The Art and Aggravation of Vetting in Post-Conflict Environments
On 22 February 2006, insurgents posing as Iraqi police officers destroyed the Golden Mosque in Samarra, one of Iraq's holiest Shi'ite shrines. The attack set off a spasm of sectarian violence that has metastasized into what some consider an intractable civil war. Since then, the insurgent tactic of infiltrating the security forces and corrupting its personnel has become almost commonplace, with catastrophic results for Iraq. The populace distrusts Iraqi security forces, coalition forces distrust their Iraqi counterparts, the Iraqi Government is viewed as increasingly illegitimate, and the country has plunged into further chaos, delaying the safe transfer of security responsibilities to Iraqi forces.
The undermining of the Iraqi police forces occurred, in part, because of negligible vetting-the investigation and selection of new recruits for the police force. Creating a professional indigenous security force is a mandatory component of any exit strategy in a costly post-conflict reconstruction mission. Yet creating such a force depends utterly on the competent vetting of candidates for that force. Failure to vet recruits to ensure they possess the "proper character" can result in the infiltration of criminals, insurgents, warlords, and other undesirables into the state's security apparatus, setting up the possibility of a coup d'etat or worse.1 This, in turn, may trigger a cycle of costly international interventions and endless peacekeeping operations. Thus, competent vetting of indigenous security forces is the linchpin of post-conflict reconstruction.
Unfortunately, no model for vetting exists, and recent efforts to establish a vetting process in Iraq and elsewhere have been ad hoc and disappointing. Nor has the situation been helped by the paucity of literature, either academic or practical, on vetting indigenous security forces: there is scant scholarship on the issue and no large-scale comparative study of vetting. That no international treaty addresses the subject reflects the relative novelty of the issue and the general lack of interest in formulating a common approach. Also, no U.S. Government, United Nations, or nongovernmental organization has written a manual on vetting, a remarkable fact given that security forces are currently being reconstituted in Iraq, Afghanistan, Liberia, Sierra Leone, the Democratic Republic of Congo, and elsewhere.
Lessons Learned After WWII
Since the end of World War II, the international community has learned many lessons about security force vetting and lustration (culling an existing security force for the best individuals while dismissing the others).
The first lesson is that the effects of vetting or lustration may be short-lived if the process is hurried or abandoned halfway through (i.e., recruit first, vet later). The largest post-WWII lustration effort occurred immediately following the war, as the Allies judged Axis leaders. In Europe, this was known as denazification, and it is estimated that 13 million Germans underwent it, 600,000 of whom were sanctioned. Separately, France purged collaborators of the Vichy regime and Italy dismissed approximately 2,000 government employees. Despite denazification, many former Nazis eventually made their way back into public service. Similarly, Italy reinstated all lustrated personnel in 1948.
A second lesson is that failure to respect the rights of individuals under review will delegitimize the process and open it up to external challenges. Following the fall of the Berlin Wall, former Communist countries passed lustration laws to drastically reduce the size of their governments, including the security sectors. In Hungary, 12,000 high-level officials were subject to lustration, although only a fraction were sanctioned for their participation in the previous regime. In Czechoslovakia, out of approximately 300,000 cases considered, 15,000 individuals were removed from office. Poland also lustrated citizens alleged to have collaborated with the secret police. All the post-Communism lustration laws of the 1990s were widely criticized for insufficiently taking into account the rights of those subjected to lustration.2
A third lesson is that a lack of political will, inadequate resources, or a poorly thought-out plan will result in failure-the task is that complex. The International Police Task Force in the Federation of Bosnia and Herzegovina was initially tasked with screening all candidates for the Federation's police forces and identifying anyone previously engaged in ethnic cleansing or other crimes against ethnic minorities. Its vetting was so ineffectual that the task was eventually transferred to the Human Rights Office in Bosnia.3
Lesson four is that process matters. A 2005 Government Accountability Office (GAO) report, Southeast Asia: Better Human Rights Reviews and Strategic Planning Needed for U.S. Assistance to Foreign Security Forces, examined U.S. security sector reform efforts to equip and train military and police forces in the Philippines, Thailand, and Indonesia from 2001 to 2004. It found "no evidence that U.S. officials vetted an estimated 6,900 foreign security trainees" as required by U.S. law.4 Worse, 32 Indonesians from a notorious special-forces police unit received training, even though the unit was prohibited by the U.S. Department of State (DOS) from receiving U.S. training funds because of human rights abuses. This undesirable outcome resulted from the lack of "clear policies and procedures for vetting foreign security forces."5 The GAO found little evidence of "ground-truthing," investigating, public records checking, consultation with victims' groups, or accounting for aliases or noms de guerre. No consolidated written policy existed to establish interagency vetting standards and procedures, sources and methods, roles and responsibilities, databases, or oversight mechanisms. Conducting Internet searches and scanning newspaper clippings-the usual expedients-is simply insufficient for complex, prolonged conflicts.
The fifth and last lesson learned is that the failure to vet recruits might help an insurgent organization penetrate state security forces. The joint Department of Defense (DOD) and DOS inspector general (IG) report on vetting for the new Iraqi Police Service (IPS) states that "recruitment and vetting procedures [were] faulty," resulting in incompetents, criminals, and insurgents joining the IPS, a problem not easily undone.6 The report also reveals that "the IG Team was told that, especially early in 2003, only a cursory background check, if even that, was conducted before policemen were trained or entered the force."7 The vetting process was stymied by a lack of public records and witnesses and by cross-cultural and language difficulties. Overall, according to the report, "the Coalition's ability to conduct thorough background checks on IPS personnel [was] severely limited."8 The report notes that inducting criminals into the IPS was a continual concern. Even more troubling was infiltration by terrorists or insurgents. Sufficient evidence was found to conclude that "such persons indeed are among the ranks of the IPS," which underscores "the need for the most rigorous possible review of each applicant's records."9
Why Vetting is Difficult
One reason why no coherent vetting policy exists is because failed and weak states are, by their nature, disordered and chaotic. Typical sources and methods used in background checks (criminal records, credit history, education records, employment history, and so on) do not exist, are not credible, or are insufficient. Even establishing identity can prove daunting, as attempts to hold legitimate elections in post-conflict states have demonstrated.
Another reason why no policy exists is the high prevalence of criminal behavior during conflicts, especially during prolonged civil wars. In such environments the number of problematic candidates will be correspondingly high. Those conducting the vetting process might find themselves rejecting most of the candidates. Lustration also might not work as a security sector reform technique because, given the high rate of crimes, remaining veterans might corrupt new recruits, thus compromising the new security force. Instead, it might be better to completely demobilize the security sector and reconstitute it.
However, reconstructing a state's security sector is dangerous. Instability and violence are never far beneath the surface in post-conflict environments, and the vetting process can easily cause dangerous ripples. In weak or failed states, a security force is often the strongest institution, and, in many cases, is or was a major contributing factor to the state's demise. Attempts at reform can result in violent reprisals against staff and supporters of reform, while investigations into war crimes might dredge up painful memories for a fragile population and possibly rekindle violence over unaddressed wrongs. The vetting process must remain absolutely unconnected to instruments of post-conflict justice such as so-called truth and reconciliation commissions. Often, security and justice are at odds in post-conflict settings.
Vetting is a highly sensitive process that invites a relapse of violence and state failure. If the vetting process fails to safeguard the identities of victims who help identify perpetrators, then those victims might be intimidated, coerced, or killed in reprisal. If the vetting process accidentally overlooks a war criminal, then all vetted individuals could be discredited and a violent backlash might occur. Additionally, wrongful denunciations of innocent individuals could generate antagonism in the community. The vetting procedure must understand these risks and remain sensitive to how the process might affect a frail society. Failure to do so could result in tensions within the new security force, a lack of public confidence in the force, and the emplacement of a force more likely to reproduce patterns of abuse.
A post-conflict environment is one of the most difficult operating environments in the world. It is almost uniformly characterized by extreme poverty and lack of infrastructure, law, and security. Simply moving cross-country can become a daunting expedition requiring robust security convoys, careful route reconnaissance, resupply points, spare vehicles, air medical evacuation support, river-crossing capabilities, a disciplined staff, and significant contingency planning. Other factors that could affect operations include institutionalized corruption, exotic diseases, prevalence of traffic accidents, a lack of logistical resupply, wild animals, and high rates of crime. One cannot assume the availability of amenities such as potable water, electricity, and shelter. A vetting staff must be prepared for a possible lack of cooperation from authorities, the novelty of the procedure for the population, the absence of precedents, and cultural misunderstandings that could prove disastrous. Consequently, the vetting program must be designed around these limitations, all of which influence morale, operations, budgeting, scheduling, and the quality of vetting.
A Vetting Model
The purpose of vetting personnel for an indigenous security force is to select individuals who will respect the rule of law and human rights norms. Vetting is often part of a larger security sector reform program to create a new force subservient to the state, not vice versa. To achieve this, the vetting staff's primary goal should be to ensure that no person of improper character is accepted into the new force. This is the raison d'etre of vetting, and it overrides all other priorities, such as an applicant's relevant experience or technical skills.
Before designing a vetting plan, practitioners must develop an end-state vision for the new force through consultations with stakeholders. Typically, the security force will be an all-volunteer force with a balanced mix of ethnicity, religion, gender, and other political categories. The goal of the recruiting, vetting, and training components of security sector reform is to achieve a force that maintains a professional ethos, respects the rule of law, cultivates public service leadership, is apolitical, and accepts civilian control with transparent oversight mechanisms. The force must be postured so that it is strong enough to defend the integrity of the nation's borders but not so strong that it threatens neighbors with its force-projection capability. Its structure, equipment, and training must be appropriate to the force's mission (for example, Liberia does not require F-16 fighter jets). Perhaps most critically, the new security force must not be so large that the government cannot pay its salaries. Such a condition is a precipitant to civil war.
In line with the end-state vision, the vetting process is not about establishing guilt or innocence, but about determining suitability for acceptance into the new security force. A vetting model must be founded on two fundamental but divergent considerations: normative issues and pragmatic concerns. The normative component concerns what to vet. In other words, what behavior, criminal or otherwise, justifies rejecting a candidate from the indigenous force? The pragmatic component examines how to vet. That is, what are the actual vetting procedures, how is a candidate's application examined, and what principles are applicable to that examination?
The grounds for disqualification are fundamentally different for each component. The normative component rejects a candidate based on credible evidence of wrongful conduct unrelated to the vetting process, such as prior crimes. For the pragmatic component, a candidate is rejected based on credible evidence of wrongful conduct related to the vetting process, such as cheating, lying, or refusing to cooperate during the vetting procedure.
Normative Component: What to Vet
What behaviors or crimes justify rejection from service in post-conflict state security forces? How do we derive these rejection standards? How do we legitimize these standards to the myriad domestic and international stakeholders?
Because each post-conflict environment is unique, we cannot decree a universal set of vetting principles. However, it is possible to develop a set of core crimes to serve as the foundation for vetting. Core crimes are wrongful acts such as genocide that justify exclusion from state security forces in most situations. Individual security sector reform programs can build on this set of core crimes to develop a tailored set of behavioral standards appropriate for each post-conflict situation. Several sources of international law exist that can inform the compilation of a set of core crimes, including international criminal law (ICL), international humanitarian law (IHL), and international human rights law (IHRL).
International criminal law. ICL is an imperfect instrument for vetting because it often requires proof of intent, which is difficult to demonstrate. As defined in the 1948 Geneva Convention on the Prevention and Repression of Genocide, the international crime of genocide requires proof that the crime has been pursued "with intent to destroy, in whole or in part, a national, ethnical, racial, or religious group."10 Such intent would be difficult to prove. The act of murder, however, is easier to prove and equally effective as a rejection criterion. In fact, any of the underlying acts enunciated by the 2002 Rome Statute of the International Criminal Court definition of crimes against humanity are sufficient for rejection from most security forces.11
International humanitarian law. IHL, also known as the law of war, provides a useful framework for vetting. Although IHL is designed to exculpate individuals from acts during war (such as killing) that would be considered crimes in civil society, this distinction is often complicated by the intrastate nature of most of today's conflicts. However, any grave breach of the Geneva Conventions or violation of the customs of war would warrant rejection from most security forces, especially if the country were party to the Conventions. Of particular relevance within the Conventions is the treatment of noncombatants.12
International human rights law. IHRL can buttress IHL, but it is too nebulous to use in developing a set of core crimes. For example, are international human rights directives or aspirations? Which human rights violations clearly justify rejection of an applicant? Some rights are too vague for a candidate's disqualification, such as violating the right "to a social and international order in which the rights and freedoms . . . can be fully realized."13 Because most current post-conflict settings are recovering from intrastate wars characterized by widespread human rights abuses over many years (Sudan, Liberia, Iraq, Somalia, Haiti, El Salvador, Cambodia, Sri Lanka, and so on), it is unrealistic to judge post-conflict populations by the same standards of reasonably functioning rule-of-law states during the same period. What would be considered a serious violation of human rights in the developed world might be overshadowed by more egregious violations committed by others during an armed conflict.
Given the ambiguities of human rights and the nature of intrastate warfare, determining what constitutes core crimes comes down to distinctions between violations that result in immediate disqualification and those that do not. Many international human rights instruments make such a distinction by differentiating between derogable and nonderogable rights. Derogable rights are rights that may be suspended by states under limited circumstances, as specified in article 4 of the International Covenant on Civil and Political Rights. Nonderogable rights cannot be restricted even in cases of public emergency. They include-
- The right to life (article 6).
- The right not to endure torture and cruel or inhuman punishment (article 7).
- The right not to endure slavery and involuntary servitude [article 8(1) and (2)].
- The right not to endure imprisonment for breach of contract (article 11).
- The right not to endure retrospective criminal legislation (article 15).
- The right to be recognized before the law (article 16).
- The right to freedom of thought, conscience, and religion (article 18).14
By analyzing international law, it is possible to derive a set of core crimes to use as a basis for most security sector reform vetting programs. These crimes constitute per se disqualifications for service in the security sector (see Table 1). In order of gravity, core crimes are unlawful killing, unlawful wounding, torturing, outrages on personal dignity, rape, and abduction or arbitrary detention.
For a vetting program to operate successfully in a post-conflict environment, the program must be perceived as legitimate. It is critical for stakeholders to find common ground on fundamental questions regarding the definitions of core crimes, the proper character for police or military candidates, and the standard of evidence necessary to reject a candidate. Although core crimes are based on international law, a country's own domestic law should be built into the edifice. Domestic penal law will generally include a number of offenses that find equivalency in core crimes, such as criminal homicide; assaults, endangering behavior, and threats; sexual offenses; and kidnapping and related offences. Combining international core crimes and domestic law will help secure legitimacy and local cooperation.
Pragmatic Component: How to Vet
Vetting candidates in post-conflict environments is extremely challenging. How should candidates be selected in a country where many individuals are victims, perpetrators, or both of human rights abuses? How can the vetting staff conduct background checks in a country that has no credible public records, a dubious justice system, weak institutions, institutionalized corruption, and a distrustful public?
Currently, there is no widely accepted methodology for recruiting indigenous security forces in a post-conflict setting. Typically, vetting is embedded within the recruiting program of a larger security sector reform effort. Most recruitment efforts include-
- Conducting a nationwide public information campaign.
- Taking applications from candidates at a recruitment center.
- Giving candidates a physical-fitness test, a functional literacy and/or aptitude test, and a medi•cal exam.
- Conducting the vetting process.
- Making a final review and judgment of a candidate's suitability for service.
- Informing rejected candidates of their limited right to review why they were rejected.
- Informing accepted candidates that they must serve the first year on a probationary status (allowing additional time for vetting, if necessary).
Recruitment begins with a national public information campaign, which serves two functions: to sensitize the populace to the reconstitution of the security force and to attract volunteers. The sensitization process helps in explaining why a new security force is needed and what its mission and principles will be (such as respecting the rule of law, human rights, and so on). Volunteers can be attracted by describing service benefits and eligibility standards and by informing interested persons how and where to apply. Planning and implementing such a campaign might be an onerous task given a lack of infrastructure, low literacy rate, diverse ethnic languages, conflict history, and general mistrust, especially the mistrust of security forces that is prevalent in post-conflict societies.
If the public information campaign is ably conducted, recruitment centers can expect many applicants. The prospect of a stable, honorable income in a poor country with high unemployment will appeal to many, and this should generate a sizable candidate pool for the security sector reform program. However, the legacy of violence will also mean that many candidates of dubious character will seek to join the new military or police force, owing to the historic relationship between power and force. As a result, the vetting process should not count on self-selection; rather, it must rely wholly on the rigors of the vetting procedure to uncover unqualified individuals.
Once candidates arrive at a recruitment center, they are systematically evaluated as efficiently as possible. Failure to pass the physical test, functional literacy test, or medical exam should result in an immediate exclusion from service without appeal.15 The vetting staff should administer tests in the order of least resource-intensive to most resource-intensive, because it is cheaper and faster to evaluate candidates' physical fitness than their literacy. By combining immediate exclusions and prioritization of resource-efficient tests, recruiters can rapidly weed out unqualified candidates. This is critical because vetting is the most resource-intensive portion of the recruitment process, and the candidate pool must be culled as much as possible early on to allow a more manageable caseload for the vetting team and to ensure higher quality vetting.
Vetting in post-conflict environments involves background checks, records checks, and publication vetting. Actual vetting begins when the investigating team (one international and one local investigator) interviews the applicant. The team should ask each candidate a standard set of comprehensive questions in order to obtain and confirm basic information regarding the identity and background information the candidate provided on the application. Following the interview, the team should conduct a background investigation to establish the overall truthfulness of the applicant's claims and to uncover any credible evidence of wrongdoing. The background check should cover such essentials as age, citizenship, schooling, work history, claimed special skills, and any documents the applicant submitted. The investigating team should also interview people who know the candidate well: people who provided the candidate's references as well as neighbors, employers, co-workers, relatives, municipal authorities, teachers, community leaders, and local religious leaders. In many cases, it will be helpful to have an applicant draw a map to his or her home and community, since street names and numbers can be rare in post-conflict settings.
Simultaneously, the vetting team runs a public-records check on the applicant. Although weak states often have few credible records, a public-records audit can prove useful to determine document fraud, criminal activity, and allegations of human rights abuses. Sources of information include old government records, domestic and international nongovernmental organizations, the United Nations, and other government sources such as U.S. watch lists. The records audit should produce a weighted index of record veracity, breadth, depth, and applicability.
Publication vetting is a direct appeal to the population to solicit local knowledge of a candidate's past wrongdoing. Safe, anonymous channels must be established so that victims can give information without suffering reprisal. In this form of vetting, the candidate's picture, name(s), physical description, place of birth, and unique recruiting identity number are publicized nationally to afford witnesses and victims an opportunity to identify undesirable candidates. Mediums for publication vetting include radio and television stations, especially those with nationwide coverage; national and regional newspaper inserts; and posters and face-books positioned at transportation hubs, commercial districts, victim centers, refugee and internally displaced people camps, and major community centers such as churches, schools, and sports stadiums. The team might also ask select members of the public who possess special knowledge of past crimes, such as solicitors, academic researchers, civil society groups, and journalists, to submit relevant information concerning the human rights records of candidates. Because publication vetting invites false accusations, the vetting staff must allocate extra time for investigating complaints. While publication vetting is resource-intensive, the cost of not engaging the public in vetting is greater, given the limitations of background checks and public records in failed states and post-conflict environments.
After gathering relevant information and records, the vetting staff must make a final judgment about the candidate. The staff should designate a joint review board or similar entity to act as the selection approving authority. The board should include major stakeholders in order to foster local ownership, imbue the process with legitimacy, and help insulate the vetting staff from culpability should a candidate be or become an insurgent.
As aforementioned, candidates can be disqualified on either normative or pragmatic grounds. The latter refers to credible evidence of wrongdoing during the recruitment and vetting process (lying, cheating, noncooperation, or other behaviors not desirable in a security force). Normative grounds-credible evidence of wrongdoing unrelated to the vetting process-include but are not limited to-
- Credible allegations of commission of one or more core crimes.
- Discovery of a criminal background or association with or direct involvement with persons engaged in criminal activity.
- Association with any party or persons wanting to do harm to or interfere with reconstruction programs.
- Involvement in financial crimes, acts of corruption, or the accumulation of significant illegal wealth, property, or possessions as a result of intimidation, corruption, the taking of bribes, smuggling in violation of international sanctions, or other illegal acts.
- Mental instability that could be a threat to the safety and security of soldiers and civilians.
- Use of illegal narcotics or other illegal drugs.
If there are allegations against the recruit, the staff must assess the gravity of any crime and the credibility of the evidence. As is the case with determining what constitutes a human rights abuse, it is inappropriate and unpractical to apply the same legal standards for developed states to failed states when attempting to determine the credibility of an allegation. The "balance of probabilities" standard, widely accepted by the European Court of Human Rights in adjudicating human rights cases, offers the best hope for a post-conflict vetting environment. In essence, the balance-of-probabilities standard is an injunction to evaluate whether an alleged offense is more probable than not. Guidelines for assessing the credibility of evidence are:
- The general trustworthiness of allegations made against an applicant (level of detail; coherence and absence of contradiction; identification of dates, locations, and circumstances).
- The general trustworthiness of the authors of the allegations (personal circumstances, general interest or involvement, link with the applicant, link with parties to the conflict, link to other applicants).
- Whether or not there are concurring allegations, especially when the circumstances of the crime's commission indicate that multiple persons witnessed the crime or its circumstances. However, the fact that an allegation comes from only one source should not be a bar to its being considered credible, particularly if the circumstances of the alleged criminal behavior make it likely that there could be only one witness.
The final review process is critical because it corrects deficiencies in vetting, maximizes probity within the process, and instills a sense of procedural justice for disqualified applicants, thereby reducing acts of vengeance against the security sector reform staff and society in general.
A Tightrope Act
Vetting is a high-profile tightrope act in which the need for individuals of proper character must be balanced against the need for skilled individuals with scarce expertise, while the rights of victims must be weighed against the interests of applicants. Also to be considered is the ethnic mix of the new security force. Because internal conflicts and civil wars are often fought between groups of different ethnicities, religious beliefs, tribes, and other nonstate identifiers, the ethnic mix of a new security force can be a particularly sensitive issue. Generally, post-conflict reconstruction seeks diversity in government and power-sharing among divided populations. A single group should not disproportionately dominate the new security sector lest it seize control of the government.
But what happens when the principle of diversity collides with other principles necessary for a competent indigenous security force? Should a vetting program lower the standards for human rights vetting for a group with a high rate of human rights violations in order to achieve diversity in the new armed forces? Should a vetting program waive the literacy requirements for groups that were unfairly denied access to education? Having multiple standards of entry among an already polarized and unstable society can have deleterious consequences for the security forces. However, for one group or only a few people to dominate the security force could prove disastrous. Achieving balance within nascent security forces is often a Hobson's choice. It is what makes vetting as much an art as a science.
1. For lack of a better term, I use "proper character" to loosely convey the qualities and ethics desirable in individuals charged with providing state security and applying force. The specificities and standards of these characteristics must be negotiated among stakeholders before vetting begins.
2. See Wojcieck Sadurski, "'Decommunisation,' 'Lustration,' and Constitutional Continuity: Dilemmas of Transitional Justice in Central Europe," EUI Law Working Papers, no. 2003/15 (2003); Roman Boed, "An Evaluation of the Legality and Efficacy of Lustration as a Tool of Transitional Justice," in Post Conflict Justice, ed. Cherif Bassiouni, (2002); Marc S. Ellis, "Purging the Past: The Current State of Lustration Laws in the Former Communist Bloc," Law and Contemporary Problems 59, no. 4 (1996): 181-196; Mark Gibney, "Decommunization: Human Rights Lessons from the Past and Present, and Prospects for the Future," Denver Journal of International Law and Policy 87 (1994).
3. Human Rights Watch, "Bosnia and Hercegovina: Beyond Restraint; Politics and the Policing Agenda of the United Nations International Police Task Force," vol. 10, no. 5(D), June 1998, www.hrw.org/reports98/bosnia .
4. U.S. law restricts the provision of funds to units of foreign security forces when the U.S. Department of State (DOS) has credible evidence that the unit has committed gross violations of human rights. Commonly referred to as the Leahy Law, this restriction first appeared in the 1997 Foreign Operations Export Financing and Related Appropriations Act (P.L. 104-208) and only applied to funds appropriated to DOS's international narcotics control program. In 1998, Congress broadened the law to apply to all funds appropriated under the 1998 Foreign Operations Export Financing and Related Appropriations Act (P.L. 105-118). In 1999, a similar provision appeared in the U.S. Department of Defense (DOD) Appropriations Act (P.L. 105-262), which applied to funds appropriated under that act. The two provisions have appeared each year since in the annual Foreign Operations Appropriations Act and the DOD Appropriations Act, respectively.
5. U.S. Government Accountability Office Report-05-793, Southeast Asia: Better Human Rights Reviews and Strategic Planning Needed for U.S. Assistance to Foreign Security Forces (Washington, DC: U.S. Government Printing Office, 29 July 2005), www.gao.gov/cgi-bin/getrpt?gao-05-793 .
6. Office of the Inspector General Report, "Interagency Assessment of Iraq Police Training," 15 July 2005, 22.
7. Ibid., 23.
9. Ibid., 25.
10. United Nations (UN), Geneva Convention on the Prevention and Punishment of the Crime of Genocide, Geneva, 9 December 1948.
11. Rome Statute of the International Criminal Court, 2187 UNT.S. 3, entered into force 1 July 2002, www.un.org/law/icc/statute/contents.htm . The United States is not a signatory to the Rome Statute.
12. UN, Geneva Convention Relative to the Protection of Civilian Persons in Time of War, 75 UNT.S. 287, entered into force 21 October 1950.
13. UN, A Universal Declaration of Human Rights, Article 28, Geneva, 1948.
14. UN, International Covenant on Civil and Political Rights, Article 4, www.hrweb.org/legal/cpr.html .
15. There are exceptions to the vetting process I describe here, but a fuller explanation and analysis of post-conflict recruitment is beyond the scope of this study.
Also available online at: