MSI Evaluation Early Warning System MSI Management Systems International EVALUATION OF COLOMBIA’S EARLY WARNING SYSTEM MSI Evaluation Team Marc Chernick, Team Leader Susan Burgerman Danilo Rojas November 3, 2004 3 Executive Summary Management Systems International was contracted to undertake a complete evaluation of the Early Warning System (EWS) currently functioning in Colombia with the support of USAID. Colombia’s Early Warning System was designed to prevent massive and systematic violations of human rights and international humanitarian law. There is currently no comparable Early Warning System operating anywhere else in the world. For USAID, as the principal force behind its inception and implementation, the EWS represents a powerful experience that might be applied to other conflictive regions. Indeed, there is a growing international interest to understand and learn from this singular Colombian experience to protect civilians from the consequences of violence and internal armed conflict. Since 2001, the Early Warning System has successfully prevented many massive violations by analyzing the conflict at the local and regional levels, anticipating threats, alerting the appropriate authorities and implementing protective measures. Yet the evaluation team found weaknesses throughout the system. Addressing these flaws could substantially increase the effectiveness of the program and further reduce human rights violations. The principal conclusions and recommendations of the evaluation are:  Colombia’s Early Warning System is an essential instrument of government policy to protect Human Rights and ensure the application of IHL. Despite the flaws discussed in the evaluation, the alerted violations did not occur after a Risk Report or Early Warning was issued in approximately 50% of the cases.  Given the current humanitarian crisis and societal demand for protection in conflict zones, it is imperative – and possible -- to make the system even more effective and improve its success rate significantly.  Although there are a variety of national and international programs in Colombia to promote democratic security and/or humanitarian assistance, there is currently no other program that can potentially respond as effectively to emergency and medium term situations where the civilian population is at risk as a result of the ongoing armed conflict.  An early warning system should have the capacity to identify both emergency threats of an imminent nature, and threats that are gathering and will likely result in violations within a [3 – 6 month] period. Each type of threat should trigger an appropriate and timely response, the nature of which will depend on the conditions of the situation. At present, the system does not function adequately in either terrain. It is often too slow and cumbersome for emergencies and is only beginning to develop more comprehensive strategies to address medium-term threats.  At present, the recommendations by the Inter-Institutional Committee for Early Warnings (CIAT) over-emphasize security responses and ongoing military operations to address 4 impending threats and under-utilize broader conflict resolution, accompaniment and development measures.  The warning and response components of the Early Warning System are not well￾integrated. Moreover, the response component headed by the CIAT is not legally constituted and has no legal authority to require compliance on the part of local civilian authorities. Local authorities are not obligated to inform the EWS/CIAT of what measures they took specifically in response and what the outcome was in terms of human rights and IHL. The informality of these arrangements lowers compliance, reduces the threshold of urgency and hinders the subsequent evaluation of the process.  The program’s future remains tentative given the limited capacity of the Colombian Government to increase its material commitment. Efforts must be made to ensure the future sustainability of the system.  There is a need to better coordinate the EWS with the work of international humanitarian and development agencies, including the Humanitarian Situation Room run by OCHA, and to involve greater participation of relevant human rights and humanitarian NGO’s. Some of the key general recommendations are: 1. The EWS must function as an integral arm of the state’s conflict prevention and human rights policies and as part of its commitments under international law to protect the civilian population from the effects of the armed conflict. a) To meet emergency needs, the CIAT should be upgraded into a fulltime 24-hour unit with the legal authority and sufficient rank to make decisions and require immediate action. The relevant local authorities should report back within a reasonable time both what actions were taken [maximum 1 week] and what were the effects of those actions [monthly during the course of the alert activation]. b) Early Warnings, issued by CIAT, should carry a great sense of urgency and should compel local decision-makers to prioritize specific security and civilian response measures to address the situation. Risk Reports issued by the Human Rights Ombudsman that are not elevated to the level of Early Warning by the CIAT should convey equal urgency but should indicate a slightly longer time frame, permitting local authorities to develop more coherent, integral and enduring responses. c) When making recommendations, CIAT needs to go beyond the limited repertoire of measures that it currently advocates. At present, these are too general and mostly pass the baton to departmental security council meetings. CIAT should recommend a variety of security measures and civilian-led initiatives to confront both emergencies and medium-term threats outlined in either the Risk Reports or Early Warnings. In general, the more imminent threats will likely require a greater mix of security measures while the medium-term threats should more fully incorporate the broader initiatives involving state 5 civilians, international community and select NGOs. Non-security measures can include: state protection through the Human Rights Ombudsman office [on the model of the Community Ombudsmen program created with the assistance of the Government of Sweden], social investment, humanitarian action, national and international accompaniment, and greater community input and participation in the protective measures. d) Since November 2002 with the creation of the CIAT, all Risk Reports and Early Warnings have been cloaked under the cover of National Security, reversing the earlier public distribution of the Ombudsman’s Early Warnings. There has been great concern among international humanitarian organizations and NGOs about the changed policy which basically removes all but a few state actors from the process. A human rights policy should be as transparent as possible. It is strongly recommended that the Risk Reports be made public and available to all concerned citizens and national and international actors. The greater openness will make the system more effective and responsive. 6 TABLE OF CONTENTS Executive Summary 2 I. COLOMBIA’S EARLY WARNING SYSTEM FOR THE 2 OF MASSIVE AND SYSTEMATIC VIOLATION OF HUMAN RIGHTS 7 A. The Dynamics of the Armed Conflict and Massive Human Rights 7 Violations, 2000-2004 B. Objectives of the Early Warning System and Anticipated Patterns of Human Rights Violations 2004-2007 11 II. THE EARLY WARNING SYSTEM – TWO PHASES 14 III. ASSESSMENT RESULTS A. MEASURING RESULTS. HOW EFFECTIVE IS THE EARLY WARNING SYSTEM? 21 B. EVALUATING THE PROCESS 25 C. PROCESS RESULTS: 1. Early Warning System: Assessing the Alert Function 26 2. Inter-Institutional Committee on Early Warnings, CIAT: Assessing the Response Function 28 D. HOW SHOULD THE EWS FIT INTO A NATIONAL HUMAN RIGHTS VIOLATION PREVENTION PROGRAM? 33 IV. PRINCIPAL CONCLUSIONS 35 V. GENERAL RECOMMENDATIONS 38 A. OTHER RECOMMENDATIONS 40 1. Organizational 40 2. Logistical 41 3. Methodological 41 APPENDIX 1: Total Risk Reports and Occurrences of Violation 2003 44 APPENDIX 2: Risk Reports and Occurrences of Violations 2004 45 APPENDIX 3: Case Study: Department of Arauca 46 APPENDI X4: CASE study: Department of Chocó 55 APPENDIX 5: Case Study: Department of Norte de Santander 66 APPENDIX 6: The Early Warning System and International Humanitarian Law 73 APPENDIX 7: List of Persons Interviewed 75 APPENDIX 8: List of Documents Consulted 82 7 I. COLOMBIA’S EARLY WARNING SYSTEM FOR THE PREVENTION OF MASSIVE AND SYSTEMATIC VIOLATIONS OF HUMAN RIGHTS Colombia’s Early Warning System was designed to prevent massive and systematic violations of human rights and international humanitarian law (IHL). It is the only such system of its kind operating anywhere in the world. Since its initial implementation in 2001, it has successfully prevented many violations by analyzing the conflict at the local and regional levels, anticipating threats, alerting the appropriate authorities and implementing protective measures. Yet the evaluation team found weaknesses throughout the system. Addressing these flaws could substantially increase the effectiveness of the program and further reduce human rights violations. a) The Dynamics of the Armed Conflict and Massive Human Rights Violations, 2000-2004 The armed conflict in Colombia, one of the oldest violent conflicts in the world, is deep￾rooted, complex and disturbingly dynamic in its geographic extension and shifts in patterns of human rights violations. The guerrillas and paramilitaries change strategies to defend or open up access to key economic resources, to control strategic military or contraband corridors, or to pursue new political-military objectives designed to influence and control civilian populations. They also adjust their operations in response to state counter-terrorism and anti-narcotics policies. Since 2000, the transformation of the conflict has been dramatic. During the presidency of Andrés Pastrana (1998-2002), the Colombian government engaged in separate peace talks with two guerrilla movements, the FARC (Revolutionary Armed Forces of Colombia) and the ELN (National Liberation Army). Reflecting the prevailing view at the time and during the previous two decades of attempted peacemaking, the rightwing paramilitary group, the AUC (United Self-Defense Forces of Colombia) was excluded from the negotiating table. The peace talks with the FARC broke down definitively in February 2002 after three and a half frustrating years, while the peace process with the ELN never got off the ground in a significant manner. Worse, throughout the period of negotiations, both the FARC and the AUC increased their military strength and expanded their attacks on the civilian population in clear violation of international humanitarian law. By the presidential elections in May, 2002, massacres, kidnappings, selective assassinations, involuntary recruitment and forced displacements of families and communities had reached new heights. The AUC continued a strategy of targeting civic activists, human rights workers, labor leaders and communities believed to support the guerrillas. Following the rupture of the peace process, the FARC announced a strategy to destabilize local governance. Local officials were told: resign or face assassination. Over the next two years, 22 mayors and 155 city councilmen were killed and several others were “disappeared”.1 1 República de Colombia, Informe annual de Derechos Humanos y DIH 2003 (Bogotá, 2004). 8 In response to the deteriorating situation of public order and the growing humanitarian and human rights crisis, the Colombian electorate in the presidential election turned to a candidate who promised a new hard-line policy against the illegal armed actors and a more robust state presence throughout the national territory. During his first two years in office, the new president, Álvaro Uribe Vélez, developed a Democratic Security Policy that increased the numbers of police and military, created a network of “informants” and returned the police to every one of Colombia’s 1125 municipalities. When President Uribe took office, over 200 municipalities were without any police presence and an additional 200 had only minimum police protection. To help consolidate these policies and the presence of state authorities at the local level, the National Police launched the Municipios y Departamentos Seguros program in March, 2004. This program is designed to complement and reinforce President’s Uribe Democratic Security Policy by developing security plans implemented by mayors and governors in coordination with the National Police. A key feature of this program has been the establishment of municipal crime and violence “observatories” that assist local officials to understand present and emerging threats.2 President Uribe’s Democratic Security Strategy has clearly impacted on the course of the war. By 2004, the number of gross human rights violations had declined compared to their height in the 2000-2002 period. Kidnappings, massacres and general homicides all showed a downward trend. Nevertheless, despite the gains made under the Democratic Security Policy, human rights violations remain at crisis levels under any comparative standard. Though violations have declined in general – depending on the specific human rights or IHL violation or on the targeted group, i.e. labor leaders, indigenous population, etc. -- overall violations have fallen mostly in relation to the peak of the 2000-2002 period but have remained at the extremely high levels of the mid-1990s. Moreover, in certain departments, including Arauca, Chocó and Norte de Santander where the MSI team visited, overall violations have increased (see Appendices 3-5). Further, even as the country struggles to address still unacceptably high levels of forced internal displacement and massacres, the Human Rights Ombudsman’s Office and human rights organizations have identified and have begun collecting data on a new phenomenon – multiple, related homicides (homicidios de configuración multiple). This new category of human rights violations appears to reflect one of the ways in which illegal armed actors have adjusted to human rights policies and monitoring. In general, human rights groups count the collective killing of three or more persons as a massacre, although there is no standard national or international methodology, and the number has changed over time and varies among organizations in Colombia. “Multiple, selective homicides” represent the collective killing of three or more individuals drawn out over a period of time – days or weeks. The murders are categorized as homicides and not massacres, thus effecting the statistical reporting. Moreover, the EWS has registered an increase in several categories, including selective homicides, forced disappearances, and “methods to terrorize the civilian population.” 2 See the Colombia Program at the Center for Latin American Studies, Georgetown University. Research: Support for design and implementation of Local and Departmental Programs of Violence and Crime prevention and reduction http://www.georgetown.edu/sfs/programs/clas/Colombia/colombia_program_research_violence.htm 9 A look at the evolution of the leading human rights and IHL violations as determined by the number of incidents and percentage of total cases that sparked an alert by the HR Ombudsman’s Early Warning System is instructive to understand the evolution of the conflict from the perspective of human rights and international humanitarian law (Also See Appendix 6) In 2002, the leading categories of human rights and IHL violations that prompted an Early Warning response were: Forced Displacements 29.5% of total warnings emitted Massacres and multiple, related homicides 24.5% Indiscriminate attacks on civilian population 20.9% * Source: Defensoría del Pueblo. Note that through October 2002, the HR Ombudsman issued Early Warnings. After this period, the system was reorganized and the HR Ombudsman issued Risk Reports. These figures are based on both categories. See section III of this report. In 2003, the leading abuses were: Forced Displacement 29.6% Massacres and multiple related homicides 23.6% Indiscriminate attacks on civilian population 21.4% Source: Defensoría del Pueblo During the first four months of 2004, the HR Ombudsman separated the category massacre and multiple related homicides and recorded a significant rise in other categories of violations, reflecting changed tactics within the conflict: Selective homicides 14.4% Forced displacement 13.4% Forced disappearance 13.4% Methods to generate terror in the population 12.4% Multiple related homicides 11.5% Massacres 9.6% Source: Defensoría del Pueblo Even as patterns of human rights violation have changed, certain abuses stand out and appear to be a harbinger of the type of atrocities that are likely to occur as the dynamic of the conflict changes once again. On June 15, 2004, one of the worst massacres in recent Colombian history occurred in La Gabarra, a remote village in Norte de Santander along the Venezuelan border. It was perpetrated by the FARC against 34 raspachines or coca leaf pickers in a farm believed to be controlled by the AUC. The FARC entered the zone with a small unit of 12 to 15 fighters. Many analysts we interviewed declared that such actions vividly portray the new face of the conflict: conflict over territory, corridors and economic resources – particularly in non￾traditional coca growing zones -- perpetrated by small armed bands that are difficult to detect. 10 Further, as the state has increased its presence in many of the central conflict zones and has actively eradicated the large concentrations of illicit crops in the southern part of the country, illicit crops, guerrilla forces and/or paramilitary groups have moved into heretofore more peripheral areas of the country (and of the conflict), particularly Chocó, the Sierra Nevada de Santa Marta, Arauca and remoter areas of the Amazon Basin. In so doing, the conflict has disproportionately affected the indigenous and Afro-Colombian populations in recent years. As measured by populations at risk in areas of massive violations, in 2003, the HR Ombudsman reported that 29.5% of all Risk Reports were aimed at protecting indigenous communities, and 12.5% were stimulated by threats against the Afro-Colombian population. The newer trends of the conflict tend to reinforce this pattern. This brief overview of new patterns of conflict indicate that despite many positive steps taken by the GOC during the past few years, Colombia continues to face a profound humanitarian and human rights crisis. Jan Egeland, the United Nations High Commissioner for Humanitarian Assistance, in a visit in May 2004, declared Colombia to be one of the worst humanitarian crises in the world, exceeded only by Congo and Sudan—though he emphasized that Colombia’s is a largely “forgotten humanitarian crisis.” He based this assertion on the more than two million Internally Displaced Persons (IDPs), the existential threat to many indigenous communities as a result of the armed conflict and the stepped-up attacks on the civilian population by the armed combatants in the pursuit of control of strategic military and economic corridors.3 The Uribe government has not ruled out negotiations with the illegal groups. In a break from the previous twenty years of negotiating strategies in Colombia, President Uribe initiated peace talks with the AUC paramilitary forces with the aim of demobilizing one of the central actors in the conflict. These talks advanced throughout 2004 and have established a goal of complete demobilization by December 2005. A ceasefire was declared and key groups and leaders were assembled under state protection -- including a suspension of arrest warrants -- in a 368 square kilometer zone in Santa Fe de Ralito (an area in the Municipality of Tierralta, Cordoba). Successful negotiations would substantially restructure the nature and scope of the conflict and would likely help consolidate the trend toward a diminution of massive violations. Further, in 2004, the government accepted an offer from the Mexican government to help facilitate a peace process with the nation’s second largest guerrilla group, the ELN. Nevertheless, as of late 2004, the peace talks with the AUC had yet to alter substantially the dynamics of the conflict. Guerrilla and paramilitaries continue to dispute territory along strategic corridors of the country; they confront each other over control of economic resources, particularly coca, and continue to seek advantage to extort other key resources such as petroleum, gasoline, gold, or agribusinesses such as African palm. Moreover, despite the declaration of ceasefire by the paramilitaries and the initiation of scheduled demobilizations of fighters, there is still little discernible evidence that regional and 3 “Jan Egeland, Subsecretario General de la ONU dice que en Colombia no puede haber victoria militar”, El Tiempo (May 8, 2004). 11 local paramilitary groups have ceased hostilities against the FARC and the communities and individuals who they believe support them. According to the EWS data, from January 1 to April 30, 2004, 46.2% of Risk Reports were in response to threats from the AUC. 30.8% from the FARC and 8.5% from the ELN. The Organization of American States signed an agreement to provide technical support for the demobilization process but has indicated that it can only verify the actions of the assembled group of fighters in Ralito; the OAS does not have the capacity or the mandate to supervise the ceasefire agreement across the national territory. In the absence of an effective verification mechanism, the violations of the ceasefire have been registered by the Early Warning System -- although the EWS has no mandate to supervise directly the ceasefire. Furthermore, absent major changes agreed to by the parties in conflict, the EWS cannot serve as a viable replacement for an effective verification mechanism. Yet the Early Warning System by registering imminent and consummated human rights violations is nevertheless sending clear warnings on the weaknesses and violations of the current ceasefire and will likely continue to do so. b) Objectives of the Early Warning System and Anticipated Patterns of Human Rights Violations 2004-2007 The Early Warning System was designed at a time when massive violations of human rights and IHL were escalating at an unprecedented pace. This escalation, as viewed from the national perspective, has been largely halted. Yet violations continue. They are occurring not only in the remoter areas of the national territory and along the country’s borders with its neighbors but increasingly in urban areas as the illegal armed actors have taken advantage of the massive IDP population and have recruited youths from many of the marginal urban barrios that have burgeoned in the country’s large and medium-sized cities. Given the embryonic nature of the state’s presence in large swaths of rural and urban Colombia, establishing state control and legitimacy will be a prolonged process. Ending the multiple armed threats will also be a protracted process. Yet the EWS has demonstrated that it is capable of adjusting to the changing nature of the conflict and to anticipate and prevent violations. It has presciently registered the dynamic changes of the armed conflict in the last few years and has issued numerous Risk Reports and Early Warnings in some of the most vulnerable areas of the country. It has increased the number of regional analysts to 214 and even designated a regional analyst for Bogotá to register the evolving nature of urban violence. One result of the direct attention paid to Bogotá was that on March 26, 2004, an Early Warning was issued for nine barrios in Ciudad Bolivar, an impoverished urban sector in southwest Bogotá of over 600,000 people plagued by a toxic mix of youth gangs, urban militias, and territorial confrontation between the FARC and AUC. The primary objective of the EWS, then, should be to keep constant vigil over all areas of the national territory, to identify current and emerging threats to human rights and IHL, and to 4 During the Evaluation Team’s visit, there was discussion of reducing this number due to budgetary realignments within USAID’s human rights budget. 12 respond immediately to imminent or impending violations before they occur. Its mandate should be viewed in the context of Colombia’s constitutional and international treaty commitments to protect human rights and to comply with international humanitarian law. [[maybe place at end] The Human Rights Ombudsman interprets the mission and objectives of the EWS in similar terms, though it proposes a broad mandate of early warning, conflict prevention, humanitarian intervention, and conflict resolution: Strategic Objectives:  Ensure the exclusion and protection of the civilian population from the armed conflict and its consequences  Promote the adoption of strategies and preventative measures to protect the civilian population and to prevent the violation of their rights and reduce to a minimum the damages caused by the internal armed conflict.  Support from the perspective of humanitarian intervention, the generation of spaces and attitudes that favor a political solution to the armed conflict. Specific Objectives:  Institutionalize an effective and reliable mechanism to manage information and conduct risk analysis in order to contribute to an optimal response and adequate attention by the State to communities at risk as a result of the armed conflict.  Promote and help coordinate cooperation between the State and communities and social organizations in order to reach a consensus on the development of action plans to prevent violations within the different regional and territorial contexts.  Alert the competent authorities about the likely occurrence of massive human rights violations in order to protect the affected population and implement the necessary actions to dissuade, control, mitigate or overcome the indicated threat (emphasis added).  Elaborate and disseminate research and studies covering diverse regional and national contexts analyzing structural aspects of the armed conflict in order to determine the vulnerability and exposure to risk that is faced by the civilian population, with the objective of formulating appropriate public policy to address the situation.5 Much of this agenda overlaps with broader programs of conflict prevention currently being developed separately by the Ministry of Interior, the Vice Presidential program on Human Rights, and within the Municicipios y Departamentos Seguros program briefly described earlier. These initiatives are still not well coordinated and need to be incorporated into a more comprehensive national conflict prevention strategy. The EWS, and the conflict prevention work of the Ombudsman’s office in general, have a unique analytical capability that can reinforce, enhance and complement these other initiatives while still maintaining the Ombudsman;s institutional autonomy. However, it is the third objective – alert the competent authorities about likely human rights violations and ensure that the necessary measures are taken to protect the civilian population -- that should be viewed as the cornerstone of an effective early warning system. As the Ombudsman states, this should be done following the principles of 5 Información general Sistema de Alertas Tempranas, Defensoría del Pueblo, Defensor Delegado para la Valoración de Riesgos de la Población Civil como Consecuencia del Conflicto Armado, 2004. 13 As such, the EWS should be viewed and utilized principally as a rapid response mechanism that can quickly alert the relevant state actors about imminent and near-term threats and which can help orient an effective state response to prevent the violations from occurring. ] [ The EWS should continue to work with all relevant state actors as well as with civil society and the appropriate international organizations involved in humanitarian intervention and the protection of human rights. It is imperative that the EWS, as a tripwire that puts into motion set of actions to prevent violations, maintain its operational autonomy and integrity and maintain its primary focus on preventing massive and systematic violations of human rights and international humanitarian law (IHL). For the government and many other state actors, human rights represent one of multiple strategic objectives within a policy of democratic security. But for the EWS to be effective, its primary and exclusive objective must be to do everything in its power within the alert and response areas of its authority to ensure that massive human rights violations do not occur. How long will it be necessary to continue operations of the EWS in light of the rapidly changing conditions on the ground? The answer is straightforward: As long as massive and systematic violations of human rights and IHL continue to occur in any one of the twelve categories recognized by the Human Rights Ombudsman, it should remain operational -- including during periods of peace negotiations, during increased security operations and/or as the state systematically implements a long-term democratic security strategy. Only when the regional and national analysts assigned to the EWS report that the major categories of human rights and IHL violations have become isolated phenomena or have been effectively eliminated should the EWS be phased out. The Early Warning System, then, will need to continue to register and respond to the evolving pattern of human rights violations. It has developed the necessary analytical capability to anticipate changing patterns within the conflict. Based on current indicators and analyses, over the next 2-3 years, likely patterns of massive human rights violations will likely include:  forced internal displacement of rural populations as the conflict and illicit crop cultivation moves into new areas;  massacres, particularly of the type enumerated in the above description of La Gabarra by small armed groups in isolated areas along strategic economic and military corridors and zones;  multiple related homicides;  terrorist actions against civilian populations;  selective homicides;  violations by urban militias, urban sicarios, or by hybrid militias/ youth gangs such as those specified in the Bogotá alerts. However, an effective EWS can mitigate substantially against these trends and significantly lower the probability of their occurrence. 14 Moreover, the EWS will need to anticipate other threats. If the peace processes and demobilizations advance in the short-term with the AUC and ELN, attacks by the FARC guerrillas against demobilized combatants and individuals tied to the former combatants should be expected. Protective measures can and should be built into the any final accord. However, partial peace agreements with these actors will introduce a new dynamic into the conflict with the FARC; the EWS should be prepared to alert and respond when necessary. Finally, although Colombia is still a long ways from a post-conflict situation, if other national experiences with internal peace processes are valid indicators, massive and systematic human rights violations will occur well into the post-conflict period as demobilized armed actors degenerate into bandits or organized criminals, or as dissident groups attempt to spoil the accords. Again, the EWS can help mitigate against these risks as well. In sum, Colombia is undergoing a very difficult transition from a country with a limited state presence in many areas of its national territory to a country with a more robust projection of state authority throughout all areas within its borders. It has also embarked on an even more arduous path of moving from a multi-polar, internal armed conflict towards the development of a secure and lasting peace. The EWS can help reduce and respond to continuing violations as the Colombian government attempts to advance and consolidate these transitions. II. THE EARLY WARNING SYSTEM – TWO PHASES The Early Warning System began operations during the first semester of 2001. The idea was first discussed in 1997 following the tragic massacre of innocent civilians by paramilitaries in a FARC-influenced region in Mapiripán (Meta). At the time, this massive atrocity represented an unprecedented projection of AUC military power far from its usual operating bases; several high ranking state agents were later charged with responsibility because of their failure to respond to warnings that preceded the attack. The central idea of a proposed Early Warning System was to create a mechanism to alert state authorities before a massive violation occurred. Following the national and international uproar in the wake of the events at Mapiripán, the Ombudsman’s office entered into discussions with UNHCHR. Throughout 1998 and 1999, the Ombudsman’s office with UN support began to closely follow the occurrences and patterns of massacres and to identify the most vulnerable and at risk communities as well as to distinguish dynamics of the armed conflict. The Human Rights Unit of the Ministry of Interior began to outline plans for an institutionalized Early Warning System. Yet an operational EWS was not implemented until 2001. The key catalyst to its development and implementation was the decision by USAID to provide critical financial assistance and technical support as part of the U.S. government’s commitment to an effective human rights program within its broader assistance package to Colombia in support of Plan Colombia. Phase I: (2001) 15 The EWS from the outset was centered in the Human Rights Ombudsman’s office and was designed to take advantage of the network of regional Ombudsmen who work with communities and state authorities throughout the country. In many cases, the HR Ombudsman represents a singularly legitimate state presence in conflict zones. From the beginning, analysts were placed in regional offices. They, in turn, were charged with alerting a team of national analysts in Bogotá of impending violations. The warnings were to be analyzed and verified in a timely fashion. If the risk was confirmed, the Director of the EWS (the Delegate Ombudsman for the Evaluation of Risk to the Civilian Population as a Consequence of the Armed Conflict) would issue an Early Warning and notify national, regional and local authorities about the situation. The regional and national analysts would then monitor the situation (See Chart 1). The system, as originally constituted, was relatively quick, agile and transparent. In March, 2001, the EWS had three regional analysts – in Cauca, Norte de Santander and Magdalena Medio. By July 2004, there were 21 regional analysts covering all areas of the national territory except the islands of San Andres and Providencia. The HR Ombudsman projects an optimal number of 34 regional analysts (See Maps ). The work of the regional analysts is overseen by a team of five national analysts, each responsible for a macro-region. The EWS is also developing a physical and digital data bank and documentation center to better support the work of the conflict analysts. Over the past three years, it has progressively developed and streamlined methodologies and procedures for the entire system. The advantage of this early model was that Delegate Ombudsman could verify information from local sources and issue warnings in a timely fashion. The Early Warnings were public and conceived within the framework of national and international human rights and international humanitarian law. They were sent to local and national authorities and were available to the local communities and to representatives of the international community involved in humanitarian assistance and the protection of human rights, such as U.N. High Commission for Refugees, U.N. High Commission for Human Rights, the International Committee of the Red Cross, as well as national and international NGOs. The disadvantage of this first model was that the Human Rights Ombudsman does not make public policy; this is a function of the national government. There is broad agreement that, as established in the Colombian constitution, the Ombudsman should not be in charge of formulating state policy and that the institution should confine itself to the protective constitutional function that it has been assigned.6 6 For further elaboration of this point, see Chapter III (p. 28) of Función Preventiva de la Procuraduría General de la Nación en materia de Derechos Humanos, (Bogotá: Procuraduría General de la Nación, 2003). Research for this Procurator’s report was coordinated by Professor Danilo Rojas, a member of the MSI evaluation team. Its recommendations were formally adopted “as part of the policy of prevention in the area of human rights that is the responsibility of Procurator General” (Resolución Nº 138 de 26 de febrero de 2003 del Procurador General de la Nación). 16 The Human Rights Ombudsman’s office understood its role in developing the EWS in this manner. The current Delegate Ombudsman who heads the EWS wrote: the EWS was conceived as an instrument to inform the “competent authorities, so that the State can implement in a coherent and coordinated manner the necessary preventive and dissuasive actions, in order to give protection to the civilian population.”7 As such, the first phase of the Early Warning System was principally hortatory. The Ombudsman could exhort but could not require the authorities to act. State authorities, particularly the army and police, were not clear how they should respond to the alerts. They did not know if they should drop ongoing operations and redirect their activities, or whether they would be held responsible for acts that occurred following a public alert. They also were unclear over the differences between the Ombudsman’s alerts and those increasingly issued by NGO’s and civic groups. Although the Ombudsman’s office made efforts to work with local authorities and security officials about the implications of the new system, the confusion and uncertainty persisted. Many of these issues came to a head following a major human rights tragedy that occurred in Bojayá, Chocó on May 2, 2002. There, while the community took refuge in a church to protect themselves from armed confrontations between the FARC and regional paramilitary forces, a stray gas cylinder – a rudimentary weapon of notorious inaccuracy and prohibited under international humanitarian law – directly hit the church and killed 119 people, the majority of whom were women and children. The cylinder was launched by the FARC. Following the massacre –one of the worst in decades and which achieved international attention – it came to light that the Ombudsman had issued an Early Warning alerting authorities to impending clashes and threats to the civilian population. Questions were raised as to why the authorities had not taken the necessary preventative measures. Other questions were raised as to whether responsible government or military officials could more easily be held criminally negligent as a result of the Ombudsman’s broadly-diffused Early Warnings. The pressure mounted from within the government and the Armed Forces to modify the system. Phase II: In response to the failures and shortcomings of the initial EWS design, in late 2002, the newly-inaugurated government of Alvaro Uribe proposed the creation of the an Inter￾Institutional Committee for Early Warnings (CIAT from its Spanish acronym) to develop a more coherent state response to the reports issued by the Ombudsman’s Office. 7 Daniel Medina González, Delegate Ombudsman for the Evaluation of Risk of the Civilian Population as a Consequence of the Armed Conflict, op.cit., p. 2. 17 CHART 1 Early Warning System Phase I (2001- November 2002) EARLY WARNING ISSUED by Human Rights Ombudsman Office – Delegate Ombudsman for the Evaluation of Risk to the Civilian Population as a Consequence of the Armed Conflict  Pubic Document Sent to Local and National Authorities Monitoring of Possible Occurrence of Massive Violations Regional Analysts – Regional HR Ombudsman Office Verification and Assessment of Information National Analysts in Bogotá Human Rights Ombudsman Office Monitoring of Situation Regional and National Analysts HR Ombudsman Office 18 CHART 2 EARLY WARNING SYSTEM Phase II (November 2002 - Present) Human Rights Ombudsman SENDS RISK REPORT TO CIAT EWS Monitoring of Possible Occurrence of Massive Violations Regional Analysts –Regional HR Ombudsman Office Verification and Assessment of Information National Analysts in Bogotá HR Ombudsman Office Monitoring and Follow-up Reports EWS Regional and National Analysts Human Rights Ombudsman Office CIAT Members (Army, Police, MinInterior, etc.) Separately Verify Risk Report CIAT meets:  Decides to Issue Early Warning  Decides Not to Issue Early Warning  In either case, Recommends Measures to be taken by Local Authorities  Early Warning or original Risk Report and Recommendations Sent to Departmental, Municipal, Military, Police, Intelligence Authorities  Generally Governor convenes a Departmental Security Council meeting Preparation of Risk Report EWS National Analysts Human Rights Ombudsman Office Governor’s office Reports back to CIAT Coordinator on Actions Taken CIAT Members monitor Situation Separately and Report Back to CIAT 19 CIAT members initially included:  Vice President’s Office – the lead agency for Human Rights policy  National Police  Army  Ministry of Interior and Justice  DAS – Administrative Security Department, the national intelligence agency combining functions equivalent to the U.S. FBI, CIA and Secret Service. Since April 2002,  the Social Solidarity Network – Red de Solidaridad Social, the presidential agency responsible for attending the Internally Displaced Persons. The Committee was established as a “working group” and is coordinated by the Ministry of Interior and Justice. Many institutions and observers, including the UNHCHR welcomed this initiative.8 The advantages of the new system were apparent. Decisions regarding protection and prevention of the civilian population would now be made by the competent authorities. Under the revised system, the Ombudsman no longer issues Early Warnings (see Chart 2). Instead, the EWS analysts and the Human Rights Ombudsman issue Risk Reports. These are no longer public and broadly circulated. They are sent directly to the CIAT whose members are responsible for verifying the information through their local representatives and contacts. It was thought that the assessment of risk would be greatly enhanced if it was verified through the intelligence capabilities of DAS, the Army, the Police and the network of local officials tied to the Ministry of Interior and Justice, the Presidential Program on Human Rights and later, the Social Solidarity Network. Following verification, CIAT convenes a meeting of its members and decides whether to issue an Early Warning. In those cases where the CIAT declines to do so, the Inter-Institutional Committee nevertheless sends the Risk Report to the competent local authorities along with a set of recommendations on what measures should be taken. Following verification, CIAT convenes a meeting of its members and decides whether to issue an Early Warning. Yet the new system quickly manifested several disadvantages. What had once been a public and transparent process was transformed into an internal state mattered cloaked under the 8 See recommendation 11 of Appendix I del Report of the United Nations High Commissioner for Human Rights on the human rights situation in Colombia (2003): 11. In this respect, the Government’s decision at the end of 2002 to establish the Inter-Institutional Early Warning Committee (or CIAT, from its initials in Spanish), inside the Ministry of Justice and Law, must be greeted with satisfaction. This committee is made up of public servants from that ministry as well as from the Ministry of Defense, the Security Forces and the Presidential Programme for Human Rights. This initiative from its very beginnings had to overcome great challenges, such as establishing its way of operating, the criteria for determining risk and the measures to be adopted for preventing an alert. It must be noted, however, that the establishment of the SAT represents a positive element to counter the difficult human rights and international humanitarian law situation.” 20 protection of national security policy. Those who petitioned for information were told that such information was protected for reasons of national security outlined in the regulation of Article 12 of the Constitution on the citizens’ rights to petition their government.9 Moreover, from the outset, there has been a lack of coordination and integration between the two component parts of the system: the EWS in the Ombudsman office, and the CIAT. The CIAT often downplays the Risk Reports issued by the EWS when its members cannot confirm the Reports’ findings or when its members are divided over their contents. Without a representative at the CIAT meetings, the Ombudsman is unable to explain or answer questions concerning the overall decision to issue a Risk Report (which for the Ombudsman’ Office means that they already have sufficient reasons to issue an Early Warning as they did under the earlier system), or at the very least to answer questions on the assessments made by the regional and national analysts. Finally, despite the presence of the Vice Presidents office and the recent addition of the Social Solidarity Network, CIAT’s deliberations, analyses, and recommendations have until now been principally weighted toward security and public order concerns, reflecting the strong presence of the Ministry of Defense (Army and Police) and DAS on the committee. Curiously, the coordination of the Committee was placed within the Public Order and National Territory Division of the Ministry of Interior, instead of its more logical placement within the Division of Human Rights. In revising the system, a breach has opened up between the Human Rights and IHL orientation of the Ombudsman’s office and the public order and security orientation of the CIAT. While the Human Rights Ombudsman issues Risk Reports within the context of IHL, the CIAT principally provides responses within the framework of maintaining public order and pursuing the wider security strategy of the government. These two spheres of public policy are not the same. One seeks to protect the civilian population from the ravages of the armed conflict; the other seeks to implement a set of military and police strategies to maintain public order. The EWS was originally conceived as an instrument of the former. There are many other state programs designed to ensure the efficacy of democratic security and public order. The separate realms of public policy can -- and should - be complementary. To be credible, however, an Early Warning Strategy for massive and systematic human rights violation must stand alone. Indeed, as will become evident throughout this analysis, military and police officials legitimately complain that the Early Warnings divert them from their essential responsibilities and that they do not have the resources or manpower to protect every community. The concern reflects the fact that in the response function of the system too much weight has been placed on the shoulders of the police and military and too little attention has been placed on other forms of state response. Though some security measures will almost always be necessary, other responses include establishing a broader civilian state presence within the at-risk communities, emergency assistance, state investment, or support for international or NGO accompaniment. Moreover, it also should be noted that the move from Phase I to Phase II did not lead to a measurable improvement in the number of violations that occurred following the issuance of an 9 See Appendix x. 21 Early Warning or Risk Report. Each model revealed certain strengths and certain deficiencies. Cumulatively, the two models appear to have had about the same impact in preventing massive human rights violations, as discussed in the next section. III. ASSESSMENT RESULTS The MSI team conducted a results evaluation to determine the degree of success the EWS has had in preventing or reducing immediate threats, and a process evaluation to assess how the system functions at each of the critical decision-making points within the system. A) MEASURING RESULTS. HOW EFFECTIVE IS THE EARLY WARNING SYSTEM? Can an early warning system prevent massive and systematic human rights violations? The answer is yes. Using data assembled by the Ombudsman office, Appendices 1 & 2 show the relationship between Risk Reports & Early Warnings and the subsequent occurrence of the alerted violations. In 2002, with the original system in place during the first ten months of the year, a total of 93 Early Warnings were issued. Of these, the reported violations did not occur in 52 cases. They did occur in 41. Although it is difficult to attribute the no occurrence to the Early Warning, since other factors may be involved, the reported violation did not occur in 55.9% of the cases. In 2003, when the new system was in effect, the Ombudsman issued a total of 84 Risk Reports. Of the 84 cases, the alerted violations subsequently occurred in 37 cases or 44% of the time. In 56% of the cases, violations did not occur, or almost the exact result of the previous year under the earlier system. During the first trimester of 2004, from January 1 to April 30th, the Human Rights Ombudsman’s office issued a total of 30 Risk Reports. As of July, 16 of the alerted violations had occurred, or in 53.3% of the cases. The EWS was successful, then, from 47.6% to 56% of the time. The data, then, indicate, that the move from Phase I to Phase II did not result in a more effective EWS. Of the 84 Risk Reports that the EWS sent to the CIAT, only 29 were converted into Early Warnings. In the remaining 55 cases, the Risk Reports and Recommendations were sent to local authorities without the Early Warning. However, the new system has potential to achieve greater results. The percentage of success improved in those cases that when the Risk Report was elevated to an Early Warning by the CIAT. The massive violations subsequently occurred in only 10 of the 29 cases where the Early Warning was issued, or in just 34.5% of the cases. However in the 55 cases that CIAT declined to emit the extra warning, violations occurred in 27 of them, or 49% of the time. 22 As such, the Ombudsman’s data show that when the CIAT simply passes on the Risk Reports, the system operates much as it did in Phase I, with similar rates of success. However, when the authorities place greater attention on the warning signals, the effectiveness of the system increases. The failure of Phase II to achieve greater results seems to be due to the over￾reluctance or reticence of the CIAT to issue early warnings. Both the Ombudsman’s office and the CIAT have been developing evaluation mechanisms over the last year, including monitoring more systematically the actions of state authorities in response to Early Warnings and Risk Reports sent to local authorities by CIAT. However, one of the difficulties in standardizing evaluations is that each entity applies different criteria of success and failure and employs different definitions of “massive violations.” This point is graphically underscored when the Ombudsman’s evaluations cited above are compared with the Annual Report on Human Rights and IHL 2003 produced by the Vice President’s office. In this report, the Vice President’s office reported that from November 2002 through the first semester of 2004, “CIAT has received, processed and attended to 106 risk reports covering some 150 municipalities, of which 56 were considered to be early warnings. Mass violations occurred only in one place (Sabaletas, Buenaventura)” 10 (emphasis added). There is a need for EWS and CIAT to standardize definitions, categories and methodologies following acceptable international norms even while recognizing the great complexity and fluidity of the boundaries between armed conflict and crime in the Colombian case. To address this problem of measurement and evaluation, the MSI Evaluation team independently examined the effectiveness of the system in three departments: Arauca, Chocó (excluding Urabá) and Norte de Santander. These departments were selected both because they represent central areas of the current and (likely) future stages of the conflict and because they represent key factors that stimulate conflict: weak state presence, large minority communities that have been over-represented in the victim population, and areas of economic production – coca, oil, gold, African palm, etc. – that have continually fueled the conflict. Between November 2002 and July 2004, these departments represented a total of 28 cases of Early Warnings or Risk Reports plus 13 Follow-up Reports. Moreover, all three showed an increase in human rights violations during the first two semesters of 2004. As a result, we were able to witness directly and up close the EWS system in action. To measure effectiveness, the team relied on data provided by official and NGO sources. These included:  All Risk Reports and Follow-Up Reports (Notas de Segimiento) for the three departments - Human Rights Ombudsman  All Early Warnings and Recommendations – CIAT  CIAT Acts of its meetings and selected correspondence between local authorities and CIAT 10 Republic of Colombia, Annual Report on Human Rights and IHL 2003 (Bogotá, 2004), p.136. 23  Preliminary Evaluations of warnings and risk reports by CIAT staff members.  Analyses and evaluations of EWS in conflict zone by Ombudsman’s office.  Evaluations of public order by the National Police.  Human rights and IHL violations registered by the Vice President’s Observatory of Human Rights  Human Rights Data Bank of CINEP, primarily through the CINEP publication and on-line data base, Noche y niebla. 11  CINEP Press Archives on Human Rights violations.  Interviews with local authorities in each department. A table of our findings in each Department measuring the effectiveness of the Risk Report, Follow-up Reports and Early Warnings issued per department can be found in the Appendices. Chart 3 presents a summary of the findings. To overcome the definitional problem, we employed the general categories and criteria of human rights violations used by the Ombudsman. We then adjusted, where possible, CINEP, Human Rights Observatory and National Police data into these categories. We did this because the Ombudsman’s categories correspond to the types of violations or prohibitions enumerated in international human rights treaties and in international humanitarian law conventions. They are also the categories that are found in the initial Risk Reports. To maintain methodological consistency, it was necessary to ascertain whether the types of violation that sparked the initial risk report did in fact occur. The categories we used, then, are:  Massacres (killings of 3 or more persons)  Multiple related homicides  Forced displacement  Forced disappearance  Forced recruitment  Seeding land mines/ landmine accidents  Restrictions on free circulation of individuals  Attacks on indispensable goods and provisions for a community  Interposing civilian populations amid hostilities  Indiscriminate attacks against civilians  Selective homicides  Methods to generate terror among the population Our finding are summarized in Chart 3: 11 CINEP is an independent Jesuit research organization. that has recently entered into an agreement to share data with the Human Rights Ombudsman office and the Vice President’s Observatory of Human Rights. 24 Chart 3 4 11 13 3 4 5 1 7 8 0 5 10 15 20 25 30 123 Total Cases 28 EWS Success 12 EWS Failure 16 (100%) (42.85%) (57.15%) Effectiveness of EWS in Arauca, Choco and Norte de Santander Nov. 2002 - July 2004 Norte de Santander Arauca Choco Of the 28 total number of Risk Report issued by the Ombudsman for these three departments, 10 were converted into Early Warnings by CIAT or 37%, roughly the same as the national proportion for 2003. For all cases with at least a Risk Report, the EWS was successful 42.85% of the time, or slightly lower than the Ombudsman’s national evaluation. However, in these three departments, the issuance of an Early Warning by CIAT did not prove to be more effective than simply passing on the Risk Report and recommendations to local authorities. Of the 10 Early Warnings, violations occurred in 6 and were prevented or did not occur in 4, or in 40%. Although we cross-checked all violations with CINEP’s human rights data base and press archives on human rights, as well as through personal interviews, we did not find significant discrepancies between the Ombudsman’s data and CINEP’s. The discrepancy between our findings in three departments and the Ombudsman using a national sample may be explained, then, by the intensity of the conflict –particularly in 2004 – in these regions and the greater difficulty of operating the EWS under these circumstances. Especially in the cases of Arauca and the Chocó, the system was significantly less effective than at the national level. 25 B) EVALUATING THE PROCESS Because the EWS and CIAT components are not functionally integrated, it was necessary to develop indicators of success or failure for each component individually. The team therefore separated the alert function (the activities of the EWS overseen by the Human Rights Ombudsman) and the response function (overseen by the CIAT). Appropriate indicators were therefore developed for each. Again, the team endeavored to the degree possible to ensure reasonable levels of data validity by cross-checking several credible data sources. In breaking down the alert and response functions of the system, we developed the following matrix of indicators for successful or unsuccessful outcomes. Although there is a sizable number of cases from the first year of operations before the current system was put in place, we have principally focused on the system as it has functioned since November 2002. Matrix of Indicators EARLY WARNING SYSTEM Alert Human Rights Ombudsman (EWS) Response (CIAT- Local Authorities) Success Regional analyst sends reliable information concerning a credible threat, verified by HRO national analysts, Risk Report (RR) transmitted in timely manner to CIAT. RR verified by government agencies, Early Warning is issued in timely fashion to local authorities, appropriate action is taken and threatened violation does not occur. Failure 1) Regional analyst receives warning but fails to communicate to HRO, and threatened violation occurs; 2) Regional information received at HRO but no RR created and threatened violation occurs. 3) No regional analyst to cover area of risk, no information to DP, violation occurs 1) IR received by CIAT, after verification process IR is passed on to local authorities without issuing Early Warning, threatened violation occurs; 2) IR received, Early Warning is issued, local officials take inadequate action and threatened violation occurs 3) IR received, Early Warning is issued in untimely manner, threatened violation occurs Cases were selected that, controlling for other factors, can illustrate success or failure in the warning stage – the initial decisions made by the local expert and the later decisions made by the Ombudsman to create a Risk Report -- and in the response stage – the decisions made by first the CIAT, then by local officials in response to CIAT recommendations. Decisions were examined against the following indicators:  Was the information received from the regional analyst externally validated, and was the threat credible (were civilian populations at risk due to the conditions noted in the verification report)?  Did the Ombudsman transmit the risk report quickly enough to allow time for CIAT verification and response?  Did the CIAT recommend actions that could, with a reasonable degree of certainty, contribute to preventing the threatened violation?  Were appropriate authorities notified early enough to take appropriate preventive steps? 26  Did local authorities comply with recommendations from CIAT members?  Did a violation occur in the municipality identified by the risk report? The cases were chosen from a national sample of all Risk Reports and Early Warnings issued. They highlight different patterns of success and failure at key decision-making points. They also illustrate trends in three pre-selected departments –Arauca, Chocó and Norte de Santander – as well as two urban cases – Ciudad Bolívar (Bogotá) and Barranquilla. We used primary documents and standardized interviews to assess the process at each stage and to examine the relationship between the analyses of EWS, the recommendations from CIAT, the actions of local officials and their impact on the ultimate outcome – the occurrence or no occurrence of the alerted violation. Interviews with the relevant regional and national officials and analysts focused on: 1) The initial situation assessment conducted by regional EWS analysts. 2) The methodologies used by national analysts to verify information and produce a Risk Report. 3) The decision by the director of the Early Warning System who is an official of the Human Rights Ombudsman’s office to issue a Risk Report (RR) and send it directly to the coordinator of the Early Warning Inter-Institutional Committee (CIAT) in the Ministry of Interior. (Since March 2004, an additional step has been added: the Director of the SAT sends the potential Risk Report to the Human Rights Ombudsman who then makes the decision to send the Report to the CIAT). 4) The decision by the CIAT to issue an Early Warning to local authorities based on internal verification by each member agency: in most cases this is done by the Ministry of Interior’s human rights office, the Armed Forces’ human rights office, the national police human rights office, the national intelligence agency (DAS), the Vice President’s human rights program, and (since April 2004) the Network of Social Solidarity (RSS), responsible for policy affecting the country’s IDP population. 5) The actions taken by local civilian, military, and police authorities in response to an Early Warning or Risk Report and in compliance with the CIAT agencies’ recommendations. C) PROCESS EVALUATION RESULTS: 1. Early Warning System: Assessing the Alert Function In Bogotá, the team met with Early Warning System director, coordinator, and national analysts on three separate occasions to discuss their methodologies and procedures for verification. In Arauca, Norte de Santander, and Chocó, the team accompanied the EWS regional analysts in their various activities, including training departmental personeros, speaking with local community organizations, and monitoring rural areas. Team members also interviewed the regional analysts in depth concerning their verification procedures in particular cases. 27 Based on these interviews, observation and available documentation, we concluded the following: Externally Validated Information EWS Regional analysts appear to follow the procedures outlined in current procedures manual: these include following the format of the “ficha de verification,” description of all available information concerning the specific threat, and consulting with at least three distinct sources to verify facts on the ground. The “fichas” contain names and identities of informants and information sources. These are not kept with the Risk Report file in the Defensoría office and by law are confidential. As a result, the team was not able to fully assess compliance in all regions. The team noted that, while regional analysts were likely to consult with non-ranking police officers in the local municipal or village station, in the cases we investigated they generally did not consult with the departmental military and police commanding officers. It was explained to us that this would be a highly visible act which would greatly elevate their already high level of personal threat. This is a serious consideration. However the absence of even a courtesy call to the ranking officers of the military has contributed to a high level of distrust of the EWS by the public forces. At the regional level, the main cause of system failure (no information, therefore no risk report, therefore no alert) appears to be the long periods of time with no coverage, especially in the most dangerous conflict zones. The entire Early Warning System depends on having a trained analyst in the field with sufficient coverage of conflict areas and sufficient consistency to build confidence in local communities. Frequent turnover and long periods without a regional analyst undermine the Early Warning System at both the alert and the response stages. National analysts follow the procedures outlined in the current procedures manual: they compare the information received from the regional analyst with previous history of conflict and violations in the area, consult other sources of information such as news media and outside database, analyze threat level, and provide all available information on Risk Report form. National analysts assigned to one of five macro-regions compare the information received from the regional analysts with other documentation and broader analyses of regional trends. One of the problems is that there exists a “flow chart” of who reports to whom as well as a format for the Regional Analyst to follow. However there does not exist a substantive manual of procedures and processes beyond this flow chart. As such, much depends on the individual initiatives and prior training of both regional and national analysts. The Credibility of Threats. Is the civilian population deemed to be at risk due to the conditions cited in the Risk Report, based on patterns of violence in the immediate area and known facts on the ground? Can the system be manipulated by one of the armed actors? 28 The team found that it is practically impossible for a “false alert” to make its way through the system. Assuming that the regional analysts follow procedures and verify information with at minimum three distinct credible sources, and that the national analysts likewise follow procedures, and that the agencies represented on the CIAT (which include the army, police, and national intelligence) also conduct internal verification, there are simply too many redundant verification processes to allow for manipulation. The system is designed to distinguish between “noise” and “real alarms.” What is possible is for one of the armed actors to commit human rights violations against a given community in order to activate the EWS and trigger a diversion of ongoing military activities. This possibility needs to be assessed by the relevant military and intelligence personnel. It falls within the category of diversionary tactics used in any war. However, due to the cumbersome and delayed procedures currently in place for issuing an Early Warning, the system lacks the agility that would be necessary to effectively divert military operations on the ground. 2) Inter-Institutional Committee on Early Warnings, CIAT: Assessing the Response Function In Bogotá, the team held an informational meeting with the CIAT, met individually with the director of the Vice President’s human rights office and the MSD consultant who serves that office as a senior adviser, the CIAT representatives from the National Police intelligence agency, the Defense Ministry’s human rights and humanitarian affairs office, the Interior Ministry’s human rights office, the DAS’s Strategic Analysis group and the MSD consultants assigned to the CIAT. We also had the opportunity to sit in on a CIAT meeting where decisions were made on whether or not to convert a group of Risk Report received from the EWS into Early Warnings. In Arauca, Norte de Santander, and Chocó, the team met with the highest-ranking officials available at the local police, army, DAS, and departmental government offices, taking care to meet with the human rights officials for each of these offices. Additionally, we systematically examined all of the Acts of CIAT meetings since its inception to better understand the decision-making process. According to the CIAT’s work plan for 2004-2005, a procedures manual, training plan for CIAT members and consultants, filing and information security protocols, and CIAT information system should all be created and implemented this year. However, these processes were only incipient at the time of this evaluation. Procedures to verify Risk Reports received from EWS In all of the cases for which documentation was provided, the human rights office of the Vice Presidency conducted its own investigation of each Risk Report, using a verification format 29 that directly refers to the location and type of threat in the Report, notes the sources that were contacted in the verification process, and their responses. The National Police investigates Risk Reports and requires its agents to provide requested information within eight days. In these cases, local agents complied fully and in a timely manner, using a verification form that directly references the Risk Report. The Army investigates Risk Reports in most cases. The request for information concerning the presence and activities of armed groups in the area is processed in two to three days, and the response references the Report by number. It should be noted that the Defense Ministry human rights officer is in the process of developing a systematic verification procedure, but it is still in the draft stage. The DAS provided intelligence materials within one week of receiving a Risk Report in some of the cases for which documentation was made available. However, the information did not respond specifically to the threat cited in the Risk Report and did not reference the Report itself. CIAT Response Time There is a fundamental problem with timing – imminent or not. The CIAT system is only effective for early warning, not for emergency response. Likewise, the Ombudsman’s EWS office lacks the capacity for emergency response. At present, neither institution has a clear method to determine the imminence of a threat. One resolution to this problem that was advocated by several United Nations agencies is to communicate Risk Reports of an emergency nature directly to Humanitarian Situation Room run by the Office for Coordination of Humanitarian Affairs, and by establishing expedited communication procedures between EWS and CIAT to expressly address emergency situations. Even when a threat is looming but not yet at emergency level, several persons interviewed voiced complaints that the EWS system and its directors are overly-cautious in issuing Risk Reports once the analysts have concluded that a threat does exist. From there, the report may languish for another 3 to 4 weeks before CIAT addresses the situation. An EWS study found that the average time between the issuance of a Risk Report and action by CIAT is 21 days. By then, the impending violations may have occurred. CIAT’s Decisions to Issue Early Warnings and Recommendations CIAT, once it receives a Risk Report, elects not to issue an Early Warning in about two thirds of the cases (65% for all cases in 2003, and 63% for cases in Arauca, Chocó and Norte de Santander between November 2002 and July 2004). Instead, the Committee sends the Risk Report directly to the local authorities and makes a set of standard recommendations. 30 CIAT has not developed a set of protocols or criteria to determine when to issue an Early Warning and when not do so, nor has it developed clear guidelines that define the significance and meaning of a declaration of Early Warning versus the sending of a Risk Report to local authorities without an Early Warning. At the CIAT meeting, there is no one from the Ombudsman’s office to explain, defend or provide further specificity where it is found to be lacking. In many instances, members disagree among themselves on the seriousness of the threat based on their own verification efforts. Yet in the cases of controversy or disagreement, there appear to be no adequate procedures to break the impasse. Many valid assessments made in the Risk Reports are incorrectly downgraded. At the same time, CIAT’s recommendations provide only general guidance. Initially, CIAT’s recommendations were quite generic, such as : “The Public Security Forces should take into account the altercations of public order that might be occurring in La Platanera and La Balsa.” 12 The principal recommendation that accompanies most Early Warnings is the following: the governor should convene a Departmental Security Council and implement the necessary measures.13 However the same recommendation is also regularly made in cases where an Early Warning is not issued. Other recommendations made by CIAT, regardless of whether an Early Warning is issued, include:  reinforce military presence  develop a security plan  attend the humanitarian needs of IDPs  continue ongoing operations  increase military/police forces on ground  provide more protection for the population  do everything possible to prevent Human Rights violations and more displacements  support the return of the Internally Displaced Persons to their communities  carry out humanitarian “jornadas”  register the population and maintain military operations  develop an integral response to the problem of illegal crops14 Neither CIAT or its member agencies have defined a set of specific measures that should be taken by local officials in the region (civilian or military) in response to the particular threat described in the either the Risk Report or the Early Warning. On the ground, there is also confusion concerning the difference between a Risk Report and an Early Warning. Both are issued by the CIAT and accompanied by recommendations from the Ministry of the Interior. When questioned, most local officials were at a loss to differentiate between the two. 12 (Acta 06/02). 13 See, for example, Actas 14/03, 16/03, 18/03, 21/03, 29/03, 33/03). 14 See the Tables of cases in Arauca, Chocó and Norte de Santander in Appendices 3-5 31 Individual CIAT member agencies do not appear to have developed protocols for authoritatively communicating response mechanisms to their officials in the region. Moreover, in most cases, the representative of each institution at the CIAT meeting is a mid-level or relatively junior official or officer. When questioned about this, CIAT members and staff assured the team that the level of seniority was not the decisive issue but rather the institutional commitment. We agree. However, we found that the decisions and recommendations that came out of CIAT meetings were not well integrated into each institutions subsequent actions and that there was little general knowledge about the workings of EWS or CIAT beyond a small group within each CIAT member institution – particularly within the Army, Police, DAS, and Social Solidarity Network. Decisions by the CIAT are transmitted by the CIAT Coordinator from the Ministry of Interior. She sends recommendations in the form of a brief letter to the governor, the local secretary of the interior, the Departmental Peace Advisor when applicable, the local military commander, and the senior local police official. However, the CIAT lacks the legal authority to enforce its recommendations. The result of the process is that most of the specific decisions are made not in the CIAT but in the departmental or local Security Councils. The Security Councils are convened by the Governor and are generally attended by senior elected and appointed civilian leaders from the region plus the senior military, police and intelligence officers assigned to the department. The active participation of senior elected authorities in the department can potentially be an advantage when the officials are sufficiently knowledgeable and supportive of the goals of the EWS. However, as of yet -- and despite some notable exceptions -- there is still not sufficient understanding or acceptance of the EWS among most local level officials, starting at the level of the Governor and local military and police commanders on down to their subordinates. The chief outcome, then, is that the human rights issues raised by the Ombudsman’s Risk Reports and the CIAT recommendations often get diluted within the broader security priorities of the moment. The most successful cases of prevention have occurred when the local civilian authorities have taken the initiative to lower tensions and protect the population by creating mechanisms of community dialogue and development backed up by appropriate security methods. One of the most successful cases occurred when the CIAT issued an Early Warning for 10 barrios in Ciudad Bolívar in Bogotá in May 2004 to defuse a growing conflict between AUC and FARC urban militias and youth gangs that had placed the civilian population at risk. The Mayor of Bogotá convened a special Security Council of key civilian, military and police officials and visibly designed programs to address the security threats and social tensions that underlay the alerts. The mayor took the lead in authorizing and implementing the following measures:  creating a ‘social plan’ for the zone,  assigning a task force of social workers, psychologist and conflict resolution specialists to the area  assigning a permanent an investigating judge (fiscal) to the zone,  increasing the police force by a 120 –man “search force” for the area 32  reinforcing the intelligence capabilities of DAS, Dijin and CTI. 15 Failed cases, such as the well-known cases of Bojayá (Chocó), Tame (Arauca) and La Gabarra (Norte de Santander) discussed in the table of cases on Arauca, Chocó and Norte de Santander in the Appendices, often represent cases where the civilian authorities failed to address adequately the warnings outlined in a series of Risk Reports, Follow-up Reports or Early Warnings and where the general response was to continue the normal conduct of military operations in the area.16 Broadening the Range of CIAT Responses Because of the way the system operates in practice, there has been an over-dependence on the deployment security personnel and military solutions to respond to threatened violations. Civilian initiatives and response mechanisms are underutilized, including community-level conflict resolution efforts; concerted attention and dialogue with community residents by elected civilian authorities, such as in the case of Bogotá; national, international or NGO accompaniment of threatened groups or individuals;17 humanitarian relief efforts; controlled evacuation; state investment and development initiatives. This over-reliance on the security forces to respond to human rights and IHL emergencies satisfies almost no one, least of all the Police and Armed Forces. It tends to collapse human rights and IHL concerns – the protection of the civilian population from the effects of the armed conflict – with the security and military strategy of the state. These should be complementary; they are not the same. The Armed Forces and Police complain because many view the Risk Reports and Early Warnings as counter-productive or unnecessary civilian interference into security affairs that interrupts ongoing military operations. If the police and military were to respond with appropriate force in each case, they would be spread exceedingly thin. As they cannot leave emergency forces in a locale permanently, their response is of necessity temporary. The result is that at times their intermittent presence creates an even worse situation for the population. Once the increased police detail or military unit leaves, illegal armed actors generally return and retaliate against civilians for inviting military/police. 15 Interviews in Ciudad Bolivar with local authorities and residents, August 19, 2004. Also see www.eltiempo.com (May 19, 2004) “Seguridad/ Bloque de Búsqueda de la Policía para combatir delincuencia, Ofensiva social en Ciudad Bolivar. 16 See Appendix 5. In the case of La Gabarra, the case of the 34 coca-pickers (raspachines) massacred by the FARC on June 15, 2004, there is a discrepancy between the EWS and the CIAT over the specification of the threatened area. The EWS had issued a Risk Report covering the municipality of El Tarra but did not single out the village (vereda) of La Gabarra. The issue raises the question of the difficulty of specifying the exact location and timing of the threats in the Risk Reports. 17 One of the most successful programs is the Community Ombudsman program (Defensores Comunitarios) of the Human Rights Ombudsman’s Office currently funded by the Swedish Government. This program sends special Community Ombudsman to live in at-risk communities. The team interviewed the Community Ombudsman for Medio Atrato (Chocó) who reported that he represented the most visible presence of the state in his community and that his very presence as an official representative of the Public Ministry – the human rights and control branch of the state -- afforded the community a degree of protection and a greater sense of security than it otherwise would have had. 33 A more integral and sustained civil and military response is needed for both imminent and intermediate-term threats. This will be even more necessary as the vulnerable populations increasingly come from two types of social contexts: 1) remote areas where the state presence is minimal and armed actors have maintained a presence, in some cases, for decades; 2) urban zones that have burgeoned with refugees from the rural violence where armed militias have mixed with local gangs. Did the CIAT recommended actions contribute to preventing the threatened violation? The data analysis provided in Section IIIa revealed something fairly startling: a simple hortatory alert by the EWS or CIAT produces demonstrable results. The average success of the EWS and CIAT since its inception is greater than 50%. Moreover, the Ombudsman’s office found that when the CIAT issued an Early Warning –thus indicating greater concern or attention – the success rate increased. This happened despite the absence of systematic recommendations or accepted protocols from local authorities on what measures should be taken in response to an Early Warning. What happens in the roughly 50% of the cases where the system fails to prevent massive violations? Why does a particular violation occur despite actions taken by local authorities? Although the evaluation team was able to bear in on a few specific cases, the CIAT does not have a methodology for following through with local officials, nor does it have a methodology for evaluating the results of actions taken in response to either Early Warnings or Risk Reports. The CIAT coordinator has discussed these methodological problems with the team. At the time of this evaluation, the CIAT was actively seeking to develop useful indicators of positive outcomes of actions taken by the state in response to risk reports, to supplement the raw indicator of whether or not a violation occurred. Doing so could substantially increase the rate system’s rate of success. One of the key obstacles to measuring the effectiveness of the CIAT is that no local government official is obliged to report to the CIAT specifically what actions were taken in compliance with CIAT recommendations. The current system of monthly public order reports submitted by the departmental governors is not designed to accomplish this task, nor is it an adequate tool for measuring the EWS response mechanism. D) HOW SHOULD THE EWS FIT INTO A NATIONAL HUMAN RIGHTS VIOLATION PREVENTION PROGRAM? An early warning system should have the capacity to identify: 1) emergency threats of an imminent nature, 2) threats that are gathering and will likely result in violations within a [3 – 6 month] period; 3) longer term threats with time to implement preventive measures. 34 Each type of threat should trigger an appropriate and timely response, the nature of which will depend on the conditions of the situation. The Ombudsman Delegate for the Evaluation of Risk in the Civilian Population as a Result of the Armed Conflict, the lead official of the EWS, defines the EWS within the broad human rights mandate of the Ombudsman’s office. The EWS has set strategic and specific objectives that include early warning, conflict prevention, humanitarian intervention, and conflict resolution: Strategic Objectives:  Ensure the exclusion and protection of the civilian population from the armed conflict and its consequences  Promote the adoption of strategies and preventative measures to protect the civilian population and to prevent the violation of their rights and reduce to a minimum the damages caused by the internal armed conflict.  Support from the perspective of humanitarian intervention, the generation of spaces and attitudes that favor a political solution to the armed conflict. Specific Objectives:  Institutionalize an effective and reliable mechanism to manage information and conduct risk analysis in order to contribute to an optimal response and adequate attention by the State to communities at risk as a result of the armed conflict.  Promote and help coordinate cooperation between the State and communities and social organizations in order to reach a consensus on the development of action plans to prevent violations within the different regional and territorial contexts.  Alert the competent authorities about the likely occurrence of massive human rights violations in order to protect the affected population and implement the necessary actions to dissuade, control, mitigate or overcome the indicated threat (emphasis added).  Elaborate and disseminate research and studies covering diverse regional and national contexts analyzing structural aspects of the armed conflict in order to determine the vulnerability and exposure to risk that is faced by the civilian population, with the objective of formulating appropriate public policy to address the situation.18 Much of this agenda overlaps with broader programs of conflict prevention currently being developed separately by the Ministry of Interior, the Vice Presidential program on Human Rights, and within the Secure Departments and Municialities program briefly described earlier. Indeed, the EWS was first conceived as part of a National System for the Prevention of Massive Human Right Violations.19 However, to date, these initiatives are still not well coordinated. The conflict prevention work of the Ombudsman’s office in general, and the EWS in particular, has developed a unique analytical capability that can reinforce, enhance and complement these other initiatives while still maintaining the Ombudsman’s institutional 18 Información general Sistema de Alertas Tempranas, Defensoría del Pueblo, Defensor Delegado para la Valoración de Riesgos de la Población Civil como Consecuencia del Conflicto Armado, 2004. 19 Daniel Medina González, Director SAT, La prevención como política pública, en Sistema de Alertas Tempranas # 1, julio 2002, p. 2. 35 autonomy. The Ombudsman should continue to develop this component as an essential element of its own work and as a resource for other state and civil society organizations. However, it is the third stated objective – alert the competent authorities about likely human rights violations and ensure that the necessary measures are taken to protect the civilian population -- that should be viewed as the cornerstone of an effective early warning system. As the Ombudsman states, this should be done following the principles of humanitarian protection, respect for the institutional authorities, impartiality, and celerity. Because the Human Rights Ombudsman is part of the control branch of the state, it does not make public policy. Its mission is to ensure that the design and application of policy conforms with the rules established in a democratic and lawful state. The EWS should be viewed and utilized, then, principally as a rapid response mechanism that can quickly alert the relevant state actors about imminent and near-term threats and which can help orient an effective state response to prevent the violations from occurring. The principal components of the EWS are the Risk Reports and the Follow-up Report issued by one of the control branches of the State, the one that possesses the most legitimacy: the Human Rights Ombudsman. From this basis, the state has an obligation to respond to imminent threats involving violations of Human Rights and IHL. This very fact justifies the existence of the CIAT and obligates it to act in as decisive a manner a possible. Yet throughout, the Human Rights Ombudsman and other control institutions within the Public Ministry at no time forfeit their constitutional mandate to supervise state actions and to officially question the behavior of the government with regards to human rights. IV. PRINCIPAL CONCLUSIONS:  Colombia’s Early Warning System is a unique, internationally-pioneering and essential instrument of government policy to protect Human Rights and ensure the respect of IHL. It is a model to respond quickly and effectively to imminent or gathering threats of massive violations before they occur. Despite the indicated flaws, the system has shown demonstrable success in responding to potential threats approximately 50% of the time.  It is imperative that the system be made even more effective and improve its success rate significantly given the current humanitarian crisis and societal demand for warning and response system in conflict zones.  Although there are a variety of national and international programs to promote democratic security and/or humanitarian assistance, there is currently no other program that can potentially respond as effectively to emergency and medium term situations where the civilian population is at risk as a result of the ongoing armed conflict. 36  At present, CIAT’s recommendations and their translation into policy by Departmental Security Council over-emphasize security responses and ongoing military operations to address impending threats and under-utilize broader conflict resolution, accompaniment and development measures in coordination with other state, international and NGO actors.  The warning and response components of the Early Warning System are not well￾integrated. Moreover, the response component headed by the CIAT is not legally constituted and has no legal authority to require compliance on the part of local civilian authorities. Local authorities are not obligated to inform the EWS/CIAT of what measures they took specifically in response and what the outcome was in terms of human rights and IHL. The informality of these arrangements lowers compliance, reduces the threshold of urgency and hinders the subsequent evaluation of the process.  The program’s future remains tentative given the limited capacity of the Colombian Government to increase its material commitment. Efforts must be made to ensure the future sustainability of the system.  There is a need to better coordinate the EWS with the work of international humanitarian and development agencies, including the Humanitarian Situation Room run by OCHA, and to involve greater participation of relevant human rights and humanitarian NGO’s. V. GENERAL RECOMMENDATIONS The EWS must function as an integral arm of the state’s conflict prevention and human rights policies and as part of its commitments under international law to protect the civilian population from the effects of the armed conflict. To do this: 1) The EWS and CIAT must function as a more integrated and more responsive system, using common methodologies of assessment and evaluations: a) To meet emergency needs, the CIAT should be upgraded into a fulltime 24-hour unit with the legal authority and sufficient rank to make decisions and require immediate action. The relevant local authorities should report back within a reasonable time both 37 what actions were taken [maximum 1 week] and what were the effects of those actions [monthly during the course of the alert activation]. b) The director of the EWS, as representative of the Human Rights Ombudsman and the EWS Analytical Team, should be present at each meeting of the CIAT in which a risk report is evaluated and recommendations are decided to further explain the reasoning that went into the report and to provide clarifications and greater detail. c) The Ombudsman should automatically send copies of the Risk Report to the Procurator General’s office so that this control branch is fully informed of human rights and IHL violations in conflict zones. Similarly, the Procurator should be kept fully apprised of CIAT’s issuance of Early Warnings and recommendations for state authorities to act in conflict zones. d) CIAT needs to develop a set of protocols and criteria for determining when Risk Reports should be elevated to Early Warnings. One way to do this would be to categorize all imminent threats as Early Warnings, while intermediate-term threats could remain as Risk Reports to be forwarded to local authorities with recommendation. e) Early Warnings should carry a great sense of urgency and should compel local decision￾makers to prioritize specific security and civilian response measures to address the situation. Risk Reports should be viewed with equal urgency. However the response mechanism would be able to allow for a slightly longer time frame, allowing local authorities to develop more coherent, integral and enduring responses. Local leaders and their staffs and should be trained to understand the differences between the two types of warnings. f) CIAT should recommend a variety of security measures and civilian-led initiatives to confront both emergencies and medium-term threats outlined in either the Risk Reports or Early Warnings. In general, the more imminent threats will likely require a greater mix of security measures while the medium-term threats should more fully incorporate the broader initiatives involving state civilians, international community and select NGOs. There is no one to one formula that matches specific responses to particular classes of violations. In each case, CIAT and local decision-makers should be able to draw from sufficiently large basket of options and tools to respond to the situation. The security measures might include: convening a Departmental Security Council, enhanced army patrols, reinforced police and intelligence presence, checkpoints, offensive military operations against illegal groups in the area with special attention to keep the civilian population out of the conflict. Other measures might include: state protection through the Human Rights Ombudsman office [on the model of the Community Ombudsmen program created with the assistance of the Government of Sweden], social investment, humanitarian action, national and international accompaniment, and greater community input and participation in the protective response measures. 38 2. Risk Reports prepared by the Ombudsman’s EWS office should be public documents: a) Since, November 2002 with the creation of the CIAT, all Risk Reports and Early Warnings have been cloaked under the cover of National Security, reversing the earlier public distribution of the Ombudsman’s Early Warnings. There has been great concern among the international humanitarian organizations and NGOs in the changed policy which basically removed all but a few state actors from the process. A human rights policy should be as transparent as possible. It is strongly recommended that the Risk Reports be made public and available to all concerned citizens and national and international actors. The greater openness will make the system more effective and responsive. The Ombudsman could place the Risk Reports on a special web site. (It should be noted that the Risk Reports do not reveal sources which are contained in a separate document – the ficha de verificación. These should remain confidential.) g) CIAT, if it so chooses, could maintain the confidentiality of its recommendations. However it is recommended that the decision to convert a Risk Report into an Early Warning should be publicized. h) The U.N. Office in Colombia for Humanitarian Assistance, OCHA has expressed great interest in participating in the EWS. It is recommended that the EWS work with OCHA to develop an emergency response for impending violations when the situation so warrants. All Risk Reports should be sent directly and immediately to the United Nations OCHA’s Humanitarian Situation Room for their evaluation. OCHA would be encouraged to verify the information and to work with CIAT and local officials in developing a rapid, non-military and complementary response. Many of these issues were being negotiated between United Nations human rights and humanitarian agencies operating in Colombia and the GOC at the time of the evaluation. Barring national security reasons, OCHA should be able to simultaneously assess the threat and work with CIAT and local officials to prevent potential violations. If greater attention equals greater effectiveness, as this study suggests, then OCHA’s involvement will likely enhance the system’s effectiveness. The GOC international cooperation agency (ACCI) is currently finalizing an accord with the Group of 24 and with other international donors including the European Commission, Japan, other nations to fund a “plan for humanitarian action”, PAH, spearheaded by ACNUR, HCHR Office in Colombia, and OCHA. A section of this accord is devoted to support for the Ombudsman’s Early Warning System. Incorporating this measure would lead not only to greater responsiveness but would open up the system to greater international financial and logistical support. At present, almost the entire weight of this system is being borne by the United States; it would be helpful to think about broadening the responsibilities. A. Other recommendations: 39 Organizational 1. The EWS needs to be better integrated into the structures of the Human Rights Ombudsman’s office both at the national and regional levels. In August 2004, the 24 regional analysts, five national analysts (there were two vacancies at the time), and the EWS Director were all employees of MSI, not the Ombudsman’s office. Only the Ombudsman Delegate was a state employee. Senior officials of the Ombudsman’s office expressed concern about hiring practices, independence and integration both in the Bogotá office and in the 24 field offices. It would be advisable to begin a process of greater responsibility and “ownership” of the EWS by the Ombudsman office and the GOC so that the EWS could survive a reduction in USAID or Plan Colombia funding. Greater organizational integration would also elevate the weight of the EWS within state councils. 2. Feedback regarding results should be shared between the two component parts of the system, EWS and CIAT. Both entities should be able to evaluate their operations by analyzing whatever reports are sent to the CIAT from local authorities, including the police and military where possible. There should be a stringent definition of the type of intelligence or information that must remain classified for security purposes. 3. The Ombudsman regional analysts should whenever possible and within reasonable levels of personal risk consult with the human rights officers of the departmental police chief and military commander. These should be considered verification sources additional to the minimum of three credible civilian sources. 4. In order to mitigate the currently high burn-out rate, regional analysts should be given a regularly scheduled one-week break with paid travel every three months. Where possible, regional analysts should be asked to commit to the job for a minimum of 18 months unless special circumstances require less. This amount of time is necessary to establish trust and relations with local communities and officials in conflict zones. Their salaries might be augmented to include health and life insurance. 5. To overcome concern on the part of the government security forces that the EWS will impede war efforts, EWS managers should share its Risk Reports and Follow-Up Notes directly with local officials responsible for security and protecting human rights, particularly within the framework of the Departamentos y Municipios Seguros program and the current GOC policy to decentralize many human rights programs from the national to the local. Doing so presents an opportunity to incorporate information concerning local civilian populations and the means to protect them directly into strategic and operational plans at the local level. 6. Outreach and publicity on the goals of the EWS is greatly needed. The FY04-05 budget includes funds for a community radio network, a video clip to be used at workshops and training programs, for the website, print publications, and a radio and television strategy. To the extent possible, outreach funds should also be provided to mount a national public 40 information campaign, based on radio and television and also on widely-distributed print materials. 1. The EWS should as soon as possible hold weekend seminars targeted to governors and other key civilian officials as well as for senior members of the Police and Armed Forces to train them on the goals and practices of EWS/CIAT. Logistical 2. Similarly, the regional analysts cannot perform their duties without means of transportation and appropriate telecommunications devices. The EWS budget should include these items. 1. An early warning system cannot function where local authorities and other sources have no telecommunications resources. The EWS has contracts with telecommunications providers (Comcel and BellSouth) to equip regional analysts. The EWS budget should also include provision of communications, including cellular phones where no regular lines exist and internet capabilities for local authorities in the urban centers. This could be accomplished in cooperation with other international donors. 2. Similarly, the regional analysts cannot perform their duties without means of transportation and appropriate telecommunications devices. Where possible, the Regional Human Rights Ombudsman office should make every effort to provide these Methodological 1. Timing: It is unreasonable to expect the system to be able to specify the date and hour of a threatened attack, but the EWS analysts should develop a more rigorous method for classifying the range of imminence, especially given the weight that the CIAT places on immediacy in the decision whether or not to issue an alert. 2. Degree of Specificity: It is unreasonable to expect the EWS system to be able to specify precisely which small village will be subject to an attack that is anticipated within a particular area. Identification of the municipality should be sufficient to allow verification by the military, police, and internal security intelligence agencies. Upon receipt of a risk report (report headers refer to department and county, and smaller units where relevant) the National Police should rigorously compare the risk report with its own more sophisticated intelligence collection in that county, looking to smaller units – corregimientos and veredas – where indicated in the report and also to neighboring population centers. In order to avoid confusion or discrepancies between the two components as to whether or not a warning was issued in advance of a particular violation, all records maintained by both the SAT and the CIAT should include measures taken by state agencies in response 41 to risk reports, as listed by IR number. All records should be capable of cross-referencing among state agencies. 3. Definition of What Constitutes a Violation: Defined too broadly: one problem encountered by the evaluation team was the tendency to take an overly broad approach to the Early Warning System, by conflating it with a national conflict prevention policy. The EWS should be one, central component of a national prevention policy, but risk reports lose their utility if they refer to longer term structural effects of the internal conflict such as the loss of livelihood and destruction of necessary infrastructure. This type of violation should be dealt with quickly and effectively by the government’s national conflict prevention strategy, but the EWS must be focused on threats to life and liberty of civilian populations in order to be effective. Standard definitions: The EWS has developed a set of parameters as to what constitutes a massive violation, using registered according to standard definitions provided in international conventions and agreements. All state agencies should standardize their registration of violations accordingly. 4. Identifying and Measuring Effectiveness: As discussed in Section III, the EWS has already developed a measure of success and failure analyzing whether an alerted violation occurred within three months of issuing a Risk Report or an Early Warning. The MSI Evaluation team used the EWS methodology and data for three departments and cross-checked it with independent data from CINEP. To do so, we used the EWS standards for defining human rights violations.20 MSD has signed an agreement with CINEP to share human rights data which CINEP collects independently. Yet surprisingly the exercise of cross-checking official data with independent data had not been done with reference to the Risk Reports. Our findings did not significantly differ from the Ombudsman’s. We found that the reported violations did not occur (success) in 42 .8% of the cases. The EWS found a slightly higher rate 47.6 to 56% The lower success rate of the system can be explained by the particular intensity of the conflict in the three departments. We propose that MSD incorporate two basic evaluation criteria, one measuring results, the other assessing process: 20 EWS categories of violations are: Massacres (killings of 3 or more persons); Multiple related homicides; Forced displacement; Forced disappearance; Forced recruitment; Seeding land mines/ landmine accidents; Restrictions on free circulation of individuals; Attacks on indispensable goods and provisions for a community; Interposing civilian populations amid hostilities; Indiscriminate attacks against civilians; Selective homicides; Methods to generate terror among the population. 42 First, as adapted from the existing EWS methodology, did the alerted violations occur within three months of issuing a Risk Report or Early Warning? Second, in terms of process, what measures were implemented by local officials, and how rapidly and effectively were they applied since the initial receipt by CIAT of a Risk Report? Measuring Results: Objective: Improve the effectiveness of the system as measured by the “no occurrence of alerted violation within three months of issuing a Risk Report.” Baseline The baseline for future analysis should be: the % of cases (of all cases where a Risk Report was issued) where the alerted violation did not occur within 3 months of issuing the Risk Report or Early Warning in 2003. (The EWS reports this figure as between 47 and 56% of all cases in 2003. MSD or a contracting team should cross-check the EWS figures by examining every Risk Report issued in 2003 and cross referencing alerted violations with EWS, CIAT, Human Rights Observatory and CINEP data banks). The result will provide the project baseline. Indicators:  specific violation(s) identified in all Risk Reports issued during 2004 (and 2005 and 2006).  data on whether alerted violation occurred (EWS, CIAT, Human Rights Observatory and CINEP data banks, adjusting different data sources and definitions of violations to meet the EWS standards). Benchmark By 2006, the EWS should achieve a success rate (no occurrence of alerted violation) of or success from the 2003 baseline of 47% - 56% to 65%-75% in 2006. The advantage of this method of evaluation is that it does not depend on the unexpected dynamics of the internal armed conflict. The universe of violations is limited to those anticipated violations that are systematic or massive and which are identified in the Risk Reports and Early Warnings. Although there is not necessarily a direct causal relationship between the actions of EWS/CIAT and the no occurrence of a violation since other factors may be involved, there is sufficient evidence to show that there is enough of a causal relationship between the operation of the EWS and the occurrence of a violation to establish discernible relations, trends and patterns. Measuring and Improving the Process 43 Once a baseline is established, the margin for improvement will depend on the increased effectiveness and celerity of the state response. CIAT is currently developing a system to evaluate state responses; if the recommendations requiring local officials to file reports on the impact of their measures are implemented, there will be further data. The EWS has also begun to assess state actions in their follow-up notes. As this data collection improves, it should be categorized and evaluated. Data should be collected on: Time between issuing Risk Report and CIAT recommendations (currently 21 days on average, which should be considered the baseline). The benchmark should be 7 days or less by the end of 2005.  Time between CIAT actions and the implementation of measures by local authorities  Relationship between specific measures and occurrence of violation. By focusing on correcting weaknesses and shortcomings revealed in these data, the EWS should show a higher rate of success (as measured by the no occurrence of the alerted violation). 44 Appendix 1 84 29 37 0 10 20 30 40 50 60 70 80 90 Number of Risk Reports Issued (EWS) Number of Early Warnings Issued (CIAT) Number of Violations that Occurred Following Risk Report or Early Warning EWS Phase II 2003 Total Risk Report (EWS) and Early Warnings (CIAT) And Occurrence of Violations Source: Defensoría del Pueblo. 45 Appendix 2 30 6 16 0 5 10 15 20 25 30 Number of Risk Reports Issued by EWS Number of Risk Reports where Violation Did Not Occur Following Report Number of Risk Reports where Violation Occurred Following Report EWS Phase II (EWS-CIAT) January 1 – April 30, 2004 Risk Reports Resulting in No Violation or Violation of Predicted Threat Source: Defensoría del Pueblo 46 APPENDIX 3: CASE STUDY: DEPARTMENT OF ARAUCA Despite the deterioration in Arauca, only __The Defensoría del Pueblo issued 7 Early Warnings and 1 Follow-Up Report for Arauca in 2002, all prior to the establishment of the CIAT. Risk Reports and __ No Risk Reports were issued for Arauca during 2003, but 4 Follow -Up Reports (Notas de Seguimiento) were issued that warned of renewed threat in specific areas: 3 concerned the municipality cited in alert #044-02 (May 2, 2002) and one concerned the municipality cited in alert #68-02 (July 26, 2002). Of the 6 Risk Reports issued by the Defensoría during the first half of 2004, the CIAT renewed an Early Warning that had been issued by the Defensoría in May 2002, for the department during the period __ to __, and only __ were and three others were made extensions of that same Early Warnings. Although these four were not instances of the CIAT issuing a new warning, they are considered to be Early Warnings issued by the CIAT. The following table attempts to trace the relationship between the reports that were issued and violations that occurred in the indicated municipalities. Information concerning violations is taken from several sources and should not be considered definitive. MSI Evaluation Early Warning System Early Warning System in Arauca, January 2002-June 2004 Date Rpt # Municipality Indicated threat CIAT response Actions taken Violations* Jan 24, 2002 7-02 level 2 Cravo Norte (municipal center, Hato Corozal-- Corralito, Puerto Florez) Threatening signs posted Jan 19 by AUC; 40 people displaced from Corralito & Pto Florez to municipal ctr Not applicable; DP alerta to: Minister Interior, Vice Presidency, Military General Command, Natl Police Director General, RSS March: Military withdrew from Cravo Norte, leaving civilians vulnerable; No other response from authorities No further IHL violations reported, successful outcome Sept 9, 03: Alert 7-02 terminated Feb 12, 2002 16-02 level 1 Tame (municipal center) Paramilitary death threats against demonstrators Not applicable; DP alerta to: Minister Interior, Vice Presidency, Military General Command, Commander Army Division II, Natl Police Director General, Commander Arauca Police 18th Brigade/ Police officers protected demonstrators; Governor negotiated peaceful end to strike; Arauca police sent two squadrons to Tame & Saravena; police investigated death threats Demonstrators not harmed, successful outcome Sept 10, 2003: Alert 16-02 terminated Feb 26, 2002 Of.No. 4020/CO￾SAT￾049/02 Follow up to #16-02 (ficha de seguimiento) AUC threats against local government officials (mayor, education secretary) who supported demonstrators As above As above Feb 26-March 15, 02: 11 targeted assassinations, failure March 12, 02 29-02 level 1 Puerto Rondón Paramilitary death threats against 23 civilians accused of collaborating with guerrilla Not applicable; DP alerta to: Minister Interior, Vice Presidency, Arauca Governor, Military General Command, Commander Army Division II, Natl Police Director General, Commander Arauca Police Army patrols in urban center & on Rio Casanare; threatened individuals given police protection; registration of newcomers to region; increased police intelligence Civilians not harmed, successful outcome Sept 11, 03: Alert 29-02 terminated March 18, 02 31-02 level 1 Cravo Norte (municipal center) Threatened AUC attack on civilians accused of collaborating with guerrilla; threat increased by withdrawal of military from municipality; guerrilla roadblocks March 16: 3 disappearances Not applicable; DP alerta to: Minister Interior, Vice Presidency, Arauca Governor, Military General Command, Commander Army Division II, Natl Police Police patrol deployed March 20 in municipal center, register newcomers to area, protect threatened individuals; governor had food flown in past roadblocks; Proc/Fiscal investigations. April 28: Inter-institutional No violations reported, successful outcome Sept 9, 03: Alert 31-02 terminated 48 Director General, Commander Arauca Police security council meeting of sectl Fiscal, regl Procurador, Commander Arauca Police, Commander 18th Brigade, Gobernacion delegates. May 2, 2002 44-02 level 1 Tame (municipal center, Betoyes, Rincon Hondo, La Holanda) Serial targeted assassinations by paramilitaries of civilians accused of collaborating with guerrilla; forced displacement resulting from confrontations among AUC, FARC, ELN April: 2 council members kidnapped by ELN Not applicable; DP alerta to: Minister Interior, Vice Presidency, Arauca Governor, Military General Command, Commander Army Division II, Natl Police Director General, Commander Arauca Police Military operations against AUC April-May 02; military/ police patrols on roads into Tame and in rural areas; special unit of Judicial Police created; Fiscal/CTI investigations & arrests; electromagnetic monitoring system set in municipal police station; paid compensation to victims; improved police contact with mayor and personeria; RSS attending to IDPs Early July: AUC death threats against labor union members; July 20-21: 2 teachers killed; July: 5 tortured, killed, dismembered failure (August-end of 2002) Aug-Dec: 56 assassinations, presumed AUC Oct 02-Jan 03: 5 explosions FARC/ELN June 28, 2002 61-02 level 1 Municipios: Cravo Norte, Puerto Rondón, Arauquita, Arauca; Entire plains region Civilian population endangered by armed confrontations FARC/ ELN/ AUC competing for control of plains region and Arauca deptal capital; FARC threats against municipal government personnel Not applicable; DP alerta to: Vice Presidency, Arauca Governor, Military General Command, Commanders Army Division II & 18th Brigade, Natl Police Director General, Commander Arauca Police Gubernatorial elections postponed; police control posted, police intelligence intensified July 22: 5 civilians dead, AUC￾army battle Aug 25: app. 100 families displaced in region Aug, Arauca: 5 found tortured and killed failure (Oct 2002-Aug 2003) Serial targeted killings: Arauquita: 35 Arauca: 86 Cravo Norte: 1 Puerto Rondon: 3 Oct 9, 2003 Alert 61-02 terminated July 26, 2002 68-02 level 1 Saravena (municipal center) Residents endangered by FARC/ELN indiscriminant attacks on urban center, military and police stations Not applicable; DP alerta to: Vice Presidency, Minister Interior, Arauca Governor, Military General Command, Commanders Army Division II & 18th Brigade, Natl Police Director General, Commander Arauca Police July 26: municipal security council— mobile military police squad transferred for two months from Arauca to Saravena to reinforce municipal police. July 30: all police stations ordered to increase intelligence gathering and coordination w/local civil and military authorities, hold weekly local security meetings, Sept: 4 killed FARC attack October: 4 targeted killings failure 49 New system Jan 15, 2003 Of.No. 1040/CO￾SAT￾10/03 44-02** Tame (municipal center, Betoyes, Rincon Hondo, La Holanda, Pueblo Nuevo, Flor Amarillo, Botalon) Ongoing serial killings, threats, forced displacements from Pueblo Nuevo, Flor Amarillo, Botalon to Tame center: Jan 3-13: 12 serial killings Jan 11: (FARC) car bomb Jan 12: (FARC) bridge blown up Jan 24: NO early warning; No recommends: notes that there is a security advisory for the region, under military control (no record of comm. w/local authorities) Rehabilitation & Consolidation Zone: ongoing increase in military/ police forces on the ground Feb 8: AUC attack on village: 1 killed, 8 kidnapped (3 released, 5 tortured and killed); February: 24 disappeared Failure March 3, 2003 Of.No. 1040/CO￾SAT￾85/03 44-02 Tame (municipal center, Betoyes, Rincon Hondo, La Holanda, El Pesebre) Continued serial targeted killings, hostage taking, Civilians caught in AUC￾FARC confrontations: Feb 16 El Pesebre battle: 16 civilians killed No response Rehabilitation & Consolidation Zone: ongoing increased in military/ police forces on the ground March 7, 03: 3 killed FARC attack; March 15-20: 4 killed AUC attack; April 1: 5 civilians killed AUC- Army battle; April 30: 4 killed unidentified armed unit failure (5 mo. period) March-July 03: 64 civilians killed in Tame area May 9, 2003 Of.No. 1050/CO￾SAT￾247/03 44-02 Tame (Betoyes, El Pesebre, La Holanda) Civilians caught in AUC￾FARC confrontations: Feb 16 El Pesebre battle: 16 civilians killed No response Increased military/police forces on the ground May: forced displacement entire village Betoyes Aug: 4 killed FARC attack failure May 9, 2003 Of.No. 1050/CO￾SAT￾248/03 68-02 Saravena (municipal center) Increase in AUC serial targeted killings: Dec 02-April 03: 37 killed No response No change recorded May: 12 targeted killings June: 4 targeted killings July: 15 targeted killings Aug: 10 targeted killings Sept: AUC units situated in main plaza of urban center; guerrilla threatening retaliatory attacks Failure Jan 30, 2004 3-04 Saravena Threats and attacks against labor, social & human rights activists and municipal council & administrators, by FARC, ELN, AUC; road blocks & robbery by illegal armed groups Feb 12: NO early warning; Recommend: comply with decisions Feb 4 meeting, continue ongoing operations. Feb 24: memo to local authorities. Feb 4, 03: Representatives from MinInterior, Presidencia, ForMin, DP, and local civil authorities met in Arauca to discuss public order and respond to the situation. No further targeted killings Successful outcome March 5, 2004 15-04 Tame (Betoyes, La Holanda, Los Aceites, Tame-Pto Rondon highway disputed by AUC&FARC; March 26: no new March 26: Governor’s office provides security guards for March 8-15: 2 ranchers killed March 13: FARC attack, 1 dead 50 Matarrala, El Plato, San Ignacio) Puerto Rondon (San Mateo, El Milagro) road blocks & extortion; June 03-March 04: 62 families displaced in area Feb 14&20: 4 killed on hwy by FARC Feb 26: 3 killed, 3 disappeared by AUC early warning, but maintain May 2002 IR#44-02 Recommend: cont. existing measures; call security council w/mayors present; develop security plan; attend to humanitarian needs of IDPs March 31: memo to local authorities mayors in high risk areas; April 29: departmental security council -- improved logistics and number of police, improved intelligence efficiency 5 injured Mid-March: 6 targeted killings March 24-28: 1 targeted killing, 1 disappearance (AUC) April 5: 1 targeted killing (FARC) April 12: & 18: 1 targeted killing (FARC), 1 disappearance May 5-17: 3 targeted killings May 19-20: 13 kidnapped & killed, 1 disappeared by AUC failure March 19, 04 19-04 Cravo Norte (Cumare, San Rafael, Corralito, Las Mercedes, Las Calcetas, Caño El Medio) FARC & ELN advancing on area to block AUC (new alliance BVA-ACCU); March 1-15: 10 families displaced March 5-15: 9 targeted killings by AUC April 5: NO early warning Recommend: cont. current measures, reinforce military presence, attend to human. needs of IDPs April 5: memo to RSS (no record of comm. to local auth.) Departmental security council convoked No further violations reported as of time of evaluation Successful outcome May 20, 2004 39-04 Tame-Fortul highway (Alto y Bajo Tamacay, Babaica, Caño Grande, San Pedro, Sta Helena, Sta Ines, Angosturas, El Triunfo, Corocito, Cravo Totumo, Cravo Corozo, Cravo Regilia, Caño Guata, Mata de Topocho, La Perla, Cravo Charo, La Soledad, Botalon, highway between Gualabao and El Mordisco) FARC, ELN, & AUC rapid build up for control of Tame￾Fortul hwy, guerrilla road block; combat near urban areas; serial targeted killings, forced displacement & disappearances May 31: no new early warning, but maintain existing alert for 3 months Recommend: convoke security council w/ relevant authorities June 1: memo re:39, 41, 45 to governor May 22: Departmental security council –Tame police unable to investigate incidents for personal security concerns; Tame inspector agrees to investigate if 18th Brigade lends aerial support. June 14: Pres. Uribe, FGN, various ministers convoke community council, discuss ineffectiveness of democratic security policy in region, commit to placing at least one officer in every village in Tame May 21: 1 targeted killing by FARC May 21: 22 displaced from Botalon following AUC massacre June 7: 1 killed on hwy, unidentified armed group Failure May 21, 2004 41-04 Tame- Pueblo Nuevo highway (Flor Amarillo, Puerto Jordan, Santo Domingo, Intensified combat for control of Tame-Pueblo Nuevo hwy; rapid deployment of AUC toward indicated villages, threatening population accused of collaborating with guerrilla May 31: no new early warning, but maintain existing alert for 3 months. Recommend: as above June 11-24: 5 targeted killings by AUC Failure May 28, 2004 45-04 Tame municipal Tame urban center controlled by AUC since 2002, guerrilla May 31: no new June 1: Governor’s human rights office creates a Municipal No further violations in Tame urban center as of time of 51 center (barrios Villadela and San Francisco) attacks to regain control; April 18: 2 targeted killings AUC May 3: FARC/ELN car bomb in Tame commercial district May 3: 8 targeted killings AUC May 14: FARC/ELN assassination of bank manager early warning, but maintain existing alert for 3 months Recommend: as above Peace Committee to find means of demilitarizing urban center evaluation Successful outcome (to date) Sources: Defensoría del Pueblo, SISAT database and archives; Ministerio del Interior, CIAT archives; MSD database; Presidential Program for Human Rights and International Humanitarian Law, “Algunos indicadores... de Arauca” * Violations of international human rights and humanitarian law committed by the armed actors that were cited in the Report, committed in the municipality listed in the Report, and within three months of the Report being issued. Note that this is an approximation – not all violations that took place during these time periods appear in the documents cited, and the sources use different definitions of violations. ** Defensoría’s Notas de Seguimiento referring to prior Alerta Temprana Extrajudicial killings in department of Arauca probably related to the armed conflict, January 2002-March 2004, by month Jan 02 Feb Mar Apr May June July Aug Sept Oct Nov Dec Jan 03 Feb Mar Apr May June July Aug Sept Oct Nov Dec Jan 04 Feb Mar 65 20 25 30 33 35 42 20 26 29 40 31 50 71 50 40 49 20 31 40 19 29 20 29 31 37 40 Source: National Police, cited by Observatorio del Programa Presidencial de DDHH y DIH MSI Evaluation Early Warning System Arauca Overview Conclusion: 1. Was the Risk Report based on reliable information concerning a credible threat? N=18 In all 18 cases, the Risk Report referred to location by specifying municipality and listing subunits. In each case the type and level of threat was demonstrated with reference to specific occurrences: rapid change in conflict dynamics, such as recent appearance or intrusion of armed actors into locations dominated by opposing armed actors; and acts of violence that were already occurring in that municipal area, such as targeted killings, forced displacements, death threats, and kidnappings. Based on what the team learned in Arauca, to verify information the regional analyst routinely consults with the local or municipal police station, government agents who cover rural areas such as the Procurador Agrícola, military officials such as the (now deceased). commander of the Móvil 5 Brigade, municipal personeros, parish priests, bus drivers, leaders of civic organizations, and local human rights organizations. 2. Was the Risk Report issued in a timely manner? We cannot assess how quickly the Defensoría del Pueblo responded to the regional analyst’s warnings by issuing a Risk Report without reviewing the dated documents (fichas de verificación) that contain the information sent to the national analysts. The Defensoría del Pueblo keeps these documents confidential in order to protect local witnesses and informants. In five cases, major violations of the type cited in the Risk Report occurred within four days of the Report being issued. This underscores the urgency of developing a method to clearly distinguish between warnings of imminent emergencies and warnings issued early enough for a preventive response. The current Early Warning System, entailing multiple layers of analysis and verification, is not designed for emergencies. In cases where the threatened violation is likely to occur in less than a week, the regional analyst should send an emergency report, which the SAT director would then forward directly, with the understanding that the information has not yet been verified, to the United Nations Humanitarian Situation Room in Bogotá, which is coordinated by OCHA. The report would be distributed only to the circle of humanitarian agencies that are members of the Humanitarian Situation Room: International Committee of the Red Cross, UN High Commissioner for Refugees, UN Office for Coordination of Humanitarian Affairs, International Organization for Migration, Unicef, Codhes, World Health Organization, World Food Program. These agencies operate in the field and have the experience and capacity to mobilize an immediate response. MSI Evaluation Early Warning System 3. Did the CIAT conduct its verification in a timely manner, to allow local authorities time to take appropriate preventive steps? N=10 No response = 3 CIAT met within one week of receiving Risk Report = 1 CIAT met between one and two weeks after receiving Risk Report = 4 CIAT met more than two weeks after receiving Risk Report = 2 CIAT met more than three weeks after receiving Risk Report = 0 Note: records indicate that written communications to local authorities are often sent more than a week after the CIAT meeting, but documentation is too inconsistent to use for evaluation purposes. 4. Did the CIAT recommend actions that could, with a reasonable degree of certainty, contribute to preventing the threatened violation? No recommendations = 4 No change (maintain existing efforts) = 1 Convoke security council = 4 Develop security plan = 1 Increase force levels = 1 Provide social services to IDPs = 2 5. Did local authorities comply with recommendations from CIAT members? Recommendation Response Convoke a security council Security council held whenever recommended Develop a security plan Plan to improve logistics and intelligence Increase force levels Convoked a security council Provide social services to IDPs RSS contacted on at least one occasion 6. Did threatened violations occur in the municipality identified by the Risk Report: a) (prior to November 2002) within three months of SAT Early Warning? N=8 Yes = 4 No = 4 b) (after November 2002) within three months of CIAT issuing recommendations without an Early Warning? N = 6 Yes = 4 No = 2 c) (after November 2002) within three months of CIAT issuing an Early Warning? N= 4 Yes = 3 (of these Early Warnings, 2 were issued less than three months prior to the time of evaluation) No = 1 (this Early Warning was issued less than three months prior to the time of evaluation) 54 ► Table Of the 18 cases of Early Warning System operation in Arauca, violations of the type indicated in the Risk Report occurred within three months in 11 cases, and were presumably prevented in 7 cases. ► Of the 11 cases in which the information was communicated to authorities in the form of an Early Warning, violations occurred in 7 cases and were presumably prevented (to date, in the most recent warning) in 4 cases. ► Of the 7 cases in which the information was communicated to authorities not in the form of an Early Warning, violations occurred in 5 cases and were presumably prevented in 2 cases. Note: Risk Reports warn of general categories of violations, likely targets, likely perpetrators, and likely locations with as much specificity as the information allows, based on patterns of violence and the changing dynamics of the conflict. They do not warn of any particular threatened event which could be demonstrated to have either occurred or not on a predicted date. Likewise, there is no test for how many violations that fall within the parameters of a Risk Report did not occur as a result of actions taken in response to the report. The anticipated violations did not occur in only 3 of the 10 cases of Risk Reports issued after the CIAT’s inception in November 2003. In addition to the problem of timeliness, the vague, under￾specified nature of the recommendations relayed to local authorities by the CIAT might partially account for the system’s failure rate at the response stage. In fact, of the three cases where no further violations occurred (to date), in one case local authorities were only advised to continue with what they had been doing, in one case they were advised to continue with current efforts and request military and police reinforcements, and in one case they were advised to convoke a security council. The same outcome (no further violations) was achieved by no recommendation as by the recommendation to hold a security council, indicating that the CIAT’s recommendations have little influence on the outcome. The CIAT member agencies need to develop more problem- and condition-specific measures to respond in situations calling for preventive (not emergency) action. They should collaborate with international humanitarian assistance agencies and other public agencies that operate in conflict zones in order to expand the range of responses well beyond military or police action. MSI Evaluation Early Warning System Appendix 4: Case Study: Department of Chocó DATE RPT # MUNICIPALITY INDICATED THREAT CIAT RESPONSE ACTIONS TAKEN VIOLATIONS March 1, 2002 22-02 Pueblo Rico & Tado Level 1 Military and Police operations to combat the FARC, ELN and AUC caused the forced displacement of over one hundred afro-Colombian families. AUC made public that they will close roads and prevent goods from coming into the area. High level EWS alert due to AUC presence. EWS warms of possible Selective Homicides and Destruction of public goods necessary for the well being of the population. Non￾Applicable. CIAT had not been formed. Departmental security council Military road checks Permanent military and police presence Some security measures like prohibiting the circulation of vehicles at night time. Selective Homicides. Destruction of public goods necessary for the well being of the population. (DP) Illegitimate tolls where food and other indispensable goods were Confiscated Oct 15 AUC kills 2. This causes the forced displacement of 170 people. (NdS) Failure Feb [?]2003 NdS As Above Illegal tolls by the armed groups block the entrance of food to the area. ELN threatened to attack the rural and urban zones of Santa Cecilia as retaliation to the attacks by the AUC. April 24, 2002 40-02 Carmen del Darien, Bellavista (Bojaya) 300 AUC soldiers were spotted near the cities of Vigia del Fuerte and Carmen del Darien. It was reported that they were making their way to Murindo and Bojaya to dispute the FARC over control of the area. High level EWS alert. EWS warns of possible Selective homicides, Destruction of public goods necessary for the well being of the population and Forced internal displacement. N/a EWStifies governmental departments that could prevent violations from taking place such as the Army, the Police, several ministries, etc. Selective Homicides. Destruction of public good necessary for the well being of the population. Forced internal displacement (DP) May 1, 2002 In combat between the AUC and The FARC, civilian Diego Luis was Shot in the crossfire. (Cinep) Failure May 29, 2002 51-02 Carmen del Atrato The ELN and the ERG threatened with invading the Municipal center. This presents a high threat to Indian leaders from the Embera Wounaan and Embera Chami communities. al N/a SAT notifies governmental departments that could prevent violations from taking place such as the Army, the Police, several ministries, etc. None (DP) Success 56 High level EWS alert due to possible massacres and selective murders, Destruction of public goods necessary for the well being of the population. Forced internal displacement. May 29, 2002 52-02 Lloro Probable violent incursion to the urban areas of Lloro by the ELN. Although the direct threat is against the Police station, civilians are likely to be hurt. Medium level EWSalert warning of possible Indiscriminate attacks; Destruction of public goods necessary for the well being of the population and Forced internal displacement. N/a EWS notifies governmental departments the Army, the Police, several ministries, etc. to ensure violations of HR do not take place Red de Solidaridad gives food and tools to the communities that were displaced. Indiscriminate attacks Forced internal displacement (DP, NdS) Aug 2, ELN takes over the town And kill the police chief (NdS) Failure Aug. 13, 2002 NdS Same as Above After the August 2 incident, regenerated fear in the population thus making probable the forced displacement of the population. N/a (By DP) Aug 13 Questioned Red de Solidaridad on what they did/are doing socially and economically to prevent future HR violations. Aug 13, Request more protection for the population to Ministry of Interior, to the Army, to the governor, and Police. ambiguous outcome/ partial success Nov 28, 2002 NdS Same as Above To CIAT, during the Aug 2 ELN take over of Lloro, there was no Army in the town. Although the army returned after the incident, they left 3 weeks a go and there are already rumors about a guerrilla attack. The SAT warns of possible displacement and selective deaths. March 17, murder of civilian (NdS) Abiguous outcome/ partial success April 10, 2003 NdS Same as Above Feb 18, combats between the Army and the ELN, once again, put the population at risk. This caused the forced displacement of communities from El Llano. SAT asks for permanent monitoring 57 along with military operatives to protect the rights of the population. NEW SYSTEM CIAT IS FORMED Nov 28, 2002 IR 84- 02 Rio Iro FARC’s 34th and the ELN have carried out incursions in the urban areas of the municipality. They accuse the population of taking sides with the paramilitaries. They have committed indiscriminate attacks, homicides and massacres. Medium Level EWS alert due to ELN presence. EWS warns of possible selective Homicides and Massacres; Indiscriminate attacks. Destruction of public goods necessary for the well being of the population. Forced internal displacement. Dec 5, 2002 No E.W. 12/22/02 The Police commander for Choco announced that a security council had been held on Dec. 13 with officials from the department in order to better understand the problem of public order in the area. The Manisalva battalion as well as the national police agreed to carry out control operations in the urban areas of the region. 12/27/03 the Secretary of the Interior announced that 3 security councils had already been held with officials and mayors from the San Juan region None (CIAT) EWS chart displacement Ambiguous outcome Nov 28, 2002 85-02 Condoto (rural zone and La Florida) Probable combats between the FARC-EP and the AUC are likely to leave civilian victims. Medium level EWS alert due to FARC and AUC presence. SAT warns of possible Selective Homicides, Indiscriminate attacks, forced internal displacement and preventing the entrance of food to the area. Dec 5, 2002 No E.W. 12/3/02 – DAS has informed government branches of the situation in the areas to prevent a situation similar to Bojaya. 12/22/02 – Police Chief held a security council with many of the state authorities. The Manosalva battalion as well as the National Police agreed to carry out control operations in the urban parts of the region. 12/27/02 – The Secretary of Government announced that the government had held three security councils thus far with high officials from the department. Forced displacement (DP) 12/3/03 Illegal tolls by the armed groups. March 10, 2003 FARC shot a civilian riding a Motorcycle (CINEP) CD NONE Conflicting info Dec 3, 2002 88-02 Jurado It is reported that the main objective of FARC and the AUC is the control of the narcotraffic in this area. The Embera Katio community is one of its principal victims. The AUC is responsible of Dec 10, 2002 Early Warning 5/19/03 – DAS informs that the locality now has the presence of the Public Force with nearly 200 men. Although they do not have the appropriate facilities in case of None (DP). None (CIAT) Note: Although there are no 58 blocking food, medicines and other vital goods from coming into the region. Medium level EWS alert due AUC presence. SAT warns of possible Massacres, indiscriminate attacks, forced internal displacement. confrontations with armed groups, since their arrival the “tranquility reigns.” Public officials have returned to carry out their functions. reported violations, the Mayor has on several occasions, been threatened telephonically by the FARC due to his public anti-leftist declarations. CD FAILURE Conflicting info May 16, 2003 38-03 Pueblo Rico, Tado (Trinidad, Aguadita, El Recreo, Parrupa High level EWS alert due to FARC, ELN and AUC presence. There is increased pressure both from the FARC as well as from the AUC. In the first 6 months of 2003, there were 6 homicides. The population has also been forced to attend promotional meetings and then been victims to forced recruiting. EWS warns of possible selective homicides, Multiple homicides, and destruction of public goods necessary for the well being of the population. May 23, Early Warning Security Council Actions from civil and military authorities have permitted the return of state presence in the area, that were formally controlled by illegal armed groups. 15 homicides forced recruiting of Katios forced displacement to St. Cecilia Failure October 9, 2003 NdS Same as above Since the alert there has been 15 Homicides of member of the Embera Katios indigenous reservation. The guerrilla has also carried out several forced recruiting operations In the area. Constant threats from the AUC Accusing the afro-Colombian community of abogado as being sympathetic towards the guerrilla has forced the displacement of nearly 200 people. The EWS warns of the possible displacement of 200 families to St. Cecilia. The EWS also warns of possible selective and indiscriminate homicides. Community was displaced. Failure June 4, 2004 NdS Same as Above The situation had settled down. 200 displaced families were planning their return on June 4, however on June 2, the ELN had confrontations with the army in Extend Early Warning 6/15/04 CIAT recommended the return of the internally displaced. CIAT also recommends to do everything possible to prevent HR Not known as of know. 59 Mumbu. This puts 79 of the returning families at immediate danger once again. EWS suggest that CIAT do everything possible to prevent more forced displacements and other HR violations. for another 3 months violations and more displacements. June 13, 2003 46-03 Bahia Solano (Cupica) The navy discussed withdrawal from the area, this increases possibility of attacks to the population from the AUC’s Bloque Cardenas and FARC’ s Front 57. There were warnings of disputes between the FARC and the AUC over the control of the area. The EWS issues a High Level alert due to FARC and AUC presence. SAT warns of possible Selective homicides and Forced internal displacement. June 27, 2003 No E.W. 7/7/03 Red de Solidaridad Social continued rebuilding homes for people in the area that were struck by an avalanche in 1998. 7/19/04 The infantry battalion #6 informed that they will stay in the area to protect the population until a police department is reinstated. After a security council, Cpt. Augusto Vidales, commander of the Marine unit present in Cupica informed that they will remain in the area. None SUCCESS Oct 20, 2003 68-03 Bojaya (Mesopotamia) The FARC and the AUC are in a continuous dispute over the lower Atrato region. This dispute has led to the homicides of several indigenous and afro-Colombian people. The armed groups have attacked the communities accusing them of taking sides with the enemy. The EWS issues a High level alert due to FARC and AUC presence. The SAT determined possible selective homicides, indiscriminate attacks, armed conflict leaving civilian victims, usage of terrorist methods for generating fear in people and forced internal displacement and destruction of civic good necessary for the well being of the population. Dec 11, 2003 No E.W. CIAT informs that the Public force present in the area should be enough to assure the protection of the population. Jan 18, 2004 Paramilitaries Destroyed the phone lines, the only type of communication available in the area. (cinep) 6/28/04 displacement of 318 families (ficha de seg) Failure Jan 28, 2004 NdS Same as Above The Cordoba battalion is entering the area. Possible combats with the AUC are likely to violate HR. Feb 5, 2004 Early Warning Feb 5, 2004 Reinforce the public force, carry out all possible actions to prevent violations of HR. Feb 5, 2004 Reported internal displacement of 30 families from Vereda de Caimanero. 60 Failure Feb 29, 2004 NdS Same as above The increase in the conflict around this area is endangering the population more and more. March 18, 2004 Extend Early Warning for another 3 months May 21, 2004 carry out all possible humanitarian “jornadas” and assist the internally displaced population.. Also carry out all necessary actions to prevent human right violations. May 15, 2004 FARC murdered farmer Horacio Chaverra for collaborating With the Colombian Army 3/19/04 displacement of 552 Embera ingineours (EL Tiempo) Failure May 17, 2004 NdS Same as above Same as above, plus possible second displacement of 1225 people. Increased possibility of indiscriminate attacks and destruction of civic goods necessary for the well being of the population. May 31, 2004 Extend Early Warning for another 3 months Continue the humanitarian “jornadas” and assistance to displaced population, and to carry out the necessary institutional measures to prevent human right violations March 6, 2004 May 28, 2004 FARC murdered Benjamin Romana For collaborating with the Colombian Army May 29, 2004 Out of fear due to the increased Violence over 130 families were Displaced from Piedra Cadela and Corazon de Jesus. On this same day 90 families from Napipi were displaced due to lack of public force in the area. There are also many reported threats from the FARC which are causing more displacem These armed groups has also on several occasions blocks the entrance of food and other necessary goods to the area. Failure July 8, 2004 NdS Same as above Due to the temporary presence of the army, it is quite likely that the FARC will retaliate after their departure. EWS warns of increased risk of HR violations, with an emphasis on forced displacement. July 12, 2004 Extend Early warning for another 3 months 7/12/04 Maintain registration and military operations in order to assure the observances of the population’s human rights. Continue the humanitarian jornadas March 19, 2004 20-04 Carmen del Atrato (Guangarales, la There has been increased FARC presence from the 34th front in the rural zones of April 5, 2004 No E.W. CIAT acknowledged presence of the FARC and AUC however they Forced recruiting by the guerilla. 61 Clara, el Lamento) the Carmen de Atrato region as a response to the Public Force’s operations. AUC troops have also been spotted at the head of the municipality this has lead to a belief that there will be conflict between the FARC and the AUC for control over the area. Medium level SAT alert due to FARC and AUC presence. SAT warns of possible selective homicides, homicides of multiple configurations, combat leaving civilian victims and forced internal displacement. security council was recommended determined that the presence is temporary and thus the population is not at imminent risk. CIAT recommends keeping all of the necessary measures needed to assure the well being of the population. Humanitarian “Jornadas” recommended. CIAT recommends to the civil and military authorities to carry out all necessary measures to assures the protection of the population May 2004, the FARC burned down The home of an indigenous family that refused to cooperate with them May 5, 2004. 14 families from the Noventa Village were displaced. Failure May 17, 2004 NdS Same as Above The increased conflict in the area puts the Population at an even greater risk. May 31, 2004 No E.W. Local public force was informed of the content of SAT’s Nota de Seguimiento so that they can take the necessary preventive measures. Forced recruiting in Fiera Failure June 23, 2004 NdS Same as Above Threats to the indigenous leader of the Fiera region, accused of collaborating with the army. Higher likeliness of forced recruiting, selective homicides, massacres and forced displacement. July 12, 2004 No E.W. No information as of now June 11, 2004 50-04 Novita ( Irabubu, El tigre, Tamana, Surama, Curundo) The FARC’s 57th front and the AUC Calima front are in dispute to control the planted illicit crops in the area. The FARC has plans to construct a back road which will allow them the commercialize coca paste with in this region of the country as well as it would allow the transportation of weaponry. The AUC has increased its offensive against the FARC in order to impede this road from being constructed. As part of this offensive by the AUC, they have threatened communities along the Tamana rivers for taking sides with the guerrilla. The AUC has carried out several selective homicides, burned down several homes, a school and a church. June 22 2004 Early Warning CIAT recommends strengthening the public forced in the area. The CIAT also recommends the implementation of effective methods for preventing massive HR violations and an integral response to illegal crops. Finally, CIAT recommends humanitarian assistance to families of Novita. No information as of now (DP). 62 They have also blocked food and food from entering the area which led to the forced internal displacement of 130 families. The EWS issues a high level alert due to FARC and AUC presence. SAT warns of possible Forced disappearances, Selective Homicides, Massacres, combats between illegal armed groups leaving civilian victims and terrorism to generate fear in the population. June 11, 2004 53-04 Medio Atrato (Bete) The FARC’s 57th and 34th fronts since the 90s have attempted to gain geopolitical control of the area. Recently they have increased their tolls in order to restrict civilian, food and goods traffic. In these tolls the FARC also carries out forced recruiting which has led to the instilled panic of the population. The EWS issues a high level alert due FARC and AUC presence. SAT warns of possible Forced disappearances, selective homicides, massacres, combats between illegal armed groups leaving civilian victims. Using terrorist methods to generate fear, forced internal displacement. July 8, 2004 No EW Due to all of the EW in the Area, the CIAT does not issue an EW. CIAT however, sends letters to the officials from the Dept. to recommend that they do everything possible to safeguard the population. No information as of now (DP). MSI Evaluation Early Warning System MSI Evaluation Early Warning System Choco Overview Conclusion 1. Was the Risk Report based on reliable information concerning a credible threat? N=25 In all 25 cases, the Risk Report or the Follow-up report referred to the location by specifying municipalities, the subunits. The level of alerts were supported with the change in the dynamic of the conflict. In some cases, the EWS issued an alert indicating the proper risk, however they decided to only rank it as Medium level alert, and a violation occurred, e.i. Lloro, and Carmen del Atrato. Each violation occurred within two months of the warning. Based on what the team learned in its visit to Quibdo, to verify information the regional analyst routinely communicated with its municipal personeros, the church, leaders of civic organizations etc. It was noted however, that the analyst had little contact with government agencies such as the police and DAS. 2. Was the Risk Report issued in a timely manner? We cannot assess how quickly the Defensoría del Pueblo responded to the regional analyst’s warnings by issuing a Risk Report without reviewing the dated documents (fichas de verificación) that contain the information sent to the national analysts. The Defensoría del Pueblo keeps these documents confidential in order to protect local witnesses and informants. In Bojaya, arguably the greatest violation of Human Rights since the EWS was formed, the SAT emitted its specific Alert over a week before the attack occurred. As CINEP has mentioned, it was the army whom decided not to act . 21 In the January 28th Follow-up report for Bojaya, a violation occurred just about a week later, with the displacement of 30 families from Vereda Caimanre on February 5, 2004. This case is quite ambiguous however, since the SAT had sent a risk report for the municipality several months before. It is the CIAT however who did not emit the Alert. The ambiguity goes even further, the day the displacements were taking place, the CIAT was discussing the case, after evaluating the follow-up report, the CIAT decided to emit an EW (68-03). Generally, Risk Reports are issued in a timely manner. 3. Did the CIAT conduct its verification in a timely manner, to allow local authorities time to take appropriate preventive steps? No response = 2 22 21 See, CINEP and Justicia y Paz, “Bojaya bajo el prisma de los medios de comunicacion,” 2002. 22 the EWS claims that they issued a follow up report to the CIAT on feb 18, 2003 that was never answered however this follow up report was not within the packet of follow up reports for choco. 65 CIAT met within one week of receiving Risk Report = 5 CIAT met between one and two weeks after receiving Risk Report = 6 CIAT met more than two weeks after receiving Risk Report = 3 CIAT met more than three weeks after receiving Risk Report = 2 4. Did the CIAT recommend actions that could, with a reasonable degree of certainty, contribute to preventing the threatened violation? No recommendations = 2 No change (maintain existing efforts) = 3 Convoke security council = 3 Develop security plan = 4 Increase force levels = 5 Provide social services to IDPs = 7 Note: these do not add up to the cases studied since on several occasions, more than one recommendation was made. 5. Did threatened violations occur in the municipality identified by the Risk Report: a) (prior to November 2002) within three months of SAT Early Warning? N=8 Yes = 3 No = 1 Ambiguous = 4 b) (after November 2002) within three months of CIAT issuing recommendations without an Early Warning? N = 8 Yes = 3 No = 1 Ambiguous / conflicting information = 2 Not known at present = 2 c) (after November 2002) within three months of CIAT issuing an Early Warning? N= 9 Yes = 5 No = 0 Ambiguous / conflicting information = 1 Not known =3 MSI Evaluation Early Warning System APPENDIX 5 CASE STUDY: DEPARTMENT OF NORTE DE SANTANDER 67 DATE RPT # MUNICIPALITY INDICAED THREAT CIAT RESPONSE ACTIONS TAKEN VIOLATIONS Jan 10, 2002 y April 16, 2002 NS de Julio 19/02, febrero 3 y abril 11 de 2003 no respondidas por el CIAT 02 El Tarra, Convencion, Teorama (La Cecilia, Aserrio) AUC threatens the population; they are with the guerrilla. This has created the municipalities like El Tarra. There is increased combat in Teorama and Tibu, specially in la Gabarra, between the FARC and the AUC, the population is at clear danger. The AUC is 20 minutes from the urban areas of the municipalities. Forced displacements are already taking place in the Urban area of Tarra SAT recommends that the state take security measures as well as humanitarian measures to prevent human right violations N/A DP. Warning fpr Policìa nacional, RSSS, Miniterior, Vicepresidencia Segunda Divisiòn y Quinta Brigada del Ejército Enero 2002: presence of army in zone (Aserrío), caused AUC to flee Julio 2002: convened Securtiy Council 19 massacres with 81 dead, since is until Sept. 2003. Forced displacement in el Tarra EWS Failure Feb. 26, 2002 019 Chitagá Possible forced displacement N/A Insufficient information No Forced Displacement EWS successful New system CIAT IS FORMED Ene. 21, 2003 005 Arboledas , correg Barrientos (Saravita, road between Arboled Increased combat between the FARC, E control of the territory. The AUC, duri three passengers of an inter-municipal bu SAT emits a high level alert, warning No Early warning on 30 Jan/ 2003 Increase the intelligence labors of the Poliice. Sent mixed patrols of police and army. Two confrontations between the Public Forces and subversives, killing The alerted attacks did not occur. Subversive action was neutralized. EWS successful 68 homicides, massacres, combats leaving indiscriminate attacks six subversives. Deactivated two explosive devices and confiscated arms. March 25, 2003 023 Cucutilla (La despensa, Santa Teresita, la meseta, la Cuchilla via Arboledas Cucutilla) Due to an strategy of expansion, it is quite possible that the FARC and the ELN will penetrate the urban parts o occur, the population will be victims of indiscriminate attacks. On the other hand, the AUC sees the rural areas thus retaliation attacks are also probable EWS emits a high level alert warning of combat leaving civilian victims, destruction of civic good necessary for the well being of the population, selective homicides, indiscriminate attacks and restriction of movement No Early warning on April 30/2003, but recommend convening Dept Security Council The Army has realized different operations in the area. The Police realized permanent patrols and checkpoints.. Confrontation between the ELN and the Public Forces, leaving three insurgents dead. Arms captured and an anti-personnel mine was deactivated. The alerted attacks did not occur EWS successful April 30, 2003 036 Hacari (municipal center) The dispute over control of the Northern part of Norte de Santander is placing the population or the urban areas of Hacari at high risk. This area is particularly strategic both from an offensive perspective as well as from a defensive, as a refuge. The SAT thus emits a high level alert. Massacres, selective homicides, attacks on civic good necessary for the well being, combat leaving civilian victims, and forced displacement are likely to occur as retaliation from the FARC due to increased AUC presence in the Area Early warning en mayo 12/03 Augmented the presence of the public force, cooperation with judicial activities to carry out searches. There were also joint operations between the police and the Army’s Ocana batallion to neutralize the threat (NdS). Hubo ayuda humanitarian por parte de l RSS April 30, large displacement (NdS) Oct 12, 2003 FARC kidnaps the Amaya brothers, who were likely Candidates for the local elections. They are released a week later (Cinep) Nov 25, 2003. AUC soldiers enter The Torrado Perez family Household, they demand all money And weapons. The AUC takes 3 Million worth of goods from their Store and 1.2 million of cash. This Causes the forced displacement of The family . EWS failure Oct 15, 2003 NSIR 036 As above There are increased AUC expansions to control ELN and FARC territory, specially towards the urban areas. The guerrilla on the other hand are making their presence better known Answer to the NS on March 2 of 2004. EW Extended for another 3 months, Maintain the operatives in the zone and designs a permanent security operation that will prevent HR violations. Jul. 28/94: IDPs return to communities as a result of confidence in the presence of the Army. 69 to show their control. In Sept 3, ELN sabotaged the Hacari police station. This is likely to create combats leaving civilian victims. since Feb 19, 2004. EWS successful May 1, 2004 031 El Tarra (municipal center) The FARC and the AUC are in dispute over control of the population, who they force to work in their illicit plantations. This dispute has afftect the social and economic areas of the community, making them dependent on these armed groups. The AUC and the FARC oftentimes place road blocks on the road that goes to Tibu, thus limiting the movement of the population. These road blocks make probable the occurrence of selective homicides, massacres, force displacement, land mines, and combat leaving civilian victims. The SAT thus emits a high level alert. No early warning on Mayo 19/ 2004, But recommends that that Oublic Forces assert greater control especially on the highway between Tibú and Convención The RSS and theCICR attend displaced persons with humanitarian May 21, Forced Displacement of 126 people. June 15, 34 coca pickers Assassinated in La Gabarra, of El Tarra EWS failure MSI Evaluation Early Warning System Norte de Santander Overview Conclusion 1. Was the Risk Report based on reliable information concerning a credible threat? N=6 In all 6 cases, the Risk Report or the Follow-up report referred to the location by specifying municipalities, the subunits. The level of alerts were supported with the change in the dynamic of the conflict. The recent arrival of the EWS regional analyst to Norte de Santander (May 25/03), de Santander (mayo 25/03), following the absence of an analyst for close to six months, prevented thorough verification of the data, information and procedures. However, the information given by the Regional Ombudsman concerning modus operandi of the previous EWS analysts – wh6 was there between November 2001 and December 2003 – coincides with the information given by analysts in other parts of the country, such as Chocó, Arauca y Cordobá: she affirmed that there was permanent communication between the analyst and the municipal personeros, the chutrch, leaders of civic organizations and, to a lesser extent with local authorities. 2. Was the Risk Report issued in a timely manner? We cannot assess how quickly the Defensoría del Pueblo responded to the regional analyst’s warnings by issuing a Risk Report without reviewing the dated documents (fichas de verificación) that contain the information sent to the national analysts. The Defensoría del Pueblo keeps these documents confidential in order to protect local witnesses and informants. The Risk Reports issued with respect to Norte de Santander permit us to distinguish between two different situations: The first is where CIAT delays providing a definitive response to the initial Risk Report, but in any case, the warned threat still does not occur;23 and the second denotes the case where CIAT responds more quickly but the alerted threat does occur in the manner that was warned of in the Risk Report.24. 3. Did the CIAT conduct its verification in a timely manner, to allow local authorities time to take appropriate preventive steps? 23 This occurred, for example, with Risk Report 036/03 that caused the Early Warning issued by CIAT for the municipalities of Hacarì in el Catatumbo. The Ombudsman sent a Follow-up Note on October 15/03 that called for reinforcing the existing measures and the CIAT responded almost five months later, on March 2/04, maintaining the Early Warning. . 24 In the case of what occurred with the Risk Report 021/04 concerning El Tarra en el Catatumbo, the Risk Report warned, among other things, of forced internal displacement. CIAT responded on May 1/04 declining to issue an Early Warning. On May 21, the forced displacement occurred. One month later, on June 15, 2004, in Tibú (La Gabarra) in El Tarra, 34 coca pickers –raspachines -- were assassinated. 71 No response = 025 CIAT met within one week of receiving Risk Report = 0 CIAT met between one and two weeks after receiving Risk Report = 2 CIAT met more than two weeks after receiving Risk Report = 1 CIAT met more than three weeks after receiving Risk Report = 1 4. Did the CIAT recommend actions that could, with a reasonable degree of certainty, contribute to preventing the threatened violation? Convoke security council = 2 Increase force levels = 3 Increase the fumigation of illicit crops = 1 5. Did local authorities comply with recommendations from CIAT members? Recommendation Response Increase fumigation of illicit crops According to the local press, since November 2003, the “third phase of fumigation in el Catatumbo” began. Increase force levels. Temporarily the Army remained in FiloGringo-La Gabarra en Tibú Convoke security council No information available 6. Did threatened violations occur in the municipality identified by the Risk Report26: a) (prior to November 2002) within three months of SAT Early Warning? N= 2 Yes = 1 No = 1 b) (after November 2002) within three months of CIAT issuing recommendations without an Early Warning? N = 3 25 It is necessary to clarify that this data was applied only to the Risk Reports. As was indicated in the Table, the Early Warning 2/02 had three follow-up Reports without an answer from the CIAT. 26 This does not include the data from what occurred with the Follow-up Report to Early Warning AT 02/02. There was no response from CIAT, or at least it was not registered statistically in this report. 72 Yes = 1 No = 2… c) (after November 2002) within three months of CIAT issuing an Early Warning? N= 1 Yes = 0 No = 1 ► Of the 3 cases of Early Warning System operation in Norte de Santander, violations of the type indicated in the Risk Report occurred within three months in 1 cases, and were presumably prevented in 2 cases. ► Of the 3 cases in which the information was communicated to authorities in the form of an Early Warning, violations occurred in 1 cases and were presumably prevented (to date, in the most recent warning) in 2 cases. ► Of the 3 cases in which the information was communicated to authorities not in the form of an Early Warning, violations occurred in 1 cases and were presumably prevented in 2 cases. 73 Appendix 6 The Early Warning System and International Humanitarian Law The Early Warning System is principally an instrument to protect human rights, ensure compliance with International Humanitarian Law and prevent massive and systematic violations against the civilian population as result of the internal armed conflict. The objectives of the EWS conform closely with the general objectives of the Human Rights Ombudsman office, an institution created in the 1991 Constitution, to defend basic human and constitutional rights of Colombian citizens and part of what is known a the Public Ministry in Colombia, a separate control branch of government to ensure basic human right, civil liberties and constitutional protections. Moreover, the idea of an Early Warning System fits into the basic commitment to human rights and IHL found in Colombia’s international treaty commitments to protect the civilian population from the affects of the armed conflict. Since 1991, these international obligations have been systematically integrated into Colombia’s national laws. The principal human rights and IHL guarantees are: the obligation to guarantee the rights all persons27, the primacy of the inalienable rights of the person, 28 the respect of human dignity, 29 the principal of pro homine (in favor of the rights of man in Colombia’s international treaty commitments)30 and especially the principal of distinction between combatants and non￾combatants. The latter, the so-called principal of distinction, is the foundation of all International Humanitarian Law. It is recognized in articles 13, 48, 50.1, 50.2 y 51 of Protocol I that was added to the Geneva Convention of 1949.31 It requires that combatants distinguish themselves from the civilian population and that they protect civilians from the consequences of military operations. Protocol I refers to international armed conflicts. However, Protocol II expressly states in its preamble that the terms of the Geneva Conventions apply to internal armed conflicts as 27 Art. 2 of the National Constitution; Art. 1 of the American Convention on Human Rights; and Art. 2 of the International Agreement on Civil and Political Rights. 28 Art. 5 of the National Constitution. 29 Art. 1. of the National Constitution. 30 Art. 5 International Agreement on Civil and Political Rights and Art. 29 American Convention on Human Rights. 31 Protocol I was ratified in Colombia through Law 11 of 1992 and became Colombian law on March 1, 1994. The four Geneva Conventions were ratified through Law 5 of 1960 and became law on May 8, 1962. In Colombia, International Humanitarian Law has been incorporated into the Constitution through Articles 93 and 214.2 . The latter clearly states that even during states of exception “in all cases the rules of international humanitarian law shall be respected.” 74 well.32 Protocol II was ratified by Colombia in 1994 (Law 171) and has been in effect since February 15, 1996. Additionally, Article 4 of Protocol II states that non-combatants, whether or not they are denied liberty, have the right to be treated with humanity and that their honor, personal integrity, beliefs and religious practices must be respected. This last norm establishes objective criteria for the application of the principal of distinction: “ as such the parts in conflict cannot arbitrarily define who is and is not a combatant, and therefore who can cannot be a legitimate military target.33 From this principle is derived the state’s obligation to protect non-combatants, that is “all persons who do not participate directly in the hostilities or who have stopped participating in them ( Art. 4, Protocol II) and if there is a doubt about the status of the person, they will be considered to be civilian (Art. 50 Protocol I). The Early Warning System has been most effective when it has been guided primarily by the normative standards of IHL. It has been least effective when it has strayed from or confused this mission with other objectives. From the earliest discussions among the Human Rights Ombudsman’s office, the United Nations HCHR, and the Human Rights Division of the Ministry of Interior, it was understood that the emerging national and international standards of Human Rights and IHL were to be the guiding force behind the implementation of an effective Early Warning System. Yet as the system has evolved, its origins based on the application of IHL have become less clear. The result is a system that has often registered impending threats but has not developed the appropriate response mechanisms to protect some of the most vulnerable populations from the consequences of armed conflict. 32 The concluding section of the Preamble to Protocol II to the Geneva Convention of 1949 state: “Recognizing that, in cases that are not foreseen under existing law, human beings remain protected under the principals of humanity and by the demands placed on the public conscience.” Protocol II reaffirms the Martens clause of the Convention on the Laws and Customs of Ground War signed in the Hague in 1907: “Populations and belligerents remain under the protection and the authority of the rights of man, in the manner that these have been established between civilized nations, by the laws of humanity and by the demands of the public conscience.” 33 Constitutional Court, sentence C-225/05 that reviewed and declared constitutional the law that incorporated Protocol II into Colombian law. 75 Appendix 7: List of Persons Interviewed Katy Boulay Analyst for Internally Displaced, Colombia Commission of Jurors Jorge Calero Operational Coordinator – Early Warning System, Onbudsmen Office Major Hugo Casas, Chief, Strategic Area – National Police Andrés Celis Director for Protection, United Nations High Commissioner for Refugees - ACNUR Tomas Concha Vice- Director of the Human Rights and ILH Program – Vice-presidency of Colombia Sandra Devia Ruiz CIAT Coordinator – Director of Territorial Affairs and Public Order - Ministry of the Interior and Justice Colonel Carlos Castillo Representative from the Ministry of Defense -CIAT Andrés Dávila Director of Justice and Security – National Department of Planning Juan Estrada Counsultant for the Subdirectory of Projects – ACCI Fredy Fajardo Rozo Administrative Representative of the DAS - CIAT Maria Flynn Director of Humanitarian Affairs, United Nations Office for the Coordination of Humanitarian Affiars, OCHA Carlos Franco Director of the Human Rights and IHL Program - Vice-presidency of Colombia Beatriz García 76 Colombia Coordinator – International Peace Brigades Fernán Gonzales Researcher - Cinep Daniel Medina Delegate Ombudsman, National Director of the EWS – Defensoria del Pueblo Darío Mejia Secretary General of the Defensoria del Pueblo – Defensoria del Pueblo Juan Carlos Mongue Coordinator, Observation Division, United Nations High Commissioner for Human Rights Office in Colombia, UNHCHR Jaime Prieto Consultant Luz Maria Salazar Consultant of the Subdirectory of Projects – ACCI Luz Amparo Salinas Abdala, Coordinator, Humanitarian Situation Room United Nations Office for the Coordination of Humanitarian Affairs – OCHA Luis Carlos Toledo Delegate for Prevention and Human Rights – Procuraduria Martha Leal, Representative of DAS to CIAT Patricia Linares Procuradora Delegate for the prevention in human rights and ethnic affairs – Procuraduria General de la Nación Raul Rosende Resident Representative Counselor - UNDP Sophie Haspeslagh Junior Analyst, Andean Region, International Crisis Group Marcus Schultze-Kraft Project Director, Andean Region, International Crisis Group Volmar Perez Defensor del Pueblo Yolanda Sarmiento 77 Sub-director of International Affairs – Fiscalia General de la Nacion Captain Ivan Macias Direction of Army Intelligence -CIAT Augusto Moscarella Ministry of Interior – Human Rights Dumar Otalora Delegate Disciplinary Counselor - Procuraduria Eduardo Junguito Advisor – Ministry of Interior and Justice Elva Beatriz Silva Fiscal Chief – Fiscalia General de la Nacion Teofilo Vasquez Researcher - Cinep Fernando Britto Procurador, Delegate for the National Police – Procuraduria General de la Nacion Francisco Santos Vice-President of Colombia Thomas Johnson, Democracy and Governance Officer, USAID-Colombia Gloria Elsa Ramirez Delegated Defensor for Collective Rights and the environment, Defensoría del Pueblo Ana Linder Coordinator of the Human Rights and IHL Program – Vice-presidency Natalie Suescun Assistant Human Rights and IHL Program – Vice-presidency Patricia Lora Communications – Human Rights and IHL Program – Vice-presidency of Colombia Germán Sanchez Human Rights Unit – Ministry of the Interior Cristina Munoz 78 Consultant CIAT Carlos Zuniga Geo-reference - EWS Martha Judo Varegas Communications - EWS Olivar Toscano Statistics and Indicators - EWS Carolina Rudas National Analyst Regions 4 & 5 – EWS, Ombudsman Office Renan Cuesta National Analyst Region 3 – EWS, Ombudsman Office Nestor Homero Cotrina Secretary of Departmental Government Harvey Suarez Morales Director, Codhes Lucia Garcia Giraldo Administrative and Finance Coordinator, Human Rights Program - USAID/MSD ARAUCA General Gustavo Matamoros, 18th Brigade, Arauca Sally Andrea Mahecha Advisor for Peace and Human Rights department of Arauca Jesus Angarito Secretary of Sindical Affairs, USO Arauca Dolka Arias Beltran Coordinator of the Territorial Unit – Red de Solidaridad Social Arauca Rodrigo Palacio Lieutenant Colonel, Police Dept. of Arauca Juilo Enrique Acosta Governor of Arauca Gloria Cultiva 79 Ombudsman for Arauca Rudht Patricia Santos Manosalva Coordinator of Human Rights – Police Department Arauca Jose Maria Bolivar General Vicar & Director of the Diocese Commission for peace of Arauca Sargeant Juan Carlos Gonzales Rivera Human Rights Official – 18th Brigade Arauca Maria Eugenia Pinto Borrego Jefe de Gobierno DNP Martín Sandoval Roso Executive Secretary – Permanent Committee for the Defense of Human rights – Arauca CHOCO Ventura Diaz Ceballos Counselor of Human Rights Lieutenant Colonel Edgar Bejarano Operation Commandant Police Dept. of Choco Luis Augusto Morales Devia Coordinator of Intelligence – DAS, Choco Hector Enrique Galvis Ruiz Sub director – DAS, Choco Fidel Leon Cadavid Bishop of Quibdo Father Franklin Renteria Valencia Dioceses of Quibdo Commission for Life, Justice and Peace Father Mario de Jesus Herrera Dioceses of Quibdo Commission for Life, Justice and Peace Luz Marina Prada, Sister Dioceses of Quibdo Commission for Life, Justice and Peace Luis Moreno Lemos Coordinador of Red de Solidaridad Social, Choco 80 Leyner Palacios Asprilla Treasurer Cocomacia - ACIA Santiago Palacios Asprilla Legal Representative - ACIA Rosendo Blandon Cordoba Treasurer Cocomacia - ACIA Jose Reyes Vice-president -ACIA Anin Mena Cordoba Secretary - ACIA William Rivas Torres Secretary - ACIA Fredy N. Palacios Secretary of the Interior, Choco Wilson D. Palacios Director of Citizen Participation, Secretary of the Interior, Choco. Alberto Achito Lubiasa President OREWA Luis Enrique Murillo Robledo Community Defender for the Medio Atrato Region - EWS Neiffy Porras Mena Regional Analyst - EWS Lieutenant Colonel Luis Fernando Velez Battalion Manosalva, Quibdo, Choco NORTE DE SANTANDER Carmen Ligia Galvis Regional Ombudsman for Norte de SantanderDefensora Luis Pérez Regional Analyst – EWS, Defensoría del Pueblo 81 Jorge Enrique Pinzòn Dueñas, Acting Mayor - Cucuta Major Juan Sandoval Commander - Maza Battalion Vitelimo Galvix Mogollon Director of DAS, Norte de Santander Carlos Coronel Director of Red de Solidaridad Social, Norte de Santander Name - Police Chief Department of Norte de Santander Martha Maldonado Representative - Juntas de Acción Comunal Jairo Sabavana Representative - Etnia Motilón Jimy Oviedo Barí y Uwa Community Coronel Marco Antonio Pedreros Commander – Department of Police of Norte de Santander Vitelmo Galvis Sectional Director - DAS Carlos Tibaduiza Sectional Director of intelligence - DAS Narda Rodrìguez Rozo, Director – Office of Human Rights of the Battalion Grupo Maza Jairo Oviedo Director - PARCOMÙN-Norte de Santander 82 Appendix 8: List of Documents Consulted Agencia Colombiana de Cooperación Internacional: “International Cooperation Strategy,” Presidencia de la Republica de Colombia, 2002. Amnesty International, “Colombia: Amnesty International is always with the victims of human rights violations and abuses, whoever the perpetrators are,” Press Release, 2004. Amnesty International, “Colombia A Laboratory of War: Repression and Violence in Arauca,” April 19, 2004. Asociación Campesina Integral del Atrato (ACIA): “La defensa del territorio: una historia de vida” Autoridades Tradicionales Wayuu de Portete. “Informe de los Alaulayuu de Portete victimas de la masacre del 18 de abril del 2004 sobre los hechos,” 2004. CIAT Actas # 1 – 60, CIAT, “Actas de reunión” para los departamentos estudiados (Choco, N. de Santander, Arauca y Ciudad Bolivar (Bta), 2002-2004. CIAT, “Monitoreo de Acciones de prevención implementadas por las autoridades” para los departamentos de Choco, Arauca y Norte de Santander. CIAT, “Verificación de Informes de Riesgo #15-04, 2004 CINEP & Justicia y Paz, Noche y Niebla #28: Julio-Diciembre 2003, Bogota: 2004 CINEP, Cien días #54 volumen 11 Bogota: Mayo 2004. Comisión de Verificación, “ Informe de la comisión de verificación de la masacre de Tame, Arauca,” 2004. Defensoria del Pueblo, “ Formatos y Fichas – SAT,” 2004 unpublished Defensoria del Pueblo, “ SAT – Documentos Metodológicos”, 2004 unpublished Defensoria del Pueblo, “ SAT – Macro proceso de Atención”, 2004 unpublished Defensoria del Pueblo, “ dialéctica del riesgo en Colombia: Perdicción y análisis “, 2004 unpublished 83 Defensoria del Pueblo, “ Entronque del sistema de alertas tempranas en la Defensoria del Pueblo,” Febrero, 2002, unpublished. Defensoria del Pueblo, “ Información Estadística del SAT: Enero 1 – Diciembre 31, 2003 “, 2004 unpublished Defensoria del Pueblo, “ Información estadística del sistema de alertas tempranas – SAT Proyección Social – Enero 1- diciembre 31, 2003,” 2004, unpublished. Defensoria del Pueblo, “ Informe Estadístico de alertas tempranas – 1/1/03 – 4/31/04,” 2004, unpublished. Defensoria del Pueblo, “ Informe Tri-mestral,” 2004, unpublished. Defensoria del Pueblo, “ Lineamiento para la elaboración de los informes estructurales “, May 2004, unpublished. Defensoria del Pueblo, “ Matriz y proyecto social,” 2004, unpublished. Defensoria del Pueblo, “ Presentación del CIAT,” 2004, unpublished. Defensoria del Pueblo, “ Presentación del SAT,” 2004, unpublished. Defensoria del Pueblo, “ SAT - IR Remitidos por Actor fuente del amenaza “, 2004 unpublished Defensoria del Pueblo, “ SAT- Proceso y Procedimientos,” 2004, unpublished Defensoria del Pueblo, “Fases de implantación regional- SAT“, 2004 unpublished. Defensoria del Pueblo, “Guía metodologica para la elaboración de informes analíticos subregionales de riesgo y vulnerabilidad“, 2004 unpublished. Defensoria del Pueblo, “Mapa de Informes de Riesgo emitidos Jan 1 – Dec 21, 2003,” 2003, unpublished. Defensoria del Pueblo, “Mapa: Informes de riesgo emitidos por municipio según mes de emisión hasta abril, 2004,” 2004, unpublished. Defensoria del Pueblo, “Mapa: Informes de riesgo emitidos por municipio según actor de la amenaza,” 2004, unpublished. Defensoria del Pueblo, “Mapa: Informes de Riesgo según municipio, según nivel de riesgo,” 2004, unpublished. Defensoria del Pueblo, “Mapa: Tercera Fase SAT – Junio 2003,” 2003, unpublished. 84 Defensoria del Pueblo, “Operative Plan - SAT ,” 2004, unpublished. Defensoria del Pueblo, “política Defensoral para la prevención de las violaciones masivas de los derechos humanos: Proyecto SAT,” January, 2002, unpublished. Defensoria del Pueblo, “Presentación del SAT” 2004 unpublished Defensoria del Pueblo, “SAT – Casos Relevantes,” 2004, unpublished. Defensoria del Pueblo: Colombia , Mapas del movimiento de los grupos armados en Ciudad Bolívar. Defensoria del Pueblo: Colombia, “Choco 2004 Comunicación SAT: Informes de Prensa” Defensoria del Pueblo: Colombia, “Defensoria delegada para la evaluación de riesgo de la población civil como consecuencia del conflicto armado: SAT, Arauca” July 7, 2004. Defensoria del Pueblo: Colombia, “Delegada para la evaluación de riesgo de población civil como consecuencia del conflicto armado”, 2003 Defensoria del Pueblo: Colombia, “Informe Estadístico del sistema de alertas tempranas – SAT, Enero 2003 – abril 30, 2004. Defensoria del Pueblo: Colombia, “Política Defensorial para la Prevención de las violaciones masivas de los derechos humanos”, 2003 Gobernación de Arauca, “Despachos del Gobernador” Gobernadores y comandantes de policía, “Manejo de la convivencia y la seguridad ciudadana: 2004,” 2004 International Crisis Group, “ Colombia’s Humanitarian Crisis” July 9, 2003. International Crisis Group, “Colombia: Negotiating with the Paramilitaries” September 16, 2003. International Crisis Group, “Hostages for Prisoners: A way to peace in Colombia?” March 8, 2004. International Crisis Group, “President Uribe’s Democratic Security Policy” November, 2003. Lind, Michael. Preventing Violent Conflicts: A strategy for preventive diplomacy, U.S. Institute of Peace Studies, Management Sciences for Development, INC, “USAID/Colombia Controller’s Office Bogota, Colombia” 2004. 85 Ministerio de Defensa Nacional, “Política de defensa y seguridad democrática,” Presidencia de la Republica de Colombia, 2003. Ministry of Interior and Justice of the Republic of Colombia, “Informe Annual de Derechos Humanos – CIAT” Naciones Unidas OCACNUDH “Recomendaciones para Colombia: 2003” March 23, Bogota. Naciones Unidas OCACNUDH “Recomendaciones para Colombia: 2004” March 23, Bogota. Observatorio Del programa Presidencial de DDHH y DIH, Vicepresidencia de la Republica “Algunos indicadores sobre la situación de los derechos humanos en la Sierra Nevada de Santa Marta”, 2004 Observatorio de los Derechos humanos en Colombia: Indicadores sobre la situación de Derechos Humanos y DIH en Colombia” May 2004. Observatorio del programa presidencial de derechos humanos y derecho internacional, “Panorama Actual del Norte de Santander” Vicepresidencia de la Republica de Colombia, 2002 Observatorio del programa presidencial de derechos humanos y derecho internacional, “Panorama Actual del Choco” Vicepresidencia de la Republica de Colombia, 2003 Oficina en Colombia del Alto Comisionado de las Naciones Unidas para los derechos humanos, “Recomendaciones para Colombia, 2004” Oficina en Colombia del Alto Comisionado de las Naciones Unidas para los derechos humanos, “Recomendaciones para Colombia, 2003” Peace Brigades International, “Presentación, Mandato y Objetivos: PBI Colombia,” 2004. Policía Nacional Dirección Operativa, “Pautas para el Manejo de alertas tempranas,” January, 2004. Procuraduría General de la Nación, Estados de Excepción, Bogota: Servigraph, 2004. Republica de Colombia, Informe annual de derechos humanos y DIH, Bogota: ABC Press, 2003 SAT, “Early Warning System: Alerts issued (Excell file)” July 9, 2004. SAT, “Mapa de alertas tempranas” updated, 2004. SAT-CIAT, “Sistema General de Alertas Tempranas: Programa de Derechos Humanos de USAID”, Junio 2004. 86 Secretaria de Gobierno de Bogota, “Balance por localidades Comportamiento del homicidio común: Primer semestre 2003-2004”, 2004 Sistema de Alertas Tempranas, :Encuentro de Socialización con autoridades regionales de Arauca, alcaldes, y personeros” July, 2004. Sistema de Alertas Tempranas, “Caso Tipo: Valledupar” 2004 Sistema de Alertas Tempranas, “Departamento del Choco: Informes sobre el departamento del Choco, Informes sobre caso tipo: Bojaya, Reportes de Prensa”, 2004 Sistema de Alertas Tempranas, “Información estadística del sistema de alertas tempranas – SAT Proyección social”, 2003 United Nations Comisión de Derechos Humanos “ Organization of the work of the session” report of the UNHC for human rights on the human rights situation in Colombia”, 2004.