Home EJIL Analysis The Security Council and Human Rights: What is the role of Art. 103 of the Charter?

The Security Council and Human Rights: What is the role of Art. 103 of the Charter?

Published on March 30, 2009        Author: 

At last week’s ASIL meeting there was a panel on whether the United Nations Security Council is bound by human rights law. The panelists (Vera Gowlland-Debbas, Graduate Institute of International and Development Studies, Linos-Alexander Sicilianos, University of Athens  & Gráinne de Búrca, Fordham University School of Law) discussed cases such as the Kadi decision of the European Court of Justice, Al Jedda (House of Lords), Sayadi (Human Rights Committee and Behrami (European Court of Human Rights). These cases have been the subject of posts on this blog (for Kadi, see here and here, for Sayadi, see here and for Behrami, see here). One of the things that strikes me about much of this discussion is the use made of Article 103 of the UN Charter. That article provides that:

In the event of of a conflict between the obligations of the Members of the United Nations under the present Charter and their obligations under any other international agreement, their obligations under the present Charter shall prevail.

Of the four decisions mentioned above, only the Al Jedda decision discusses and applies Art. 103. According to Lord Bingham,

The central questions to be resolved are whether, on the facts of this case, the UK became subject to an obligation (within the meaning of article 103) to detain the appellant and, if so, whether and to what extent such obligation displaced or qualified the appellant’s rights under article 5(1) [of the European Convention on Human Rights]. (para. 26)

The House of Lords held that the Security Council authorisation to detain the appellant did indeed bring Art. 103 into play (on the theory that Art. 103 also extends to authorisations) and that rights under the ECHR were qualified to the extent that they conflicted with that authorisation. Some have criticised the ECJ in Kadi  and the Human Rights Committee in Sayadi for not evening mentioning Art. 103 and for failing to take the Al Jedda approaching (for some more discussion of this issue see here and here).

However, the role of Art. 103 is often overplayed in these debates concerning the conflicts between Security Council obligations and human rights law.  There are 2 overlapping questions here: (i) Is the Security Council bound by human rights norms when it acts (eg in combatting terrorism, imposing sanctions or in authorising action in peacekeeping or peace enforment)?; (ii) are States bound to apply Security Council decisions that may conflict with the human rights obligations of those States?.  Art. 103 does not and cannot answer the first question. Art. 103 should not be regarded as the starting point in answering the second question. Furthermore one may not even reach Art. 103 in answering that latter question.

The question whether the Council is bound by human rights norms relates to the issue whether there are limits to the powers of the Council. Art. 103 does not purport to deal with the limits to the powers of the Security Council. The answers to those questions are to be found elsewhere. Firstly, in the fact that the Council is an organ of a bigger organization (the United Nations) which is itself a subject of international law. Secondly, and more specifically, I  (see here) and others have argued, the Council is, as a general matter, bound by human rights law. This is because the Charter says so. Art. 24(2) of the Charter provides that in discharging its duties “the Security Council shall act in accordance with the Purposes and Principles of the United Nations.” Those purposes include (in Art. 1(3)) achieving international cooperation in promiting and encouraging respect for human rights and fundamental freedoms [see also Art. 55(c)]. There is a question as to which human rights obligations are binding on the organization and on the Security Council (see below) but unless one ignores the Charter there ought to be no doubt that the Council is bound by human rights law.

Whether States are bound by Security Council resolutions that are inconsistent with human rights obligations of those States is a matter that might at first sight appear to be governed by Art. 103. Afterall, it appears af if there is a conflict of obligations here and apparently we should refer to Art. 103 to resolve that conflict. As a general matter, Art. 103 applies to obligations derived from Security Council resolutions (see para. 42 of the ICJ’s provisional measures order in the Lockerbie case) and therefore obligations under those resolutions would prevail over inconsistent obligations under another treaty. However, since the Security Council is bound by human rights norms, any particular resolution that fails to respect the limits to the powers of the Council would be ultra vires the Council. That then leads to the question whether ultra vires  resolutions create any obligations for UN members. For it is only if they create obligations that one gets to Art. 103. On the contrary, if an ultra vires  is void then it creates no obligations for members.  If so, Art. 103 does not apply. There is simply no conflict of obligations to which that provision would apply.

So when faced with a Security Council decision that appears to be inconsistent with human rights norms, the key questions are these: (i) Is the Security Council decision consistent with limitations on the powers of the Council; (ii) If the Council has acted beyond  its powers what are the legal effects of the illegal decision. This second question is an area on which there hasn’t been enough analysis. Are ultra vires decisions void or they voidable only? Unless one can answer this question, one does not get to Art. 103. My own view (as expressed in my Chapter on “International Organizations” in Evans (ed.) International Law ) is that adopting the view that illegal decisions of international organizations are voidable only (i.e valid until there is a decision that it is illegal) is problematic because there are few (and in the case of the Council, no) organised procedure by which a competent body can decide that decisions of the Council are illegal. To take the view that illegal decisions are voidable is to say that they stand, unless by accident, there is review or perhaps unless the Council changes it mind. This is clearly unsatisfactory and the better view is that ultra vires decisions–but not those merely suffering some minor procedural defect–are a nullity. As Judge Morelli said in the Certain Expenses case:

In the case of acts of international organizations . . . there is nothing comparable to the remedies existing in domestic law in connection with administrative acts. The consequence of this is that there is no possibility of applying the concept of voidability to the acts of the United Nations. If an act of an organ of the United Nations had to be considered as an invalid act, such invalidity could constitute only the absolute nullity of the act. In other words, there are only two alternatives for the acts of the Organization: either the act is fully valid, or it is an absolute nullity, because absolute nullity is the only form in which invalidity of an act of the Organization can occur. (1962, ICJ Reps. p. 222)

Thus, where a decision is illegal, a State is free to depart from it. There is no obligation on it to comply with such a decision. Would this be a useful position to adopt with regard to Council decisions? The danger of adopting it is that the entire collective security system may unravel if States were free to pick and choose which resolutions they wish to follow. However, there is always the risk that the decision might later be found to be lawful and the non-compliant State in breach of its obligations. Also, it is important to reaffirm the presumption that Council decisions which are directed at the fulfilment of the purposes of the organization are valid. Thus, the burden of proof is on the State arguing otherwise.

The other question that is often analysed insufficiently is: which human rights are binding on the Council? The effect of the Charter is that the Council is bound by human rights but the question is which ones. I have argued that the Council and the Organization are bound by those rights that are proclaimed in instruments  adopted within the Organization. Those instruments are to be regarded as an elaboration of the rights provided for in the Charter. In fact they were adopted in order to fulfill the mandate with regard to human rights provided for in the Charter. But, arguing that the Organization and the Council are bound by these human rights does not mean the Council is or ought to be bound by them when acting under Chapter VII. The ICCPR (and other treaties) provide for derogations in time of public emergency which threaten States. An analogous position must apply to the Council. If States can derogate from their obligations in order to safeguard domestic peace and security, the Council ought to be able to derogate in order to safeguard international peace and security. By definition every Chapter VII resolution implicates international peace and security. Thus, when one thinks about human rights norm that bind the Council when acting under Chapter VII one ought to think about non-derogable rights. One also ought to think about how derogable rights are to be interpreted when derogations have been made.

In the Al Jedda case – the one case where Art. 103 was applied – the House of Lords ducked all of these issues and simply applied that provision. This was poor analysis by the House of Lords and it appeared that it simply wished to take the line of least resistance and was not prepared to engage with the difficult but important issues that arise in this area.

Print Friendly

4 Responses

  1. Nehal Bhuta

    Hello Dapo

    Thank you for your post. I too attended that panel, and it struck me that there were several different methods available to think about this, and none is self-evidently superior to the other.

    One is the one proposed by you: a “competence” perspective which avoids or tries to minimize the possibility of direct conflict between Security Council Chapter VII resolutions and human rights obligations by arguing that the Security Council is not empowered to pass resolutions of which violate (certain) human rights.

    This approach has the merit of arguing for a Security Council that is required to respect certain limits in its conduct, and leaving it open to states to argue that they are not bound because those limits have been exceeded. One concern for me is that Article 1(3) is a rather vague provision, and it is not clear that, on a textual analysis, it requires strict compliance with human rights law. The obligation is “to achieve cooperation in … promoting and encouraging respect” for human rights. This seems like a rather attenuated obligation, not the kind that self-evidently gives rise to bright lines concerning de jure competence.

    Now, I am not arguing that an interpretive practice strengthening our reading of Article 1(3) should not be encouraged. Indeed, it seems sensible, and I see the separate opinions of Rodley and Iwasawa in the HRC’s Sayadi decision as moving in that direction. But we should understand this as one possible interpretive strategy, among others that might be available. This might also help clarify which rights we think are those that the Security Council cannot transgress.

    But this then raises a further question: who decides? The ICJ has proved reluctant to address this issue head on, and instead we have other bodies (some regional, some treaty-based) each developing a particular jurisprudence, and each responding to specific mandates and concerns (ECtHR, Hum Rts Ctee, ECJ, House of Lords). I think what we may witness is, at least initially, a period of divergent and potentially inconsistent interpretations, perhaps followed by a period of greater harmonization and dialogue.

    In this transitional period, it seems to me that the role of scholars and scholarly debate will be very significant: in articulating and rationalizing the different methods and interpretive practices that emerge from different institutions, and in elaborating them systematically in order to clarify and debate their strengths and weaknesses.

    I saw the ASIL panel as an interesting example of this, with Grainne de Burca providing an analytical account of the state of interpretation, Alexander Sicilianos arguing for an approach which minimized “conflict” rules and maximized articulation between different regimes, and Vera Gowland-Debbas emphasizing the need for progressive and dynamic interpretation – with reference to the larger context of developments in the international legal order – when approaching the fairly sparse text of the Charter.

  2. Dapo Akande Dapo Akande


    I agree with you that the obligation in Art. 1(3) does not at first sight appear to be a particular strong one. However, I think there are two points that could be made in support of the view that the obligation on the Security Council to respect human rights is more potent than first appears:
    (i) that obligation is contained not in Art. 1(3) but rather in Art. 24 which says that in discharging its duties the Council “shall act in accordance with the Purposes and Principles of the United Nations.” So the Council has an obligation to take actions which accord with the purposes of the Organization and must not take actions which would detract from those purposes.
    (ii) a violation of human rights norms is probably the clearest example of not encouraging respect for human rights. It is difficult to see how the Organization would be promoting or encouraging respect for rights/norms that it itself violates. The reasoning here is similar to that with regard to the responsibility of States under the Genocide Convention. Recall that the Genocide Convention does not provide an explicit obligation on States not to commit genocide but does contain an explict obligation to prevent it. Nevertheless, the ICJ held in the Bosnian Genocide Convention case (2007) that the obligation to prevent includes an obligation not to commit. So too here, an obligation to encourage respect for includes an obligation not to violate.

  3. Dapo Akande Dapo Akande


    Just one more thing. I think one reason why we are seeing different, and apparently divergent, answers in this area is because the courts and tribunals are answering different questions. The two questions I address in my post are related but not the same. And that is even if one confines the analysis to international law. But as you say, different courts responed to different mandates and they may not even get to international law.

  4. Nehal Bhuta

    Hi Dapo

    I think these are both plausible arguments. I would point out though that Article 1(3) does not simply require promotion and encouragement of human rights; it says that one purpose of the UN is to achieve international cooperation to promote and encourage (etc). So the obligation is once removed from a direct obligation to promote and encourage, and could be read as largely hortatory rather than delimitative (even if read by reference to Article 24).

    I think the 2 questions you address in your post are two sides of the same coin, in the sense that they (currently) arise out of the same set of problems, and represent two ways of slicing that particular problem. The concrete problem is, what to do in respect of a binding SC resolution that on its face seems to require acting in a way that is inconsistent with other binding human rights obligations, and in light of the Article 103 obligation? There are 4 possible responses:

    1. The Resolution is ultra vires through some kind of interpretive exercise involving Article 24 (say, the Resolution explicitly authorizes indefinite detention of named individuals without trial), so we need not apply it (presumably it is the state which determines this in an act of auto-interpretation and would make its views known to other states). (Here, one question is: if a state does act according to the Resolution, is it also excused from state responsibility?)

    2. The Resolution can be read in a way that does not produce direct conflict with a human rights obligation binding on the state, and thus the state can and should implement the resolution in ways consistent with both the resolution and the relevant human rights treaty (Sayadi, esp. Sep. Ops. by Rodley and Iwasawa, and, on one interpretation, the ECJ in Kadi). No need to resort to Article 103 here, but also no need to claim that the resolution is ultra vires: one simply applies a special rule of interpretation which may be justified by virtue of Article 1(3) and Article 24.

    3. The Resolution is in direct conflict with a binding human rights treaty obligation, and Article 103 applies (Al-Jedda on the court’s reading), but the extent of permitted deviation may be constrained by principles such as proportionality and necessity (although which entity should apply these principles is not clear to me: is it the state, or the Security Council? Maybe both, the former in its implementation and the latter in terms of what it can demand intra vires in respect of compliance.)

    4. The Resolution requires conduct which violates a binding human rights obligation and leaves little or no discretion in implementing in a manner inconsistent with a binding human rights treaty norm which is perhaps less serious than rights relating to property or person (say, right to privacy, violated by an obligation to turn over surveillance information however obtained by the state). Here, we face the problem of what kinds of rights violation might lead to a resolution being ultra vires. How do we begin such an inquiry? How do we evaluate the gravity, relative importance and weight of the particular right?