Security sector reform, including vetting of security institutions

The present report is submitted by the Special Rapporteur on the promotion of truth, justice, reparation and guarantees of non-recurrence to the General Assembly in accordance with resolution 18/7 of the Human Rights Council. The activities undertaken by the Special Rapporteur from July 2014 to June 2015 are set out in his most recent report to the Human Rights Council (A/HRC/30/42).

That report addresses the fourth pillar of the mandate, guarantees of non-recurrence. The present report is properly conceived as a part of the framework presented therein. In it, the Special Rapporteur concentrates on certain preventive measures associated with the reform of the security sector, in particular the vetting of security institutions. He also discusses defining the different roles of the police, the military, and the intelligence services; rationalizing security institutions; narrowing military jurisdiction; strengthening civilian control and oversight over the security sector; and eliminating military “prerogatives”.

Without attempting to provide a detailed summary of the earlier report, given that the two reports are part of the same overall framework, some background is called for. The notion of guarantees of non-recurrence has received significant, albeit sporadic and uneven, attention in various instruments and documents. It has been said that a commitment to non-recurrence is part of a proper understanding of remedies and, moreover, that it is entailed by a commitment to rights.

The first position became enshrined in the Basic Principles and Guidelines on the Right to a Remedy and Reparations for the Victims of Gross Violations of Human Rights Law and Serious Violations of International Humanitarian Law, adopted by the General Assembly in resolution 60/147 on 16 December 2005. According to the Principles and Guidelines, guarantees of non-recurrence are one of the forms of reparation (remedy) to which victims are entitled (see para. 23 of resolution 60/147).

More generally, it has been argued that a commitment to adhere to a right involves making an effort to ensure that its violation ceases and is not repeated. The duty to prevent recurrence is hence closely linked to the obligation of cessation of an ongoing violation. On that basis, “guarantees serve a preventive function and may be described as a positive reinforcement of future performance”.¹

1. Commentary of the International Law Commission on article 30 of the Draft Articles on State Responsibility, Official Records of the General Assembly, Fifty-sixth Session, Supplement No. 10 and corrigenda (A/56/10). ↩

In its general comment No. 31 (2004) on the nature of the general legal obligation imposed on States parties to the Covenant, the Human Rights Committee held that the purposes of the International Covenant on Civil and Political Rights “would be defeated without an obligation integral to article 2 to take measures to prevent a recurrence of a violation of the Covenant. Accordingly, it has been a frequent practice of the Committee … to include … the need for measures, beyond a victim-specific remedy, to be taken to avoid recurrence of the type of violation in question”. The same practice can be observed in the jurisprudence of both regional and national courts.² In the landmark case of Velásquez Rodriguez v. Honduras, the Inter-American Court argued on the basis of the idea that States, in addition to an obligation to respect rights have an obligation to ensure them, that the commitment to rights includes the obligation “to organize the governmental apparatus, and in general, all the structures through which public power is exercised, so that they are capable of juridically ensuring the free and full enjoyment of human rights”.³

2. See A/HRC/30/42 for further examples and elaboration. ↩
3. OAS/ser.L/V/III 19, doc. 13, app. VI (1988), para. 166. ↩

Conceptually, there is a difference between guarantees of non-recurrence and the remaining three core elements of a comprehensive transitional justice approach, namely, truth, justice and reparation. While those three elements refer to measures, guarantees of non-recurrence refer to a function that can be satisfied by a broad variety of measures. The foundational texts already demonstrate that variety, pointing to, inter alia, reforming institutions, disbanding unofficial armed groups, repealing emergency legislation incompatible with basic rights, vetting security institutions and the judiciary, protecting human rights defenders and training members of the security institutions in human rights.

The core function of guarantees of non-recurrence is preventive in nature. It is one to which truth, justice and reparation are themselves supposed to contribute: criminal justice mainly through deterrence and affirmation of norms; truth commissions through disclosure, clarification and the formulation of recommendations with a preventive intent; and reparations by strengthening the hand of victims to claim redress for past and future violations and to enforce their rights more assertively.

Because guarantees of non-recurrence are a function that can be satisfied by diverse measures, there is no such thing as a general non-recurrence policy that will be equally effective in all contexts. An effective policy designed to prevent systemic violations will need to adjust form to function and choose the proper measures.

In his recent report to the Human Rights Council (A/HRC/30/42), the Special Rapporteur sketched a framework of what an actionable non-recurrence policy could include. After highlighting the significance of security (equitably distributed), of legal identity (a right which in turn is a gateway to other rights) and of economic conditions that, among other things, allow those who leave power to find a livelihood (for otherwise, as we see often, the stakes of losing power are so high that no one is willing to leave office), the report refers to a range of institutional interventions that serve a preventive aim. That range includes initiatives from the immediately actionable to long-term and complex measures that require the coordination of multiple institutions. Part of the point of presenting such a range of interventions is to emphasize that, even under inauspicious circumstances, there is something that can be done to promote non-recurrence.

In the domain of institutional reforms, the report includes as possible elements of a non-recurrence policy the ratification of international human rights treaties; legal reforms to incorporate international criminal types and to solve issues relating to the statute of limitations and retroactivity; a review of emergency, security and counter-terrorism laws, in order to make sure that they are compatible with human rights and that they do not provide incentives for the violation of rights; judicial reforms to strengthen both the internal and external independence of judges; constitutional reforms to remove discriminatory provisions and to introduce mechanisms of inclusion; the adoption of a bill of rights; strengthening of the separation of powers; the establishment of a constitutional court; and, at the limit, the adoption of a new constitution (or of transitional constitutional arrangements).

However, in part because in some of the countries where mass violations are more likely to take place, State institutions may be weak, ineffective, inaccessible, corrupt or captured and partly because even in those places in which institutions are not beset by any of those problems, guaranteeing non-recurrence should not be reduced to a technical issue of “institutional engineering”, in his report, the Special Rapporteur also emphasized the importance of interventions in the societal sphere, particularly at the level of civil society, and in the sphere of culture and of personal dispositions — interventions that have received virtually no attention in discussions about non-recurrence.

While discussions about transitional processes have recognized the contributions of civil society organizations to redress in terms of advocacy, evidence gathering, monitoring functions and reconciliation initiatives, the preventive potential of civil society has not been paid sufficient attention. At the most general level, the argument is that if terror works by virtue of its “disarticulating power”, its capacity to isolate people from one another and hence hamper resistance (see A/68/345), conversely, one can argue that the “power of aggregation” of civil society lends it a preventive function, i.e. that where people are not alone, victimization is less likely.

In his recent report to the Human Rights Council (A/HRC/30/42), the Special Rapporteur claimed that for preventive purposes, strengthening civil society was important. However, because civil society cannot be reduced to non-governmental organizations (NGOs), further work needed to be done to find effective ways of bolstering other kinds of organizations that had in fact played important (but largely unexamined) roles in past transitions, including religious organizations and trade unions. That would involve strengthening protection of the freedoms of expression, association, religion, etc.

Finally, the Special Rapporteur also drew attention to the preventive potential of interventions in the sphere of culture and of personal dispositions. Although it is well understood that overcoming the legacies of mass violations requires some interventions at the level of individuals, including trauma counselling and psychosocial support, the preventive dimension of those and other interventions has not received sufficient attention.

In his report, the Special Rapporteur stated that reform of the security sector, in particular vetting, would be the subject of a report of its own. It is to that topic that the remainder of the present report is devoted. Security sector reform is a vast and highly developed field of policy and practice. In the present report, the Special Rapporteur does not in any way pretend to cover or do justice to the subject. Rather, he describes certain aspects of security sector reform that he considers particularly relevant to prevent recurrence. The United Nations has adopted policies on the human rights screening of its personnel and on assessing the human rights record of non-United Nations security forces that it intends to support. While he cannot elaborate on this in the present report, the Special Rapporteur stresses that the reputation and effectiveness of the United Nations significantly depends on the integrity of its personnel and on the consistency of its support for international human rights standards.

In writing his report to the Human Rights Council (A/HRC/30/42), of which the present report is the companion, the Special Rapporteur was acting on the conviction that institutional reforms are an important part of a non-recurrence policy, but only one part. Similarly, in the case of the present report, the Special Rapporteur would like to acknowledge the potential of vetting programmes, but also thinks that those discussions (again, such as they have been) have crowded discussions about guarantees of non-recurrence in the transitional justice sphere and that, even in relation to preventive measures in the security sector, the range of interventions needs to be broadened.⁴ It is not just that vetting programmes are significantly harder to implement in transitions which differ from those in which an authoritarian regime collapses, but that, even in regard to vetting, it is worth emphasizing the importance of civil society. However, the fundamental point is that it is important to think about prevention in terms that go well beyond vetting, as important as the latter might be.

4. Three out of the six principles for the protection and promotion of human rights through action to combat impunity, which address guarantees of non-recurrence, deal with vetting (see E/CN.4/Sub.2/1997/20/Rev.1, principles 40-42). See also S/2004/616, paras. 52 and 53, and A/62/659-S/2008/39. ↩

Vetting can, in fact, make an important contribution to transitions, provided that it is meaningfully differentiated from purges. Vetting, as the term has come to be used, far from being the name for massive dismissals on the basis, for example, of mere membership in a party or organization or, even less of ascriptive factors, denotes a formal process to screen the behaviour of individuals and assess their integrity on the basis of objective criteria, so as to determine their suitability for continued or prospective public employment (see S/2004/616, para. 52). On the basis of that distinction between vetting and purging, the rightly criticized “de-Baathification” programme in Iraq does not even count as a vetting programme.⁵

5. In May 2003, the Coalition Provisional Authority dissolved the Iraqi armed forces, security services, party militias and other organizations close to Baathists and dismissed senior members of the Baath Party, leading to well-known consequences resulting from the increased enrolment of security-trained officers in the insurgency and the significant weakening of the capacity of the Government to provide security and protect the rule of law. See Coalition Provisional Authority, Order Nos. 1 and 2 on de-Baathification of Iraqi society and the dissolution of entities. ↩

As with other transitional justice measures, when implemented as a part of a comprehensive transitional justice policy, vetting can offer recognition to victims, foster civic trust, contribute to social integration or reconciliation and strengthen the rule of law (see A/HRC/21/46). In contexts in which it is unlikely that all of those who are responsible for human rights violations will face criminal punishment, vetting is a means for addressing part of the “impunity gap”. Making sure that violations do not remain unaddressed and that they are not considered to be inconsequential provides a measure of recognition to victims. It spares them (and others) the indignity of having to show deference to those who have violated their rights and who are still in positions of authority (for example carrying government-issued arms and bearing the insignia of the State), when they seek State services to which they are entitled. Vetting can induce trust, not just by “re-peopling” institutions with new faces, but thereby demonstrating a commitment to systemic norms governing the hiring and retention of employees, disciplinary oversight, prevention of cronyism and so on.⁶ By removing from security institutions those who are responsible for serious violations, which are normally socially divisive, vetting can make a contribution to social integration. Finally, because in the end vetting is guided by the ideal of guaranteeing the conditions under which citizens can relate to one another and to the authorities as equals, it can contribute to strengthening the rule of law.

6. See Pablo de Greiff, “Vetting and transitional justice”, in Justice as Prevention: Vetting Public Employees in Transitional Societies, Alexander Mayer-Rieckh and Pablo de Greiff, eds. (New York, Social Science Research Council, 2007). For recent empirical evidence see also Cynthia M. Horne, “Lustration, transitional justice and social trust in post-communist countries. Repairing or wresting the ties that bind?” Europe-Asia Studies, vol. 66, No. 2 (February 2014). ↩

In the context of a report on non-recurrence, however, there are three arguments more closely related to prevention that explain why, despite its lacklustre history, vetting remains attractive to transitional justice policymakers. First, under the right circumstances, vetting could be an “enabling condition” of other transitional justice measures. In other words, vetting is a mechanism to manage spoilers in a transition. The idea is that if institutions that were responsible for violations can be vetted early on in a transition process, other transitional justice measures will work better. In Morocco, for example, civil society organizations have claimed that the refusal of some segments of the security sector to cooperate with the Instance équité et réconciliation was due in part to the presence of officials who should have been vetted, and that the truth-telling functions of the Instance, including access to archives and testimonies, would have been more effective if they had been removed. Obviously, the argument is not restricted thematically to truth-telling, just as it is not restricted geographically to Morocco. During his visit to Uruguay, the Special Rapporteur received information about the lack of vetting of the military that has facilitated the preservation of an esprit de corps, which has hampered collaboration by the military with the judicial investigations of violations that took place during the dictatorship more than 30 years ago (see A/HRC/27/56/Add.2).

There is a closely related argument that is important in the context of an analysis of effective prevention tools: vetting is important because it can help to dismantle networks of criminal activity which, quite aside from resisting the implementation of transitional justice measures, may destabilize the transition and eventually pose a serious threat to democratic institutions and to the rule of law. In the light of recent events in Guatemala, in which a former General turned politician (chief of military intelligence and member of the Kaibiles, an elite force with a spotty human rights record during the conflict), who was elected President in November 2011 and resigned when Congress removed his immunities in September 2015 to face charges of corruption, it is difficult not to think of the positive difference that vetting would have made in a country not only with high levels of corruption, but with current murder rates that in some cities match or even top those that occurred during the years of conflict (see A/HRC/28/3/Add.1).

Indeed, it is the preventive potential of vetting, a potential that comes from its capacity to dismantle networks of criminality, which offers both the strongest justification for it and the best explanation of how it works.⁷ It is certainly better than arguments for the deterrent effects of vetting; attributing strong deterrence powers to vetting is not particularly plausible. If, for example, more severe forms of punishment (including the death penalty) in well-established, efficient, and expeditious legal systems have a dubious deterrent value, it is not far-fetched to think that deterrence is not the strongest defence of vetting in the midst of all the uncertainties of a transitional situation. Two main forms of uncertainty are relevant in that context: the first concerns the course of the transition. It is rarely a foregone conclusion, especially at the pre-transitional stage and even in the early stages of the transition, that a new Government intent on vetting will be able to consolidate its hold on power and implement a vetting programme. That uncertainty may weaken the deterrent potential of any proposed measure (and may actually generate incentives to block the transition). The second has to do with uncertainties about the probability of those responsible being subject to any justice measure. It has been observed in the literature on criminology that beyond a certain threshold, what ends up having greater deterrent potential than the length of sentence, is the degree of certainty over being apprehended, tried and convicted and hence having to serve a sentence of some kind. In countries where the security and justice sector have been left in disarray, that is highly unlikely. It is doubtful, then, that vetting has a significant deterrent power.

7. For further elaboration of the argument see de Greiff, “Vetting and transitional justice”, in Justice as Prevention: Vetting Public Employees in Transitional Societies. ↩

However, it makes sense to think that vetting does have a preventive function — and prevention is not the same thing as deterrence.7 Rather than focusing on the possible reactions of individuals to particular measures, as the deterrent argument does, the claim for vetting operates at a different, more structural level. That argument takes vetting to be something akin to an anti-mafia measure. Thus, its purpose is not primarily to send signals to individuals (i.e., that they will lose their jobs and social standing if they engage in certain behaviour), but to disable structures within which individuals (who, by the way, may have refrained from criminal activity were it not for those structures) in fact carried out criminal acts. Vetting, then, is defended on the grounds that it may prevent the recurrence of violations, not necessarily because the sanctions it metes out (loss of job, public prestige, etc.) are sufficient to deter individuals, but because it dismantles networks of criminal activity, even if it does not reach each and every participant in activities that violate the rights of others. That also means that the focus of vetting measures can be strategically targeted, a point that will be examined below.

In the preceding paragraphs, the Special Rapporteur has described the potential of vetting and sought to capture some of the reasons why it has occupied so much space in the discussions about transitional institutional reforms and guarantees of non-recurrence. It goes without saying, however, that that potential is not always actualized.

Indeed, the challenges faced by vetting are severe. Among transitional justice measures, vetting is arguably the one that has lent itself more frequently to political manipulation. The following factors may explain why that is the case: post-conflict or transitional prosecutions are infrequent to begin with and they generally involve sufficient due process guarantees as to raise a shield against easy political manipulation; and truth-telling exercises, particularly by way of truth commissions, are also extraordinary events and the publicity that surrounds them is one among many factors that have kept such commissions from falling into the trap of partisan politics. Vetting, on the other hand, usually involves thousands of people, the process more often than not takes place with little public scrutiny, is usually carried out either by institutions close to the executive or by the very institutions whose members are being vetted. Perhaps as a result, they offer weak procedural guarantees and, contrary to even the best outcome of prosecutions or truth-telling, what hangs in the balance is (some degree of) control of public institutions — a strong incentive to engage in it for partisan political purposes.⁸ Vetting can affect the distribution of power, in that certain political groups may be disproportionately affected by a vetting process, while other groups may gain influence as a result of the process. Rather than establishing accountability for past abuses, a vetting process can be manipulated for ongoing political purposes. In the political isolation process in Libya, the disqualification criteria were so vague that they led not only to the removal of Gaddafi-era officials, but also resulted in marginalizing some members of one side of the political spectrum (including those appointed through national elections), thereby contributing to the destabilization of an already fragile transition.

8. For a review of the politicization of lustration processes in Central and Eastern Europe, see Cynthia M. Horne and Margaret Levi, “Does lustration promote trustworthy governance? An exploration of the experience of Central and Eastern Europe”, in Building a Trustworthy State in Post-Socialist Transition, Janos Kornai and Susan Rose-Ackerman, eds. (New York, Palgrave Macmillan, 2004). ↩

The political manipulation of vetting, using it as a way of, for example, hammering political opponents, already involves the crossing of a threshold of political opposition to vetting that many countries never overcome. A transitional government that is composed of representatives from various political factions with different agendas, or one that suffers from limited legitimacy may find it hard to implement a politically sensitive vetting process; or, as is more often than not the case, senior security officers and politicians are able to maintain their positions of power after the end of the conflict or the authoritarian regime and successfully resist the establishment of a vetting process.

Although many countries could be used as examples of political resistance to vetting, Nepal provides a recent and continuing illustration. Several commitments were made in the 2006 Comprehensive Peace Accord, the 2007 interim constitution and subsequent agreements to pursue accountability for serious abuses committed during the 10-year armed conflict. Various domestic civil society organizations and international actors repeatedly called on the Government, the army and other security institutions to put in place a vetting process. In a ruling in 2012, the Supreme Court of Nepal recognized vetting as a measure of transitional justice and ordered the Government to develop a new law on vetting and adopt temporary guidelines for vetting public officials when making new appointments, promotions or transfers until the law on vetting had been adopted.⁹ Despite repeated commitments and demands, those instructions have not been followed and senior security and government officials have so far effectively resisted not only criminal prosecutions but also the establishment of a vetting process.

9. See, for example, Sunil Ranjan Singh v. the Government of Nepal (case No. 067/2067). ↩

Assuming that countries can overcome political resistance to vetting and manage to avoid its political manipulation, they must still face a very tough set of challenges stemming from the complexity of the issue.

To begin with, there is a great deal of complexity in the very design of a vetting procedure. The following (partial) list of design variables around which complex decisions need to be made, illustrates the challenges:
(a) Vetting programmes have differed in terms of their targets. Choices have to be made both about the institutions where vetting will be applied and the positions within those institutions which will be subject to screening;¹⁰
(b) Programmes also differ in terms of screening criteria, i.e. precisely what kind of violations and abuses the system is designed to cover;
(c) Different decisions may also be taken about the types of evidence admissible in the process and importantly, about the criterion for making determinations;¹¹
(d) Not all programmes are the same in terms of the sanctions they impose — even dismissals can take place in many different ways (starting in a relatively mild way by giving people the opportunity to resign without disclosing their participation in behaviour considered abusive). Vetting sanctions can involve different degrees of publicity and prospective limitations in seeking employment in various sectors in the future;
(e) Finally, but importantly from a rule-of-law perspective, programmes differ in their establishment of review or appeals mechanisms.

10. More countries have established vetting programmes for sectors of the armed forces and police service than for the judiciary. Transitional justice measures implemented in Greece and Germany also included the vetting of universities. Proposals to vet sectors of the news media and other professions were not uncommon in the aftermath of the transitions in Eastern and Central Europe and of course, many countries including all of those mentioned above have used vetting processes for some electoral and other government offices. ↩
11. As administrative procedures, vetting programmes do not apply the standard of “beyond any reasonable doubt”, but use the “balance of probabilities” or similar concepts. ↩

Quite apart from the challenges that arise at the design stage, there are also formidable challenges at the implementation or operational stages. The truth is that in most settings, but particularly in post-conflict contexts, vetting programmes operate under conditions that would challenge any system. The programmes are supposed to screen very large numbers of people on the basis of a variety of criteria (not always completely settled once and for all) and on information that is rarely complete and of which the reliability is often questionable. They also almost always take place under a great deal of time pressure.

The case of Liberia, where the National Transitional Government with the support of the United Nations Mission in Liberia (UNMIL) conducted a vetting of the Liberia National Police and other civilian security agencies between 2004 and 2006, illustrates some of the challenges which characterize vetting efforts, particularly in post-conflict and weakly institutionalized settings.¹² Despite significant efforts, only limited information on the involvement of serving officers in human rights abuses during the years of conflict was obtained: criminal records had been largely destroyed; information contained in reports of non-governmental organizations, in the files of the Special Court for Sierra Leone and the Sierra Leone Truth and Reconciliation Commission and in the databases of the disarmament, demobilization, rehabilitation and reintegration process was usually not recorded in a manner which made cross-referencing of names possible; visits by UNMIL police and the background section of the Liberia National Police to the home communities of serving officers did not yield much information, owing to the enormous displacement that had taken placed during the conflict and security concerns felt by community members; and the publication of the names of serving officers in national newspapers resulted in only four complaints,¹³ which were found on investigation to be false.¹⁴

12. The role of UNMIL included assisting with the reform and restructuring of the police and other civilian security institutions, including determining the composition, selection and vetting of members of the new national police (see Security Council resolution 1509 (2003) and S/2003/875). ↩
13. It should be noted that no pictures of the officers were published, which made the identification of abusive officers more difficult, that Liberia has a high illiteracy rate and the circulation of national newspapers is not more than 20,000 and very limited outside Monrovia. ↩
14. That is not to say that the vetting process was inconsequential. In fact, around 60 per cent of Liberia National Police officers were disqualified, but only a few were removed on human rights grounds. Most of the dismissals resulted from a failure to meet educational or other criteria. ↩

The difficulties in gathering information pertaining to human rights criteria are shared across a variety of cases. Between 1999 and 2002 the United Nations Mission in Bosnia and Herzegovina (UNMIBH) vetted more than 23,000 police officers in the country on a variety of criteria, including human rights. UNMIBH established a complex system to collect information on the background of police officers (see Security Council resolution 1088 (1996)). It had access to and searched the databases of the International Criminal Tribunal for the former Yugoslavia, maintained field offices throughout Bosnia and Herzegovina and collected information from victims and witnesses on abuses dating from the time of the armed conflict. The mission also systematically searched reports issued by national and international governmental and non-governmental organizations and, finally, questionnaires completed by officers during their registration by UNMIBH sometimes contained relevant information on their background at the time of the armed conflict. On that basis, around 200 serving police officers were identified as having questionable backgrounds, triggering further investigations by UNMIBH to substantiate the allegations. In the end, as a result of the comprehensive data collection and investigation process, some 60 out of a total of around 23,000 police officers were removed from service on human rights-related grounds. Despite the enormous investment made by the mission, the number of removals based on misconduct at the time of the armed conflict was small in comparison, not only to the total of around 500 officers removed on other grounds, but also to the magnitude of the crimes committed during the armed conflict and the key roles held by uniformed actors in the commission of those crimes.¹⁵

15. See Alexander Mayer-Rieckh, “Vetting to prevent future abuses: reforming the police, courts, and prosecutors’ offices in Bosnia and Herzegovina” in Justice as Prevention: Vetting Public Employees in Transitional Societies, Alexander Mayer-Rieckh and Pablo de Greiff, eds. (New York, Social Science Research Council, 2007). ↩

In general and particularly from a preventive standpoint, the numbers of people screened are by themselves not indicative of success (or failure). As argued before and, in the context of a prosecutorial strategy, elsewhere (see A/HRC/27/56), a crucial preventive aim is the dismantlement of networks, rather than the wide distribution of sanctions. The problem, however, is that most (although not all) vetting efforts in the security sector have not been targeted but general and in that sense not particularly strategic. Against that background, low numbers do acquire a more revealing character and what they reveal is not a picture of huge success.

The difficulties having to do with information do not pertain only to availability or reliability (“supply” issues, as it were). They also have to do with a design or “demand” issue. Vetting processes screen individuals on a variety of criteria, some of them having to do with “capacity” (qualifications, training) and some of them with “integrity” (human rights compliance, financial probity) apart from other demographic criteria (gender, geographical provenance, ethnicity, religion). In a perfect world, of course, there is an incentive to vet people in accordance with the most comprehensive set of screening criteria. In the real world, where information is in short supply, every addition of a criterion to the screening set generates costs. Thus, to illustrate the point, the vetting of the police in Bosnia and Herzegovina was based initially on criteria that included registration, age, nationality, training, performance, conduct, criminal record and not having committed atrocities during the armed conflict. In the middle of the process, a new criterion was introduced to verify whether police officers had illegally occupied the housing of refugees or internally displaced persons. As a result, supplementary information had to be sought from more than 23,000 officers and the legality of the housing situation of each and every officer had to be determined. That led to the identification of close to 8,000 illegal occupations by police officers, but it also further extended the duration of the process and contributed to the difficulties in completing it on time (see S/2002/1314).

Complexity in itself is not the main problem; what is problematic is the tendency to design vetting programmes without any sense of the conditions, the fulfilment of which is necessary for achieving the tasks in hand, including the availability of evidence, or even the magnitude of the load that such programmes will have to bear.¹⁶ When design gets so far ahead of possibilities and ignores so many constraints, the probabilities of success diminish.

16. In that sense, the problem is analogous to the challenges raised by truth commission mandates that depart from criteria of “functional adequacy” (see A/HRC/24/42). ↩

Political opposition to vetting can be addressed both by trying to strengthen incentives (demand) for it and also by trying to weaken opposition to it. On the first point, it has been a constant factor in the work of the Special Rapporteur to emphasize that almost without exception, it has been the tireless work of civil society organizations that explains the relative success of transitional justice. It is no different with regard to vetting. Ultimately, the political costs and risks involved in dislodging from positions of power those who bear arms will not wreck a vetting programme only if the incentives for establishing one are strong enough to overcome the disincentives. The moral convictions of individual political leaders are rarely sufficient. Resistance must be met by other sources of demand and those usually come from civil society. However, civil society organizations are much less familiar with vetting alternatives than with other transitional justice measures. Hence, there is also less activism in favour of it. Strengthening the capacity of civil society in relation to vetting can therefore catalyse support in its favour.

In certain contexts, perhaps there is no way of sufficiently quelling opposition to vetting. Other strategies for bringing a certain degree of control over membership in security institutions are called for. There are situations in which even the basic question of who is part of a security institution has no clear answer. In the Democratic Republic of the Congo, for instance, vetting the police was a possible reform activity that was regularly discussed following the establishment of the Transitional Government in 2003.¹⁷ However, in addition to political and internal police opposition to vetting, the institutional conditions for vetting were not in place. While the Police nationale congolaise was nominally unified, the police remained internally divided and former warring factions continued to interfere with policing. Estimates of the number of police in the country varied between 80,000 and 110,000 officers. The human resource management system of the police had not been maintained and basic information on police officers did not exist. Police officers did not possess identification cards and the public could not always determine who was a police officer and who was not. Moreover, most immediate efforts at police reform focused on preparing the police to provide effective security for the elections in mid-2005. As a result, plans for putting in place a vetting process were not pursued. Instead, police reform efforts concentrated on establishing a registration process to determine the number of police officers and define the parameters of a future personnel reform process, including vetting.

17. Other security institutions were also considered for vetting. In particular the members of the defence forces were to be vetted in the context of the brassage process, the integration of former warring armed groups into the Forces armées de la République démocratique du Congo (FARDC). ↩

Similarly, in Burundi, the 2000 Arusha Peace and Reconciliation Agreement dissolved all law enforcement agencies and integrated their members into a single institution, the Police nationale de Burundi, with over 17,000 officers. But the exact number of officers “inherited” from previous agencies, their level of education and their professional experience could not be determined owing to the swift process of integration of various groups of officers with hugely diverse backgrounds. As a result, the Ministry of Public Security had difficulty in controlling and managing police personnel, which nurtured a culture of lawlessness and prevented the police reform process from moving forward. In 2008, the Ministry launched a census and identification process to register and issue service cards to all police officers. Following the process, every officer had to be in possession of a service card and carry a visible badge. A public information campaign informed citizens that they could ask an officer to identify him- or herself. The census and the requirement to carry identification established a measure of public accountability in that it enabled citizens to identify police officers and attribute police activities to specific officers. The process also made it possible to identify and sanction persons who gave the appearance of being police officers but were not. In addition, ghost officers who were on the Ministry payroll, but did not exist, could be identified and their salary payments could be terminated. The database of the census and identification process became the basis for the newly developed human resources management system and other reform efforts (see A/HRC/30/42/Add.1).

There are many contexts in which a great deal of political capital is misspent in discussions about vetting, when it is plain that long before an effective vetting programme can be put in place, it is crucial to take measures as basic as establishing a registration census for the members of security institutions. That is not only less likely to awaken opposition, it is a sort of “gateway” condition for discipline and oversight of the security institutions.¹⁸

18. Not unlike the way that legal identity is a sort of “gateway” right for citizens, which enables access to other rights (see A/HRC/30/42). ↩

One way to diminish opposition to vetting in some contexts, although not all, and at the same time reduce the demands on the system (including by reducing the number of people to be screened) is to target the programmes more narrowly, or to roll them out incrementally. In Kenya, the National Police Service Commission was mandated in 2011 to vet all 80,000 police officers in the country.¹⁹ The Commission took considerable time to develop the regulatory framework for the vetting process and set up the vetting secretariat. The vetting process was launched in December 2013. After one year of operation, the Commission had completed the vetting of the 198 most senior police officers, around 0.25 per cent of the total number of officers to be vetted, while the vetting of the next level had just started. Of the 198 senior officers, 17 were found unsuitable. The Commission was criticized not only for what the public considered the low number of removals, but also for its slow implementation rate. The delays were caused by the complexity of the vetting procedures, operational and security challenges, internal management issues and funding shortages. Had the programme been designed and implemented in a stepwise fashion, perhaps at the very least expectations would have been managed more effectively.

19. See National Police Service Act (2011), part II, section 7, and National Police Service (Vetting) Regulations (2013). ↩

Where to start a vetting process that targets some but not all ranks in an institution depends on circumstances and there may be some contexts in which that will not work. One can imagine certain places where focusing first on categories of officials who are likely to put up the greatest resistance, so that spoilers are removed early on, would work well. In others, by contrast, fine-tuning a vetting strategy first, accumulating experience and information and letting other reforms of the security sector take root before addressing the most controversial cases might be more beneficial. Surprisingly, the alternative of targeted vetting with different incremental paths has received little systematic attention. In other words, just as prosecutions in a transitional setting may be strengthened by the articulation of a prosecutorial strategy of the sort advocated by the Special Rapporteur in an earlier report (A/HRC/27/56), vetting may profit from the design of a corresponding vetting strategy.

Such a strategy may seek to diminish resistance, not merely through “targeting” but also through the adoption of “soft lustration” or “indirect vetting”. Soft lustration as it was applied in Poland in the late 1990s was a form of vetting that only punished a “lustration lie”: officials who were truthful about their collaboration with the secret services were not sanctioned. Only those officials who made untruthful lustration statements lost their jobs.²⁰ Of course the system did not apply primarily to members of the security institutions, its primary target was individuals in political office and in some sectors of the civil service. Forms of “soft lustration” for members of the security institutions would need to be modified in order to account for the special duties and responsibilities of members of the security sector. Soft forms of vetting, which also include measures that allow people to resign their positions quietly can adopt forms that warrant more attention, especially if combined with other measures which may facilitate other justice aims, such as truth-seeking or other forms of institutional renewal. They may also be combined with limited forms of vetting, such as merely screening new appointments, promotions or transfers.

20. Polish lustration law adopted on 11 April 1997. See also Roman David, Lustration and Transitional Justice: Personnel Systems in the Czech Republic, Hungary, and Poland (Philadelphia, Pennsylvania, University of Pennsylvania Press, 2011). ↩

“Indirect vetting” is the name that can be used for strategies that create incentives for abusive officials to vacate their positions. At the greatest extreme of indirection and perhaps no longer, strictly speaking, a form of vetting, given the generality of the incentives (which therefore apply across the board and no longer differentiate on the basis of individual behaviour), one finds, for instance, either incentives to take early retirement or, as in the case of Spain after the transition, a reduction in the mandatory retirement age (to 65 years) and the creation of the “active retirement” regime. As a result, most generals who served during the authoritarian regime had to retire but kept their honours and were given responsibilities outside the military chain of command, such as research assignments or forestry management. Measures of that sort enable generational renewal in the security sector without generating the same sort of opposition as screening measures that target individuals (see A/HRC/27/56/Add.1).

Argentina ended up, not entirely by design, with an especially interesting form of “indirect vetting”. As a result of legislation to increase both the transparency of, and the participation of civil society in, parliamentary debates, civil society organizations gained authorization to make submissions to debates concerning promotions to the highest ranks in the security sector. That is perhaps the paradigmatic case of a strategy that creates incentives for retirement: candidates to promotion to the highest ranks who had murky pasts could be sure that NGOs would make submissions to the debate that could not be ignored. Given that the enabling legislation was not even related to vetting but to congressional procedures, no one could complain too much. However, since beyond a certain rank it is difficult to stay in the security sector if one is not going to seek a promotion, the system did provide strong incentives for people to retire and it has been used to very good effect by human rights NGOs in Argentina. So, although quite “indirect” it is nonetheless an effective form of vetting.²¹ A lot of room is still open for policies of that sort.

21. See Valeria Barbuto “Strengthening democracy: ‘impugnación’ procedures in Argentina”,in Justice as Prevention: Vetting Public Employees in Transitional Societies, Alexander Mayer-Rieckh and Pablo de Greiff, eds. (New York, Social Science Research Council, 2007). ↩

While the Special Rapporteur is convinced that there is plenty of room for creative thinking about ways of overcoming or mitigating the challenges posed, particularly (although not exclusively) when a conception of vetting that was designed for collapsed post-authoritarian transitions is applied in post-conflict settings, the truth is that for many reasons thinking about non-recurrence in the context of the security sector should not be reduced to vetting. There are other preventive measures that he would like to highlight in the present report.

In a good number of the countries where atrocity crimes have taken place, legal texts are ambiguous concerning the various roles of the different parts of the security sector. In still more countries, the military have been deployed in counter-insurgency and counter-terrorism operations within their borders, for riot control and in crime control operations, such as in the war on drugs. Countries seeking to establish a preventive strategy would do well to make sure that their constitutions clearly establish a difference between the external defence function of the military (and the strict conditions under which it can be used internally) and the internal security functions of the police. The precise functions and limits of both internal and external intelligence services should also be specified.

Failing to establish those distinctions can lead and has often led to the politicization of the armed forces, exposes them to corruption, distorts the understanding of their proper social role, puts them in situations for which their training is unsuited and can create an incentive for acquiring interests (including economic interests) that are foreign to their role. Ultimately, such situations may result in a “militarization” of politics. Additionally, that has as a consequence the corresponding weakening of the police service whose training and equipment, although in theory best suited for internal security purposes, suffers, relegating them to a place in which corruption and other forms of criminality become attractive, thus deepening a spiral of dysfunction easy to see in many police services in countries in which the military has a bloated mandate.

During periods of conflict and also under authoritarianism there is often a proliferation of security bodies (including intelligence services) with partially overlapping mandates and often crossing lines of command. A rationalization of the security sector in order to streamline both services and lines of command and to strengthen civilian oversight to guarantee compliance with constitutionally defined functions, is a critical part of a long-term prevention policy. In the present report, the Special Rapporteur cannot delve in more detail into that sort of reform but he wishes to highlight its importance.

Similarly, during times of conflict and under authoritarianism, a proliferation of non-State armed groups is often observed. Depending on the concrete circumstances, various measures may be taken to prevent their continued involvement in human rights abuses, including their disbandment, the disarmament, demobilization and social reintegration of their members, or vetting and the integration of their members into statutory security institutions.

The rationalization of the security services would be greatly enhanced, and this would have significant preventive potential, by tying expenditures in security to objective risk assessments, which could be the subject of public discussion (compatible with legitimate needs for confidentiality). Sophisticated risk assessments (including for some security factors) are done on an ongoing basis by both finance and insurance firms. Considering the vast expenditures in the area of security, their opportunity costs (including foregone investments in the social sector that may address some of the factors with which violations correlate), the tendency to use security budgets to hide incorrect appropriations and illegal expenses and the track record of human rights violations of bloated intelligence and security forces, greater public scrutiny of the relationship between security expenditures and real risk (referring in this context to the security needs of the population, particularly marginal and other vulnerable groups) should be promoted.

The Special Rapporteur reiterates that international law clearly establishes that the jurisdiction of military courts is restricted to military functions, that is, to trying members of the military and for military offences, such as insurrection, only (see, for example, A/61/384 and E/CN.4/2006/58)²² and that trying civilians in military courts or trying members of the military for crimes other than military crimes contravenes international law, in itself constitutes a violation of rights and, furthermore, is one of the most prevalent ways of shielding violations (see A/HRC/30/42). That is a serious problem in an extensive list of diverse countries including Cambodia, Colombia, Egypt, Nepal, Mexico and the United States of America. A non-recurrence policy should prioritize narrowing the jurisdiction of military courts to military personnel for offences against military discipline only and ensure that the jurisdiction of ordinary courts prevails over that of military courts to conduct investigations and prosecute alleged offences involving human rights violations, including when the alleged acts were committed by military personnel (see A/68/285).

22. See also Juan Carlos Gutiérrez and Silvano Cantú, “The restriction of military jurisdiction in international human rights protection systems”, International Journal on Human Rights, vol. 7, No. 13 (July 2010). ↩

The overall aim of security sector reforms in post-authoritarian or post-conflict situations is not necessarily best captured in terms of diminishing the size or costs of security institutions, even less diminishing their strength. The aim is rather to professionalize those institutions, preventing the power that inevitably derives from a monopoly of legitimate large-scale force from spilling over into other domains, politics and the economy included, and primarily, securing civilian control and oversight over them — that control and oversight falling, of course, under the law and in full compliance with all relevant international standards.

Civilian control is strengthened by establishing a truly civilian ministry of defence (which is not the same as appointing a civilian minister) responsible for the administrative aspects of the defence sector and, more importantly, representing the head of State, and also for defining defence policy, possibly within the framework of a national security council, with the advice of members of the armed forces. The ministry should be in charge also of drafting the military budget, establishing personnel policy and the general oversight of the armed forces. Additionally, effective parliamentary oversight of all relevant aspects of the military (policy, budget, acquisitions, career paths, etc.) and other oversight, monitoring and accountability mechanisms (such as effective internal disciplinary mechanisms, independent human rights bodies, ombudsperson offices, judicial oversight and informal oversight mechanisms, including the media and human rights NGOs) should be an important part of a preventive strategy. Experience suggests that civilian oversight is more effective if it is provided through multiple channels.

One of the legacies of the involvement of the military in functions that are not their own are “prerogatives”, which often include control over various aspects of politics and the economy. They include “tutelary powers”, such as the authority to appoint members of the legislature (25 per cent in Myanmar, for example), which, in conjunction with super-majority requirements, guarantees the military a veto power over law-making and constitutional reforms. Those powers may also include outsized participation in security councils, with more than advisory powers, giving the military effective control over crucial issues, sometimes including when to declare a state of emergency (with the corresponding suspension of, and threats to, rights);²³ guaranteed financial resources with little oversight or transparency; and loosely-to-not regulated opportunities for military industries and for members of the military to participate in businesses. Those prerogatives, “authoritarian enclaves” or areas of autonomy beyond control, should be eliminated as part of a preventive strategy. They weaken civilian oversight and control of the armed forces, are detrimental to economic development, subvert democracy and the rule of law, and undermine transparency in governance.

23. The 1980 constitution in Chile gave half the seats in the powerful National Security Council to members of the military. Reforms in 2004 made the Council an advisory body, convened at will by the President. See, for example, David Pion-Berlin, “Defense organization and civil-military relations in Latin America”, Armed Forces & Society, vol. 35, No. 3 (April 2009). ↩

One of the reasons why even militaries that have not held political power have managed to carve out so many significant prerogatives or “reserve domains” is that there is relatively speaking little technical competence in civil society as regards defence policy issues. The Special Rapporteur urges States and international organizations to invest in the development of such capacities. Meaningful civilian oversight over the military depends on decisively disputing its alleged monopoly of expertise in defence issues, including strategy, budgeting, armaments, etc. A policy to prevent human rights violations would be significantly strengthened if it paid attention to the conditions under which effective civilian oversight of military forces could take place. The informed involvement of civil society actors in a security sector reform process significantly shapes its direction and scope.

In his most recent report to the Human Rights Council (A/HRC/30/42), the Special Rapporteur noted that institutional reforms are an important part of a non-recurrence policy, but also emphasized that other types of initiatives, including strengthening civil society and those targeting culture and individual dispositions, could make important contributions to preventing a recurrence of violations. Similarly, in the context of the present report, the Special Rapporteur would like to acknowledge the potential of vetting programmes, but is also of the view that discussions about vetting have crowded the field of non-recurrence. It is important to situate vetting within the broader framework of security sector reform and to think about prevention in terms that go well beyond vetting, as important as the latter might be.

The Special Rapporteur calls upon States, in the aftermath of conflict and/or repression, to adopt comprehensive transitional justice policies, including the removal from security institutions of individuals responsible for human rights violations. Vetting can make a significant contribution to transitions. As with other transitional justice measures, when implemented as a part of a comprehensive transitional justice policy, vetting can offer recognition to victims, foster civic trust, contribute to social integration or reconciliation and strengthen the rule of law. As a mechanism to manage spoilers in transitions, vetting can also be an “enabling condition” of other transitional justice measures. More broadly, vetting can contribute to prevention, because it can help to dismantle criminal networks that may destabilize the transition and pose a serious threat to democratic institutions and the rule of law.

The Special Rapporteur also calls attention, however, to the severe challenges faced by vetting in transitions. Political opposition is regularly severe enough to prevent the establishment of vetting processes. Vetting can also be manipulated for ongoing political purposes, rather than establishing accountability for past violations. In addition, there are formidable operational challenges that arise at the implementation stages. Commonly in a vetting process, very large numbers of people are supposed to be screened under a great deal of time pressure, on the basis of a variety of criteria that are not always clearly defined, requiring background-related information that is difficult to find and substantiate, and usually with limited resources and inadequate capacities. The consequences of a vetting process may also generate considerable challenges affecting the environment, including further weakening of already fragile institutions, making it more difficult for them to deliver their services and creating a pool of disenchanted individuals who may destabilize the security situation.

The Special Rapporteur urges States and international organizations to invest in the development of civil society capacities in vetting and other areas of security sector reform. Generally, civil society organizations are much less familiar with vetting and security sector reform than with other transitional justice measures. Military and other security actors usually provide little space for civil society in a domain they consider their prerogative. Informed civil society actors can, however, not only counter resistance to vetting and advocate for reform, but also lend technical know-how on vetting that rarely exists in transitions. Civil society organizations can also play an important role in collecting and analysing background information on officers of the security institutions. Such information constitutes the foundation of any vetting process, but is often not available at all or not available in the right form. The Special Rapporteur encourages the training of civil society organizations on how information should be collected and organized so that it can be used in the vetting process. Of particular importance in this context is detailed information on perpetrators of human rights violations.

The Special Rapporteur calls upon States in transition to respond flexibly to operational challenges and opposition to vetting and, while insisting that individuals implicated in serious human rights violations ought to be removed from or not be integrated into the security sector, to adopt approaches to vetting that stand a realistic chance of being implemented. In contexts in which it is not clear who is part of a security institution and who is not, and in which basic information on personnel is not available, reform efforts may initially focus on a personnel census rather than on vetting. A census is not only less likely to awaken opposition, it is also a sort of “gateway” condition for discipline and oversight in the security sector and an institutional condition for other reforms, including vetting.

The Special Rapporteur encourages States in transition to adopt vetting strategies that, akin to prosecutorial strategies, respond to the characteristics of weakly institutionalized settings and to the likelihood of strong political opposition to vetting; to the insufficient capacities to remove all officers involved in human rights violations; to the potential security threats posed by a pool of officers who have been removed from their posts as a result of vetting; and to the potentially negative repercussions of a failed vetting process that not only misses its stated objectives, but may legitimize abusive officers and further erode public trust. One strategic option might be to adopt a “targeted vetting” approach that focuses on some ranks or certain units in an institution. Rather than covering all personnel, targeted vetting aims to dismantle criminal networks, remove spoilers, or empower units within the institution that have a specific responsibility to combat criminal activity, such as an internal discipline function, and ensure accountability, thereby contributing to preventing recurrence. Variations of “indirect vetting”, such as creating incentives (both positive and negative) for retirement or resignation, may be adopted in contexts in which more traditional forms of vetting are effectively opposed.

The Special Rapporteur calls upon States designing a vetting process to limit the number of criteria applied and to focus primarily on criteria related to the integrity of the personnel, particularly their human rights record. Vetting in transition can contribute to and prepare for, but cannot substitute for and should not be confounded with, regular human resources management in security institutions. The aims of vetting in the context of transitional justice relate primarily to the integrity of personnel and less to their competence. The type and number of criteria also significantly affect the duration of the process. Whereas ideally vetting may aim at the highest levels of the competence and integrity of personnel, a protracted process absorbs too many resources, is difficult to sustain and will hardly be able to achieve its goals, particularly that of contributing effectively to dismantling criminal networks within a reasonable period of time. Hence, vetting should aim primarily at removing those individuals who have committed the most serious violations and at ensuring at least minimum levels of integrity of the personnel concerned.

In the present report, in addition to vetting, the Special Rapporteur calls attention to several other institutional reform measures in the context of the security sector, including defining the different roles of the police, the military and the intelligence services; rationalizing security institutions; narrowing the jurisdiction of military courts; strengthening civilian control and oversight over the security sector; and eliminating military “prerogatives”.

The Special Rapporteur calls upon States emerging from conflict or authoritarianism to define in their constitutions the roles and functions of the armed forces, the police, the intelligence services and other security institutions, and to clearly differentiate between the external defence functions of the armed forces and the internal security functions of the police. In addition, measures need to be taken to prevent the continued involvement of non-State armed groups in human rights violations, including their disbandment; the disarmament, demobilization and social reintegration of their members; or vetting and the integration of their members into statutory security institutions.

The Special Rapporteur also strongly encourages States to strengthen civilian control over the military by establishing a functional civilian ministry of defence and to put in place effective oversight, monitoring and accountability mechanisms in the security sector. He notes that civilian oversight is more effective if it is provided through multiple channels.

The Special Rapporteur calls attention to the “prerogatives” of the military, which are a common legacy in countries in transition and include control over various aspects of politics and the economy. He urges States, as part of a preventive strategy, to eliminate such areas of military autonomy.

The Special Rapporteur calls on States with military courts to restrict their jurisdiction solely to members of the military for offences against military discipline.

The Special Rapporteur also notes that security sector reform is a vast and highly developed field of policy and practice and that significantly more work needs to be done to explore the preventive dimension of security sector reform. The Special Rapporteur calls for more discussion and constructive dialogue between the transitional justice and security sector reform communities to overcome misconceptions and explore synergies leading to coordinated actions.