked machismo and female sub- missiveness to the role of stepfathers, who often created discord within the home and committed both physical and sexual abuse. With the arrival of new father figures in the home, new rules were laid down for children, which mothers rarely opposed. As a consequence, children invariably felt rejected. Intergenerational Transfer of Abuse A common perception was that adults who were abused as children abused their own children. In Embudo, Bogota, a 43-year-old male com- munity leader referred to sexual abuse or rape as "hereditary rape," stat- ing, "He who was raped gets converted into a rapist." Rape was also closely associated with drug use. Other Types of Violence In communities affected by high levels of political violence, assassination of family members can lead to family disintegration and conflict. In Colombia Chiquita, Aguazul, a focus group identified how political vio- lence undermines the institution of the family. Family members of two SOCIAL VIOLENCE IN THE FAMILY AND HOUSEHOLD 35 Figure 3.1. Time Lines Showing Changes in Intrafamily Violence, 1970-1999, Prepared by Three Groups of Adults in 14 de Febrero, Bogota, El Arca, Cali, and Cachicamo, Yopal I) Drawn by: Mixed Sex Group of Seven Adults (35-60), 14 de Febrero, Bogota g High 4 Total c ~~~~~~~~unemployment- lack of .,~ New home education owners arrive 0 *Low 1971 1975 1980 1985 1990 1995 1999 2) Drawn by: Group of Four Women (35-50), El Arca, Cal i Fathers Much Because fathers More in the uc hom are n the unemployment- home c hom dueto can'tgive s: unemployment- education or they let out their food to E Regular i anger on children children Parents o maintain Little 1987 1989 1991 1993 1995 1997 1999 3) Drawn by: Mixed Sex Group of Four Adults, Cachicamo,Yopal I0 4 Intrafamily violence ° |------- Drug addiction ' 5 4 b _*--: * Lots of money c 970 1975 1980 Pe9 l Bo * Lots of fighting _ --* Lots of prostitution A +~~~~~~~~~~~~~~~~~ Lots of drugs 1970 1975 1980 198S 1990 1995 1999 36 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Figure 3.2. Causal Impact Diagram of the Effects of Political Violence on the Family in Colombia Chiquita, Aguazul, prepared by a Mixed Group of Four Adults Weare lef with th Separation of Fero0oiia desire for revenge brothers and sisters subversion abandonment o Hatred of \ / ~~~~~~~How a violent act military institutions a~~~~~~ffects the family Sale of property at cost in order to leave loneliness of widows Cildren lef wtout Trauma in 0 the protection of sons and daughters } parents due to politcal subvrsion women whose brother had been killed by the guerillas or paramilitary had joined the army and the paramilitaries in search of revenge. The two women noted that their families were torn apart by the assassinations in their families. Both women felt hate and fear of all armed groups and worried about their own safety and that of their children. Their emotions caused intense conflict within their households, sometimes erupting in violence (figure 3.2). Consequences of Intrafamily Violence Intrafamily violence has several serious consequences. It erodes the social capital endowments of households, increases the level of violence outside the home, and erodes the human capital of young people. Erosion of Social Capital Endowments within Households Intrafamily violence undermines the way households function internally in terms of creating norms, values, and trust (social capital). Family dis- integration, in turn, increases the level of violence within the home. SOCIAL VIOLENCE IN THE FAMILY AND HOUSEHOLD 37 Family relations provide a primary source of cognitive social capital (that is, values, norms, attitudes, and beliefs). Family disintegration thus severely erodes social capital endowments. In Colombia Chiquita, Aguazul, for instance, one group identified "lack of family communica- tion" as the key problem in their community and showed how lack of family communication appears to both generate violence and exacerbate disintegration of the home (figure 3.3). Erosion of Social Capital In discussing strategies to cope with intrafamily violence, people noted that their response was often to remain silent or not get involved because of the problems that result when people try to help. Two women from 14 de Febrero, Bogota, noted that intervening in domestic arguments gener- ated fights among neighbors: "Fights between neighbors arise because one gets involved so that the mother is not hit, or so that the mother does- n't massacre her children. Then one is treated like someone who inter- feres." They reported that they were actively excluded and stigmatized by neighbors when they tried to stop the abuse. Figure 3.3. Causal Impact Diagram of Causes and Effects of Lack of Family Communication in Colombia Chiquita, Aguazul, Analyzed by Ten Girls (aged 9-14) Lack of c'srass fmily communication I / Family is Children suffr destoye Hoei fiise _ Vilne 38 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Increase in Intrafamily Violence outside the Home Violence that occurs within the home can lead to violence in the public sphere. According to the urban poor, violence permeates the entire spec- trum of social relations within their communities, the critical nexus being households and families. In Jeric6, Bogota, a group of women drew a problem tree through which they noted that a major consequence of intrahousehold violence was "total violence-war," which would create a "country without future." According to them, violence would lead to Figure 3.4. Problem Tree of Intrafamily Violence in Jeric6, Bogota, Prepared by a Group of Adult Women (aged 28-38) Intrafamily violence (a> 0 , 0 SOCIAL VIOLENCE IN THE FAMILY AND HOUSEHOLD 39 further family disintegration, drug addiction, prostitution, mental prob- lems, and the abandonment of children (figure 3.4). Children and young people experiencing violence in the home-either as victims or as witnesses to violence between their parents-often join gangs and become involved in drug-related crime and violence. With trust in the home environment severely eroded, children spend long peri- ods of time with their friends in the street. A focus group in Colombia Chiquita, Aguazul, revealed how violence against children, coupled with parental abandonment, had far-.eaching effects on children in later life. According to the group, violence toward children can lead to such extreme responses as children killing their step- fathers, children becoming thieves and living on the streets, and children killing people outside the family (figure 3.5). Several focus groups discussed how violence within the home also leads young women "to the streets." While some young women joined Figure 3.5. Causal Impact Diagram of Causes and Effects of Violence against Children in Colombia Chiquita, Aguazul, Identified by a Mixed Group of Four Adults exam Drunkenness an doned from among parents without food Lack of care Houses Ill-1- Putting aong mothers inpo retet children 'Al cmuiisb down cilde XL Ktehers with bliged wordsobietowr Children left Abandonment E at home alone b ohr paren ta s reponsibiliy by parents| _ |management | Violence against children - o ence agairi t crenMothers less t7han c 15 years old 0 N S E Due to Children Maltreated Q maltreatment become chilrn Childry en r idrn U children grow violent kil heteyr abno E to kill thieves and w ce hungy the tehm N their own streete c step-fathers children pol a dlsd E S 40 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Figure 3.6. Causal Impact Diagram of the Causes and Effects of Intrafamily Violence in P6rtico, Medellin, Prepared by a Mixed Group of Six Primary School Teachers s rseexuan s n ea ans in pn (Robb moryn fe t -oe mter In Rosario, Gir\n, a focus group of young women noted howm et gi" (hack s asof love o dmad angsaa lntrafamilyViolencexperceorpra erlysxa actviy ofe reulin inS prenacy Insm ae,ia nyung peop rostitutess well. with nothin C chilre sand Youg P gangs and used drugs, the more usual pattehr was for them to engage in sexual relations at an early age, often resultta in pregnancy. In all of the communities it was common for 14- to 15-year-old women to be mothers. In Rosario, Giron, a focus group of young women noted how some street girlsn (muchachas vagas) took drugs and participated in male- dominated gangs as a result of violent experiences or parental conflict. Drug use and gang membership were seen as having led these girls into early sexual activity, often resulting in pregnancy. In some cases, it was noted that they became prostitutes as well. Erosion of Human Capital of Children and Young People Violence in the home influences the behavior of children and young peo- ple, which in turn affects their educational attainment and acquisition of skills. Intrafamily violence thus erodes human capital of future genera- tions, something teachers in the cornmunity emphasized. A group of 37 SOCIAL VIOLENCE IN THE FAMILY AND HOUSEHOLD 41 Box 3.3. The Effects of Intrafamily Violence Identified by Primary School Teachers Seven primary school teachers 30 primary school teachers in 14 de Febrero, Bogota in P6rtico, Medellin * Lack of discipline * Low self-esteem * Bad vocabulary * Emotional instability * Aggression with other pupils * Verbal and physical aggression * Introverted children * Lack of respect * Low academic achievement * Aggressive games * Sexual abuse of girls * Antisocial behavior * Unhelpful attitude * Loneliness * School desertion * Absenteeism * Low academic achievement * Lack of conflict resolution * Abandonment of the home * Disarticulation within the community teachers from 14 de Febrero, Bogota, and P6rtico, Medellin, identified 13 problems affecting the community, including aggression in the class room, low educational attainment, antisocial behavior, and absenteeism or dropping out of school altogether (box 3.3). According to these teach- ers, sexual abuse and rape are the most important factors affecting young people's human capital. Dropping out of school was a common outcome of intrafamily violence among children. Low educational attainments and limited employment opportunities meant that many young people turned to gangs, drugs, or, in the case of young women, to early sexual activity (figure 3.6). Chapter Four Economic Violence and Drug Consumption Drug-related problems were a major concern in most of the communities studied, with drugs linked to 21 percent of all violence-related problems. The importance of drugs varied across communities. In Rosario, Gir6n, for example, drugs were perceived to be related to one-third of violence problems. In contrast, in Colombia Chiquita, Aguazul, drugs were not perceived to affect the community at all (table 4.1). Alcohol-related prob- lems were perceived as significant in only five of the nine communities, representing an average of 2 percent of all violence-related problems (see annex B, table B-1). Although alcohol was a causal factor in other types of violence, especially of intrafamily violence, community members did not perceive it as significant in itself. Table 4.1. Drug-Related Violence as Proportion of All Violence- Related Problems (percent) Violence-related Drug-related violence as problems as a percent a percent of all violence- Community of all problems related problems Embudo, Bogotf 58 27 14 de Febrero, Bogoti 63 28 Jeric6, Bogotf 49 13 Amanecer, Bucaramanga 52 20 Rosario, Gir6n 35 33 El Arca, Cali 41 22 P6rtico, MedeUin 21 23 Cachicamo, Yopal 34 10 Colombia Chiquita, Aguazul 23 0 Total 43 23 Source: 159 focus group listings of general problems. 43 44 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Drugs represented 15 percent of all types of violence across the nine communities, with Rosario, Gir6n, and 14 de Febrero, Bogota, the most affected. The figure understates the actual extent of drug-related vio- lence, however. In Embudo, Bogota, for example, where drug use is widespread, fights made up 5 percent of the violence-related incidents cited. Many of these fights were related to drug consumption and sale. Types of Drug-Related Problems People in the communities identified 24 types of problems related to drugs (box 4.1). These problems ranged from the presence of drug Box 4.1. Types of Drug-Related Problems Identified in Nine Communities * Presence of drug addicts, including female addicts * Presence of viciosos (people with vices, usually referring to drugs) * Robbing for drugs * Killing for drugs * Drug use among children * Parents forcing their children to obtain drugs * Fathers under the influence of bazuco raping their daughters * Children under the influence of bazuco having sexual relations with their mothers * Mothers under the influence of drugs raping their sons * Violence by women anxious to get drugs * Drug addicts making others leave the basketball court * Drug addicts mistreating people in the community * Sex offered in exchange for drugs * Husbands killing wives who don't give them money for drugs * Police belief that everyone is a drug addict and/or desechable (disposable) * Young women forced to smoke drugs against their will * Smoking and sale of drugs * Drug trafficking * Youth under the influence of marijuana * Many people using marijuana * Easy access to drugs * Death from drugs Source: 159 focus group listings of general problems. ECONOMIC VIOLENCE AND DRUG CONSUMPTION 45 addicts to murder. The most common problem identified was "there are lots of drug addicts" or "lots of people with vices," a term in Colombia that refers to drugs. In all communities with high levels of drug use the most commonly cited consequence was robbery. In Rosario, Gir6n, a 13- year-old girl noted, "The people who take marijuana rob, they rob from houses." Drugs were also associated with other types of violence, includ- ing intrafamily and sexual violence. Young women frequently blamed drug addicts for rapes in their barrio, both in the street and at home. Drug consumption was also linked to death. Some people killed to obtain drugs or while under the influence of drugs; others were killed by the physiological or psychological effects of drug use. In Embudo, Bogota, a male addict noted, "We were all born to die." A focus group of drug addicts-a man and woman who had just been released from prison and two other men who robbed to feed their drug habits-described the inevitability and acceptance of death (figure 4.1). Drug consumption in most communities was a relatively new phe- nomenon. While community members noted that alcohol consumption had negatively affected their barrios for as long as they could remember, drug consumption began mainly in the 1980s, with marked increases in the 1990s. Alcohol was associated with intrafamily violence; drugs were closely linked with insecurity. Drug use had become more widespread over time. A timeline drawn in 14 de Febrero, Bogota, showed how in the past insecurity was linked with police harassment in the neighborhood Figure 4.1. Causal Impact Diagram of Causes and Effects of Death due to Drug Consumption in Embudo, Bogota, Prepared by Four Drug Users ea ou ine ua iaR a meone doesn't wcontrol "etween ien s) how to smoke d oyfe ~~~ Cause~~~~~~~as _ ~~~~~~What you do to DEATH | oavi i to avoid death . . | ~~~~~~~~~you have to trick it." / aton| "Te dead to teE What do you feel? grave and the living _ toted 46 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION and a policy of "social cleansing." But since 1990 there has been an increase in the perception of insecurity as a result of the increased levels of drug consumption (figure 4.2). Before 1990 drugs tended to be con- sumed in the home or "hidden from view," while in the past few years the number of "open" drug users (using drugs in public view), has increased considerably. Attitudes toward drug traffickers were mixed. In Embudo, Bogota, it was noted that the sale of drugs decreased following the death of Pablo Escobar. In both Medellin and Cali, however, the drug cartels were often seen in a very positive light. P6rtico, Medellin, for example, benefited from a basketball court built with money donated by Pablo Escobar. The period during which the Medellin cartel was most active (the early 1980s) was viewed favorably because of its economic benefits. As one man in the community pointed out, "When the drug traffickers left, the situation became very difficult. Everyone was left worse off." In Cali the econom- ic fortunes of the El Arca community declined with the capture of Los Rodriguez of the Cali cartel. As one community leader stated, "The drug traffickers helped us to a certain extent. They produced an increase in employment and allowed the people to build their own homes." Figure 4.2. Time Line of Changes in Drug Consumption and Insecurity in 14 de Febrero, Bogota, Drawn by a Mixed Sex Group of Six Adults I0 (High) E U Police install burglar alarms and introduce Insecurity rises "social cleansing" as a result of increase in drug consumption Founding of c boorrin o Reduced insecurity because of burglar alarms and "social cleansing" 0 ) I I I l I 1971 1975 1980 1985 1990 1995 1999 ECONOMIC VIOLENCE AND DRUG CONSUMPTION 47 Types of Drugs Available and Tolerance Levels A variety of drugs was available in low-income communities (box 4.2). Marijuana was most commonly mentioned, followed by bazuco and pericola, both forms of crack cocaine that are usually smoked rather than injected. Drug consumption is so common in poor urban communities in Colombia that many young people immediately identified themselves in terms of whether or not they consumed drugs. Those who did not con- sume drugs called themselves sanos (healthy) or zanah6rios (healthy). Drug users were referred to by nonusers as drogadictos (drug addicts), viciosos (people with vices), marihuaneros (marijuana smokers), colinos, or sopladores (drug addicts). Estimates of drug use varied widely across communities (table 4.2). In 14 de Febrero, Bogota, where tolerance of drug use is moderate, res- idents were resigned to the fact that 40-50 percent of their community used drugs. A group of seven community leaders (six women and one Box 4.2. Types of Drugs Available * Cocaine * Marijuana * Marimba (type of marijuana) * Pericola (type of cocaine) * Bazuco (type of cocaine sometimes mixed with ground brick or broken glass) * Crack * Pepas (generic name for tablets, usually uppers) * Piola (tablets, usually uppers) * Ruedas (tablets for epilepsy) * Roche (tablets for epilepsy) * Exstasis (Ecstasy) * Boxer (type of glue) * Antiseptic alcohol (mixed with Coca-Cola-or other soft drink- known as chamberlai) * Gasoline * Tiner (paint thinner) * Escama de pezcado (cocaine-based fish flakes) * Bicha (joint of marijuana) Source: Focus groups in nine communities. 48 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION man) estimated that there were 500 users of whom only 50 used drugs openly. The remaining users, known as "closed" users, were usually adults, who consumed drugs in their homes. The "open" users were often also drug distributors for the "closed" users. Drug addicts from the community estimated that there were 30-40 hard-core users and 100 young people who took drugs on a less regular basis, getting the regular users to buy drugs for them. In P6rtico, Medellin, where the level of tolerance and visibility of drug use was highest, 60 percent of the population was estimated to smoke marijuana. This marijuana use was viewed as a hobby, and a way to "relax, sleep, pass the time, and keep your mind elsewhere." One young man stated that everyone took drugs-children as young as 8 or 10, boyfriends and girlfriends, and parents and relatives. Although consumption of marijuana was high, bazuco was rarely consumed, because it was associated with the "war" from which the city is just emerging. Table 4.2. Tolerance and Visibility of Drug Consumption in Four Communities, Identified by Focus Groups Percentage of population consuming drugs! Level of age range Community tolerance of users Visibility Main type of drug 14 de Moderate 40-50 percent/ Visible among Marijuana and bazuco Febrero, young people young people, among young people, Bogota hidden among marijuana and cocaine adults among adults P6rtico, High 60 percent/ Visible Cocaine, marijuana Medellin young people and roches and adults Cachicamo, Very low Floating popu- Visible Marijuana and bazuco Yopal lation of 3-4/ young people Rosario, Very low 5 percent/ Visible Marijuana and bazuco Gir6n young people Source: Various focus groups in nine cormmunities. ECONOMIC VIOLENCE AND DRUG CONSUMPTION 49 Characteristics of Drug Consumers Young people, primarily young men, were reported to make up the bulk of drug consumers, with women representing about a third of consumers in those communities in which estimates were obtained (see table 4.2). Most users were identified as young, with the majority between 15 and 30 (table 4.3). All communities noted that parents often used drugs. Children as young as 8 also reportedly used drugs, with different drugs taken at different ages. Children began with marijuana at the age of 8, moving on to gasoline and glue at the age of 12. The use of bazuco came at the age of 14, with perico, the most expensive drug, used later. There were marked generational differences between drug and alco- hol consumers. According to a group of young women from Amanecer, Bucaramanga, drug use was associated primarily with the young, while alcohol use was linked with middle-age and elderly people (box 4.3). Alcoholics were also more likely than drug users to be male. Drug users were perceived as more dangerous to society than alcoholics because of their loss of control when under the influence of drugs as well as their stealing to feed their habits. At least in the public sphere, alcoholics were better able to control themselves and were not associated with acts of delinquency. The Cost of Drugs Drugs were noted to be considerably less expensive than alcohol (table 4.4). One joint of marijuana, for example, cost 300-500 pesos ($.19-$.30)-much less than a small bottle of beer (800 pesos, or $.50). A bottle of brandy cost more than a gram of the most expensive hard drugs (cocaine and perico). A gram of bazuco was not much more expen- sive than a bottle of beer. Table 4.3. Distribution of Drug Consumers by Age in 14 de Febrero, Bogota, Identified by Seven Community Leaders Estimated number Age range of consumers Percentage of total 10-20 years 100 20 21-30 years 250 50 31-40 years 100 20 Over 40 50 10 Total 500 100 50 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Box 4.3. Characteristics of Drug Addicts and Alcoholics Reported by Five Young Women Drug addicts Alcoholics * Act without thinking * Despite everything, they * Think only about doing know what they're doing bad things * Mainly fall asleep under * Act without conscience the effects of alcohol * Rob from houses * Fall over * Are dangerous * Mainly middle-age and * Are mainly adolescents elderly * 7 out of 10 are male * 9 out of 10 are male Source: Five women ages 12-16, in Amanecer, Bucaramanga. Several focus group members also estimated how much they spent a day on drugs (box 4.4). In 14 de Febrero, Bogota, a young male addict reported spending 10,000-20,000 pesos ($6.20-$12.50) a day on drugs. Three other users reported spending an average of 15,000 pesos ($9.40) a day, with a low of 5,000 ($3.10) when their daily earnings were low. In P6rtico, Medellin, nine young men from the Los Muchachos gang esti- mated that they each smoked 10 marijuana joints a day, at a cost of 3,000-5,000 pesos ($1.90-$3.10). Table 4.4. Costs of Drugs and Alcohol in 14 de Febrero, BogotA, and P6rtico, Medellin (pesos) Type of drug or alcohol 14 de Febrero P6rtico Marijuana 300-1,000 per joint 300-500 per joint Bazuco 1,000-2,000 per gram Perico 5,000 per gram 4,000-8,000 per gram Pepas 1,500 per tablet Cocaine 4,000-8,000 per gram Antiseptic alcohol 800 per half bottle Beer 800 per small bottle Spirits 6,000 per bottle of rum 9,000 per bottle of brandy Note: Exchange rate: US$1 = 1,604 pesos. Source: Six focus groups in both communities. ECONOMIC VIOLENCE AND DRUG CONSUMPTION 51 Box 4.4. Household Drug Expenditures in P6rtico, Medellin, Reported by Two Adult Groups The household expenditures shown below were provided by Elvira, a moth- er from P6rtico, Medellin, in a focus group discussion. Her household com- prises nine people, including Alfonso, a painter, who earns 350,000 pesos ($218) a month and spends 540,000 pesos a month on alcohol, drugs, and gambling. The household's monthly shortfall of 744,300 pesos ($464) is made up by Alfonso working extra hours or asking for salary advances. Cost per month Cost per month Percent Expenditure item (pesos) ($) of total Alcohol, drugs, gambling 540,000 336.7 49.3 Food 400,000 249.3 36.6 Transport 95,800 59.7 8.8 Services 50,000 31.2 4.6 Education 8,500 5.3 0.8 Total 1,094,300 682.2 100.0 The household expenditure patterns shown below are aggregated estimates from a focus group of four women (19-40) and two young men (17 and 23). Expenditures on drugs-74,000 pesos a month ($46)-exceeded expendi- tures on education. Drugs and alcohol combined were the second largest expenditure item after food. Cost per month Cost per month Percent Expenditure item (pesos) ($) of total Food 200,000 124.7 30.7 Services 100,000 62.3 15.3 Rent and credit 85,000 53.0 13.0 Transport 60,000 37.4 9.2 Education 50,000 31.2 7.7 Bazuco 40,000 24.9 6.1 Parties 40,000 24.9 6.1 Alcohol 30,000 18.7 4.6 Cocaine/perico 20,000 12.5 3.1 Gambling 12,000 7.5 1.8 Marijuana 8,000 5.0 1.2 Pepas 6,000 3.7 0.9 Total 651,000 405.9 100.0 Source: Group of four women ages 25-40 and a mixed group of adults ages 17-40. 52 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Causes of Drug Consumption Drug consumption among young people, the primary users, was close- ly linked with intrafamily violence and conflict (see chapter 3). The problem tree drawn by community mothers from Rosario, Gir6n, focused on nonviolent conflict within the home that could cause young people to turn to drugs (figure 4.3). Key factors they cited included "lack of comprehension of parents toward children," "lack of interest" on the part of parents, and "lack of love and dialogue within the fami- ly." Young men were more likely than young women to turn to drugs and gangs when family life becomes untenable. Young women tended to get involved in early sexual relationships, often becoming pregnant or turning to prostitution.5 Figure 4.3. Problem Tree of Drug Addiction in Rosario, Gir6n, Prepared by Seven Community Mothers (Madres Comunitarias) Street Rp violence Rape EFFECTS Social Physical Theft (Branches) deterioration Bad examples set, 1/~ \\ I I II / / mistreatment Prostitution of children Drug PROBLEM \ Addiction (Trunk) Lack of Interest or comprehension passtime by parents of of parents children Lackof interest in oneself and CAUSES \ Bad company work so that it's difficult to (Roots) / and lack of develop self-esteem /9 recreation Lack of education Too much freedom and training without control I Lack of love and dialogue within the family ECONOMIC VIOLENCE AND DRUG CONSUMPTION 53 Another important cause of drug consumption was peer pressure, which was noted by children as young as 10. In Jeric6, Bogota, for instance, a group of four 11-year-old girls said that they had been pres- sured to smoke cigarettes and then use drugs (starting with Boxer glue). Often, gangs of drug pushers stood outside the school gates trying to entice students into trying drugs by offering the first consignment free of charge. Pressure also came from friends. In all communities malas amistades (bad friendships) or malas companias (bad company) was cited as a cause of drug use. Finally, some parents not only provided bad examples but actually taught their children to smoke marijuana. A group of three adults in El Arca, Cali, noted that with so many parents using drugs, children learn to accept drug consumption as normal. Consequences of Drug Consumption Lack of physical and human capital was both a cause and an effect of drug use. Community members often blamed the drug problem on lack of recreational opportunities, especially sports and leisure facilities. It is important to note, however, that both basketball courts and football fields were among the most popular sites for consuming drugs. Some people suggested that it was the lack of organized sports clubs or youth clubs that generated boredom and discontent among young people, lead- ing them to take drugs. With unemployment one of the major problems in all of the commu- nities, people become desperate and turn to drugs as a way to fill their time or soften the edge of their despair. As figure 4.4, drawn by three Communist Party members who founded 14 de Febrero, illustrates, stress related to having nothing to do can lead people to take drugs. In their highly politicized view, lack of employment was also linked with lack of workers' conscience, lack of professional training, and unfair labor laws. Lack of education was also cited as both a cause and effect of drug con- sumption. Drug consumption tends to erode physical and human capital. When young people become involved in drugs, they usually leave school, according to the focus groups. Once they become involved in drugs, they are rarely able to secure employment other than drug dealing and rob- bery. Young people claimed that they were increasingly succumbing to drugs because of lack of alternatives. The combination of deteriorating physical and human capital meant that the quality of the labor force with- in these communities was extremely low. In all communities, it was mainly teachers who linked drug consump- tion among students with gang membership. One teacher complained 54 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Figure 4.4. Causal Impact Diagram of Causes and Effects of Drug Addiction in 14 de Febrero, Bogota, Analyzed by a Group of Three Male Community Leaders Iduc wb°< \ Neo-liberal policies t tach o e u caiton G e m ere for living . 8 > < ~~~~~ADDITO Et Lakaf need special training. The pandillas (gangs) are veryfrigtnmenigsta"t fctioens erroen mos severe fo yon epe employndm u ustice onr wst \ncrease in fear and insecurity i c t. e were _afa ofes that teachers were unable to cope with these types of problems: bWe are not trained to deal with drug addiction nor with gang culture. For tlds we need special training. The pandillas (gangs) are very frightening." Thus while drugs affect all strata of society in the nine communities, the ralni- fications were most severe for young people. A second itportant consequence of drug consumption was the increase in fear and insecurity in the communities. People were afraid of drug addicts, both because they were perceived as muggers and robbers and because drugs are iwegal and therefore feared, at least by older peo- ple. "[Drugs] create fear" was a constant rejoinder. Discussing marijuana users, two women and a man from Rosario, Gir6n, noted, "People can't go out in the evening .... One can't send a girl or even a boy out alone because they'll get caught-up with them." In many communities, people feared areas where drugs were sold or used. The most commonly cited dangerous places in 14 de Febrero, Bogota, for example, were the basketball courts, where young people consumed drugs; the banks of the river, where young women had report- edly been raped by drug addicts; a park, which was dark and attracted ECONOMIC VIOLENCE AND DRUG CONSUMPTION 55 Figure 4.5. Map of Fear in 14 de Febrero, Bogota, Drawn by a Group of Six Young Women (aged 12-22) Parkb- ouide ______ D It's very isolated DN7 Main street 4- Street of crack < \ \ IE(robbery, drug sales, \ < \ > \t I ~~~~~~~~~~~drug consumption) +XV \ < ~~~~~~~~Barrio 14 de Feb rero Key ,, Shaded areas deroe\ \\> dangerous places \, Everything is dangerous because if s very dark drug addicts; a street on which crack was sold; and a house from which drugs were sold (figure 4.5). These patterns were found in other commu- nities as well. The reduced mobility as a result of fear eroded trust and collabora- tion-cognitive social capital-within the community. A woman from 14 de Febrero, Bogota, noted that "before," people were more united and "one didn't see young people smoking [marijuana] because there was more communication." In contrast, "now," there was no collaboration, people no longer respected one another, and members of the community were afraid to confront others about the problems facing the barrio. There was also a strong gender dimension to fear. While men of all ages were afraid of robbery and mugging, levels of fear among women of all ages were higher and related to fear of attack, especially rape, by drug addicts. Chapter Five Unemployment, Displacement, Area Stigma, and Economic Violence Colombia's current economic crisis has increased unemployment, which community members perceived as a major cause of violence and insecu- rity. As a man in Cali pointed out, "In a poor country with hunger and without work, there's no peace." Unemployment was a major contribut- ing factor in many types of violence, and it was closely linked to the large-scale displacement of people fleeing political violence in rural areas. This displacement exacerbated unemployment in urban areas. "Area stigma"-sometimes referred to as "bad reputation"-attached to low-income communities was also perceived as a major obstacle to secur- ing employment. Nature and Scope of Unemployment In five of the nine communities studied, unemployment was the leading problem cited under the rubric of lack of physical capital. Unemployment constituted 13 percent of all problems in P6rtico, Medellin, and 11 per- cent in 14 de Febrero, Bogota; Cachicamo, Yopal; and Colombia Chiquita, Aguazul. Unemployment represented about a third of all problems relat- ed to the lack of physical capital (table 5.1). Estimates of the rate of unemployment in the sample communities were very high. In Colombia Chiquita, Aguazul, a group of three women and one man estimated that 90 percent of the barrio was unemployed. In P6rtico, Medellin, a group of two men and six women estimated that 70 percent of the population had no work. Similar figures were estimated in El Arca, Cali, where a group of three men estimated that unemployment stood at 80 percent. Estimates of unemployment by age reveal that younger people have higher rates of employment than older people (table 5.2). Unemployment among all age groups is very high, however, with two of three 15- to 25-year-olds unemployed. 57 58 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Table 5.1. Relative Importance of Lack of Physical Capital and Unemployment (percent) Lack of physical capital Unemployment as a percentage as a percentage of of problems related to Community all problems lack of physical capital Embudo, Bogota 15 33 14 de Febrero, Bogota 23 46 Jeric6, Bogota 33 19 Amanecer, Bucaramanga 21 30 Rosario, Gir6n 27 24 El Arca, Cali 28 32 P6rtico, Medellin 37 34 Cachicamo, Yopal 34 33 Colombia Chiquita, Aguazul 36 30 Total 28 31 Unemployment also varies by gender, with unemployment higher among men than among women. It is easier for women to find informal sector work, and in many communities women were the primary income earners. As one man from El Arca, Cali, pointed out, "Today there are more women working and they are maintaining us." Another man explained that he was now taking care of the home, cooking, cleaning, and caring for his children. Types of Employment Available Employment opportunities lacked stability in all nine communities. Men tended to work in construction, either as laborers or in skilled trades; as guards; as drivers; or in a host of informal sector activities, particularly Table 5.2. Unemployment by age range, El Arca, Cali Age range Percentage unemployed 15-25 66 26-35 77 36-45 78 46-60 82 Over 60 50-100 Source: Six focus groups in El Arca, Cali. UNEMPLOYMENT, DISPLACEMENT, AREA STIGMA 59 street vending. Women were employed as community mothers, as domestic servants, or in a range of informal activities, such as selling cooked food and dressmaking. Many people made ends meet through rebusque (literally, careful searching), engaging in different work activities every day. One man from Jeric6, Calixto, who earned his living from rebusque pointed out, "I'm not ashamed to say that I would sell anything, even a pregnant woman.... I have to get enough to eat every day." Child labor was also noted in most communities. In Jeric6, BogotA, one community leader estimated that 20 percent of all children worked, sell- ing newspapers and lottery or bus tickets. Even higher proportions were noted in P6rtico, Medellin, where a group of four women and two men estimated that 60 percent of 10- and 14-year-olds worked, usually engag- ing in some form of selling activity (table 5.3). Types of Violence Caused by Unemployment High levels of unemployment and economic hardship are closely linked with various types of violence. In some cases, violence is a survival response to unemployment; in other cases, it is the outcome of despera- tion and frustration created by lack of economic opportunities. As sug- gested by a group of adolescents in 14 de Febrero, Bogota, violent crime is often a last-resort survival strategy and often an integral element of rebusque. The most common type of crime committed was theft, especial- ly street robbery and robbery of trucks delivering soft drinks. One man from Jeric6, Bogota, noted: "If they don't let us work, we are pushed into robbing. You can't let yourself die from hunger, and less so your children." Other survival strategies include prostitution (mainly among women but also among men) and small-scale drug deal- ing (figure 5.1). The frustration of being unemployed also led to violence. Drug use was frequently cited as a way of dealing with the lack of work. To pur- chase drugs, unemployed people often robbed. Unemployed men tended to feel inadequate when they could not fulfil their role of breadwinner, often turning to alcohol and intrafamily abuse to vent their feelings. The national recession has forced more and more workers to become casual workers. The resulting instability has increased insecurity and vio- lence. In 14 de Febrero, Bogota, a group of young men who had worked intermittently over the year in construction activities noted that they usu- ally worked four months a year (June-July and November-December). During the slack periods, they hung about making money from illegal means, usually by robbery. ON Table 5.3. Economic Activity Matrix by Age Group, P6rtico, Medellin Item 0-10 11-14 15-25 26-35 36-45 45-55 Over 55 Number in group 396 500 700 500 300 400 300 Z Number of people 100 300 200 150 150 100 50 o0 who work (all boys) (all boys) (50% male, (50% male, (60% male, (60% male, (60% male, 50% female) 50% female) 40% female) 40% female) 40% female) t Type of activity * Selling sweets * Selling - Stealing * Construction * Construction * Construction . Construction * Street vending * From age 13, * Construction * Dressmaking * Street vending * Street vending * Street vending * Recycling stealing * Street vending * Domestic * Domestic * Domestic * Domestic 0 * Domestic servant servants servants servants ur servant * Begging 0 -TI Estimated monthly < earnings in pesos 0 (in dollars) 30,000 ($18.7) 40,000 ($25) 180,000 ($112) 240,000 ($150) 240,000 ($150) 240,000 ($150) 200,000 ($125) z Note: $1 = 1,600 pesos. nm Source: Focus group of women (ages 19-56) and men (ages 27-34). > X to x oz UNEMPLOYMENT, DISPLACEMENT, AREA STIGMA 61 Figure 5.1. Causal Impact Diagram of Causes and Effects of Unemployment in P6rtico, Medellin, Prepared by a Mixed Group of Seven Adults 90X ~~~~~~~~ Delinquency 7?he.e ~ E" an Unemployment) Since che new more l to get president came employment ~~emplImen than /X;;> 214 Can t find A ~~t oo Causes of Economic Violence Unemployment in the communities studied was tied to a variety of fac- tors, including the national recession, local constraints, legislation regu- lating the informal sector, political violence in the countryside, area stig- ma, and the collapse of the oil industry. Economic violence was tied to all of these factors. Changing Economic Circumstances Colombia's economy is in recession. In the first quarter of 1999 (when the study was conducted), GDP was projected to contract by 4-7 percent. The worst-affected sectors include manufacturing, construction, and retail- ing, all concentrated in urban areas. Urban unemployment in the seven largest cities reached 19.5 percent in March 1999, up from 14.4 percent the previous year, according to the national statistics department. Urban unemployment was highest in Medellin, where 22.6 percent of the eco- nomically active population had no work (Oxford Analytica 1999). 62 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION In all of the communities the national economic recession as well as changing political circumstances were viewed as causes of violence and insecurity. An 18-year-old construction worker from Rosario, Gir6n, explained that because of the economic crisis, there was no work and therefore no money. With no money, there was little food, especially in urban areas. In addition, people, especially young people, had nothing to do and got involved with bad company, which led to drug use and rob- bery. These sentiments were repeated throughout the nine communities (figure 5.2). Constraints Faced by Local Economies Local urban economies experienced the economic crisis in different ways. In Bogota, especially in 14 de Febrero and Jeric6, both located close to large industrial zones, residents complained about factory closures and layoffs. Discussing the bankruptcy of the local factories in 14 de Febrero, Maritza, the treasurer at a communal action group, predicted that the future held only further economic crisis, war, and violence if nothing was done to deal with unemployment. In her words, "A member of the fam- ily who has no work and has to respond to their obligations often has to turn to robbery." Figure 5.2. Causal Impact Diagram of Causes and Effects of Unemployment in El Arca, Cali, Analyzed by a Group of Three Male Adults (aged 20-30) \ \ \ /EconoicrTornspor (quirements-probems problem \scmk Economic , (CTranspor agith problemsbJa< problems - w~5iscrrmatior UNEMPLOYMENT, DISPLACEMENT, AREA STIGMA 63 Figure 5.3. Time Line of Changes in Employment and Insecurity, 1994-1999, in Colombia Chiquita, Aguazul, Drawn by a Mixed Group of Eight Adults 10 No deaths in A lot dAnO the first month ~~ o~~~~ of '99 -f ~~~~~insecurit e - o~~~~~~~~~~ ~Employment 50% employed --E 50% unemployed I' y t' More or less calm o,"0 ^ 20% employed Litde 80% unemployed 1994 1995 1996 1997 1998 1999 A timeline from Colombia Chiquita, Aguazul (figure 5.3), illustrates the complex relationship between insecurity and employment in oil-rich areas. Between 1995 and 1997, the level of insecurity increased as people arrived from outside the area in search of work in the oil industry. Although employment was high, increased guerilla and paramilitary activity linked to corruption and extortion meant that insecurity rose. After 1997, as demand for labor in the companies declined, people lost jobs that paid two and a half times what they could eam elsewhere ($312 a month in the oil industry versus $125 in farming). The loss of employ- ment created resentment among the population and increased the level of crime. With fewer sources of extortion among the oil companies, the paramilitaries and guerillas turned on the local residents instead. In Cali and Medellin unemployment was associated with the collapse of the construction industry, once associated with the drug cartels. The cartels had laundered large amounts of money through the construction industry which, in turn, employed large numbers of the urban poor. The national economic downturn and the capture of leading drug cartel fig- ures have caused the construction industry to contract. According to a 64 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION group of middle-age men in El Arca, Cali, the employment situation would improve 100 percent if the Rodriguez brothers were released from prison. With the capture of the Rodriguez brothers, "money did not move." The main effects were an increase in armed attacks, the pro- liferation of gangs, a 100 percent increase in insecurity, and a rise in homicides. Legislation Regulating the Informal Sector Municipal legislation has recently been passed restricting street vending in public areas. Although the intent of the law was to curb the sale of drugs, the effect has been to increase unemployment in some communi- ties. Ironically, the law has led to increased insecurity, according to resi- dents. One woman from 14 de Febrero, Bogota, said that the main solu- tion to reducing violence was to rescind this law and permit street ven- dors to work. She noted that 25 percent of the working population in the barrio worked as traders and that many were forced to rob and engage in illegal activities because of the restrictions imposed by the mayor. Political Violence in the Countryside Communities perceived political violence in the countryside as a major cause of the economic crisis. Many felt that the government was allo- cating scarce resources to the army and the war rather than to the econ- omy. A 13-year-old boy in P6rtico, Medellin, commented, "The only people that are paid are the soldiers, and when they are killed the mother is left rich." The most important effect of the civil war on poor urban areas has been the influx of people fleeing political violence in rural areas. The phenomenon is occurring throughout the country, with the highest lev- els of flight in Uraba, Magdalena Medio, and Ariari. Recent estimates by the World Bank suggest that 1.2 million people in Colombia had fled their home areas by 1998 (World Bank 1999). Displaced populations were identified in all of the study communities except Embudo, Bogota. The highest proportions of displaced persons were found in Rosario, Gir6n, where about 50 percent of the population had fled from other areas-mainly the Magdalena Medio-and in Colombia Chiquita, Aguazul, where 50-60 percent of the population were displaced. A group of six men and two women in Colombia Chiquita, Aguazul, iden- tified three types of displaced people in the barrio: people who had fled political violence in the rural areas of Casanare, people who had fled political violence in other departments, and people who sought employ- ment with the oil companies. Displaced people are in desperate need of employment. Two couples who had fled political violence in San Vicente ChucurA, leaving their UNEMPLOYMENT, DISPLACEMENT, AREA STIGMA 65 small farm, aspired to nothing more than "finding a fixed job that paid the minimum wage." They survived by eating unwanted vegetables they collected from the Centro de Abastos. Most of the displaced people in Rosario, Gir6n, felt grateful to have escaped political violence and found the barrio very peaceful. One displaced woman was relieved to be able to live without constant fear because "in C6sar they kill many people for doing nothing, only for the sake of killing." The influx of displaced people affects employment of the local popu- lation, who blame the newcomers for putting pressure on the labor mar- ket. In Cachicamo, Yopal, displacement was perceived as a major factor in unemployment, exacerbating an already difficult economic situation. "No work for those who arrive, nor those who are already here," com- mented one resident (figure 5.4). Although displaced people were not necessarily actively excluded within their new communities, they were blamed for contributing to unemployment. In P6rtico, Medellin, unemployment was also perceived to be tied to racial discrimination. A group of four Afro-Colombian men and one woman who had migrated from the coast to find work in the city blamed their unemployment on various forms of discrimination. These included Figure 5.4. Causal Impact Diagram of Causes and Effects of Unemployment in Cachicamo, Yopal, Prepared by a Mixed Group of Five Adults Employed in the Administrative Department for Security *Attacks Deinuec Asrsamults overnment_ ou j Rape provideEsecuritor tio o u Extortion services /|*Fraud / \ * ~~~~~~Kidnapping / ; _ *~~~~~~~~~~Murder Displacement - nemplowoyrrvemorfontos < contryside to city j Neighborhood committees 2 7 9 2 2 7 3 1 5 38 (21.5) > Religious groupsa 4 5 7 3 2 1 3 3 5 33 (18.6) Women's and childcare organizations 4 3 7 5 2 3 3 1 4 32 (18.6) Youth, sports, and recreational organizations 0 4 3 1 2 7 3 1 1 22 (12.4) Organizations for elderly people 0 4 1 0 0 3 1 1 0 10 (5.6) Subtotal 14 33 36 11 8 27 14 15 19 177 Table 6.1. cont. 14 de Colombia Embudo, Febrero, Jeric6, Amanecer, Rosario, El Area, P6rtico, Cachicamo, Chiquita, Type of institution Bogota Bogota Bogota Bucramaonga Gir6n Cali Medellin Yopal Aguazul Total Service delivery organizations Social service organizations 3 9 3 5 1 12 4 11 6 54 (27.8) State/government 1 4 7 5 5 0 2 12 9 45 z organizations (23.2) o NGOs 3 2 6 1 0 9 0 10 3 34 x (17.5) Statesecurity/ 2 3 0 2 2 2 1 5 5 22 justice institutions (11.3) Private sector 1 0 0 1 1 1 0 5 8 17 o organizations (8.8) Productive service 1 0 0 1 4 1 1 1 3 12 ° organizations (6.2) < Drug rehabilitation centers 3 0 2 1 0 4 0 0 0 10 ° (5.1) z z Subtotal 14 18 18 16 13 29 8 44 34 194 m Total 28 51 54 27 21 56 22 59 53 371 z Note: Figures in parentheses represent percent of total. x n a. Religious groups refers to churches and prayer groups only. Religious organizations that provide social services are included c in the appropriate service delivery category. Source: 92 focus groups. 0 z Table 6.2. Perceived Importance of Social Institutions Identified by Community Members, by Type 14 de Colombia Embudo, Febrero, Jeric6, Amanecer, Rosario, El Arca, P6rtico, Cachicamo, Chiquita, m Type of institution Bogotd Bogotd Bogota Bucaramanga Giron Cali Medellin Yopal Aguazul Total , 0 Social service 8 20 4 7 3 17 13 21 16 109 ) organizations (15.5) Neighborhood committees 2 26 10 5 12 21 6 4 12 98 > (13.9) State/government 3 4 8 6 7 0 2 33 19 82 organizations (11.7) Violence-related groups 5 11 13 0 0 11 6 15 10 71 (10.1) State security/justice 3 7 0 4 2 4 2 22 21 65 institutions (9.2) Women's and childcare 7 3 8 7 7 7 5 1 12 57 organizations (8.1) Religious groups, 4 10 10 3 6 1 3 8 10 55 (7.8) NGOs 4 2 7 1 0 19 0 13 3 49 (7) 0Ti Table 6.2. cont. 14 de Colombia Embudo, Febrero, Jerico, Amanecer, Rosario, El Arca, P6rtico, Cachicamo, Chiquita, Type of institution Bogotd Bogotd Bogotd Bucaramanga Gir6n Cali Medellin Yopal Aguazul Total Private sector organizations 1 0 0 1 5 3 0 14 17 41 (5.8) Youth, sports, and 0 5 3 2 4 11 3 1 3 32 recreational (4.6) Z Organizations for 0 7 2 0 0 5 1 2 0 17 0 0 elderly people (2.4) ° Drug rehabilitation centers 7 0 2 1 0 5 0 0 0 15 m (2.1) M Productive service 1 0 0 1 4 1 1 1 3 12 H organizations (1.7) 0 Total 45 95 67 38 50 105 42 135 126 703 0 Note: Figures in parentheses represent percent of total. 0 a. Religious groups refers to churches and prayer groups only. Religious organizations that provide social services are included m in the appropriate service delivery category. Z Source: 92 focus groups. X 0 z PERVERSE SOCIAL CAPITAL 77 Trust in Social Institutions Study participants were asked to indicate whether they viewed each institution positively (interpreted as indicating a high level of trust) or negatively (interpreted as indicating a low level of trust). Among mem- bership organizations, women's and childcare groups received the high- est percentage of positive rankings (88 percent; table 6.3). Neighborhood committees also received high ratings (75 percent). Among the service delivery organizations, drug rehabilitation centers and NGOs were the most highly trusted, followed by social service delivery organizations, a category that includes primarily schools and health centers. The least trusted institutions were those associated with perpetrating or preventing violence. Eighty-two percent of respondents viewed vio- lence-related groups unfavorably, and half of all respondents lacked con- fidence in the state security and justice institutions (the police, the army, the Administrative Security Department, and the judicial system). Table 6.3. Trust in Social Institutions Identified by Community Members (percent of respondents) High level Low level Type of institution of trust of trust Membership organizations Women's and childcare organizations 88 12 Youth, sports, and recreational organizations 86 14 Religious groups 79 21 Neighborhood committees 75 25 Organizations for elderly people 67 33 Violence-related groups 18 82 Total membership 64 36 Service delivery organizations Drug rehabilitation centers 100 NGOs 91 9 Social service organizations 87 13 Private sector organizations 72 28 Productive service organizations 67 33 State/government organizations 70 30 State security/justice institutions 49 51 Total service delivery 74 26 Source: 92 focus groups. 78 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Characteristics of Productive and Perverse Social Institutions The study distinguished between productive and perverse institutions. Productive institutions aim to provide benefits in order to improve the well-being of the community. Perverse institutions benefit their members but are usually detrimental to the community or society at large (see Rubio 1997). Most productive institutions were linked to the state. These included central government institutions, such as the Instituto Colombiano de Bienestar Familiar (Colombian Institute of Family Welfare), local govern- ment organizations, and the communal action committees, which received some funding from local governments but retained automony in their decisionmaking. An institutional map drawn by 20 residents of P6rtico, Medellin, reveals that the majority of the institutions identified by this group were influenced by the state (figure 6.1). Both the school and the health center Figure 6.1. Institutional Mapping of P6rtico, Medellin, Prepared by a Group of 20 Children, Adolescents, Adults, and Elderly People Municipal Health System of Paren~ache tenter (+) Pe A at They meet with Action t (+) J the people, but (risks) Block not with (+) otwithorgomizat _/ worga nizations uc ac / \Communit eaiyWmn 'AC \ / ~~~~~~~~~~~~~~~~~Program \/Co nity mothers ( Notes: \ + Relationship exists ~ CCommunity A Want to have relationship center + Positive evaluation s Size of circle represents level of importance PERVERSE SOCIAL CAPITAL 79 Box 6.1. Perverse Social Institutions in the Nine Barrios by Type of Violence Type of violence Institution Social Parche: A place or group of teenagers that meet to con- verse, drink, or consume drugs. May also involve acts of crime and violence. Usually based on flexible association and spontaneous congregation. Male-dominated but may have female members. Combo: A place or group that commits acts of crime and violence. Less open and more organized than a parche. Male-dominated but may have female members. Gallada: A gang of primarily male teenagers or adolescents that congregates in a parche. May commit crimes and acts of violence. May form into a pandilla if an identity and symbols are developed. Pandilla: A gang with a strong intemal organizational structure. Uses symbols and markings to denote gang identity. Comprises mainly men (usually in their early 20s) involved in delinquency, territorial disputes, and drug use. Economic Raponeros: Petty thieves or snatchers who steal from peo- ple in the streets. Operate individually or in groups and mainly comprise male children and adolescents. Atracadores: Thieves armed with guns or knives. Usually mug people in the streets, although some groups special- ize in particular types of attacks, such as attacks on taxis. Less organized than ladrones. Ladrones: Generic name for thieves. Some groups of ladrones specialize in particular types of theft or robbery. Apartamenteros: Thieves who specialize in theft from apartments. Banda: Group of male delinquents organized to commit crimes, primarily robbery and other acts of violence. A banda may offer its services to others. May specialize in a particular good, such as jewelry. Oficina: Group of organized male drug dealers or busi- nesspeople that hires others to commit acts of crime or violence. 80 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Box 6.1. Cont. Type of violence Institution Political Encapuchados: Literally, hooded people. Generic name for those who commit acts of crime and violence. May be sicarios, militia members, or guerillas. Sicarios: Paid assassins, usually contracted to kill for revenge. Often linked to social cleansing. Milicias populares: Organized militias that commit vio- lence, mainly through use of delinquents. Usually infor- mal protection/justice forces. Some linked to guerillas. Grupos de limpieza social: Social cleansing groups. Also known as paperos in Cali, rayas in Bogota, and capuchos in Medellin. Highly organized, male professional killers. Target groups are delinquents, beggars, drug addicts, petty thieves, street children, and prostitutes. Usually have links with the police, the military, or state security forces. Paramilitares/pdjaros: Paramilitary organizations known locally as the "birds." Usually linked with the extreme right. Includes a range of civil defense groups funded by landowners, emerald magnates, and drug traffickers and thought to be linked to the military and govermment. Ostensibly aim to protect civilians and eradicate the two main guerilla organizations. Highly organized, male- dominated structure. Guerilla/gatos: Guerilla organizations known locally as "cats." Includes the FARC (Fuerzas Armadas Revolucionarias de Colombia-Armed Revolutionary Forces of Colombia), a pro-Soviet guerilla group, and the ELN (Ejercito de Liberaci6n Nacional-National Liberation Army), a pro-Cuban guerilla group. Highly sophisticated organizations that control large areas of the national territory. were state run. The Communal Action Committee, the community moth- ers, and the FAMI Program (Family, Mother, and Child Program) received funds from the municipality. Indeed, the only institution not linked with the state was the Los Muchachos gang, which was seen as an PERVERSE SOCIAL CAPITAL 81 informal protection group. Los Muchachos was perceived as the most important perverse social institution in the community as well as the most influential. However, its only formal ties with other community organizations was through guarding the community center run by the Communal Action Committee. Gangs and Other Violent Organizations Respondents identified 15 types of organized groups that fostered per- verse social capital. Most of organizations were cohesive hierarchical units headed by a leader and governed by internal rules to which their members strictly adhered. Box 6.1 lists each of these groups by type of violence. Much of the violence committed by these groups was social in nature, involving the pursuit of power. As one community leader in Jeric6, Bogota, noted, the reasons for joining a gang are "often not economic. It's the desire for leadership, the force of power." Groups involved in economic violence were involved mainly in rob- bery and delinquency. Some groups, such as the apartamenteros, who steal from apartments, specialize in certain types of crime. Others, such as the bandas and oficinas, were organized based on Mafia-style structures. The bandas sold their services, while the oficinas contracted to others to com- mit crimes for them. The least organized group was the raponeros, usual- ly street children or adolescents involved in "snatching" in the streets. Organizations that commit political violence often had links with guerilla groups. In El Arca, Cali, for instance, some militias were closely associated with the National Liberation Army (ELN). A community leader noted, "The militias have the same components as the guerillas. The state views them as guerillas, the barrio views them as militias ... the militias have more power here than the guerillas." Such groups were widely known and identified by all members of the community, includ- ing children (figure 6.2). The most extreme violence-related organizations, the paramilitaries, or pajaros, and guerilla groups, or gatos, had the greatest influence on com- munities.6 In Colombia Chiquita, Aguazul, the paramilitaries and gueril- las dominated the daily lives of the population. Between 1996 and 1999, 2 community leaders and 12 other community members were killed, 2 people disappeared, and 10 families that had been threatened fled the barrio. One community leader noted, "The people have become accus- tomed to take away their dead." By 1999 the main threat was the para- militaries. Every Friday at 7 p.m. the carros lujosos (literally "luxury cars") arrived and terrorized the population. People suspected of talking to or having links to the guerillas were threatened and sometimes assassinat- ed. In El Arca, Cali, a variety of perverse institutions dominated the insti- 82 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION tutional landscape (figure 6.3). The militias were identified as the most influential force. Perverse Institutions Involved in Drug Sale Organizations selling or distributing drugs were critically important in the communities. These networks were extremely hierarchical, with pat- terns of authority rigidly and violently enforced. The most complex network of drug vendors and their accomplices was found in Embudo, Bogota. According to one woman, networks were Figure 6.2. Drawings of Guerillas by a 13-year-old Girl in El Arca, Cali / Y)<7 t;W , -1 Note: Ancillo ayudenme - help me. ELN, FARC - guerilla groups (see text). PERVERSE SOCIAL CAPITAL 83 Figure 6.3. Institutional Mapping of El Arca, Cali, Prepared by a Mixed Group of Four Community Leaders Note: Size of circle denotes importanceiofainstitution supportive/roles. Each street invariably hadtwo m cabezas (h members ), male Poremlitrpular miliias s asaro d (ea w a (clerks) who sold the d Biblia \ re c who\didno pa up on ctime Thetaquillero receivdunity 1 i0ia stero \ /Primarya~Pr \ / Center) radio station they sli2hos(o as tuaschool bu () ( ($18.0) aday. aquileroemplyedainl womena and cldencaulok outse,urity f of ce ns o ar o st \ ( leansing (odls Note: Size of circle denotes importance of institution. organized vertically, with authority maintained through violence (figure 6.4). Men held the positions of authority, with women usually playing supportive roles. Each street invariably had two male cabezas (heads), who lived outside the barrio. The cabezas distributed drugs through three male or female intermediaries know as jibaros (peasants), who also lived outside the conmmunity. Jibaros rigorously maintained their authority, killing those who broke their contracts. Jibaros controlled taquilleros (clerks), who sold the drugs to users. Taquilleros rarely consumed drugs or committed murder, but they did employ force-punching consumers who did not pay up on time. The taquilleros received 10 percent of what they sold in 24 hours (known as a tum), eaming about 30,000 pesos ($18.70) a day. Taquilleros employed mainly women and children as look- outs, advising them of police movements or the arrival of strangers. 84 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Figure 6.4. Drug Sale and Distribution in Embudo, Bogota, Analyzed by a 34-year-old Woman Has the largest amounts of drugs HEAD (male) Doesn't live in the zone 2nd-tier seller w o a a 4------- ~~~~~~~Men andwoe-ofn adon't live in the barrio F~~~~ -g 4- < \Advise when there are police or strangers - > a a \ mainly women and children Someone who sells, Sells a friend or relative of a jibaro - honest Drugs were sold in certain areas referred to as ollas (literally, stew-pot). Ollas were usually rented rooms within larger dwellings where drug sale and consumption occurred. Two female sex workers from Embudo, Bogota, noted that ollas were linked not only to drugs but also to attacks, murder, sexual violence, and the abandonment of children (figure 6.5). Often run by jibaros, ollas were also used to sell stolen goods, and many served as brothels for adult and child prostitutes. Causes of Perverse Social Institutions The prevalence and importance of gangs, militias, and other violent orga- nizations are an important indicator of the high level of distrust of state security and justice systems. This is closely associated with the long his- tory of internal conflict in the country. Fifty-one percent of respondents distrusted such organizations as the police, the military, the Administrative Security Department (which includes the secret police), and judicial institutions. Members of communities noted that informal, illegal organizations filled the gap left by lack of state intervention or the inability of the state to protect poor urban communities. PERVERSE SOCIAL CAPITAL 85 Figure 6.5. Function of an Olla in Embudo, Bogota, Drawn by Two Female Sex Workers (aged 35 and 48) Sell bozuco (Colombian crack) Destroys people Where children O LAAttacks are abandoned One enters fully Kill people for clothed and leaves the smallest thing without clothes Lots of people meet Drugs sold and consumed Perceptions of the Police Force The least trusted institution was the police force, with people in all com- munities holding negative views of the police. A man in P6rtico, Medellin, thought that the police were untrustworthy and exacerbated bad situations. As a result, "If I see someone getting killed here and the police arrive and ask questions, I say nothing. Here it's better that way." The police were often identified as major perpetrators of violence and were held in contempt in most communities. They were also perceived as having little authority. In Amanecer, Bucaramanga, a group of young men described the rela- tionship between the police and the residents of the barrio as highly antagonistic (figure 6.6). According to these young men, the police enter the barrio only to harass the population. They make about 10 rounds a day, constantly asking for papers from people they know. The police were perceived as highly corrupt and interested only in obtaining money through bribery and extortion. Occasionally, they arrive and fire shots to frighten residents. As a result, many children were afraid of them. People in Embudo, Bogota, held the most extreme views of the police and perceived them as major perpetrators of violence who promote rather than curtail the sale and consumption of drugs. According to com- munity members, the police run a well-organized system of extortion 86 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Figure 6.6. Perceptions of the Police Amanecer, Bucaramanga, Identified by a Group of Five Young Men (aged 14-23) It's not worth calling The police don't the police because care about robberies they do nothing in homes poThe police know o t s a Patrols arriveso who the thieves D d a to annoy phe are, yet accuse \ p / barrio ther socent cesig POLICE weapons of te barria and barrios believe that the police immediately release people afte arre t money i Ts cy take away They ocly see i oo ple and then whaus gooa for them. from drug dealers, resorting to violence when necessary. Eight-man police patrols work three shifts a day, each officer receiving 2,680 pesos ($1.70) per shift. Drug dealers and consumers usually met the police's demands because the police threatened them with violence or death through social cleansing. Perceptions of the Judicial System and the Government Distrust of the judicial system was also widespread. Most people in the barrios believeat the police immediately release people after arrest and that even if a case gets to court, justice is not served. The local justice institution, House of Justice in Bucaramanga, Santander, located near Amanecer, is illustrative. This was located within the barrio in an effort to improve access to conciliation and legal services for low-income peo- ple. Figure 6.7, drawn by the institution's director and two of his func- tionaries, shows the functions and components of the program in terms of their institutional links with other entities. Figure 6.8, drawn by a woman from the community, reveals a lack of knowledge of the pro- gram, associating it only with the police, a banking facility, the family PERVERSE SOCIAL CAPITAL 87 Figure 6.7. Institutional Links of the House of Justice in Bucaramanga, Identified by the Director and Two Functionaries Communal Action Committees Departmental Assemblies (in a range of communities) R Cross Municipal Councils ecreCommuityariommu Secretaria de Desarollo Soaoll| \ mother // (eretary of Social Development) | (Secretary of Youth) | _ t > ~~~~~~National Police l s CASA ~DE JUSII _4(House of Justice) Secretaria de Gobierno (Secretary of Y)ecretary of Governme USAID//\\ / / \ \\i ~~~~~~~~~Secretaria prhvada / / \ \l ~~~~~~~~~~(Private Secretary) l ^/ *¢ \ \-1 ~~~~~~~~~Secreai de Planeaa' on NGOs Judicial power (Secretary of Planning) -ALOC -Judicial council -Horizonte -SuperiorTribunal Schools -Centro JuvenilAmanecer -Escuela de Paz LIoptal -Fundacion Esruauar -Escuela FeyAllegria -Escuela Son Cristobal commissary (where domestic violence cases can be reported), and judi- cial services. A group of young men who knew more about the institution held pre- dominantly negative views, noting, "No one trusts the House of Justice.... It is the same as the police; it plays the same role." The men went on to say that the only justice in the barrio was ley de defensa (law of defense), the use of arms and force. The concept of justice by force was mentioned in most of the communities, where it was referred to as the "law of the strongest" and the "law of knives." In all cases, it related to the lack of alternatives in the absence of faith in the state judicial system and the presence of high levels of impunity. All perverse organizations were associated with the use of arms and informal justice systems, particularly social cleansing. In El Arca, Cali, for instance, a member of the M-19 guerilla group who had been "reinserted" into the community described how justice by force had become more violent over time. In the past, organized conflict- resolution procedures based on a system of warnings and community 88 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Figure 6.8. Perceptions of the Function of the House of Justice in Amanecer, Bucaramanga, Identified by an Adult Woman Judicial cases (my brother has used it) Family Commissary l Police -/ CASA DE JUSTICA a (House of Justice) r Bn meetings were used. For the most part, only informers were killed. Now, he reported, militias and bandas killed as a way of sending mes- sages to the community. Underlying the rapid growth in perverse social institutions was the perception that the power of the government as an effective law enforce- ment and welfare institution was declining. This was graphically illus- trated by an 11-year-old boy in Jeric6, Bogota, who identified the gov- ernment as a "killer" and "squeezer," diminishing in its importance and presence to the point of being a mere skeleton (figure 6.9). Perverse social institutions perpetrating political violence generated greatest fear in the frontier communities of Cachicamo, Yopal, and Colombia Chiquita, Aguazul. Although a minority of community mem- bers condoned the actions of these groups, most people feared them. In Colombia Chiquita, Aguazul, in particular, paramilitary organizations were associated with a range of negative consequences that undermined cognitive social capital (figure 6.10). Fear of reprisals from paramilitaries or guerillas made it impossible to talk openly in the community. As a result, solidarity was eroded and distrust created. This in turn created a society in which people had to look out for themselves to survive. Perceptions of the military were mixed in this community. Although rela- tions with the army have deteriorated, as the army was perceived to be doing little to prevent political violence, it was nevertheless looked to as a source of protection. PERVERSE SOCIAL CAPITAL 89 Figure 6.9. Depiction of the Government by an 13-year-old Boy in Jeric6, Bogota ((6U9 < ~ ~~~~~~~~~~~~~~ X 0'2 NO6t Note: Lo mato el gobierno: the government kills; Gobierno exprimi- dor: government squeezer. Reconstruction of Productive Social Institutions and Female-Dominated Organizations Which community organizations are the best vehicles for rebuilding cog- nitive social capital? As table 6.3 shows, the most trusted institutions were women's and childcare groups. In five of the eight communities in which these organizations existed, all were evaluated positively (100 per- cent trust level). Childcare organizations consist largely of state-run com- munity homes, while women's organizations include small NGOs and locally formed women's self-help groups. 90 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Figure 6.10. Causal Impact Diagram of the Effects of Paramilitary Groups on Colombia Chiquita, Aguazul, Analyzed by a Mixed Group of Four Adults Comunty Homesmen up(to five)years.>TheFamily, Mothe tra nP s (Poga ) pvderetforegnn ad Kla iniomen. T tion existedinall of the communities except aicam uopa meramte ithes cs a t hoes As mantry asheigto nin hoe functvional inmeach comniy Community Homes Two main types of cinsieth communi.es provide care. Social Welfare Homes (Hogares de Bienestar Social, HOBIS) provide childcare for children up to five years. The Family, Mother, and Child Programs (Programa Familia, Mujer, e Infancia del Instituto Colombiano de Bienestar Familiar, FAMI) provide care for pregnant and lactating women. These organiza- tions existed in all of the communities except Cachicamo, Yopal. muhe community homes are run entirely by women known as com- munity mothers, who operate these childcare groups from their own homes. As many as eight or nine homes function in each community. People trusted the community homes because they are operated by local women hiring inside the community. Overall levels of trust were high among all groups of people, even young men, who identified com- munity mothers as extremely important and a positive force in the com- munity (figure 6.11). PERVERSE SOCIAL CAPITAL 91 Figure 6.11. Institutional Mapping of El Arca, Cali, Drawn by A Group of Five Young Men = negative~ ~~~~N oo a wa Women's organizations were found to be extremely Martants c inter violence n Colombiaachiqut Ag u fr euousp o men and women identified the Women's ~iassocain(lolwme' self-help group) as the most imp~Loraterantdoiieisiuto ntecm i (h hinfluen wa Terronzed r it mot er information Social cweansing the h group/police k i p I positive --=negative Women's Organizations Women's organizations were found to be extremely important institu- tions in some communities, especially those experiencing high levels of violence. In Colombia Chiquita, Aguazul, for example, a group of adult men and women identified the Women's Association (a local women's self-help group) as the most important positive institution in the com- munity (figure 6.12). The only other influential organization was the Communal Action Cormmittee, which the focus group believed had lost sight of its origins and no longer kept its promises. Focus groups described the Women's Association as "successful," "trustworthy," and "well-functioning." One woman, who had been dis- placed from another department because of violence, decided to form the organization after experiencing violence at the hands of the paramili- taries and guerillas in Aguazul. The organization's aim was to reunite a 92 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Figure 6.12. Institutional Mapping of Colombia Chiquita, Aguazul, Drawn by a Mixed Group of Seven Adults Amanecer Municipal Assistance National System Foundation to Agriculture and of Education and (BP) (+) Fishing (+) Other churches (+) Learning (+) Catholic church (+) Guerillas Government (--) armilitaries - national Communal Action Committee (AC) (+) 0 OM, ~~~~~~~~~~Pramilitaries - Women's Association h Casanare (- Munipality (+/-) \N t \ V S (negative in the past) SportsX\ M committee (+) Perenco (oil company) (+) British Petroleum (-) National Army (+/-) (+) positive Police (positive in the past) (-) negative (+/-) neutral The size of triangle denotes perceived importance. community whose social fabric had been destroyed by violence by train- ing women to maintain their families, mainly by setting up their own small businesses. In this woman's view, women were more likely than men to be heard by the authorities because "there is always truth and sin- cerity in the eyes of women." The organization also managed to change the image of the barrio in the eyes of the authorities, who no longer view it as "full of thieves, drug addicts, and guerilla members." Despite the obstacles that faced her, the woman felt confident about the future. "I continue to lead the association without worrying what the macho men think. Maybe that's why they call me the mad woman." Chapter Seven Youth, Exclusion, and Violence Young people are at the apex of the nexus of violence, social exclusion, and the creation of perverse social institutions. If interventions to reduce violence are to be sustainable in the longer term, they must address the problem of youth exclusion. Perceptions of Exclusion Young people, particularly men, felt excluded, both from particular groups and organizations within their communities and from their com- munities themselves. Many young people complained that "no one takes us into account" and repeatedly mentioned rejection (rechazo). A group of young men in 14 de Febrero, Bogota, highlighted the following sources of exclusion. * Parents, especially mothers, who they claimed tried to prevent them from having friends and going out. * The Communal Action Committee, which never listened to them. * The police, who harassed them daily and engaged in social cleansing (killing young people in the barrio on the pretense that they are drug addicts). "They look at you as the worst. They don't come in peace but to fight." * The community in general, who blamed them for the drug problem in the barrio. As one man noted: "When the people begin to talk badly of you ... they invent that you are involved with vices [drugs]." Several young women also complained of discrimination. A group of five 14- to 17-year-old girls from Rosario, Gir6n, identified their main problem as "people talking badly of young people" and especially of young unmarried women who did not work. They discussed how a group of elderly women in the barrio called them muchachas vagas (lazy, corrupt young women), implying that they took drugs and hung around with men. 93 94 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Box 7.1. Exclusion of Young Women in Rosario, Gir6n The group listed stereotypes used in the community, which they felt contributed to social exclusion. Although they saw themselves as muchachas alegres, they were constantly called vagas, which they viewed as discrimination. muchacha vaga Muchacha alegre Muchacha sana (lazy, corrupt girl) (happy girl) (healthy girl) * Unemployed * Happy * Works * Spends all day in * Uncomplicated * Studies the street * Wears sporty clothes * Stays home * Smokes marijuana * Wears tennis shoes * Wears clothes similar * Wears tennis shoes * Wears necklace to her mother's (with friends' * Should not have a names on it) boyfriend (although * Has boyfriends may have one in secret) Source: Five young women. Young women felt that people in the community labeled young women as either muchachas vagas or muchachas sanas (healthy girls), refus- ing to acknowledge young women such as themselves, muchachas alegres (happy girls) (box 7.1). Although they admitted that "we are not nuns," they resented being blamed for things they did not do. They also resent- ed the fact that young men in the barrio who engaged in some of the same behaviors were not condemned. Causes and Effects of Exclusion A major cause of exclusion among young people was high levels of inter- generational conflict. Channels of communication between parents and their children were broken-often as a result of witnessing or being vic- timized by intrafamily violence resulting in family disintegration and feelings of isolation. As children turned away from their parents, the older generation began to blame young people for the ills of society. This intergenerational conflict was especially widespread in 14 de Febrero, Bogota, particularly among male community leaders of the Communist Party-influenced Provivienda committee, whose sons reject- ed the principles of collective action. Indeed, one group of three j6venes sanos (healthy young people), including the son of a community leader, disparaged communism, stating that individualism was much better. YOUTH, EXCLUSION, AND VIOLENCE 95 Some sons of Communists had become involved in drugs and gangs as way of rejecting their fathers' teaching. This type of conflict was not confined to 14 de Febrero or to fathers. In Jeric6, Bogota, a group of five female community leaders noted that some gang members were the sons of leaders. One woman whose son was in a gang pointed out, "I work for the community, and my son damages the community." Intergenerational conflict contributed to exclusion by the community as a whole. In many cases, young people withdrew from mainstream society, exacerbating their exclusion. In 14 de Febrero, Bogota, a group of male rock or punk musicians who dressed in black, shaved their heads, and espoused an ideology of anarchic communism were acutely aware of their exclusion. In their words, "The people don't like us. We are looked on as drug addicts and are associated with satanic cults." These young men gained strength from their music, embracing their exclusion, and what they call the intolerance of the community, as a coping strategy. For many young people another response to exclusion was drug use or involvement in crime and violence, often through gang membership. A group of six 14- to 18-year-old men in Rosario, Gir6n, identified the fact that "we're not recognized or taken into account" as a major issue (figure 7.1). They associated their exclusion with lack of educational and Figure 7.1. Problems Faced by Young People in Rosario, Gir6n, Identified by a Group of Six Young Men (aged 14-18) rcgized or *$ tkn into / 9o \ T / ( X~~~~~~~~~~~~~~~~~~~~r :1es \ 4~~~~~~~~~~~~~~~~~~~~~~~~n fuur 0~~~~~~~~~~~~~~~~~~~~~~i YOTHp 96 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Figure 7.2. Causal Impact Diagram of Causes and Effects of Groups of Male Youth on Street Corners in P6rtico, Medellin, Prepared by a Mixed Group of Eight Adults (all aged over 40) Economic Because capacity Lack of schools of want They don't Because of like to study conflict 4 A4 ~Am'-ong them Lack of education Lack of Friendships Rejection of male youth l \ t s t ; \ /p~~~~~~~~~~roblems Lack of repression (lots of freedom) Groups of MaLleYouth on Street Corners / T~~~~~~~~~~~~~~~~~hey create ---Tconflicts Some people are afraid With With the Stay at gangs community home Danger that Drug Children arent they are killed addiction allowed out Isolation from neighbors and from community in general recreational opportunities, noting that football tournaments in the barrio were organized either for children or adults. Exclusion led to drug use and loitering, as well as the lack of a promising future. Older people identified similar problems faced by young people in the barrios. In P6rtico, Medellin, a group of eight adults analyzed the problem of young men on street corners, providing a sophisticated account of the nexus of violence, exclusion, and social capital (figure 7.2). The causal fac- tors they identified included rejection, lack of education and employment opportunities, family problems, and the influence of friends ("bad com- pany").Young men got involved in drugs and gangs, which led to vio- lence. As a result, the community became afraid of them, contributing to the erosion of cognitive social capital. Exclusion was often exacerbated as a result of involvement in violence. For instance, a member of the Los Muchachos gang in P6rtico, Medellin, YOUTH, EXCLUSION, AND VIOLENCE 97 Figure 7.3 Effects of Addressing the Problem of Youth Exclusion in Amanecer, Bucaramanga, Identified by Two Young Men (aged 14 and 16) We would help in the programs We would make good friends We would be more sharing Everything would be great with all the neighbors Win the affection of the people Respect people Actions to be taken for us being taken into account Don't be bitter Do wherever possible to get on well with everyone around us noted that he felt rejected by the community because people ran away when he went by their houses. A drug addict from 14 de Febrero, Bogota, said he felt "rejected because of drugs" and wanted to give up drugs, have a family, and become part of the community again. Young men and, to a lesser extent, young women destroyed trust in their communities by generating fear within their communities and failed to lay the groundwork for social capital. Indeed, young people's exclusion from the functioning of the community undermined the cre- ation of ties and networks within communities-except those that devel- oped within perverse institutions. Two young men from Amanecer, Bucaramanga, who felt acutely excluded from their community, outlined what actions needed to be taken to include young people in the community (figure 7.3). They felt that if affection and respect toward young people were established, there would be more sharing, and community programs would be more efficient. As it stood, they did not participate in any community activi- ties because of the conflict between young people and the rest of the population. Chapter Eight Community Perceptions of Solutions to Violence Strategies for coping with violence can be divided into four types: avoid- ance, confrontation, conciliation, and other strategies (Moser 1996, 1998) (table 8.1). Most people reported adopting avoidance mechanisms, with "ignoring the situation" the most prevalent. Fear of retribution, powerlessness, and fear of exacerbating the situation prompted these responses. Particularly marked was the silence surrounding intrafamily violence (see chapter 3). Neighbors and friends did not intervene in this type of conflict, and fam- ily members maintained silence. One young man from P6rtico, Medellin, noted that he watched television when his father beat his mother, main- ly because he felt he could do little to help. Responses to political violence were also dominated by silence. In communities in which the level of political violence was high, "keeping one's mouth shut" was often neces- sary to avoid being killed. Another important avoidance strategy involved changes in mobility patterns. Many people avoid places where drug addicts or gangs congre- gate, changing their walking routes or staying home in the evening. Two j6venes sanos (healthy young people) from 14 de Febrero, Bogota, noted how they avoided drug addicts by taking a long route home. Many peo- ple also restricted their movements after dark, locking themselves in their homes in the evening when the incidence of violence increased. Women reported being particularly afraid to go out in the evenings. A group of young women from Rosario, Gir6n, identified the danger of going out at night as a major restriction in their lives. Some, for example, wanted to continue their education at night school but were afraid to leave their homes in the evening. Other community members noted that they avoided speaking to peo- ple involved in violence-related activities. Young people avoided falling into bad company or getting involved with friends who would lead them astray, particularly with respect to drugs. People avoided being robbed by not carrying valuables. A group of three adults from El Arca, Cali, 99 100 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Table 8.1. Strategies for Coping with Violence (percent of total) Percentage of total Strategy strategies cited Avoidance strategies Avoid bad company 4 Avoid people involved in violence-related activities 7 Remain silent/ignore situation 21 Avoid dangerous areas 11 Don't go out at night 10 Flee from attackers 11 Don't carry valuables 3 Avoid gossip 2 Leave barrio 2 Lock house/put bars on windows 4 Confrontation strategies Confront person causing problem 2 Carry weapons 2 Use violence 2 Conciliation strategies Develop relations with people involved in violence 2 Turn to religion; pray for those involved 2 Other strategies Report violence to family members or teachers 5 Report violence to police 2 Submit, cry, abort (in the case of rape) 5 Total 100 Note: Violence-related activities include drugs, insecurity, intrafamily violence, perverse social institutions (gangs, militias), rape, robbery, murder, and fights in the street. Source: 133 focus group listings. noted that one had to go out "without rings, without a watch, without luxury shoes, and without brand name clothes." Another basic avoidance strategy was to flee from thieves or gangs. Flight was also significant as a way of coping with intrafamily violence for both victims of and wit- nesses to intrafamily violence. Confrontation strategies were much less common than avoidance strate- gies because of fear of violence. Indeed, the few community members who cited confrontation as a response were either gang members or drug users. According to a member of the Los Muchachos gang in P6rtico, Medellin, "killing them first" was a way of dealing with murder in the COMMUNITY PERCEPTIONS OF SOLUTIONS TO VIOLENCE 101 community. A few people suggested that "hitting back" was an appro- priate way of dealing with intrafamily violence. Conciliation strategies were rare. In a few cases, young people got to know drug addicts or gang members, both to protect themselves and to make the addicts and gang members feel less excluded. Most people, especially women and the elderly, were afraid of these groups, however. Some said that they prayed for people comrnmitting acts of violence. The other mechanisms for dealing with violence were to report conflicts to the authorities or to family members or teachers (in the case of children and young people). Few people felt that reporting violence solved the problem, however. Most people felt that the only thing to do was to sub- mit. A group of four 11-year-old girls in Jeric6, Bogota, noted that they were afraid to tell their parents about rape for fear of rejection and could talk about it only with their friends. Their main strategy was to have an abortion and "be brave and deal with it as it comes." Three women in Colombia Chiquita, Aguazul, summarized many of the coping and conciliation strategies, including forgiveness and trusting in God (figure 8.1). With alcoholism a major problem in the community, they also suggested that the bars be closed early to prevent intrafamily violence and street fights among drunks. As this community was severe- ly affected by political violence, confrontation was not an option for fear of assassination. Interventions to Reduce Violence People in the communities had a variety of ideas for reducing violence. Almost half of the solutions were associated with social capital (tables 8.2 and 8.3). Within this category, the promotion of family values and dia- logue between family members and community members was the most important strategy (see table 8.3). Improving human capital represented almost a third of the interventions cited, while increasing physical capi- tal represented just a fifth of all proposed interventions. Increasing Social Capital Social capital interventions were the most important type of intervention in seven of the nine communities. Productive social capital interventions (building trust, integrating young people into society, and reforming pol- icy) outnumbered perverse social capital interventions (social cleansing, harsher police behavior) by a factor of three to one. The most important productive solutions were the generation of dialogue among both family and community members, the promotion of family values, and the build- ing of trust. In Jeric6, Bogota, for example, a family felt that the most appropriate solution to the problem of insecurity was "to unite the com- 102 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Figure 8.1. How to Cope with the Problem of Insecurity in Colombia Chiquita, Aguazul, Prepared by a Group of Three Adult Women (aged 25-63) Keep your mouth that you see Trust in God Go to bed at 6 pm \ / | How to cope with Close yourself the problem of insecurity in your house / ~~~~~~~~~~~~~~~~and lock up Know how to forgive Close the bars Aandon your plot early in the evening of land because of fear munity to make everyone alert" (figure 8.2). Education in conflict resolu- tion techniques was suggested in 14 de Febrero, Bogota, and P6rtico, Medellfn. Community organizations were an important solution for address- ing violence. Some community members suggested increasing the resources of existing organizations, such as the Communal Action Committees. Others suggested seeking external assistance in order to establish new organizations, which would be run and controlled from within the barrio. A group of seven community mothers in Jeric6, Bogota, identified a series of solutions to the problem of violence. They stressed the impor- tance of bottom-up solutions, noting, "Peace is not [attained by] throw- ing resources around without building projects from below, from fami- lies and community organizations." Their proposed solutions involved building both structural social capital (through community-based orga- nizations) and cognitive social capital (by building trust among those involved in violence) (box 8.1). Almost a third of the social capital interventions involved repressive activities, such as social cleansing. Community members who endorsed Table 8.2. Interventions for Reducing Violence, by Type of Capital and Community (percent of total) ) 0 14 de Colombia Embudo, Febrero, Jeric6, Amanecer, Rosario, El Arca, P6rtico, Cachicamo, Chiquita, z Intervention Bogotd Bogota Bogota Bucaramanga Gir6n Cali Medellfn Yopal Aguazul Total 4 Increase physical capital 16 17 17 23 28 21 11 20 45 21 Create jobs 5 6 4 9 8 9 11 6 24 8 Providehousing 3 0 2 2 0 5 0 0 3 1 r Improve infrastructure 0 1 11 8 0 5 0 2 0 3 Build more prisons 0 6 0 2 0 0 0 2 0 2 0 Increase household security (locks, bars) 0 2 0 2 8 2 0 0 0 2 0 Implement land reform cn in rural areas 2 1 0 0 4 0 0 6 12 2 ° Improve urban planning 6 0 0 0 0 0 0 0 0 1 Othera 0 1 0 0 8 0 0 4 6 2 6 Increase human capital 30 35 23 14 20 45 35 39 34 31 Z Improve academic and - vocational education 13 9 15 5 12 22 17 24 9 14 0 Provide more drug and alcohol rehabilitation 10 10 0 2 4 5 3 2 3 5 ° Provide drug and sex education 2 4 4 0 0 2 10 0 7 3 Provide self-esteem training 5 2 2 2 0 2 0 4 3 3 Provide sports and rec- reational opportunities 0 3 0 5 4 2 0 2 9 3 Improve health care 0 2 0 0 0 10 0 4 0 1 Provide conflict resolution education 0 5 2 0 0 0 5 0 0 1 , Provide family counseling 0 0 0 0 0 2 0 2 3 1 Table 8.2. cont. 14 de Colombia Embudo, Febrero, Jeric6, Amanecer, Rosario, El Arca, P6rtico, Cachicamo, Chiquita, Intervention Bogotd Bogota Bogota Bucaramanga Gir6n Cali Medellin Yopal Aguazul Total Increase social capital 54 48 60 63 52 33 54 41 21 48 Productive social capital C Promote family values t and dialo ue among > family and community Z members 20 11 21 12 4 17 32 7 3 14 0 Endorse effective peace 0 process 0 0 0 0 0 0 0 2 6 1 Help community organi- zations get started 10 7 2 9 20 0 0 6 0 6 n Develop programs to I integrate young people 7 into society 0 1 11 7 0 2 7 0 0 3 z Build trust 5 2 13 9 12 2 7 12 12 7 u) Reform the police 0 7 4 9 8 2 0 0 0 4 0 Perverse social capital Tighten police control 0 overcommunities 3 8 4 5 4 2 0 0 0 4 Increase social cleansing 11 6 4 7 4 5 7 0 0 6 Increase military protection/strengthen z state presence 4 6 0 5 0 2 0 12 0 4 0 Provide arms to citizens 0 0 0 0 0 0 0 2 0 1 m x Total 100 100 100 100 100 100 100 100 100 100 a. This includes solutions such as reporting violent incidents to authorities (Gir6n), leaving the country (Cachicamo), disarming o offenders (Cachicamo), and prohibiting sale of alcohol (Colombia Chiquita). Z Source: 133 focus group listings. COMMUNITY PERCEPTIONS OF SOLUTIONS TO VIOLENCE 105 Table 8.3. Interventions for Reducing Violence, by Type of Violence (percent of total) Intervention Percentage of all solutions Increase social capital 48 Productive 35 Perverse 13 Increase human capital 31 Increase physical capital 21 Total 100 social cleansing groups had lost faith in the police and army and believed that social cleansing represented the only hope for dealing with violent offenders. As a community leader from El Arca, Cali, noted, "Often the people are in favor of social euthanasia because the state doesn't respond." Figure 8.2. Interventions for Reducing Violence in 14 de Febrero, Bogoti, Identified by Seven Young Men (aged 12-15) r o problems \ ( ~~~~~~~~~~~~~~~~~Most prisons) Rehabilitation \ o center \/ + { ~~~~~Don't hit small center of Immediate children Assistance (CAl) - oie -int e 106 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Box 8.1. Seven Community Mothers' Approach to the Problem of Violence What do they do? Solution . Recognize violent people. * Create schools for parents on . Listen to them. peace building. * Make them participate in * Generate income. peace building. * Bring public and private resources * Do not reject them. within the community together, * Pacify them. without institutional corruption. . Provide moral support. * Increase participation of "Orientation." community organizations in decisionmaking. * Ensure that peace begins in the family, in the nursery, and in the school. While social cleansing was a repressive social capital solution in response to the absence of the state, the other types related to greater presence of the state security forces to deal with violence. In Cachicamo, Yopal, many people endorsed a greater military presence coupled with state provision of arms to help ordinary citizens defend themselves. Two small-scale cattle farmers called for "paid official protection from the state" as well as training the army in international humanitarian laws and human rights. Men favored repressive interventions much more often than women. Among the female-only and male-only focus groups that discussed inter- ventions, only one group of women favored social cleansing. Of 11 sin- gle-sex focus groups that suggested various types of repressive solutions, nine were male. Increasing Human Capital Education and drug and alcohol rehabilitation centers dominated the human capital interventions proposed by community members. In El Arca, Cali, human capital solutions were viewed as the most significant (see table 8.2). Also notable was the need for drug and alcohol rehabilita- COMMUNITY PERCEPTIONS OF SOLUTIONS TO VIOLENCE 107 tion in Embudo and 14 de Febrero, Bogota, and for drug and sex educa- tion interventions in P6rtico, MedellIn. Increasing Physical Capital Increasing physical capital was less important than increasing social or human capital in most of the nine communities. Such solutions were per- ceived as critical in Colombia Chiquita, Aguazul, however, where they represented almost half of all interventions. They were also important in Rosario, Gir6n, where they represented 28 percent of all solutions. Better Table 8.4. Recommended Interventions, by Type of Violence (percent) Physical capital Human capital Social capital Type of violence intervention intervention intervention Economic violence Drugs 20 32 48 Insecurity 26 17 57 Gangs 17 36 47 Robbery 28 16 56 Delinquency 14 43 43 Social violence Alcoholism 0 0 100 Fights 20 0 80 Murder 19 13 68 Violence against women 18 55 27 Domestic violence 18 45 37 Rape 9 39 52 Verbal abuse 11 22 67 Disputes between neighbors 14 0 86 Killings 14 29 57 Political violence Police abuses 0 50 50 Hooded gunmen/militias 0 0 100 Intimidation 50 0 50 War 43 30 27 Total Source: Focus groups. 108 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION community infrastructure, especially community centers, was also per- ceived as important. Interventions by Type of Violence Community members proposed different types of interventions to differ- ent types of violence. Social capital interventions were suggested most often to address economic and social violence; physical capital interven- tions were suggested most often to address political violence (tables 8.4 and 8.5). This was especially marked in Colombia Chiquita, Aguazul, where job creation was viewed as the most important solution to politi- cal violence. More than half the interventions proposed for reducing violence against women involved human capital. These interventions involved consciousness-raising activities for women as well as training to help women find jobs that would allow them to leave their husbands. Training and workshops for male perpetrators of violators were also suggested. Probably the most critical issue raised by local communities was their recognition that dealing with the continuum of violence in their commu- nities required implementing a variety of solutions simultaneously. This was illustrated by a group of members of the Los Muchachos gang, who identified a range of impacts, strategies, and solutions to socially moti- vated murder (box 8.2). Constraints and Recommendations The perceptions of the poor can help policymakers formulate public pol- icy. Local communities identified three national-level constraints to solv- ing the problems of violence: * The extensive nature of political violence. Negotiating peace among the guerrillas, paramilitaries, and other violence groups was perceived as Table 8.5. Recommended Interventions, by Type of Violence (percent of all interventions) Type of violence Physical capital Human capital Social capital Economic 22 26 52 Social 16 34 50 Political 40 27 33 COMMUNITY PERCEPTIONS OF SOLUTIONS TO VIOLENCE 109 Box 8.2. Impact, Strategies, and Solutions to Gang Murder Suggested by Three Young Gang Members in P6rtico, Medellin Impact What do you do? What would you * Shock and restraint. * Talk with the young like to do? * Personal insecurity. men, searching for * Maintain dialogue * Lack of trust among the why and for with other gangs. friends. what. * Try to make sure the * Negative psychological * Confront the young people keep and moral effects. problem. busy in their free time. * Negative atmosphere * Make sure the * Establish educational in the barrio. problem doesn't or night center in the * You look after your- get worse. school. self more. * Try to solve the * Create employment. * Community is problem in the shocked-no one friendliest way. goes out into the * Try to prevent the street, children problem from aren't allowed out. affecting the whole community. * Secure justice. an important precondition for the success of many other violence reduction interventions. * The problem of displaced people, which affects the daily lives of all com- munities. Local people perceive newly arrived displaced people as competitors for employment and income-generating opportunities, provoking economic and social violence. Even the basic needs of many displaced people are not being met, highlighting the importance of a policy and program agenda on displaced people. * Unemployment. All of the communities were concerned about govern- ment reforms necessary to pull the economy out of the worst recession in decades. While poor urban communities recognized that violence is not simply a consequence of the downturn in the economy, they understood that high levels of unemployment and stagnant local economies exacerbate political, economic, and social violence. 110 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Local community recommendations for reducing violence can be sum- marized in terms of seven priorities: * Create job opportunities in the formal, informal, and self-employment sectors. * Attack the problem of drug consumption. The high level of drug con- sumption was one of the most important concerns in most communi- ties. A comprehensive strategy must include drug (and to a lesser extent alcohol) consumption, both prevention and rehabilitation. Implementing such a solution requires collaboration among educa- tion, health, social welfare, and other sectors as well as between the government and NGOs. * Reduce society's tolerance for intrahousehold violence. All communities acknowledged that violence begins at home. Dealing with high levels of physical and mental abuse requires a strategy that encompasses both prevention and rehabilitation. Many agencies have already strug- gled long and hard to address different aspects of intrahousehold vio- lence. A holistic approach, with extensive interagency, media, and NGO collaboration, may help meet the demand from communities to address this issue. * Rebuild trust in the police and judicial system. The severe lack of confi- dence in the government's capacity to provide adequate police or judi- cial protection fosters the development of alternative informal justice systems and social cleansing and raises the level of fear and insecuri- ty. Despite extensive measures to address this issue (through, for instance, the introduction of local houses of justice), fundamental mea- sures are still needed to rebuild trust at the local level in order to elim- inate informal justice and social cleansing. a Strengthen the capacity of community-based membership organizations, par- ticularly those run by women. Childcare organizations run by women in the community and neighborhood associations are highly trusted by people in the communities. Both types of organization need to be strengthened to help reduce violence. * Target intervention for young people. Involving young people in the community was perceived as critical to overcoming violence-related problems (table 8.6). The most frequently cited intervention was workshops and talks with young people, usually about drugs and self-esteem. Establishing rehabilitation centers for gang members was also perceived as an important step toward reducing violence. One innovative solution to address the inclusion of young people into society suggested by community members was the development of small locally based community centers for young people, based on COMMUNITY PERCEPTIONS OF SOLUTIONS TO VIOLENCE 111 Table 8.6. Community Solutions to Including and Excluding Young People (percent of total) Percent Percent of all of all Interventions for solutions Interventions for solutions including youth cited excluding youth cited Offer workshops and talk 24 Have the police chase them. 5 to young people. Integrate young people 14 Don't give any charity to 5 into society. drug addicts. Provide rehabilitation for 14 Remove them from the barrio. 5 gang members. Advise young men of 10 Ignore them. 15 dangers, establish dialogue with them. Increase job opportunities 8 Kill them all. 15 for all young people. Provide young men with 8 Clean the streets of sources 5 vocational and technical of aggression. training. Employ gang members 6 Send them to the military. 10 and drug addicts. Provide more education 4 Don't mix with them. Mix 10 for teenagers. only with "healthy people." Provide psychological help/ 4 Do not get involved with 5 therapy. drug addicts. Develop relationships with 4 Build houses to displace young drug-addicted men. drug-addicted men. 5 Give them work in 2 Keep young men occupied 15 community-related type somewhere else. jobs. Acknowledge drug addicts 2 Don't allow children from 5 as human beings. other barrios here. Total 100 Total 100 Source: 155 focus group listings. 112 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION the "community mother" model and run directly by local men, women, and young people. People living in poor urban communities recognize that the problem of violence is so complex that it requires cross-sectoral solutions. They also recognize that local ownership is crucial if the sense of fear, power- lessness, and lack of trust is to be overcome. Given the conventional sec- tor divisions in line ministries and NGOs alike, developing such solu- tions is likely to prove challenging. Notes 1. Development practitioners use participatory urban appraisal as a research tool for sharing local people's knowledge and perceptions with outsiders. Chambers (1994a, 1994b, 1994c) provides comprehensive detailed reviews of the participatory rural appraisal approach. The techniques used in that methodology are also applicable in urban settings and are used here. The World Bank's Participation Sourcebook (World Bank 1995) includes participatory appraisals as one of the techniques currently being integrated into the Bank's operational work. The methodology has already been incorporated into several recent Bank studies, including country poverty assessments in Zambia, Ghana, and South Africa (World Bank 1994). The first study on urban poverty and violence using this methodology was undertaken in Jamaica in 1996 (Moser and Holland 1997). 2. For both translation and production purposes, the diagrams have been transferred into computerized form. 3. One of the participatory urban-appraisal ground rules is ownership of visu- al outputs through named acknowledgment. Given the nature of the issue being studied, however, anonymity was considered essential. 4. The terms "intrafamily" and "intrahousehold" are used interchangeably in this report. 5. Prostitution among young men was also noted in P6rtico, Medellin. A group of four teachers noted that when young men wanted to buy something or find money (which could be used for drugs), they sold sexual favors. 6. The term pajaro dates back to the Liberal/Conservative conflict that result- ed in La Violencia (1949-53). It purportedly refers to the way the Liberals threw the Conservatives out of airplanes, making them "fly." 113 Annexes Annex A. Participatory Urban Appraisal Tools Used in Study Annex B. Importance of Social Institutions Annex C. Problems Identified by Community Members 115 116 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Annex A. Participatory Urban Appraisal Tools Used in Study Participatory urban appraisals are usually based on focus group discus- sions, semistructured interviews, direct observation, ethnocentric histo- ries, and biographies and case studies of community members. Discussions with community members take place in local community centers, communal buildings, or in the street (on street corners, at football or basketball fields, outside people's houses). Various tools were used to gather information on the following themes: community characteristics; the history of the community and violent events; general problems and types of violence as perceived by communities; poverty; well-being and violence; changes in levels of vio- lence over time; social capital and exclusion; causes and consequences of different types of violence; social capital and mapping social institutions; and strategies and solutions for dealing with and reducing violence (Moser and Mcllwaine 1999). The specific tools used in the study includ- ed the following: * Matrix of general data. * Matrix of social organization. * Listing of general problems. * Ranking of general problems (scoring, onion diagram, flow diagram). * Listing of types of violence. * Ranking of types of violence (scoring, onion diagram, flow diagram). * Map of institutional relationships. * Preference matrix on social institutions. * Participatory map of the community. * Participatory map of secure and insecure places. * Matrix on history of the community. * Matrix on trends of general problems. * Matrix on trends of types of violence. * Daily, weekly, monthly, and annual time lines. * Causal flow diagram on types of violence and/or other problems. * Problem tree. * Listing of strategies to cope with violence. * Diagram of strategies to cope with violence. * Listing of solutions to reduce violence. * Diagram of solutions to reduce violence. * Drawings. Annex B. Importance of Social Institutions z z Table B.1. Importance of Social Institutions (percent of all institutions cited) 14 de Amanecer, Colombia All Embudo, Febrero, Jeric6, Bucara- Rosario, El Area, P6rtico, Cachicamo, Chiquita, commu- Problem Bogota Bogotd Bogota manga Gir6n Cali Medellin Yopal Aguazul nities Neighborhood committees 2 26 10 5 12 21 6 4 12 14 (98) Women's and childcare 7 3 8 7 7 7 5 1 12 8 groups (57) Institutions for the elderly 7 2 5 1 2 2 (17) Youth, sports, and 5 3 2 4 11 3 1 3 5 recreational organizations (32) NGOs 4 2 7 1 19 13 3 7 (49) Violence-related groups 5 11 13 11 6 15 10 10 (71) State security/justice 3 7 4 2 4 2 22 21 9 (65) Social services 8 20 4 7 3 17 13 21 16 16 (109) Drug rehabilitation 7 2 1 5 2 (15) Table B.1. cont. 14 de Amanecer, Colombia All Embudo, Febrero, Jeric6, Bucara- Rosario, El Arca, Portico, Cachicamo, Chiquita, commu- Problem Bogotd Bogotd Bogotd manga Gir6n Cali Medellin Yopal Aguazul nities State/government 3 4 8 6 7 2 33 19 12 institutions (82) Religious groups 4 10 10 3 6 1 3 8 10 8 > (55) Z Private organizations 1 1 5 3 14 17 6 o (41) 0 Productive services 1 1 4 1 1 1 3 2 (12) Total number of (45) (95) (67) (38) (50) (105) (42) (135) (126) (703) institutions Note: Figures in parentheses represent number of institutions. Source: 92 institutional listings, institutional mapping/Venn diagrams, and institutional preference matrices. z r z Annex C. Problems Identified by Community Members > z Table C.1. Problems Identified by Community Members (percent of all problems cited) .x 14 de Amanecer, Colombia All Embudo, Febrero, Jeric6, Bucara- Rosario, El Arca, P6rtico, Cachicamo, Chiquita, commu- Problem Bogota Bogota Bogota manga Gir6n Cali Medellin Yopal Aguazul nities Violence 58 63 49 52 35 41 21 34 23 43 Drug use 16 18 6 10 12 9 5 3 0 9 Insecurity 11 12 21 10 3 8 3 7 4 9 Robbery 5 11 3 14 5 13 0 6 1 6 Intrafamily violence 5 4 10 3 4 3 2 2 4 4 Loitering 0 4 0 6 9 1 3 1 1 3 Fights 5 6 2 2 0 1 0 2 1 2 Gangs 0 2 6 1 0 3 6 1 1 2 Killing 11 2 0 1 0 0 2 2 3 2 Rape 5 1 0 1 1 0 0 0 2 1 Alcoholism 0 4 0 2 0 3 0 1 1 1 Prostitution 0 0 0 1 1 0 1 2 0 1 Paramilitary groups 0 0 0 0 0 0 0 1 5 1 Threats 0 0 0 0 0 0 0 5 0 1 Guerrilla 0 0 2 0 0 0 0 1 0 0 Lack of physical capital 15 23 33 21 27 28 37 34 36 28 Unemployment 5 11 6 6 7 9 13 11 11 9 Lack of public services 0 2 13 9 9 7 7 8 11 7 Poverty 5 8 10 1 2 5 6 5 6 5 Table C.1. cont. 14 de Amanecer, Colombia All Embudo, Febrero, Jeric6, Bucara- Rosario, El Arca, P6rtico, Cachicamo, Chiquita, commu- Problem Bogotd Bogota Bogota manga Gir6n Cali Medellin Yopal Aguazul nities Housing problems 5 1 2 1 2 3 9 7 5 4 Transport problems 0 1 3 4 8 4 3 3 3 3 Lack of social capital 16 8 10 19 9 18 14 17 13 14 > Lack of unity 0 2 6 10 5 3 13 12 0 5 Z Absence of the state 0 0 3 1 2 2 2 2 11 3 8 0 Discrimination/ stigma 5 2 0 3 1 4 0 3 0 2 t Corruption 11 3 0 1 0 0 0 0 3 2 Distrust of police 0 2 0 3 1 9 0 1 0 2 n t'il Lack of human capital 10 6 8 6 16 11 27 12 17 12 - Lack of education 5 0 6 4 10 7 13 6 7 6 0 z Lack of health services 5 0 0 0 4 3 6 4 6 3 0 0 Lack of recreation 0 4 2 2 2 0 8 1 4 2 Hunger 0 2 0 0 0 2 1 1 0 1 0 Lack of natural capital 0 1 0 2 13 2 1 3 10 3 River (flooding) 0 1 0 1 13 1 0 1 7 2 Environmental hazards 0 0 0 0 0 1 1 1 1 1 Erosion 0 0 0 1 0 0 0 1 1 0 Z 0 Natural disasters 0 0 0 0 0 0 0 1 0 m Total 100 100 100 100 100 100 100 100 100 100 ) Source: Focus groups. 0 Bibliography Baulch, B. 1996. "The New Poverty Agency: A Disputed Consensus." IDS Bulletin 27 (1): 1-10. Chambers, R. 1992. Rural Appraisal: Rapid, Relaxed and Participatory. IDS Discussion Paper 311. Brighton. - . 1994a. "The Origins and Practice of Participatory Rural Appraisal." World Development 22 (7): 953-69. - . 1994b. "Participatory Rural Appraisal (PRA): Analysis of Experience." World Development 22 (9): 1253-68. . 1994c. "Participatory Rural Appraisal (PRA): Challenges, Potentials and Paradigms." World Development 22 (10): 1437-54. 1995. Poverty and Livelihoods: Whose Reality Counts? IDS Discussion Paper 347. Brighton. Chemick, M. 1997. "Changing Perceptions of Violence in Colombia and Its Implications for Policy." Washington, D.C. Coleman, J.S. 1990. Foundations of Social Theory. Cambridge, Mass.: Harvard University Press. Collier, P. 1999. "Social Capital and Poverty." Social Capital Initiative Working Paper 4. Washington, D.C.: World Bank. De Haan, A. 1998a. "'Social Exclusion': An Alternative Concept for the Study of Deprivation." IDS Bulletin 29 (1): 10-19. .1998b. "Social Exclusion in Policy and Research: Operationalising the Concept." In J. Figueredo and A. de Haan, eds., Social Exclusion: The Way Forward in Policy and Research. Geneva: International Labour Organization. De Haan, A., and S. Maxwell. 1998. "Poverty and Social Exclusion in the North and South." IDS Bulletin 29 (1): 1-9. Faynzylber, P., D. Lederman, and N. Loayza. 1998. "What Causes Violence Crime?" Washington, D.C.: World Bank, Office of the Chief Economist, Latin America and the Caribbean. Feldman, T.R., and S. Assaf. 1999. "Social Capital: Conceptual Frameworks and Empirical Evidence: An Annotated Bibliography." Social Capital Initiative Working Paper 5. Washington, D.C.: World Bank. 121 122 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Fukuyama, F. 1995. Trust: The Social Virtues and the Creation of Prosperity. London: Penguin Books. Gore, C., with contributions from J.B. Figueiredo and G. Rodgers. 1995. "Introduction: Markets, Citizenship and Social Exclusion." In G. Rodgers, C. Gore, and J.B. Figueiredo, eds., Social Exclusion: Rhetoric, Reality, Responses. Geneva: International Institute for Labour Studies/United Nations Development Programme/International Labour Organization. Groothaert, C. 1998. "Social Capital: The Missing Link?" Social Capital Initiative Working Paper 3. Washington, D.C.: World Bank. . 1999. "Local Institutions and Service Delivery in Indonesia." Local Level Institutions Working Paper 5. Washington, D.C.: World Bank. Habermas, J. 1972. Legitimation Crisis. Beacon Press. Harriss, J., and P. De Renzio. 1997. "An Introductory Bibliographic Essay. 'Missing link' or Analytically Missing? The Concept of Social Capital." Journal of International Development 9 (7): 919-37. Hobsbawn, E. 1999. On the Edge of a New Century, Free Press. IILS/UNDP (Intemational Institute for Labour Studies/United Nations Development Programme). 1996. Social Exclusion and Anti-Poverty Strategies. Geneva: International Institute for Labour Studies. Krishna, A. and N. Uphoff. 1999. "Mapping and Measuring Social Capital: A Conceptual and Empirical Study of Collective Action for Conserving and Developing Watersheds in Rajasthan, India." Social Capital Initiative Working Paper 13. Washington, D.C.: World Bank. Lederman, D., N. Loayza, and A.M. Menendez. 1999. "Violent Crime: Does Social Capital Matter?" World Bank, Poverty Reduction and Economic Management Network, Washington, D.C. Moser, C. 1996. Confronting Crisis: A Comparative Study of Household Responses to Poverty and Vulnerability in Four Urban Poor Communities. Environmentally Sustainable Studies and Monograph Series 8. Washington, D.C.: World Bank. . 1998. "The Asset Vulnerability Framework: Reassessing Urban Poverty Reduction Strategies." World Development 26 (1): 1-19. Moser, C., and J. Holland. 1997. "Urban Poverty and Violence in Jamaica." World Bank Latin American and Caribbean Studies Viewpoints. Washington, D.C.: World Bank. Moser, C., and C. Mcllwaine. 1999. "Participatory Urban Appraisal and its Application for Research on Violence." Environment and Urbanization 11. Narayan, D. 1997. Voices of the Poor: Poverty and Social Capital in Tanzania. Environmentally and Socially Sustainable Development Studies and Monograph Series 20. Washington, D.C.: World Bank. BIBLIOGRAPHY 123 . 1999. "Complementarity and Substitution: The Role of Social Capital, Civic Engagement and the State in Poverty Reduction." Washington, D.C.: World Bank, Poverty Reduction and Economic Management Network, Poverty Group. Narayan, D., C. Chambers, M. Shah, and P. Petesch. 1999. "Global Synthesis: Consultations with the Poor." Washington, D.C.: World Bank, Poverty Reduction and Economic Management Network, Poverty Group. North, D. 1990. Institutions, Institutional Change, and Economic Performance. New York: Cambridge University Press. Olson, M. 1982. The Rise and Decline of Nations: Economic Growth, Stagflation and Social Rigidities. New Haven, Conn.: Yale University Press. Oxford Analytica. 1999. [www.oxan.com]. Portes, A. 1998. "Social Capital: its Origins and Applications in Modem Sociology." American Review of Sociology 24 (1): 1-24. Putnam, R. 1993. Making Democracy Work: Civic Traditions in Modern Italy. Priceton, N.J.: Princeton University Press. Putzel J. 1997. "Accounting for the 'Dark Side' of Social Capital: Reading Robert Putnam on Democracy." Journal of International Development 9 (7): 939-49. Ravaillon, M. 1992. Poverty Comparisons: A Guide to Concepts and Measures. Living Standards Measurement Study Working Paper 64. Washington D.C.: World Bank. Rodgers, G. 1995. "What is Special about a Social Exclusion Approach?" In G. Rodgers, C. Gore, and J.B. Figueiredo, eds., Social Exclusion: Rhetoric, Reality, Responses. Geneva: International Institute for Labour Studies/United Nations Development Programme/ International Labour Organization. Rubio, M. 1997. "Perverse Social Capital: Some Evidence from Colombia." Journal of Economic Issues 31 (3): 805-16. Serageldin, I. 1996. Sustainability and the Wealth of Nations: First Steps in an Ongoing Journey. Environmentally Sustainable Studies and Monograph Series 5. Washington, D.C.: World Bank. Serageldin, I., and Steer A., eds. 1994. Making Development Sustainable: From Concepts to Action. Washington, D.C.: World Bank. Silver, H. 1995. "Reconceptualizing Social Disadvantage: Three Paradigms of Social Exclusion." In G. Rodgers, C. Gore, and J.B. Figueiredo, eds., Social Exclusion: Rhetoric, Reality, Responses. Geneva: International Institute for Labour Studies/United Nations Development Programme/ International Labour Organization. Solimano, A. (ed.). 1998. Social Inequality: Values, Growth and the State. The University of Michigan Press. 124 URBAN POOR PERCEPTIONS OF VIOLENCE AND EXCLUSION Solimano, A. (ed.). 2000. Essays on Peace and Development: the case of Colombia and International Experience, Conflict Prevention and Post Conflict Reconstruction Series, The World Bank, Washington, DC. Solimano, A., E. Aninat, and N. Birdsall, (eds.). 2000. Distributive Justice and Economic Development, The University of Michigan Press. Uphoff, N. 1997. "Giving Theoretical and Operational Content to Social Capital." Ithica, N.Y.: Cornell University, Government Department. Vanderschueren, F. 1996. "From Violence to Justice and Security in Cities." Environment and Urbanization 8: 93-112. Woolcock, M. 1998. "Social Capital and Economic Development: Toward a Theoretical Synthesis and Policy Framework." Theory and Society 27 (2): 151-208. World Bank. 1995. "World Bank Participation Sourcebook." Environment Department Papers. Washington, D.C. . 1997. Country Assistance Strategy for the Republic of Colombia. Washington, D.C. . 1998. "Violence in Colombia: Towards Peace, Partnerships and Sustainable Development." World Bank Sector Study. Washington, D.C. . 1999. "Violence in Colombia: Building Sustainable Peace and Social Capital." Environmentally and Socially Sustainable Development Sector, Management Unit Report 18652-CO. Washington, D.C. World Bank. 2000. Violence in Colombia: Building Sustainable Peace and Social Capital, The World Bank, Washington, DC. * *-ZgSida THE WORLD BANK 1818 H Street, N.W Washington, D.C. 20433, U.S.A. TELEPHONE: (202) 477-1234 FACSIMILE: (202) 477-6391 TELEX: MCI 64145 WORLDBANK MCI 248423 WORLDBANK INTERNET: www.worldbank.org E-MAIL: books@worldbank.org 11131 9 780821 347317 ISBN 0-8213-4731-4