Report: The Rise and Fall of 'False Positive' Killings in Colombia: The Role of U.S. Military Assistance, 2000-2010

Tambien en español: Falsos Positivos” en Colombia y el papel de la asistencia militar de Estados Unidos, 2000-2010

Mapa Interactivo del registro de ejecuciones y unidades responsables

No More False Positives in Colombia!

Take action: Contact the U.S. State Department

Read our recommendations
for the United States

Download the full report
(PDF, 1.9 MB)

View all maps from the report

Donate to support FOR’s
work in Colombia

Leer el informe en español (PDF7MB)

In this report, we document and analyze the impact of the billions of dollars that the United States spends each year to strengthen the military and police forces of other countries. The study focuses on Colombia, which U.S. officials cite as a model they are proud of and exporting to Mexico and Central America. Critics say it’s an example of U.S. aiding gross abuses and war.

Neither has compiled comprehensive evidence. So how would we know? This study takes on this question, relevant to U.S. policy from Afghanistan and Israel to Egypt and Honduras.

You can read the executive summary below, browse our recommendations, or download the full report (PDF, 1.9 MB).

In the report, we identified Colombian Army officers with the largest number of extrajudicial executions under their brigade command, and found that four of six with the most such killings under their command were subsequently promoted to command the Army. This includes current Army commander General Jaime Lasprilla Villamizar, who oversaw at least 75 killings of civilians when he led the brigade in southern Huila Department in 2006–2007.

Instead of being held accountable, officers like Lasprilla have received extensive U.S. assistance and risen in rank. Yet the Leahy Law requires Washington to suspend military assistance to foreign units whose commanders have committed serious abuses, unless those responsible are brought to justice. Even if he was not personally involved, General Lasprilla oversaw a pattern of more than six dozen civilian killings that is a gross human rights violation.

Please take action by calling on the United States to press for justice for the crimes committed under General Lasprilla’s command, and until then to comply with the Leahy Law by suspending U.S. assistance to the Army that General Lasprilla commands. Send a message to the U.S. State Department today!

Executive Summary

This study advances a methodology for assessing the human rights impact of U.S. military assistance, by looking in depth at a discrete set of human rights violations carried out by the Colombian armed forces. It seeks to understand what role, if any, U.S. security assistance from 2000 to 2010 played in increasing or decreasing the commission of “false positive” killings — extrajudicial executions by the Colombian Army of civilians claimed to be have been killed in combat.

1848 The United States has trained, armed and equipped the Colombian armed forces at a cost of $6.8 billion since 2000. Supporters cite aid to Colombia as proof that U.S. security assistance promotes human rights improvements. Critics look at the same case and assert that U.S. assistance tacitly encouraged or empowered the Colombian forces to carry out grave human rights abuses. Neither side presents systemic evidence.

The Colombian case is important not only because of the amount invested and the divergent claims about outcomes, but because of the renewed emphasis in U.S. military strategy on “building partner capacity,” reflected in $25 billion in foreign security assistance in 2012. Evaluating the human rights outcomes of this strategy and spending in Colombia is thus critically important for populations throughout the world.

Our findings for Colombia call into question much of the assumed wisdom about military assistance programs. We offer detail about the Colombia experience to model methods of evaluation, show what data is needed, and to demonstrate the importance of post-assistance assessments that may surprise participants, policy-makers, and critics. We found that increased information collection and analysis that would facilitate such assessments is possible.

Based on data on 5,763 reported executions in Colombia and extensive documentation of U.S. assistance to the Colombian military, we found a positive correlation between the units and officers that received U.S. assistance and training, and the commission of extrajudicial killings. A statistical analysis of 1,821 of these executions where responsible units were directly identified showed that Army brigades that received a moderate as compared to low level of U.S. assistance correlated to ten more executions per brigade in the two years following assistance. While the analysis does not show that U.S. aid specifically caused or encouraged executions, it casts strong doubt on claims that U.S. assistance improved human rights performance.

1849 Army brigades that received high levels of U.S. aid did not show a statistically robust different number of executions, compared to those receiving little aid. Those receiving high levels of U.S. assistance, such as the Counter-Narcotics Brigade, may have been subject to closer scrutiny for their human rights conduct than units receiving significant but less intensive assistance. Statistical ambiguity may reflect the small sample of high-aid units, or the difficulty of identifying units responsible for executions, especially in areas where mobile units operate; much U.S. assistance focused on mobile brigades.

To understand how widespread the practice of “false positives” was, we analyzed the prevalence of the practice of extrajudicial killings across Colombian Army brigades. We found that in 2007, at least one execution was directly attributed to 99 of the Army’s 219 combat battalions and mobile brigades. We identified 117 brigade commanders and 184 battalion and mobile brigade commanders, whose units were directly identified as committing one or more extrajudicial killing between 2000 and 2010.

A question raised by this analysis is the role of military leadership and institutional culture within problematic army units. Three of the five brigade commanders with the largest number of executions reported by soldiers under their command were promoted to command the Army, including the current Army commander, General Jaime Lasprilla Villamizar. General Lasprilla is the active duty officer with the largest number of executions reportedly committed by soldiers under his brigade command, and he received substantially more U.S. training and assistance than his peers.

1850 The provision of intelligence capacity and products is rarely evaluated for human rights outcomes. We found an example of U.S. intelligence assistance to Colombia that, though on its own may not have violated human rights, supported units that had adopted a strategy conducive to extrajudicial killing.

Measurements of the impacts of military assistance should evaluate the performance of individual officers in whom much assistance is invested, and the disproportionate role that commanders have in the performance of troops under their command. These officers and commanders, after receiving U.S. military training, transfer to other units that may or may not receive U.S. assistance, but are impacted by the training provided to individual officers.

U.S. military training, especially at the Western Hemisphere Institute for Security Cooperation (WHINSEC), is practically a required step for the promotion of a Colombian Army officer. Of the 25 Colombian WHINSEC instructors and graduates of the Command and General Staff course from 2001 to 2003 for which any subsequent information was available, 12 of them — 48% — had either been charged with a serious crime or commanded units whose members had reportedly committed multiple extrajudicial killings. A random sample of 25 Colombian officers from approximately the same period showed a better human rights record than the WHINSEC cohort: we identified four (16%) who subsequently led units with multiple extrajudicial executions under their command. Another three officers led units with one execution under their command.

1851 Colombia may have a “systemic [human rights] problem across the country” that is reflected in the high percentage of WHINSEC instructors and graduates implicated in crimes, as one U.S. military officer suggested. If that is the case, it raises a question of how the disproportionate representation of Colombian instructors at WHINSEC, or U.S. funding of Colombian training of other nations’ security forces, contributes to increased respect for human rights.

The number of executions in Colombia fell dramatically after 2008. Evidence indicates that diverse pressures — from civil society, the office of United Nations High Commissioner for Human Rights, and civilians within the Colombian government who engaged the Army, were key to changing an institutional environment that sanctioned, if not encouraged, extrajudicial killings.

If reducing impunity is important to deterring serious abuses, we must consider the United States’ key role in two changes in the Colombian judicial system. The U.S. gave substantial material support for the Human Rights Unit of the Attorney General’s Office, which made significant judicial advances in prosecuting executions by 2013 (although less than 10% of reported executions had advanced to trial). But the U.S. also funded and promoted the implementation of the new accusatory justice system, which has had poor results in prosecuting those responsible for committing executions.

1852 While other branches of Colombian security forces also have committed abuses, this report focuses on assistance to and rights violations committed by the Army, which was a focus of U.S. assistance, and in which human rights concerns had been identified at the time assistance was expanded. The report also examines “false positives” and U.S. aid in greater detail in several regions of Colombia.

This evaluation is not possible without detailed data about human rights violations, U.S. assistance, and Colombian military structure and leadership, compiled by U.S. government agencies, human rights organizations, and private researchers. For more effective assessments, greater transparency about who receives assistance, and the amount and nature of what they receive, and tracking of the human rights performance of those recipients are needed.

Recommendations to U.S. policymakers regarding U.S. security assistance

TAKE ACTION: Help implement these recommendations by contacting the U.S. State Department today!

  1. 1853 Legislators should require that U.S. security assistance be subject to rigorous assessment in order to evaluate the outcomes of such spending. As part of this effort, Congress should provide the mandate and resources needed to fully and independently evaluate the human rights outcomes of security assistance.
  2. Inputs to help assess human rights outcomes must be solicited from outside the security assistance bureaucracy.
  3. Policymakers, training managers and critics should not accept at face value “assumed wisdom” about military assistance. Honest evaluation of military assistance can yield significant and surprising lessons on the impacts of such programs.
  4. The United States must monitor the assignments and human rights performance of officers, soldiers and units after they receive U.S. assistance.
  5. Special attention should be directed to outcomes of assistance to units with histories of gross human rights abuses.
  6. The conduct of recipients of training assistance — not their record of promotions — should be the measure of success in human rights outcomes.
  7. Human rights performance by recipients of all assistance (not just human rights training) must be a fundamental metric for evaluating the results of such assistance.
  8. Systematic information about the prosecution of government officials for the commission of human rights abuses should be gathered, and the rates of impunity should be considered an important human rights indicator.
  9. Intelligence assistance — both intelligence products as well as capacities — should be considered part of assistance that is evaluated for human rights outcomes.
  10. In order to benefit from input into evaluation from diverse stakeholders, information identifying the recipients of military assistance should be declassified, unless specific information demonstrates serious risks from disclosure.
false-positives-2014-colombia-report.pdf1.92 MB