International Responsibility in United Nations Peacekeeping Operations: Command and Control Arrangements and the Attribution of Conduct
Article 5 of the International Law Commission’s Draft Articles on the Responsibility of International Organizations prescribes the use of the ‘effective control’ test to determine the attribution of conduct of United Nations peacekeeping forces. A close examination of UN command and control arrangements, however, suggests that art 5, as it is currently understood, may not allow for the comprehensive attribution of conduct as it does not fully take into account the complex arrangements governing the employment of military contingents contributed by UN member states to UN peacekeeping operations. In this regard, the premise upon which art 5 is based may need to be revisited and a new approach considered, to ensure that the conduct of such forces is accurately imputed and that responsibility is correctly attributed to the actors concerned.
Article 5 of the International Law Commission’s (‘ILC’) Draft Articles on the Responsibility of International Organizations (‘Draft Articles’) utilises the ‘effective control’ test to attribute the conduct of state organs placed at the disposal of international organisations (‘IO’s).2 Although only adopted provisionally by the ILC, art 5 — or the principle it reflects — has been invoked by a number of international and national tribunals in recent cases pertaining to the conduct of international military forces in peace operations. These have included the European Court of Human Rights in Behrami and Saramati,3 the House of Lords in Al Jedda4 and the District Court in The Hague in HN v The Netherlands,5 giving art 5 significant importance in terms of its legal application. As peace operations continue to expand in scale, scope and complexity, the issue of the attribution of the conduct of international military forces deployed in such operations is likely to continue to feature before such tribunals. This is particularly due to the corresponding increase in the range of situations in which force will be employed as well as the intensifying attention in recent years on violations of international humanitarian law (‘IHL’) and international human rights laws (‘IHRL’) and other misconduct by such forces.
In this regard, this commentary seeks to reflect on the efficacy of art 5 in the attribution of conduct — one of the two constitutive elements of international responsibility — with respect to UN peacekeeping operations (‘PKOs’), focusing in particular on its correlation with UN command and control arrangements.6 The discussion will focus solely on the attribution of the conduct of formed ‘blue-helmet’ military contingents in UN PKOs, and not that of UN military observers, civilian police or civilian personnel, which have a different status as experts or officials on mission and are governed by different rules and arrangements, for instance, in terms of privileges and immunities.7 Non-UN led peace operations will be addressed only in so far as their command and control arrangements provide relevant insights, although the conclusions suggested in this commentary are also relevant to such peace operations. The commentary begins with a review of the law governing the attribution of the conduct of peacekeepers (Part II) as well as the UN’s stance on international responsibility (Part III), before moving on to consider command and control arrangements in the current UN peacekeeping context (Part IV) and prevailing thinking and practice on the attribution of conduct (Part V). Finally, it discusses whether the UN can really have ‘effective control’ of peacekeepers in the sense envisaged by art 5 (Part VI) and if there is a need for a new rule or approach for attributing the conduct of peacekeepers (Part VII).
The Law on Responsibility of International Organisations and the Attribution of Conduct of Peacekeepers
The law on the responsibility of IOs is yet to be fully developed, although some progress has been made by bodies such as the International Law Association (‘ILA’) and, crucially, the ILC.8 The ILC has to date provisionally adopted the Draft Articles9 as well as a complete set of Draft Articles on State Responsibility,10 some of which also touch on the responsibility of IOs. Among all of these draft articles, art 5 of the Draft Articles (as the ILC explicitly indicated in its ‘Commentary’) is the provision most relevant in terms of imputing the conduct of UN peacekeepers.11 Article 5 sets out the principle governing the attribution of conduct, which includes acts and omissions,12 of organs placed at the disposal of an IO by a state.13 Peacekeepers, as members of the military forces of states, are undoubtedly elements of state organs, but are placed at the disposal of the UN by their states.14 Article 5, which was adopted in June 2004 (and its ‘Commentary’ in August 2004), provides that:
The conduct of an organ of a State ... that is placed at the disposal of [an] international organization shall be considered under international law an act of the latter organization if the organization exercises effective control over that conduct.15
As the ILC’s ‘Commentary’ made clear, the ‘effective control’ test is not applied generally to the overall conduct of the organ, but rather to each specific unlawful act, in order to verify if the act in question of the organ was performed under the control of the IO or the sending state.16 If the unlawful act was executed on the instructions of the sending state, the conduct should be attributed to the state.17 If carried out on the direction and control of the IO, the act should be imputed to the IO. Although art 5 has only been provisionally adopted by the ILC, the rule that it espouses — the application of the ‘effective control’ test to determine the imputability of conduct — is based on a principle held by several scholars, and IOs such as the UN.18 The ‘effective control’ test itself, as a test for the attribution of conduct, has been affirmed by a number of tribunals, including the ICJ in cases such as Nicaragua19 and Bosnian Genocide.20 Article 5, or the principle it reflects, has also been invoked in a number of recent cases, such as Behrami and Saramati, Al-Jedda and HN v The Netherlands, altogether giving it significant weight.
The position that the ILC has taken in drafting art 5 is interesting in view of the UN Secretariat’s position on the legal status of peacekeeping forces. The Secretariat stated in February 2004 (in response to an ILC invitation to comment on the issue of the responsibility of IOs) that ‘[a] UN peacekeeping force established by the Security Council or the General Assembly is a subsidiary organ of the UN’.21 The ILC had already dealt with the conduct of organs of an IO in art 4 of the Draft Articles, which provides that:
The conduct of an organ ... of an international organization in the performance of functions of that organ ... shall be considered as an act of that organization under international law whatever position the organ ... holds in respect of the organization.22
Evidently, the ILC felt that a separate article needed to be drafted to take into account the sort of situation that UN peacekeepers and other similar state organs find themselves in, and that art 4 would not apply, notwithstanding the UN’s view that UN PKOs are subsidiary organs of the UN. In the ILC’s view, art 5
deals with the ... situation in which the lent organ or agent still acts to a certain extent as organ of the lending State ... This occurs for instance in the case of military contingents that a State placed at the disposal of the [UN] for a peacekeeping operation, since the State retains disciplinary powers and criminal jurisdiction over the members of the national contingent.23 It added that ‘[a]ttribution of conduct to the contributing state is clearly linked with the retention of some powers by that state over its national contingent and thus on the control that the state possesses in the relevant respect’.24 It also cited the fact that peacekeepers, in some instances, persisted in seeking instructions from their governments before executing UN orders, and that governments remained responsible for enforcing compliance by their peacekeepers with certain convention obligations, to illustrate that states could continue to exercise some control over their lent organs.25 Nonetheless, as the provisional adoption of art 5 illustrates, these considerations have patently not stopped the ILC, scholars and jurists from assuming that the UN had the authority and ability to exercise ‘effective control’ over peacekeepers, even as they believed that the UN would not have the same degree of control and powers over peacekeepers as it would over its subsidiary organs — a critical issue which will be addressed in later parts below.
The UN and International Responsibility
The UN has, in general, assumed responsibility for the activities of UN peacekeeping forces. The UN perceives its international responsibility to be an aspect of its international legal personality and its capacity to bear international rights and obligations. It has noted that the principle that a breach of an international obligation attributable to the IO entails the responsibility of the IO and its liability in compensation is widely accepted.26 Thus, the UN Secretariat has stated that:
As a subsidiary organ of the United Nations, an act of a peacekeeping force is, in principle, imputable to the Organization, and if committed in violation of an international obligation entails the international responsibility of the Organization and its liability in compensation.27
Along this line, on several occasions — such as with respect to the actions of peacekeepers participating in the UN Operation in the Congo (‘ONUC’) and the UN Peacekeeping Force in Cyprus — the UN acknowledged its international responsibility for the conduct of these peacekeepers.28
The Secretariat did clarify that the international responsibility of the UN is conditioned upon the assumption that the operation in question is executed under its ‘exclusive command and control’ and hence considered a subsidiary organ of the UN.29 That is, only PKOs conducted under UN command and control have the legal status of a UN subsidiary organ.30 In other operations, such as joint operations, responsibility lies where ‘effective command and control’ or ‘operational command and control’ is vested.31 Presumably, the Secretariat is referring to the ‘effective control’ test in the latter.32 The Secretariat does not appear to have considered the prospect of dual or joint command and control, whether in UN PKOs or in joint operations. This is perhaps due to the alien nature of such a notion in military operations, given the concept of unity of command (that is, a single chain of command).
Critically, the UN Secretariat does not seem to accept — at least overtly, as the ILC has — the possibility that peacekeepers may not always act under UN direction and could sometimes act upon national direction. As indicated earlier, the ILC seemed to suggest in its ‘Commentary’ to art 5 that while the UN assumed that it had command and control of UN peacekeeping forces, this was, in reality, not always the case. The Secretariat has expressly noted that:
military personnel placed by Member States under UN command although remaining in their national service, are for the duration of their assignment to the force, considered international personnel under the authority of the United Nations and subject to the instructions of the Force Commander ... [and] bound to discharge their functions with the interest of the United Nations only in view.33
In this context, the UN has assumed responsibility for the activities of peacekeepers ‘in the performance of their duties’, which in contrast to the ILC’s position, suggests that the UN believes that the principle captured in art 4 of the Draft Articles — that conduct can only be attributed to the IO when the organ acts in ‘the performance of the functions of that organ’ — is the applicable principle.34 Similarly, the UN has accepted responsibility for the ultra vires acts of peacekeepers in ONUC and the UN Emergency Force who opened fire and killed civilians without any orders to do so.35 This practice is reflected in art 6 of the Draft Articles, which states that conduct that exceeds the authority of the organ or contravenes instructions, remains attributable to the IOs so long as the acts are performed in the organs’ official capacity — when the organ ‘acts in that capacity’.36 Conversely, the UN has stated that it has no liability for the ‘off duty’ conduct of peacekeepers, whereby it considered ‘off-duty’ to mean that the peacekeepers were acting in a ‘non official/non operational capacity’.37 Such conduct would not be attributable to the UN.
Consequently, it would appear that acts of sexual exploitation and abuse, currently a major UN concern, or other forms of misconduct such as human trafficking, would generally not incur the responsibility of the UN given that these acts cannot be considered part of the official functions of the UN.38 This would be in line with past practice such as when the UN’s responsibility was not invoked when a member of the UN Interim Force in Lebanon allegedly smuggled explosives into Israel for the Palestine Liberation Organization’s use.39 In such cases, the onus falls on the relevant TCC to exercise its jurisdiction in respect of any crimes committed by a member of its contingent as well as other acts of misconduct that do not amount to crimes. This is made clear in the latest revised draft Model MOU between the UN and TCCs governing contributions to UN PKOs.40 A TCC’s international responsibility could then potentially be engaged if it does not exercise jurisdiction, possibly in breach of its international obligations under IHL and IHRL and the memorandum of understanding governing its contributions to the PKO, among others.41 On a related note, while the Model MOU provides that the TCC will be liable for claims, if loss, damage or injury arose from gross negligence or wilful misconduct of the personnel provided by the TCC, this is not understood to suggest that the UN is not responsible for such conduct, but only that the UN has the right to recover the claims it paid out from the TCC.42
Command and Control in the Current UN Peacekeeping Context
As some writers have pointed out, the command and control structures of UN PKOs are straightforward in theory, but seldom so in practice.43 In an UN PKO, all components of the mission — military, police and civilian — come under the ‘operational authority’ of the Head of Mission, typically a Special Representative of the Secretary-General, but in some cases a Force Commander or a Chief Military Observer.44 More pertinently, peacekeepers contributed by TCCs come under the ‘operational control’ of the Force Commander, the head of the military component of a PKO. As the Department of Peacekeeping Operation’s (‘DPKO’) latest Capstone Doctrine document — that provides the doctrinal foundation for UN PKOs and to which all subordinate directives, guidelines and standard operating procedures, among others, must conform — notes: ‘In the case of military personnel provided by Member States, these personnel are placed under the operational control of the United Nations Force Commander or head of military component, but not under United Nations command’.45
The UN Secretariat is careful not to assert that it has command of peacekeepers, as the laws and national policies of many TCCs forbid command of their armed forces to be handed over to a foreign commander.46 For instance, Canadian law does not permit national command of Canadian forces to be handed over to a foreign commander at any time, but permits ‘operational control’ to be vested in a foreign commander.47 In practice, this restriction on the part of TCCs is circumvented by having the Force Commander’s orders transmitted through the NCC who actually ‘commands’ the military contingents contributed.48 The NCC is normally the most senior officer from the TCC on the ground, who in some cases serves as the Commanding Officer of the contingent and in other cases serves on the Force Commander’s staff or in other staff capacities within the PKO. They have the authority to represent their contingent on all matters pertaining to their contingent.
There has been some confusion over these concepts over the years. An earlier version of the Model MOU between the UN and TCCs on the contribution of personnel referred to the contributed personnel being ‘under the command of the UN’ and that the UN had ‘full authority over the deployment ... and direction’ of these personnel.49 This perhaps led to the Special Committee on Peacekeeping Operations, a UN committee of mainly TCC representatives, asserting in 1995 that:
the authority of the United Nations Force Commander is based on the concept of operational control, in which, inter alia, the tasks, as well as duration and zones of deployment of contingents have been fixed by agreement between troop contributing countries and the Secretary-General … 50
In a document to TCCs released in October 2001, the UN clarified that ‘operational control’ in UN terminology should be understood as:
The authority granted to a military commander, in United Nations Peacekeeping Operations, to direct forces assigned so that the commander may accomplish specific missions or tasks which are usually limited by function, time, or location (or a combination), to deploy units concerned and/or military personnel … United Nations Operational Control includes the authority to assign separate tasks to sub units of a contingent, as required by the operational necessities, within the mission area of responsibility, in consultation with the Contingent Commander and as approved by the United Nations Headquarters.51
The glossary of the DPKO website notes, additionally, with respect to the
definition of ‘operational control’ that:
it is a more restrictive level of authority ... a commander cannot change the mission of those foreces [sic] or deploy them outside the area of responsibility previously agreed to by the troop-contributing country without the prior consent of this country; further [the commander] cannot separate contingents by assigning tasks to components of the units concerned.52
The training materials for a UNITAR course, ‘Commanding United Nations Peacekeeping Operations’, for potential leaders in PKOs also observe that:
issues pertaining to the employment of national contingents in peacekeeping missions are resolved through mutual consultation between the contributing countries and the United Nations ... .[The UN] might accept some restrictions on the use of these contingents without compromising the operational effectiveness of the mission ... . Normally, such restrictions result from national concerns about force protection or the safety and security of nationals in accordance with national policy.53
In the current version of the Model MOU between the UN and TCCs on the contribution of peacekeepers and resources to a PKO, references to command and control have been altogether dropped in favour of mission-specific guidelines for TCCs that are annexed to the memorandum of understanding, which specify that the Force Commander has operational control over contributed military units.54 It has been pointed out that the Model MOU now exemplifies the ‘logistical heritage and character’ of the UN–TCC relationship, as opposed to a command and control arrangement.55
In effect, it would appear from these statements and clarifications that the current command and control arrangements in place are that:
Military contingents are deployed to PKOs only after their specific tasks and their zones of deployment have been negotiated and agreed upon between the TCC and the DPKO;
Within the mission, the Force Commander cannot change the specific tasks and zones of deployment that have been agreed upon and assign new tasks and redeploy the contingent, without the consent of the TCC, usually through the NCC. Typically, the NCCs would have been provided with specific guidelines by their government on what can be agreed and would need to be referred back to their governments.
For instance, one can imagine that a TCC may stipulate to the UN that its peacekeepers may not be deployed to quell civil disturbances, which may be politically sensitive, or that its peacekeepers, which may be assigned to protect certain key installations, may not be redeployed to serve on combat patrols where the rules of engagement (‘ROEs’) are more robust. TCCs could also provide their NCCs with instructions that they only agree to their contingent being redeployed in certain ‘safe’ areas, but would need to seek approval from national authorities for redeployment to other ‘high risk’ areas. The conclusion that there are restrictions placed on the employment of peacekeepers is also borne out by a number of other sources.56 Of course, it can be presumed that not all TCCs impose restrictions on the employment of their contingents or provide guidelines to their NCCs; one can envisage some TCCs providing the UN with a carte blanche to deploy their contingents as it wishes.
As one would expect, the fact that the UN accepts restrictions on the employment of the peacekeepers contributed is not trumpeted. The UN, in essence, finds itself in a ‘catch 22’ situation. If it does not allow restrictions to be placed on the employment of their contingents by typically risk-averse TCCs, it risks deterring TCCs from contributing at all. If it allows such restrictions, the operational effectiveness of a PKO may be compromised. Hence, the UN seems to have tacitly accepted, resignedly, that TCCs will often impose such restrictions and micro-manage their contingents on the ground to safeguard their interests and protect their peacekeepers. At the same time, given the difficulties that such restrictions pose, the UN has declared that it cannot accept restrictions by TCCs that will compromise the mission; that unity of command is critical to the effective functioning of a PKO; and that TCCs should not provide, and NCCs should not abide by, any national directions, in the hope that TCCs will minimise restrictions on the employment of their contingents. As the training materials for the UNITAR course illustrate:
[the UN] might accept some restrictions on the use of these contingents without compromising the operational effectiveness of the mission ... [a]ny restriction placed by a TCC on the deployment location and the tasks that may be undertaken by a national contingent or asset create the greatest difficulties in a mission ... 57
Various UN documents, including the Guidelines for TCCs — UNMIL, typically stipulate that NCCs and personnel serving in PKOs should not accept instructions from sources external to the UN and should perform their duties with the interests solely of the UN in view.58 These documents, however, are only exhortatory: even, for example, the Guidelines for TCCs — UNMIL that are annexed to the legally binding memorandum of understanding between the UN and each TCC. The Model MOU states amorphously that the contribution of the TCC ‘shall be governed by the general conditions set out in the relevant annexes’.59 The use of the term ‘guidelines’ suggests that the Guidelines for TCCs — UNMIL are non binding, and are distinct from legally binding ‘conditions’.
Not surprisingly, despite these stipulations, there have been a number of reported instances over the years in which peacekeepers persisted in seeking national direction before executing the orders of the Force Commander. Governments would provide instructions to their peacekeepers that were inconsistent with the orders of the Force Commander, particularly when the potential to use force or the risk was high.60 It does not appear that the UN took forceful action against the TCCs concerned in almost all such instances, other than to remonstrate with the TCC. As a last resort, it seems that the UN can only ‘name and shame’ the TCC and withdraw the contingent concerned, which it often does not want to do, given political sensitivities and manning constraints on the ground in the PKO.
In practice, the UN deals with the problem by encouraging the Force Commanders to involve the NCCs in every aspect of operational planning and decision-making, particularly where their respective contingents are concerned.61 It would follow that the respective NCCs are informally sounded out as to the employment of their contingents, and their agreement sought, prior to formal orders being issued, particularly if the redeployment entails significant risks or political ramifications. The expectation is that the NCC would quickly consult with their government, if required. Issues that cannot be resolved between an NCC and the Force Commander in such a manner would be quickly referred back to the UN headquarters for the TCC and the DPKO to resolve.62 This avoids a situation where an NCC would openly disobey the Force Commander’s orders. Having said this, the UN recognises that the Force Commanders would have to tread a fine line between constructive consultation and allowing the consultation to develop into unproductive negotiations with the NCC and the TCC on the ‘nuts and bolts’ of the employment of their contingent.63
In recent years, there have also been increasing calls by TCCs and experts for the UN to closely consult the TCCs in the formulation of mission mandates, concepts of operation and ROEs, among others.64 It would appear that the UN Secretariat and the UN Security Council have, at least, paid significant lip service to these calls — the Secretariat perhaps much more so than the Security Council, which remains the key decision-maker in terms of mission mandates — although the consultation process could be further improved.65 Crucially, TCCs now have a greater voice in the management of PKOs, particularly in the development of the concept of operations and the ROEs, which are under the purview of the Secretariat. In any case, even if the UN does not take into account the TCCs’ inputs, TCCs can limit the tasks that their peacekeepers perform and elect to provide separate national ROEs to their peacekeepers (although the latter is likely to be over the protests of the Secretariat). For this reason too, the Secretariat is more likely to listen to TCCs, as it does not want to end up with unrealistic concepts of operations or ROEs that cannot be effected.
Prevailing Thinking and Practice on the Attribution of Conduct
The preceding discussion calls into question the UN’s position that it has exclusive command and control of its PKOs, and raises interesting questions about whether the UN can even be considered to ever have effective control of peacekeepers. Most scholars and jurists generally hold that peacekeepers are, in general, under the effective control of the UN during the period that they are placed at the disposal of the UN, except for that rare instance when a TCC may override the Force Commander’s instructions and assume effective control.66 Some scholars, on the basis of TCCs retaining criminal jurisdiction and having conventional or customary law obligations, also suggest that TCCs remain attributable and responsible, perhaps concurrently with the UN, for violations committed by their peacekeepers. Others simply note that dual or multiple attribution such as on the basis of instructions jointly issued is possible.67 Actual instances of dual or multiple attribution, however, appear to be few and far between, with the attribution of conduct to only one entity continuing to be the prevailing practice. It does not appear that scholars or jurists have considered the possibility of:
TCCs negotiating, prior to deployment, the specific tasks and zone of deployment of their peacekeepers, and the placing of restrictions upon the employment of peacekeepers;
the intimate and structured ‘consultations’ that continuously take place between the Force Commander and the NCCs on the employment of their contingents; and
the TCCs having significant input into the development of concept of operations and ROEs; and their implications in terms of the effective control of the contingents; whether in the case of UN PKOs or peace operations in general.
Behrami and Saramati provides a case in point. An argument was made by the applicants that NATO Troop-Contributing Nations (‘TCN’, in NATO parlance) concerned had exercised ‘significant power’ over their Kosovo Force (‘KFOR’) contingents — on the bases of TCNs retaining criminal jurisdiction, and imposing national ROEs, among others — and therefore, that the conduct ought to be attributed to the TCNs.68 However, the European Court of Human Rights found that the TCNs did not exercise effective control — although it observed that NATO did not have exclusive command and control since the TCN retained some authority over (for example, criminal jurisdiction) and obligations (for example, material provisions) vis-à-vis their contingents — as it did not find any suggestion of actual TCN orders interfering or undermining NATO’s operational command and control.69 Unfortunately, both the applicants and the European Court of Human Rights did not delve further into the issue of command and control of the KFOR contingents.
Nor was this issue addressed by the European Court of Human Rights in Banković v Belgium,70 where it held that the case was inadmissible on the basis of a lack of a jurisdictional link, and it was not necessary to examine other admissibility issues. The applicants here had argued in their application that the target lists for Operation Allied Force had been forwarded to each NATO member country on a daily basis and that each country had the power to approve or object to any target on the list.71 Similarly, in the Legality of the Use of Force cases at the ICJ, the Court did not comment on the issue of whether the NATO countries concerned were individually or concurrently responsible or whether only NATO as an entity was responsible, finding instead that it had no jurisdiction because what is now Serbia and Montenegro had no bases for access to the Court.72
Had the European Court of Human Rights and ICJ delved into the issue of command and control, the issue of admissibility notwithstanding, they may have proffered other conclusions. As one expert had noted in testimony before the US House of Representatives in 2007 (although it appears that he underestimated the influence of TCCs in UN PKOs):
NATO ... procedures offer troop contributing members much greater day-to-day influence over the use of their contingents than do those of the UN. ... Member governments consequently have a greater voice in operational matters. ... National caveats limiting the types of missions to which any one member’s troops may be assigned are a fact of life in all coalition operations, but have lately proved even more pervasive in NATO than UN operations.73
The Netherlands Ministry of Foreign Affairs also stated that:
the Netherlands was fully involved in the decision-making process regarding all aspects of the aerial operation, the formulation of the political objectives of the aerial campaign, the establishment of the operational plan on which the campaign was based, the decision concerning the beginning and the end of the operation and the decision concerning the beginning of the various stages.74
In addition, the US had bilaterally settled the claim arising from the bombing of the Chinese embassy in Belgrade, even though the act was ostensibly attributable to NATO.75
Can the UN Really Have ‘Effective Control’ of Peacekeepers?
What conclusions might have been derived, had scholars and jurists delved into the issue of command and control? As one scholar has observed, the adoption of the ‘effective control’ test in art 5 does not fully reflect the complexities of the current practice, given the possibility of, inter alia, the dual or multiple attribution of conduct.76 TCCs can scrutinise and consent to the minutiae of the employment of their peacekeepers prior to deployment; provide input to the development of concept of operations and ROEs; and, through the NCCs, can disagree with (or have to agree to) any changes in employment of their peacekeepers in-mission. This suggests that peacekeepers are not under the effective control of the UN, but are perhaps under the dual or joint control of both the UN and the TCC. In such circumstances, the UN and the TCC could conceivably be deemed to be acting jointly and it would follow that the conduct of peacekeepers ought to be imputed to both the UN and the TCC. Having said this, to suggest (based on the other extreme) that the TCC should be considered to have retained exclusive control of its peacekeepers at all times and that all conduct should be attributed to the TCC would be too much of a stretch. The TCC could perhaps be considered more of a ‘co-conspirator’, given that the concept of operations, tactical instructions and so on originated from the UN. The ‘effective control’ test would only apply to peacekeepers in exceptional circumstances such as when the TCC provides national direction contrary to, or in the absence of, the UN’s instructions.
What is key is that a TCC can, at any time, decline to take on a task. Where the TCC has consented, even tacitly, would instructions to its contingent from the Force Commander (after all, which are typically transmitted through the NCC) not be considered joint instructions, and both the UN and the TCC considered to be acting jointly? If so, the resulting conduct should be attributed to both the UN, as the originator of the instructions, and the TCC, for having concurred in the instructions; or to the UN for having perpetuated the act, and the TCC, for omitting to prevent or not go along with the act. This is particularly so when the TCC or NCC consented to the tasking fully aware that the assigned task, if mishandled, may lead to violations of IHL or IHRL, but did not take manifest steps to prevent such violations. This scenario, however, should not be confused with the scenario envisaged in art 25 of the Draft Articles.77 There is an obvious difference between the two scenarios, given that the act envisaged by art 25 is distinctly wrong. Acts in the scenarios envisaged in this discussion are unlikely to be decidedly wrongful, although they could be wrongful, if handled wilfully or negligently.
The assumption that host countries, in consenting to a PKO in their territory, consent to bear, at least in part, any liability as a result of that presence (for example, in instances where the PKO may exercise elements of the host government’s authority such as in seizing private property) provides a useful parallel.78 Similarly, it may be reasonable for TCCs, in consenting to the instructions of the Force Commander, to bear at least some of the responsibility for the conduct of their peacekeepers, especially since the final onus in terms of the duty of care and vigilance is on the officers and non-commissioned officers in the contingent — and not a UN official — to ensure that their troops do not violate any laws in carrying out the instructions. Likewise, it is submitted that the matter of TCCs retaining jurisdiction over the conduct of their peacekeepers is also relevant, despite what appears to be the prevailing view that a TCC’s retention of jurisdiction does not affect the ability of the UN to exercise effective control of peacekeepers.79 Effective control only has ‘teeth’ and is realistic when the entity exercising effective control has the real authority and means to exercise it over the subjects that it controls.80 The numerous instances of peacekeepers taking directions from home countries over that of the Force Commander’s instructions illustrates that the UN has no real authority or means to control the peacekeepers, absent the TCC’s concurrence.
Attributing the Conduct of Peacekeepers — Determining a Suitable Approach
The ILC Special Rapporteur on the Responsibility of International Organizations, Giorgio Gaja, appears to have taken these considerations somewhat into account, suggesting that the issue is one of the ‘extent of effective control’. His proposal for the conduct to be attributed ‘to the extent that the organization exercises effective control over the conduct of the organ’ was, however, not taken up by the ILC.81 Special Rapporteur Gaja explicitly suggested with respect to the Bankovićand Legality of the Use of Force cases that ‘one envisageable solution would be for the relevant conduct to be attributed to both NATO and to one or more of its member states, for instance, because those states contributed to planning the military action or to carrying it out’.82 Likewise, the UN seemed to have recognised this principle quite early on, with the UN Secretary-General recommending in 1996 that the General Assembly consider recognising the concurrent responsibility of TCCs for violations of IHL by its peacekeepers, given the exclusive criminal jurisdiction of TCCs and their obligation to ensure respect for IHL.83 This would presumably also apply to instances of IHRL violations. However, this was not taken up by the General Assembly.84 Special Rapporteur Gaja added, with regard to the Secretary General’s recommendation, that besides joint attribution of the same conduct, one could attribute the wrongful act to either the TCC or the UN, and omission of any preventative measures to the other.85
That said, the ILC did not dismiss the possibility, in general, of the dual or multiple attribution of the same conduct, as acknowledged by a number of scholars.86 Indeed, the ILC’s ‘Commentary’ on the Draft Articles and on the Draft Articles on State Responsibility both briefly but unambiguously stated that dual or multiple attribution of conduct was possible.87 Yet, its ‘Commentary’ on art 5 of the Draft Articles, as well as the final phrasing of art 5, seemed to leave little leeway for dual or multiple attribution based on dual or joint control of the conduct concerned. This appears to suggest that when it comes to organs placed at the disposal of international organisations that still acted to a certain extent as organs of the contributing state, the issue is nonetheless seen to be one of decisive attribution to either the contributing state or the receiving organisation. This construction is seen, for instance, in the judgment of the District Court of The Hague in HN v The Netherlands, where the Court, after considering the principles reflected in the Draft Articles on State Responsibility and the Draft Articles, held that the attribution of the conduct of the Dutch Battalion serving with the UN Protection Force to the UN excluded attribution of the same conduct to the Netherlands.88 The European Court of Human Rights in Behrami and Saramati also did not consider the possibility of the dual or multiple attribution of KFOR’s conduct to the UN as well as to NATO and its TCNs, even as it took into account art 5 of the Draft Articles and its ‘Commentary’ and considered whether the TCNs could have exercised effective control.89 The European Court of Human Rights intimated in its judgment that conduct would be attributed to the TCNs in such an event, rather than the UN.90
All things considered, Special Rapporteur Gaja’s proposal to attribute conduct based on the ‘extent’ to which effective control is exercised by each party would seem to provide for the most flexibility in terms of covering all likely scenarios that could arise. This approach would allow for dual or multiple attribution in cases involving peacekeepers acting under the dual or joint control of the UN and TCCs, peacekeepers acting under the effective control of their sending states, as well as more straightforward instances of other organs placed at the disposal of an international organisation such as the example of the Pan American Sanitary Conference cited by the ILC.91 In considering the ‘extent’ of control, some scope is assumedly available to take into account various factors, such as whether the parties exercised due care in preventing wilful or negligent actions or omissions and the degree to which they did, thus permitting the exhaustive attribution of conduct.
While there have been other tests devised, such as the International Criminal Tribunal for the Former Yugoslavia’s ‘overall control’ test and the European Court of Human Rights’ ‘ultimate authority and control’ tests, these tests — like the ‘effective control’ test — cannot fully reflect the realities of command and control arrangements in UN PKOs and do not allow for the accurate and comprehensive attribution of conduct.92 These tests are also not uncontroversial, having come under criticism from jurists and scholars alike.93 The ‘overall control’ test simply lowers the threshold for attribution from ‘effective control’ of the conduct in question to ‘overall control going beyond the mere financing and equipping of such forces and involving also participation in the planning and supervision of military operations’.94 It is not apparent how this would facilitate the dual or multiple attribution of conduct. Likewise, the ‘ultimate authority and control’ test would purely lay the conduct of peacekeepers at the door of the UN, given that UN PKOs are typically established on the basis of a mandate from the UN Security Council.
The notion that dual or multiple attribution of conduct to two or more parties leads to the joint or several responsibility of these parties for the same wrongful act is well accepted, as Special Rapporteur Gaja noted.95 This principle has been captured in art 47 of the Draft Articles on State Responsibility and art 51 of the Draft Articles, which both deal with the situation where there is a plurality of responsible parties and the extent to which the parties are responsible.96 In its ‘Commentary’ on art 47, the ILC referred to — as one example of a range of circumstances where several states could be responsible for the same act — the case of two or more states acting jointly to carry out a wrongful act.97 Certainly, this would apply equally to cases where states and international organisations act jointly, which, as clarified earlier, is one circumstance of concern to our discussion. The ILC also cited, among others, the Corfu Channel Case,98 where the ICJ did not suggest that Albania’s responsibility was diminished, let alone precluded, as a result of the concurrent responsibility of a third states, to illustrate that the responsibility of each party could be determined individually, based on its own conduct and its own international obligations.99 Specific to IOs, the ILA also proposed in its Final Report on the Accountability of International Organisations — which sought to promulgate recommended rules and practices on this subject (many of which reflect existing rules of international law) — that the
responsibility of an IO does not preclude any separate or concurrent responsibility of a state or of another IO which participated in the performance of the wrongful act or which has failed to comply with its own obligations concerning the prevention of that wrongful act.100
The dual or multiple attribution of conduct, leading to joint or several responsibility of the parties concerned, would be in line with the aims of international responsibility, which is to prevent breaches of international law through deterrence.101 By attributing responsibility where responsibility actually lies, this ensures that the actors concerned exercise due precaution in making sure that breaches do not occur.102 Attributing responsibility to the UN alone, for instance, may encourage more reckless and negligent behaviour on the part of the TCCs, since they may not bear responsibility for the wrongful conduct of their peacekeepers. Concerns have been raised, for one, that the decisions of the European Court of Human Rights in cases like Behrami and Saramati could undermine IHRL, as it would be difficult to hold TCCs accountable for the transgressions of their troops during peace operations.103 Such thinking is assumedly behind the approach taken by the UN in revising the Model MOU on TCC contributions to strengthen the onus on the part of TCCs and NCCs for policing the conduct of peacekeepers and the recommendation of the Secretary General for the General Assembly to recognise the concurrent responsibility of TCCs for violations of IHL by its peacekeepers. Conversely, ‘piercing the veil’ to fully lay responsibility at the door of TCCs alone may encourage the UN to push the limits in terms of pursuing riskier tactical operations (which some TCCs may well go along with) or TCCs to micro manage every aspect of a PKO or not contributing altogether to PKOs, since they now bear responsibility.104 Such outcomes are also clearly undesirable. Having dual or multiple attribution of conduct to the UN and the TCCs concerned, leading to the concurrent responsibility of entities involved, may strike a healthier balance.
Overall, the preceding discussion suggests that a case may be made that art 5 of the Draft Articles, as it stands and is currently understood, may not allow for the full and precise attribution of the conduct of peacekeepers. As the facts intimate, TCCs can retain significant control over the operational employment of their peacekeepers, from determining the exact terms of their deployment, including the tasks that the peacekeepers may perform and the zone of deployment, to holding a veto over any changes to these terms by the UN. In this sense, it follows that it would be more appropriate to consider peacekeepers to be under the dual or joint control of the UN and their TCC. Attribution of the conduct of peacekeepers should therefore be to both entities, and accordingly, both would assume shared responsibility for the conduct, according to the degree to which they contributed to the conduct. Special Rapporteur Gaja’s proposed approach for conduct to be attributed to the ‘extent’ that an organisation exercises effective control would appear to best accommodate this and other likely scenarios envisaged by art 5. The importance of having accurate and comprehensive attribution of conduct and the correct balance of responsibility cannot be understated, given that the attribution of responsibility to only one of the entities responsible could lead to behavioural externalities and other undesirable consequences.
1 * BSc (Hons) (LSE); MA (Columbia); MIL (ANU). Former Head, International Policy Branch, Ministry of Defence, Singapore.
2 ILC, Report of the International Law Commission Fifty-Sixth Session,UN Doc A/59/10 (2004) 99 (‘2004 Report’).
3 Behrami v France and Saramati v France, Germany and Norway (2007) 45 EHRR SE10 (‘BehramiandSaramati’).
4 R (on the Application of Al-Jedda) v Secretary of State for Defence  1 AC 332 (‘Al Jedda’).
5 Case No (Zaaknummer) 265615/HA ZA 06-1671 (District Court of The Hague, Netherlands, 10 September 2008).
6 The second constitutive element of international responsibility — the breach of an international obligation — will not be examined in this commentary, which presumes that both international humanitarian and human rights law apply to UN PKOs, at the very least based on customary international law. Scholars and authorities generally agree on this point, although there are contentions over issues such as the extent to which it applies, whether the onus is on states or the UN (or both) to ensure compliance, and who bears responsibility. See, eg, Human Rights Committee, General Comment No 31: Nature of the General Legal Obligation on States Parties to the Covenant’, UN Doc CCPR/C/21/Rev.1/Add.13 (29 March 2004); UN Secretariat, Secretary-General’s Bulletin: Observance by United Nations Forces of International Humanitarian Law, UN Doc ST/SGB/1999/13 (6 August 1999); Interpretation of the Agreement of 25 March 1951 between the WHO and Egypt (Advisory Opinion) ICJ Rep 73, 90–1; Ray Murphy, ‘Legal Framework of UN Forces and Command and Control of Canadian and Irish Forces’ (1999) 4 Journal of Armed Conflict Law 41; Dale Stephens, ‘The Lawful Use of Force by Peacekeeping Forces’ (2005) 12 International Peacekeeping 157; Jaume Saura, ‘Lawful Peacekeeping: Applicability of International Humanitarian Law to United Nations Peacekeeping Operations’ (2007) 58 Hastings Law Journal 479; Daphna Shraga, ‘UN Peacekeeping Operations: Applicability of International Humanitarian Law and Responsibility for Operations-Related Damage’ (2000) 94 American Journal of International Law 406.
7 Where this commentary refers to ‘peacekeepers’, it should be assumed that this refers to members of such military contingents serving in UN PKOs.
8 See, eg, ILA, Accountability of International Organisations (2004)