Protection of Civilians Symposium: The Obligation to Protect Civilians     

[Siobhán Wills is a Professor of Law at the Transitional Justice Institute, Ulster University, Northern Ireland. This post is a part of the Protection of Civilians Symposium.]

In 2014 the UN Office of Internal Oversight Services published an ‘evaluation of the implementation and results of Protection of Civilians mandates in United Nations peacekeeping operations’ which:

noted a persistent pattern of peacekeeping operations not intervening with force when civilians are under attack…Partly as a result…civilians continue to suffer violence and displacement in many countries where United Nations missions hold protection of civilians mandates.

One of the many questions arising from this report is whether UN peacekeepers that have a mandate authorizing them to use force to provide protection are legally obliged to take action to protect civilians that are being attacked or about to be attacked.

My chapter in Protection of Civilians (eds Haidi Willmot, Ralph Mamiya, Scott Sheeran, and Marc Weller), argues that peacekeepers do have protection obligations but they are not obliged to use force to protect civilians even if the Security Council resolution mandating the mission authorizes them to do so, and even if the mission is well positioned to be able to use force in the particular circumstances. This is not a shortcoming in the legal regulation of peacekeeping. If peacekeepers routinely avoid using force to protect civilians when to do so would save lives, host state residents are unlikely to hold the mission in high regard; but to legally oblige a commander to use force to protect civilians on every occasion on which it would be feasible to do so, would likely cause serious operational problems and might prompt attacks on civilians in order to tie up mission personnel.

Traditionally peacekeeping mandates have been regarded as having a powers-creating character that does not create any obligation to act. Some peacekeeping mandates contain paragraphs that it would be difficult to interpret as merely powers-creating. For example S/Res 1996 authorized the UN Mission in South Sudan (UNMISS) to ‘report immediately’ gross violations of human rights to the Security Council. It makes no sense to interpret ‘report immediately’ as purely powers-creating unless the addressee is normally prohibited from reporting immediately – otherwise the adverb ‘immediately’ would be redundant.  But, such inconsistences notwithstanding, Security Council resolutions as a whole do not appear to be drafted with a view to creating obligations. The phrase ‘all necessary means,’ which is used to authorize force, is a euphemism and hence inherently opaque as to the scope and nature of the obligations that follow from its use. But the fact that a Chapter VII mandate authorizing use of force to protect of civilians does not create an obligation to use force (even if it is the only means of saving lives), does not mean that the mission has no obligation to provide protection pursuant to the ‘all necessary means’ authorization set out in the mandate.

Although the UN is not a party to any international treaties, it has a duty to uphold the human rights principles promulgated through the human rights regime it has created. At a minimum the UN is bound by peremptory norms, by its Charter, and by its own human rights undertakings as reflected in the resolutions and bulletins it has promulgated concerning itself. Troop-contributing states are directly bound by the human rights treaties to which they are party but a troop contributing state’s positive obligations under those treaties only apply extraterritorially to the extent that the state’s jurisdiction extends to the situation, which for the most part is based on whether the state has control over territory or over persons. But the human rights obligations of the UN itself (as distinct from the obligations of its member states) are not limited by territory – it has none. Provided the UN’s protection activities do not exceed the authorization set out in the mandating resolution, the Security Council mandate counters any ‘sovereignty’ based objections from the host state.

The Aide Memoire to the Secretary-General’s Human Rights Up Front (HRUF) Plan of Action states that  ‘[p]rotecting human rights is a core purpose of the United Nations and defines our identity as an organization.’ The HRUF Plan of Action ‘is designed primarily for settings where the UN does not have a political or peacekeeping mission’ but ‘its spirit can and should also be applied to “mission settings”’. The HRUF Plan of Action states that the UN will ‘put the imperative to protect people, wherever they may be, at the heart of UN strategies and operational activities.’ This is a clear positive commitment on the part of the UN to protecting human rights and protecting people from gross violations of those rights. In the case of Chapter VII mandated peacekeeping missions this commitment is usually backed up by Security Council authorizations to use ‘all necessary means.’ S/Res1894 states that ‘mandated protection activities must be given priority in decisions about the use of available capacity and resources.’

The existence of a protection of civilians mandate gives rise to the presumption that the mission is aware that civilians are at risk. In the event of a mission’s failure to protect, it will not be enough for the UN to say after the event that it tried to uphold the commitment it has made to ‘protecting people wherever they may be:’ it must be able to show it by producing evidence of the protection plan drawn up in response to the know threat to civilians and the efforts made to implement it. UN missions, as traditionally conceived, were thought of as having predominantly negative obligations, essentially to do no harm – and therefore they had considerable flexibility within the confines of its mandate as to how that mandate should be implemented, and also as to how decisions relating to the mandate are made, documented, and communicated. But an obligation to protect cannot be achieved simply by refraining from action. In order to fulfil a positive obligation to protect, in a situation where it is known that that attacks on civilians are likely, the mission must assess the probability, seriousness, and location of likely attacks; draw up plans to counter the risks to civilians that it has concluded are likely; document those plans; and pass that information up the chain of command to a level of seniority that can take responsibility for approving the effectiveness of the plans in light of the scale of likely harm. The mission is not obliged to use force; but it is obliged to carry out its protection plans, unless there is some intervening reason that renders the original response plan ineffective or harmful. Therefore, if the UN undertakes to protect people from violence and the Security Council has mandated the mission to use ‘all necessary means’ to provide that protection, that undertaking should shape the way the mandate is carried out both at the macro level (e.g. initial assessment of risks and operational planning in light of them) and at the micro-level (e.g. documenting and reporting of protection plans and of the steps taken to implement them at local level), and the continuous updating of protection plans in light of the continuous assessment of risks, assessed at both local and general level.