Introduction
The policies and actions of international organisations (IOs) and their partners increasingly affect the lives and rights of individuals and may thus – unintentionally or purposefully – violate core international norms.Footnote 1 Prominent cases are rights violations by United Nations (UN) peacekeepers and their cooperating partners, for example through the sexual exploitation and abuse of women and children, the organisation of human trafficking networks, or the failure to protect civilians from systematic killings.Footnote 2 In other issue areas too, the impact of transnational authority on human rights has become obvious, such as in the implementation of economic programmes funded by the World Bank or of the austerity measures on the demand of the European Union (EU) and the International Monetary Fund (IMF) in response to the economic and financial crisis.Footnote 3
As a result of the growing concern about how to constrain the authority of IOs in order to protect human rights, accountability has become increasingly prominent in International Relations (IR) scholarship on international organisations.Footnote 4 However, traditional accountability mechanisms, often derived from the national (democratic) context, are difficult to achieve in global governance.Footnote 5 Instead, alternative forms of accountability have evolved that include a broader audience than just the institutions’ member states as accountability holders, for example courts, other (regional) organisations or non-governmental organisations (NGOs).Footnote 6 These forms of ‘pluralist accountability’, as I call it, have become a recent trend that deserves further attention in research.
So far, we know little about the nature of, the reasons for, and the effects of pluralist accountability. Therefore, this article establishes a conceptual framework for analysing the various forms of third-party accountability and asks under which conditions we can expect pluralist accountability to evolve in response to human rights violations by IOs and their partners. In particular, I argue that two factors facilitate the development of pluralist accountability. First, the competition for status and resources encourages third parties to act as accountability holders towards IOs and their implementing partners. This argument is based on the research on norm entrepreneurship and regime fragmentation, which has demonstrated the dynamics of competition among NGOs, between NGOs and intergovernmental organisations, and also between courts.Footnote 7 Second, the implementing actors or the mandating authority are inclined to recognise third parties as accountability holders if they are socially vulnerable. This argument is derived from the research about states’ compliance with human rights norms.Footnote 8
In order to illustrate the argument, I provide empirical evidence from a comparative analysis of peace operations in Bosnia Herzegovina and Kosovo, which were implemented by a variety of different actors, such as regional organisations, national, or private actors. The selected cases cover operations that have been alleged to violate different human rights, such as the right to physical integrity, through sexual exploitation, abuse and human trafficking, or the right to due process of detainees. In response to these human rights violations, pluralist accountability evolved; however, not in all of the cases. The findings show that while a competitive environment motivates third parties to act as accountability holders, the vulnerability of the implementing actors, and the mandating authority with regard to human rights demands is crucial for the recognition of pluralist accountability holders.
This article is structured as follows. In the next section, I present the empirical phenomenon of pluralist accountability, outline the theoretical concept used for the analysis, and situate it in the existing literature. Thereafter, I develop the theoretical framework for analysing the conditions for pluralist accountability in global governance. This is followed by an empirical illustration of pluralist accountability in response to human rights violations in peace operations in Bosnia Herzegovina and Kosovo. The concluding section summarises the main argument and discusses the generalisability and the implications of the findings for existing and future research.
Pluralist accountability: Empirical phenomenon and concept
IOs are increasingly scrutinised by a set of different actors with regard to their compliance with human rights. The European Court of Justice’s Kadi decision of 2008 is an important example of how a regional court reviews an IO’s impact on individual rights, in this case the UN Security Council’s sanctions policy.Footnote 9 A similar attempt was undertaken in 2013 by two non-governmental organisations (NGOs) who filed a complaint in the Southern District of New York Federal Court against the United Nations Stabilization Mission in Haiti (MINUSTAH). They claimed that the operation has caused serious harm to the Haitian population by spreading a cholera epidemic, which resulted in the deaths of at least 8,000 people.Footnote 10 This growing trend of third parties, be it international or regional courts, international or regional organisations, or NGOs, to hold IOs accountable for human rights can also be identified in global economic governance and global health governance.Footnote 11
This article puts forward the concept of pluralist accountability to conceptualise and explain this empirical phenomenon. Accountability is defined as the relationship between two actors by which one actor (the accountability holder) sets the standards for another actor’s behaviour, provides a form of review, and sanctions the other (the accountee) if its actions deviate from the standards. Therefore, accountability is conceptualised in terms of three functions of standard-setting, monitoring, and sanctioning in line with existing research.Footnote 12 Through standard-setting, the accountability holder prescribes the norms that the accountee should respect throughout the implementation.Footnote 13 While a delegation contract usually prescribes the tasks to be fulfilled, holding someone accountable for human rights violations implies specifying which norms have to be respected in the course of the implementation. Monitoring defines activities that review whether these standards are respected in the implementing process. Hereby, the accountability holder establishes mechanisms for reporting and surveillance, such as spontaneous on-site visits.Footnote 14 Accountability finally also implies that the accountability holder sanctions the accountee if the latter proves to disrespect the standards set.Footnote 15 This can be normative-political sanctioning, through measures of ‘naming and shaming’, or material sanctioning, such as cutbacks in the budget, the revision, or ultimately the withdrawal of the agent’s mandate.Footnote 16
Depending on which actor exercises the three core accountability functions of standard-setting, monitoring and sanctioning, we can distinguish different types of accountability. In vertical accountability, which has also been named internal or shareholder accountability, the accountability holder is the mandating authority.Footnote 17 In contrast to that, pluralist accountability is characterised by third parties – such as international courts or regional organisations – who act as accountability holders. In pluralist accountability, the accountability holder is independent of the relationship between the mandating authority and the implementing actor. Pluralist accountability holders thus may be transnational, national, non-state or judicial actors who, individually or together, engage in standard-setting, monitoring and sanctioning. This type has also been characterised as external, diagonal, horizontal, or non-institutionalised forms of accountability.Footnote 18 If third parties act on behalf of the ‘real’ accountability holder, that is, the affected people, this has been termed ‘surrogate accountability’.Footnote 19
What are the reasons for choosing the term ‘pluralist’ to describe standard-setting, monitoring, and sanctioning by third parties? First, the existing terms bear connotations that limit the analytical focus to a specific actor type. External accountability, as well as surrogate accountability, imply that affected people are directly involved in the accountability process, which is not always the case in the phenomenon this study intends to describe. Horizontal or diagonal accountability imply that the accountability holder is situated either on the same or on a different level in relation to the implementing actor, which poses limits to which actor can be regarded as an accountability holder. Second, the choice for the term pluralist accountability is inspired by recent debate in International Law scholarship on the relationship between constitutionalism and legal pluralism in light of an increasing fragmentation of international law.Footnote 20 The concept of legal pluralism has been used in particular by scholars in the tradition of the ‘New Haven School’ to highlight the increasingly plural character of actors involved in law-making in contrast to the classical view that emphasised the role of sovereign states for the creation of law.Footnote 21 Similar to the fragmented layers of international law, I consider the term ‘pluralist accountability’ to reflect best the fragmented and heterarchical character of accountability relationships of interest for this study. This concept allows us to analyse the accountability mechanisms enacted by a variety of actors, which would otherwise be overlooked when focusing on traditional conceptions of accountability. It is important to note, however, that the conceptual approach of this article presents a social-political understanding of accountability. It thus relies on a much broader conceptualisation than pure legal accountability, while it does also include legal aspects of accountability.
Traditional vertical accountability implies that standards-setting, monitoring, and sanctioning are undertaken by the delegating authority, that is, the IO’s member states. Pluralist accountability, to the contrary, is characterised by standard-setting by external third parties outside of the original delegation relationship, for example, norm entrepreneursFootnote 22 or other standard-setting actors, such as NGOs, courts, or experts commissions. In the area of peacekeeping, the International Committee of the Red Cross, for example, has become a very important standard-setter with regard to the implementation of operations. In pluralist accountability, third parties empower themselves to collect information and monitor compliance of the agents’ behaviour and provide a forum for review in which the accountee justifies its behaviour to the general public, other external actors, or the mandating authorities. Despite the fact that the most harmful and material sanctions, that is, the termination of the delegation relationship, can be exerted only in vertical accountability, material and normative sanctions can also be imposed by third parties. Pluralist legal theorists, for example, have highlighted the increasingly important role of courts for sanctioning.Footnote 23
The ideal-type complete pluralist accountability as defined in this study encompasses all three functions of standard-setting, monitoring, and sanctioning exercised by third parties. Pluralist accountability remains incomplete if only one or two of the three functions are performed by the accountability holder.Footnote 24 However, standard-setting, monitoring, and sanctioning do not have to be undertaken by one single third party; instead, these three functions can also be exercised by several third parties who all act as accountability holders with regard to the same implementing actor. Of course, in reality, this ideal-type distinction is not always so clear-cut; the third parties that act outside the delegation relationship may be institutionally part of the mandating authority. These mixed forms can be described as ‘delegated accountability’, whereby the mandating authority delegates the exercise of core accountability functions to external third parties that may be pre-existing or even newly created for this purpose. Examples for delegated accountability are the World Bank Inspection Panel that reviews complaints from third parties regarding the Bank’s projects relating to the Bank’s own standards or the UN Office of Internal Oversight Services (OIOS), which is institutionally part of the UN Secretariat, yet can undertake independent monitoring, but has few independent sanctioning powers.
Conditions for pluralist accountability in global governance
Both vertical and delegated accountability have increasingly been studied, such as in recent studies on IOs’ relations with non-state actors and the literature on judicalisation.Footnote 25 In contrast to that, this article focuses on pluralist accountability. In order to understand the conditions for pluralist accountability, I rely on a theoretical framework that looks at both the ‘supply side’ and the ‘demand side’ of accountability.Footnote 26 What are the ‘push factors’ that motivate external actors – NGOs, courts, or regional intergovernmental organisations – to exercise accountability functions? What are the ‘pull factors’ that make the implementing actor or the mandating authority recognise them as accountability holders? I argue that a specific opportunity structure characterised by a competitive environment among third parties and certain characteristics of the implementing actor or the mandating authority, in particular their vulnerability regarding certain norms within a specific community, foster the development of pluralist accountability.
Competition
Research on norm entrepreneurship or regime fragmentation suggests that the particular competition that characterises the institutional context of third parties is an important condition for their participation in pluralist accountability.Footnote 27 This competition among third parties can take place not only between NGOs, but also between NGOs and intergovernmental organisations or between courts, who all conceive themselves as guardians of human rights.Footnote 28 Each organisation in a competitive environment strives to survive and to preserve its reputation by developing a ‘niche capacity’. External third parties such as civil society, courts, or intergovernmental organisations consider themselves as accountability holders to compensate for lacking or ineffective vertical accountability. Regarding the competition between NGOs, this has been described as the ‘independent global civil society effects’ of human rights regimes.Footnote 29 Exercising core accountability functions thus enables an organisation to develop a distinguished profile to separate itself from other concurring parties in the field of human rights. This process of profiling can result both from a rational cost-benefit analysis of how the exercise of accountability could possibly lead to greater material resources and from normative considerations that aim to promote human rights in new areas. Thus, pluralist accountability is enacted if there is an institutional competition that motivates third parties to act as accountability holders (push factor).
This condition requires the following observable implications. First, we expect the institutional environment to be populated by actors who all conceive themselves as norm entrepreneurs in the transnational system and compete with each other about resources and reputation in the same issue area. Moreover, if there is another actor perceived to have an equal standing or if an actor has newly entered the issue area and still has to develop its standing, this can also be taken as an indicator for competition. These indicators can be found in the mandates of the potential accountability holders and also in their discourse directed towards perceived competitors or towards those providing them with resources.
Vulnerability
The research on accountability emphasises the importance of the mutual recognition between the accountability holder and the accountee.Footnote 30 This implies that the standards set by the accountability holder are regarded as legitimate by the accountee, that the accountee subjects itself to a mechanism of review provided by the accountability holder and accepts the other’s sanctioning authority. If we follow this line of reasoning, it is only logical to assume that the characteristics of the implementing agent have an influence on the accountability type. The crucial question therefore is: When do actors recognise third parties as relevant accountability holders and allow them to undertake standard-setting, monitoring, and sanctioning activities?
As the subject of recognition has not yet been treated by the research on accountability, I refer to an argument that has recently been made in the literature on compliance. Even if the underlying logic of action might be different, the conditions for compliance might also be applicable to the question when actors are more likely to recognise accountability holders. According to the authors of the ‘spiral model’, material and social vulnerability are two important scope conditions for actors to comply with human rights norms.Footnote 31 Especially, social vulnerability is highly important as to when actors recognise third parties as accountability holders. Being socially vulnerable implies that the reputation and the social status of an actor within a specific community depend on its respect for certain norms. In contrast to that, an actor is less vulnerable when it can refer to a powerful counter-norm to justify its behaviour.Footnote 32 Based on this argument, several authors explain how the US government was able to undermine the prescriptive status of the anti-torture norm regarding detainees during the Bush era after 2001 through ‘discursive legitimation’.Footnote 33 The US government was able to invoke a powerful counter-norm, namely the fight against terrorism, for the justification of torture of detainees in the denial phase of the spiral model and thus reduced its social vulnerability.Footnote 34
Borrowing this argument, I posit that an actor’s vulnerability or invulnerability influences whether it recognises third parties as accountability holders. Hence, as a second hypothesis, I assume that the implementing actor is more inclined to recognise external accountability holders if it is highly vulnerable in terms of its social status and human rights reputation (pull factor). Based on this logic, I expect the degree of vulnerability of the mandating authority in the delegation relationship fosters the development of pluralist accountability. If the principal, in particular the immediate principal, is vulnerable to human rights demands, it is more inclined to recognise third parties as external accountability holders for the implementing agents. This recognition can in turn indirectly influence the implementing actor’s inclination to recognise external accountability holders.
A common strategy to decrease vulnerability in peacekeeping is to invoke a dilemma between security and human rights. In this context, human rights are conceptualised as private or individual interests that might oppose the general public interest of security and stability as promoted by the peace operation. The main observable implications for the construction of this dilemma are indicators that reflect arguments of exceptional circumstances and operational necessity, which limit the power of human rights within peacekeeping.Footnote 35 Such dilemmas are not exclusive to the peacekeeping domain, but also exist in other issue areas, for example with regard to individual versus group rights or economic development and human rights.
The argument and alternative explanations
Both conditions have independent effects on the evolution of pluralist accountability. The competition among third parties influences the motivation of third parties to act as accountability holders. The vulnerability of the implementing actors instead fosters the recognition of pluralist accountability by the implementing actors and the mandating authority. The argument of this study thus contains a combined theoretical approach, considering both the ‘demand side’ in terms of the institutional opportunity structure and the ‘supply side’ in terms of specific actor characteristics. However, two potentially relevant alternative explanations from other important research strands need to be taken into consideration when examining the conditions for pluralist accountability.
First, we could also expect path dependence to explain the turn to pluralist accountability.Footnote 36 Path dependence would imply that the principal’s past decisions, for example when mandating the agent to implement a task, influence the nature of the accountability regime. Path dependent effects could also influence a third party’s decision to act as an accountability holder. To control for possible path dependent effects, the analysis compares where the same institution acts as the mandating authority in the delegation relationship. I evaluate in each case through process-tracing whether the prior accountability regime at the outset of the delegation contract has had an impact on the nature of the enacted accountability after human rights violations have become apparent.
Second, researchers have identified the rise of a general norm in terms of more democratic global governance.Footnote 37 Based on this assessment, we should also expect a possible trend towards pluralist accountability mechanisms, where legitimate governance depends on the approval by transnational societal actors that share a common concern about compliance with human rights regarding the implementation of IO policies.Footnote 38 In contrast to that, however, the same normative trend also has increased expectations that IOs ensure that they respect the rule of law. This, in turn, results in the reinforcement of vertical accountability mechanisms in the context of a development towards a global administrative law.Footnote 39 Therefore, while the concern for greater human rights protection certainly increases the demand for accountability in general, the norm for a more participatory global governance does not allow us to draw any unambiguous conclusions about the type of accountability and the particular conditions for pluralist accountability.
Research design, case selection, and methods of data acquisition
The field of peacekeeping is particularly apt to illustrate the complexity of today’s global governance and the challenges for accountability. Peace operations today are characterised by multiple delegation steps that involve prolonged delegation and multiple parallel authorities. Moreover, peacekeeping operations involve both formal and informal acts of delegation, which is rather common in global governance. The privatisation of peacekeeping, as it happened in the operation in Bosnia, has increased since, and thus continuously poses challenges for accountability.Footnote 40 Peacekeeping also very pointedly illustrates how global governance activities can infringe upon international norms, in particular through violating human rights.
As there is no systematic dataFootnote 41 on alternative accountability forms, I illustrate the argument with evidence from a qualitative comparative case study on the development of pluralist accountability in response to human rights violations in UN peace operations in Bosnia Herzegovina and Kosovo. Both operations mandated by the UN Security Council between 1996 and 2004 were accused of human rights violations committed by international staff, such as human trafficking and the violation of the right of detainees.Footnote 42 I hereby focus on the violation of rights that are no longer legally disputed, such as the right to physical integrity and the right to fair trial and due process, which should make accountability more likely to occur. Most importantly, these operations display a clear variance with regard to whether or not pluralist accountability evolved within the same country. The case selection thus follows the logic of variation on the dependent variable and juxtaposes two pairs of cases that vary with regard to the development of pluralist accountability. In Bosnia, pluralist accountability evolved in response to the involvement of private security actors and the UN-mandated International Police Task Force in human trafficking. In the same country, however, no pluralist accountability evolved in response to the violation of the rights of detainees by the NATO-led Stabilization Force (SFOR). In Kosovo, pluralist accountability can be identified in the case of detentions by the UN international administration (UNMIK), but not in the case of detentions by the NATO-led operation KFOR (Table 1).
Table 1 Selected cases.
Each case study contains an analysis of the accountability regime at the outset of the operation, in order to control for the potential influence of path dependency. I then examine each pair of cases through the method of process-tracing as to how the two conditions, that is, competition and vulnerability, influenced the evolution of pluralist accountability. The analysis of causal mechanisms has been considered as particularly suitable for theory-building, in particular the development of middle-range theories.Footnote 43 Regarding the first condition, competition, I examine how the process of profiling contributes to the rise of pluralist accountability. The analysis of the second condition, vulnerability, investigates how the process of recognition leads to the acceptance of pluralist accountability holders.
The empirical data are gathered from a variety of sources. Primary documents, made available by the different UN bodies, such as the UN Security Council and the Secretariat, and the individual operations, are one important source. They include official resolutions, meeting protocols, draft resolutions, reports, commissioned studies, and internal documents as far as access was provided. Further evidence is gathered from media and NGO reports. To complement this documentary analysis, I draw on a number of qualitative expert interviews with present and former UN staff, member states representatives and members from regional international organisations (for example, the Council of Europe), research institutes, and non-governmental organisations. Based on these various sources, this study provides an original comprehensive picture of accountability dynamics in peace operations.
Accountability for human rights violations in the context of the operation in Bosnia: Human trafficking and detentions
After the Dayton Peace Agreement of 1995, the UN Security Council mandated the UN Mission in Bosnia and Herzegovina (UNMIBH), an International Police Task Force (IPTF), and the NATO military operation (SFOR), to promote stability and human rights in the postconflict environment.Footnote 44 Besides this complex network of implementing actors, the operation in Bosnia was the first peacekeeping operation to employ a security company, Dyncorp, to provide personnel for SFOR and the police force IPTF.Footnote 45 With the arrival of peacekeeping forces, Bosnia turned into a flourishing business place for prostitution and trafficking within Europe.Footnote 46 A significant number of IPTF monitors and SFOR private contractors frequented brothels where trafficked women were held, protected the owners from raids and sometimes even purchased trafficked women themselves on the pretext of freeing them.Footnote 47 This stands in stark contrast to the UN’s core objectives and the Security Council’s mandates for the operations in Bosnia. In addition to that, the NATO-led operation SFOR was accused of violating the rights of detainees, in particular due process rights such as the right to counsel and independent review.Footnote 48
The accountability framework at the outset of the operations was rather limited. The applicable UN Security Council resolutions 1035 and 1088 did not mention any human rights standards that should guide the implementing actors’ behaviour. A Code of Conduct for international civilian police actors that had been introduced by the preceding operation UNPROFOR in 1993 did not play any role for the successor mission UNMIBH and its IPTF, let alone contractors.Footnote 49 The Special Representative of the Secretary-General (SRSG) in Bosnia remained the most important channel for the Security Council to review and monitor IPTF behaviour.Footnote 50 The possibilities for vertical sanctioning were limited by diplomatic and functional immunity regulations.Footnote 51 First attempts to enhance vertical accountability were undertaken by two ‘whistleblowers’ within UNMIBH, who publicly exposed the behaviour of their colleagues and demonstrated how the mission leadership covered up their investigations.Footnote 52 Senior officials within the IPTF and UNMIBH, however, asked the private security company to end their contracts and thus put vertical accountability to a halt.
Instead, pluralist accountability evolved with other actors outside of the original delegation relationship enacting accountability functions. The local office of the High Commissioner for Human Rights (OHCHR) in Bosnia set up a Joint Trafficking Project and established human rights standards for IPTF and SFOR personnel.Footnote 53 It systematically collected information about the relationship between trafficking and the international presence and hired the ‘whistleblower’ Kathryn Bolkovac to investigate the involvement of IPTF officers. Most importantly, the office also engaged in sanctioning and supported Bolkovac in a court trial against Dyncorp.Footnote 54 When the British court decided that the dismissal of Bolkovac was unlawful, the decision also publicly exposed the role of the UN leadership.Footnote 55 This prompted further attempts of review by the US Congress through hearings and legislative acts.Footnote 56
Acting as an accountability holder became an important mechanism of profiling for the OHCHR. The organisation resulted from the Vienna Conference on Human Rights in 1993, where NGOs lobbied for a more effective supranational institution monitoring the respect of states for human rights. Within the Geneva-based human rights institutions, however, the OHCHR as a new institution still had to find its niche, especially in light of the notorious budget and personnel shortages that accompanied the office from the beginning.Footnote 57 Concurring institutions were, in particular, the human rights offices within individual peace operations and the International Organization for Migration (IOM).
Despite the presence of a human rights office within UNMIBH, the OHCHR also maintained a separate office in Bosnia. This parallel maintenance of OHCHR field offices and human rights offices in peace operations has often raised concerns about efficiency. Later initiatives sought to integrate the human rights offices in the different UN field presences.Footnote 58 This separate mandate, however, enabled the OHCHR field office to act as an independent accountability holder. It could thus monitor the UN peacekeeping presence and report directly to the headquarters in Geneva without having to subordinate itself to the UNMIBH leadership. Acting as an independent monitoring body, the OHCHR field offices could demonstrate their relevance and even provide the Geneva ‘human rights branch’ with more leverage and a distinct profile vis-à-vis the New York peacekeeping department.
Competition also took place between the OHCHR and the IOM, as the latter already had the issue of human trafficking in its portfolio, and became particularly strong at the headquarters level in Geneva.Footnote 59 In order to outline its distinctive profile, the OHCHR adopted a human rights based approach to distance itself from the IOM’s focus on the repatriation of trafficking victims. The High Commissioner Robinson supported the local Bosnian office with a trust fund, to ‘provide emergency support to trafficked women as well as to provide grants to local NGOs working in this area’, which exactly characterised what the IOM was also doing in Bosnia. Robinson further strongly encouraged the office in Bosnia to expose the involvement of IPTF and UNMIBH members in human trafficking.Footnote 60 In light of this institutional competition, the office in Bosnia was eager to demonstrate that its role in the fight against human trafficking was more important and more successful than what was being done by the IOM.Footnote 61 By setting standards for international personnel and monitoring their compliance, the OHCHR thus developed a distinct profile as an accountability holder with a core focus on human rights that distinguished itself from the IOM.
While the competitive environment was important for the rise of accountability holders, the vulnerability of the implementing actors and the mandating authority with regard to human rights demands was crucial for the recognition of pluralist accountability holders. Especially the vulnerability of the IPTF and the UN as a whole was particularly high, as the UN’s reputation in Bosnia had already suffered because of UNPROFOR’s failure to protect civilians from the massacre in Srebrenica. Vis-à-vis the Bosnian police forces, IPTF officers were expected to act as role models in terms of their respect for human rights.Footnote 62 Despite the UN’s strategy to decrease its vulnerability by covering up and blame-shifting, this was finally also recognised by the UN itself: ‘Peacekeepers’ use of trafficking victims for sexual and other services has been a source of major embarrassment and political damage to U.N. PKOs [… and] can taint entire missions.’Footnote 63
In contrast to the case of human trafficking, pluralist accountability failed to evolve in response to allegations of torture and the violation of the right of due process of detainees by SFOR. Some attempts at standard-setting and monitoring were undertaken by NGOs and the Ombudsman for Human Rights in Bosnia.Footnote 64 In addition to that, the Bosnian Human Rights Chamber, the Constitutional Court, and the Bosnian Supreme Court issued decisions in which they sanctioned the detention practices of SFOR, even if SFOR was not the appellant of the court file.Footnote 65 However, SFOR for the most part simply ignored the domestic legal sanctioning exercised by Bosnian courts and continued its detention practices without modification.Footnote 66
Why did pluralist accountability fail to evolve in the case of detentions by SFOR? First, there were no concurring transnational non-governmental or intergovernmental human rights organisations that competed to protect the rights of detainees. The OHCHR had already made the fight against human trafficking the priority of its work in Bosnia, even if it occasionally criticised the lack of respect for the rule of law when detainees were transferred to SFOR custody.Footnote 67 Other organisations such as the OSCE or the Council of Europe, who played a significant role regarding detentions in Kosovo (see below), showed no ambition to act as accountability holders with regard to SFOR detentions.Footnote 68 Some degree of competition can only be identified within the domestic realm. The newly established Bosnian courts needed to develop a distinct profile in the fresh institutional landscape after the Dayton Agreement and to prove their independent standing vis-à-vis the Bosnian government in terms of a democratic separation of powers and the protection of constitutional rights. The courts decided to do this by criticsing the excess of authority of the international presence and the lack of protection provided by the Bosnian state.Footnote 69 Thus, while on the international level no significant movement advocated for the rights of detainees in Bosnia, the competition among judicial actors nevertheless stimulated attempts at pluralist accountability at the domestic level.
The second condition for pluralist accountability, the implementing actor’s vulnerability with regard to human rights demands, remained rather low in this case as SFOR was very successful in keeping its detention practices immune to human rights demands. The Security Council in its Resolution 1031 required the SFOR contributing states to cooperate with the International Criminal Tribunal for the former Yugoslavia (ICTY), which had been established for the prosecution of war crimes.Footnote 70 When detention practices increased after 1997, NATO’s cooperation with the court thus provided the argumentative basis for refusing to recognise accountability holders that would monitor and sanction SFOR detention.Footnote 71 As the international community had pressured SFOR to cooperate with the ICTY in arresting war criminals, and with the failure of the international community to hinder the massacre of Srebrenica still in everybody’s mind, nobody dared to question detentions in relation to their compliance with human rights.Footnote 72 SFOR was further able to make itself invulnerable to any human rights demands by invoking security as a counter-norm, and could thus refuse to recognise accountability holders that attempted to monitor and sanction its detentions practices.Footnote 73 Detentions were thus regarded as part of the military task necessary for a successful implementation of the mandate, in contrast to the understanding that prevailed prior to 1997, which considered detentions to be part of a civilian mandate to be fulfilled by the Bosnian authorities.Footnote 74
To sum up, the two operations in Bosnia illustrate the development of pluralist accountability. In the first case, the competitive environment of the OHCHR in Geneva and in Bosnia stimulated the office to develop a distinct profile as an accountability holder vis-à-vis the international presence and its involvement in human trafficking. The increasing vulnerability of the UN and its IPTF officers motivated them to recognise pluralist accountability holders. In the second case, to the contrary, there was no competitive environment that could have stimulated external third parties to act as accountability holders for the rights of detainees. Only within the domestic context, there were some elements of competition: Bosnian courts issued decisions on detentions by SFOR to demonstrate their independence and to develop their profile in a newly created legal landscape. Nevertheless, the detaining authorities managed to preserve their invulnerability with regard to human rights demands by framing detentions in the context of an enhanced cooperation with the International Criminal Tribunal for the former Yugoslavia and by invoking the counter-norm of security after 2001. As a consequence, SFOR detentions in Bosnia remained mostly unchallenged with regard to their compliance with human rights standards.
Pluralist accountability regarding UNMIK and KFOR detention practices in Kosovo
The history of international territorial administration culminated in the transitional administration in Kosovo. Under Resolution 1244, the UN Security Council established the UN Mission in Kosovo (UNMIK) with a mandate that included a wide range of state-building tasks.Footnote 75 In addition to that, the Security Council mandated NATO to deploy a military force, which became operation KFOR. Both operations were equipped with a high degree of authority, which had a direct impact on the rights and lives of individuals. The violation of the rights of detainees was the most direct result of this authority.
Between 1999 and 2005, several types of arrests and detention were carried out by UNMIK and KFOR. As the civilian authority in Kosovo, the UN Special Representative of the Secretary-General issued executive orders to detain arrested individuals who, according to international law, would otherwise have had to be released.Footnote 76 KFOR detentions were, in comparison with UNMIK detentions, much higher in number. More than 3,500 individuals had been detained in extra-judicial detentions between 2001 and 2003 at Camp Bondsteel, a detention facility run by the US contingent.Footnote 77 Both UNMIK and KFOR detentions were accused of violating international human rights standards as outlined in the International Covenant on Civil and Political Rights (Art. 9) and the European Convention on Human Rights (Art. 5). Besides the right to information about the reasons for the detention, the right to compensation in the case of illegal detentions and the right to a timely judicial review, especially the lack of the right to habeas corpus, that is, the right to challenge the legality of the detention before a judicial body, prompted major concerns.Footnote 78
Resolution 1244 provided only a very limited accountability framework regarding detentions. The Security Council broadly stated that protecting and promoting human rights was part of UNMIK’s main responsibilities and requested regular reports from the Secretary-General, but did not outline specific standards for the mission’s own behaviour or for KFOR.Footnote 79 More concrete standards for UNMIK’s behaviour were outlined by the Secretary-General, who adopted a bulletin on the ‘Observance by United Nations Forces of International Humanitarian Law’, which guaranteed the International Committee of the Red Cross (ICRC) access to detention facilities.Footnote 80 For the first time, legal standards were outlined in a binding document that addressed the behaviour of the UN’s own personnel. The Secretary-General further determined that ‘[i]n assuming its responsibilities, UNMIK will be guided by internationally recognised standards of human rights.’Footnote 81 The general monitoring of human rights in Kosovo was undertaken by the OSCE.Footnote 82 However, these monitoring activities were meant to be directed regarding national actors and former conflicting parties, not the international partner institutions. Vertical sanctioning however was restricted by immunity regulations that applied to UNMIK and KFOR personnel.Footnote 83
This limited vertical accountability regarding detentions spurred strong criticism. Substantial critique came from the Ombudsperson Institution in Kosovo (OIK), which was established in 2000 to review the actions of UNMIK.Footnote 84 The OIK was a result of pluralist accountability exercised by the wider transnational community, in particular the OSCE and the Council of Europe. Although the OSCE had not been mandated to monitor UNMIK or other parts of the international administration in Kosovo, it became an important accountability holder regarding UNMIK’s compliance with human rights norms when undertaking detentions.Footnote 85 Together with the Council of Europe, the OSCE established the Ombudsperson Institution in Kosovo, which was mandated to take complaints and conduct investigations of human rights abuses.Footnote 86 Funded by OSCE member states, the OIK was asked to regularly report to the OSCE Permanent Council. Thus, UNMIK was independently monitored with respect to a concrete set of human rights standards.Footnote 87 When the OIK was turned into a national institution, the Council of Europe pushed for the establishment of a Human Rights Advisory Panel. Contrary to the original proposal, however, the panel remained mainly advisory in its character.Footnote 88
In addition to that, the OSCE office in Kosovo established the OSCE Legal Systems Monitoring Section (LSMS) as a reviewing body with a ‘quasi-independent role to monitor the observance of human rights in Kosovo’.Footnote 89 OSCE staff investigated conditions of detainees in UNMIK facilities and reviewed the executive orders issued by the SRSG. Moreover, the OSCE LSMS developed a forceful mechanism for normatively sanctioning UNMIK’s detention practices through the regular publication of reports and reviews in surprisingly frank language.Footnote 90 It thereby established a link between UNMIK’s detention practices and the organisation’s legitimacy both from a human rights law perspective and the viewpoint of the governed people.
Why did the OSCE on its own initiative decide to become an accountability holder towards UNMIK in the case of detentions? Within the complex structure provided by Resolution 1244, UNMIK, the EU, and the OSCE operated in an environment of institutional competition for resources and recognition. Therefore, the OSCE in Kosovo had to develop competences that rendered its role distinct from the other organisations. Relying on its mandate from the OSCE Permanent Council, the OSCE Mission in Kosovo used its expertise to sharpen its profile by acting as an accountability holder regarding UNMIK’s detention practices. The OSCE LSMS’ outspoken criticism, however, also implied the danger of becoming marginalised within UNMIK’s policymaking process.Footnote 91 The establishment of an ombudsperson institution was therefore an important means for the OSCE to demonstrate its distinctiveness without marginalising itself.Footnote 92
The development of pluralist accountability regarding UNMIK detentions furthermore depended strongly on UNMIK’s vulnerability regarding human rights demands. The Special Representative of the Secretary-General defended UNMIK’s detention policy by pointing out that there was a trade-off between human rights and security in circumstances of emergency, when ‘international human rights standards accept the need for special measures’.Footnote 93 He thus attempted to invoke the counter-norm of security as a shield against claims requesting UNMIK to protect the rights of detainees. After the early successes in establishing a stable security environment in Kosovo, however, it was difficult for UNMIK to uphold this argument credibly, since there was no real emergency any more.Footnote 94 Instead, UNMIK’s own priorities of promoting the rule of law and justice throughout Kosovo made it increasingly vulnerable to human rights demands. This growing vulnerability eventually led the organisation to recognise pluralist accountability exercised by the OSCE and the Council of Europe and to agree to establish the OIK. By the time the OIK was turned into a purely national institution and lost all competences over UNMIK, however, renewed violence between Serbs and Albanians rendered the provision of security the overarching norm that determined UNMIK’s success in Kosovo.Footnote 95 This enabled UNMIK to downsize the Council of Europe’s proposal of a Human Rights Advisory Panel after the nationalisation of the OIK.
In the case of KFOR detentions, no comparable system of pluralist accountability evolved, despite the same competitive environment. Like UNMIK, KFOR was confronted with attempts by the OSCE and the Council of Europe to exercise pluralist accountability.Footnote 96 The OSCE argued that KFOR’s mandate could only be fulfilled ‘through compliance with international human rights standards’.Footnote 97 In its public reports, the OSCE went so far as to demand that ‘KFOR cease[d] detentions forthwith and officially renounce[d] its authority in this area.’Footnote 98 This was a strong claim from an organisation that was an implementing partner of KFOR. However, KFOR made itself invulnerable to human rights demands by successfully invoking security as a counter-norm. The KFOR commander determined in directive no. 42 of 9 October 2001 that his ‘authority to detain is a military decision, not a judicial one’.Footnote 99 He argued that extra-judicial detentions were a necessary means for the successful implementation of the security-related aspects of its mandate provided by the Security Council Resolution 1244.Footnote 100 KFOR thus managed to refuse pluralist accountability that would not be under KFOR control.Footnote 101
In sum, the analysis of the two operations in Kosovo has revealed important insights into the conditions for the development of pluralist accountability, in which actors outside of the original delegation relationship enacted accountability functions. The competitive institutional environment stimulated third parties to sharpen their profiles by engaging in standard-setting, monitoring, and sanctioning activities. In the case of UNMIK, the OSCE and the Council of Europe acted as accountability holders and established the Ombudsperson Institution. However, no similar accountability regime evolved in the case of KFOR because KFOR successfully invoked security as a counter-norm and thus made itself invulnerable to human rights demands. This did not work in the same way for UNMIK. The discrepancy between its mandate and its detention practices made it impossible for UNMIK to prioritise security over human rights, at least in the years before the outbreak of violence in 2004. UNMIK’s high degree of authority, which effectively allowed the SRSG to run the whole country on its own, increased its vulnerability. Acting like a government entailed certain human rights responsibilities – this was the argument of third parties, who thus forced UNMIK to recognise external accountability holders.
Conclusion
This article addresses the empirical phenomenon that actions and policies of IOs are increasingly scrutinised by third parties. This is a sign that human rights violations committed by IOs and their partners do not go unnoticed. In this article, I developed the concept of pluralist accountability in order to systematically assess the various standard-setting, monitoring, and sanctioning activities of third parties after allegations of human rights violations by international organisations and their implementing partners have become public. Given the complexity of global governance, this new analytical tool is necessary to understand the accountability dynamics in the absence of traditional, vertical accountability. I argued that two factors contribute to the development of pluralist accountability, namely the competitive environment of third parties and the vulnerability of the implementing actor or the mandating authority.
The empirical analysis provides first empirical insights into pluralist accountability in UN-mandated peace operations, an issue area which most pointedly reveals the new intrusiveness of complex global governance for individual citizens. In particular, the analysis has shown how pluralist accountability evolved in the cases of human trafficking in Bosnia and UNMIK detentions in Kosovo and how it failed to evolve in the cases of SFOR detentions in Bosnia and KFOR detentions in Kosovo. The competitive environment stimulated the newly established Bosnian OHCHR to become an accountability holder vis-à-vis the international personnel involved in human trafficking. Similarly, the competition between regional institutions and the UN in Kosovo, in particular, made the OSCE and the Council of Europe push for the establishment of an Ombudsperson Institution and act as accountability holders towards UNMIK. The vulnerability of the implementing actors and the mandating authority with regard to human rights demands in both cases contributed to the recognition of pluralist accountability. In the cases of the violation of the rights of detainees by SFOR and KFOR, to the contrary, competition was rather weak as there were no concurring transnational non-governmental or intergovernmental human rights organisations that competed to promote the rights of detainees. In addition to that, the two operations made themselves invulnerable with regard to human rights demands and thus hindered the evolution of pluralist accountability.
The juxtaposition of the two detention cases demonstrates that the nature and character of human rights violations does not make any difference for the evolution of pluralist accountability. While this might be different for vertical accountability,Footnote 102 the cases did not provide any systematic evidence that certain types of human rights violation, for example, due process rights, are more likely to be responded to by pluralist accountability. Instead, we can see a clear variance in the outcome variable that could not be attributed to different types of rights violations but could be traced back to the two conditions of competition and vulnerability. Furthermore, the analysis has also shown that pluralist accountability is not automatic if vertical accountability is lacking. The evolution of pluralist accountability in all the cases has been independent from the accountability regimes that existed at the outset of the delegation relationships. This should reassure those who consider non-vertical accountability to be only the second-best option and who fear that pluralist accountability might replace important vertical accountability mechanisms.
The argument of this article bears several implications for contemporary IR research, in particular the literature on delegation, IOs, and their institutional environment. Most of the delegation literature so far has focused on the first delegation step, whereby IOs implement the rules for cooperation agreed upon by their member states.Footnote 103 Only recently have researchers started to analyse further delegation steps, where IOs turn from agents into autonomous principals. The results of this study demonstrate the implications of increasing IO authority for accountability: on the one hand, greater authority has created new dilemmas for accountability; on the other hand, greater authority enables IOs to engage in pluralist accountability vis-à-vis each other. This article thus takes forward the contextualisation of IOs beyond their relationships with member states and speaks to the results of recent research on the interaction of IOs with non-state actors and regional organsations, which demonstrate that the institutional environment in which IOs operate is too important and too varied to be reduced to state-IO relations only.Footnote 104 Pluralist accountability goes hand in hand with the trend of the ‘opening up of IOs’ in certain steps of the policy cycle.Footnote 105 Recognising them as accountability holders might be one way to ensure IOs interact with outside actors.
Finally, this article paves the way for several strands of future research. First, given that this article focused on pluralist accountability, further research is needed to compare different accountability types. In order to develop an encompassing theory of accountability in global governance, we need to know more about the conditions for each accountability type and the circumstances under which they enhance, compensate, complement, or necessitate each other. Second, the conditions for accountability also need to be examined in other issue areas of global governance. Possible future studies could, for example, look into the accountability regimes that evolved in response to the World Health Organization’s emergency governance or new modes of governance as public-private partnerships. Third, this article raises questions regarding the impact of pluralist accountability on the legitimacy of IOs. In the scholarly discussion on IO legitimacy and legitimation, accountability is often advocated as a means to enhance legitimacy.Footnote 106 So far, however, it remains unclear what the impact of different accountability types on legitimacy is. Last but not least, it is important to highlight that pluralist accountability does not imply that there will be fewer human rights violations. This article has analysed the conditions under which pluralist accountability evolves; however, the question of what extent pluralist accountability can effectively constrain the actors held accountable remains outside of the scope of this article. Which accountability type provides the most effective human rights protection therefore remains an open question to be dealt with by future research.
Acknowledgements
I would like to thank Monika Heupel, Thomas Risse, Michael Zürn, and my former colleagues at the WZB ‘Global Governance’ unit, as well as Jonas Tallberg and his colleagues at Stockholm University for their helpful comments on various drafts of this article. I also thank the two anonymous reviewers for their questions and comments as well as Hannah Bliersbach and Niels van Doesum for their editorial support. Last but not least, my thanks go to my interview partners for their willingness to share their insights.
