home       biography       publications       speeches       organisations       images       @contact

The French Veto Restraint Proposal: Making it Work

Panel Presentation by Professor the Hon Gareth Evans to International Conference on Limiting the Use of Veto at the UN Security Council in the Case of Mass Atrocities,  Sciences Po, Paris, 21 January 2015


If one is serious about the international community’s responsibility to protect those at risk of mass atrocity crimes, one has to be serious about “capricious use of the veto, or threat of its use … in cases where quick and decisive action is needed to stop or avert a significant humanitarian crisis”.  That was the view of the International Commission on Intervention and State Sovereignty (ICISS) which I co-chaired in 2001, and we reported accordingly in our final report. [1] Having been stimulated to focus on the issue, I am happy to acknowledge, by the path-breaking proposal for a voluntary ‘code of conduct’ of Hubert Vedrine at a roundtable here in Paris earlier that year, we recommended that:

The Permanent Five members of the Security Council should agree not to apply their veto power, in matters where there vital state interests are not involved, to obstruct the passage of resolutions authorizing military intervention for human protection purposes for which there is otherwise majority support. [2]

This was not the first attempt in the UN’s history to constrain or limit in some specifically defined way the P5’s veto power. That honour belongs to my Australian predecessor H.V. Evatt, the first President of the General Assembly, who argued, passionately but unsuccessfully, at the 1945 San Francisco Conference that the veto should be excluded from all arrangements relating to the peaceful settlement of disputes. [3] But it was the first time the idea of the responsibility not to veto in the case of mass atrocities was put publicly on the table.

Since then the idea has developed plenty of further traction at the governmental level, albeit not so far securing the passage of any UN resolution or P5 commitment.  It has been taken up by the High Level Panel on Threats Challenges and Change in 2004, of which I was also a member [4]; the ‘S5’ group of small states in 2006 and again in 2012 [5], and the 22-member Accountability, Coherence and Transparency (ACT) Group, established in 2013 to follow up its work; [6] the bipartisan US Albright-Cohen Genocide Prevention Task Force in 2008 [7]; Ban Ki-Moon’s report to the General Assembly in 2009 on Implementing the Responsibility to Protect [8]; and of course now the new French initiative, announced by President Hollande in the General Assembly on 24 September 2013 [9] with the detail being fleshed out by Foreign Minister Fabius in the New York Times a few days later. [10]

So how do we go about converting these prescriptions, for which so much international momentum has already developed, into practical reality? What will it take to persuade the other four P5 members – all of whom have so far expressed either caution or outright opposition – to accept the French proposal, or something like it? How can we achieve the core objective of  significantly raising the political cost of  a permanent member capriciously using its veto in a mass atrocity case?

I think there are three sets of conditions – constituting ten specific factors in all - which will need to be satisfied: first, what might be described as general scene-setting conditions, creating a general sense of international pressure for change; second, clarifying conditions, or modalities, identifying clearly what P5 are, and are not, being asked to sign up to; and third, protective conditions, giving P5 members some sense of reassurance that they will not be doing themselves any harm by agreeing to voluntary restraint.

Scene-setting conditions

  • General international agreement (not the same as unanimous consensus!) that there is a powerful case to be made – both ethically, and in terms of the proper discharge of the Security Council’s functions under Article 24 of the UN Charter – for the veto not being used to stop effective Council action for which there is otherwise the necessary nine-vote majority in favour. Evidence of such agreement has been steadily accumulating with 65 different states from all regions of the world so far going on record since 2008 in various UN forums in support of veto restraint. [11]
  • General international agreement that the issue is real, in the sense that lack of veto restraint is actually inhibiting effective Security Council action for which there is otherwise the necessary majority support. There should not be much doubt about this given that there have been six clear cases since the R2P norm was embraced in 2005 when the veto has been employed to block resolutions dealing with situations that could reasonably be described as mass atrocity crimes, [12] quite apart from the “silent veto” cases (the most familiar of which being Kosovo in 1999) where a veto has its effect simply through being threatened, and never actually exercised.
  • General international agreement that a purely voluntary commitment is worth having. Clearly there is no reasonable prospect for the indefinitely foreseeable of formal Charter amendment on any issue remotely as sensitive as the veto, and the only option is some form of gentleman’s agreement. While I cannot help but recall in this context the remark of a former Australian Prime Minister, Ben Chifley, that “the trouble with gentleman’s agreements is that there aren’t enough bloody gentlemen”,  there does seem to be already a large measure of acceptance that, for all its obvious weaknesses, voluntary restraint would be a big step forward.

Clarifying conditions

  •  Agreement on an initiating mechanism. Nothing more formal is here required than a document agreed to by all members of the P5, publicly released, and tabled for information,in one form or another, in the Security Council and General Assembly.
  • Agreement on how the veto restraint agreement should be described. The object here should be to avoid any language that might be seen by any of the P5 as over-formal or over-restrictive, which might be the case even for the familiar expression “code of conduct”. My instinct would be simply label the document “Agreement relating to the Use of the Veto”. 
  • Agreement on the kinds of cases to which the veto restraint agreement applies. The 2013 Hollande/Fabius language requires simply “a mass crime”; other  possible formulations are the High Level Panel’s “cases of large-scale human rights abuses”, the S5’s “genocide, crimes against humanity and serious violations of international humanitarian law”, the US Genocide Task Force’s “cases involving ongoing or imminent mass atrocities”; and Secretary-General Ban Ki-Moon’s “situations of manifest failure to meet obligations relating to the responsibility to protect, as defined in para 139 [of the 2005 World Summit Outcome Document]”.

Recognizing how hard-fought and hard-won was the argument about the reach of R2P in the lead-up to the World Summit, my strong instinct would be to ensure that the language used resonates with the  “four crimes” language used in 2005 rather than use any terminology with potentially broader application. That does not necessarily mean replicating the exact 2005 words.  ‘Ethnic cleansing’ is not a separately recognised crime, but is rather a way or carrying out both  ‘genocide’ and ‘crimes against humanity’ and need not be specifically spelt out.  In the case of war crimes, which can technically be committed by just one individual against another, it would be desirable to make explicit what is presently implicit, that these need to be on a larger scale (as genocide and other crimes against humanity including ethnic cleansing must be as a matter of law).

It would also probably be desirable to limit the veto restraint agreement to situations where there is some real immediacy and urgency about the situation. Bringing these thoughts together, ossible language might, accordingly, be “situations where populations are experiencing, or at imminent risk of, genocide, crimes against humanity or major war crimes”.

  • Agreement on a trigger process for determining when an appropriate case for applying the veto restraint agreement has arisen.  To avoid unproductive and time-consuming procedural wrangling, it would be highly desirable, as the French proposal recognizes, to have some trigger mechanism outside the P5 itself to determine when a case has arisen that requires the application of veto restraint. France proposes that “at the request of at least 50 member states, the United Nations Secretary-General would be called upon to determine the nature of the crime. Once he had delivered his opinion, the code of conduct would immediately apply.”  Other external trigger mechanisms that have been proposed over the years include certification that the case is an appropriate one by the International Court of Justice, a two-thirds majority of the General Assembly, the High Commissioner for Human Rights, the Secretary General’s Special Advisers on the Prevention of Genocide and R2P, and regional organizations.

The overwhelming need here, given that many lives may be immediately at stake, is for a process which is speedy (which determination by the ICJ manifestly would not be), but at the same time gives P5 members some confidence that an objective determination is being made by individuals or an institution in which they have trust, and also provides some evidence that the issue is one giving concern to a wide cross-section of the international community. My instinct here is that the best solution, meeting these needs, may be to require a double trigger, with the relevant process requiring:

a) a certification by the  Office of the UN Secretary General’s Special Advisers on the Prevention of Genocide and R2P, which has the resources, expertise and credibility to make a quick objective determination of the nature and gravity of the situation,  that the situation is a proper one for application of veto restraint (for example, if my own suggested language were to apply, the certification would be that a particular situation involves a population “experiencing, or at imminent risk of, genocide, crimes against humanity or major war crimes”); and

b) a signed statement from at least 50 members of the General Assembly (including at least five members from each of the five recognized regional groups (African, Asia-Pacific, Eastern European, Latin American and Caribbean) and Western European and Others) that they believe the situation in question is of concern and satisfies the veto restraint requirement. 

If it were thought necessary to have a further formal mechanism for bringing these statements to the Council’s attention, that role could be played by the Secretary-General, if necessary exercising his authority under Article 99 of the Charter. But it may not be wise to give the Secretary-General a direct ‘certification’, as distinct from transmission-mechanism, role. Given the Secretary-General’s ongoing role as a diplomatic broker, it might be preferable to avoid putting him or her in the position of making a quasi-judicial determination, as the present French proposal would require.

  • Agreement on the kinds of Security Council action to which the agreement applies. It would be possible to limit veto restraint just to resolutions “authorizing military intervention for human protection purposes” as ICISS originally proposed, or as the US Genocide Prevention Task Force put it a little more broadly, “resolutions instituting and/or authorizing peace operations in situations where mass atrocities or genocide are imminent or underway”. But experience, makes clear that vetoes cast or threatened against resolutions merely condemning particular behaviour, or foreshadowing the application of measures falling well short of military coercion, can have a devastating practical impact: the failure of the Security Council for so long to even condemn what was happening in Syria undoubtedly helped to precipitate and prolong the carnage that we have there witnessed. My instinct here is that the agreement should apply to any Security Council action at all taken in response to situations where populations are experiencing or at imminent risk of genocide, crimes against humanity or major war crimes.

Protective conditions

  •  Agreement that a P5 member is not bound by the veto restraint agreement in certain circumstances. The French proposal is that restraint not apply “where the vital national interests of a permanent member of the Council were at stake”, mirroring  escape-route language in the original ICISS and High Level Panel reports. While having no obvious ethical appeal, it is probably, as M. Fabius acknowledges, necessary as a matter of political realism – and probably more acceptable to potential hold-outs on the P5 than alternative proposals like that of the US Genocide Prevention Task Force that there be no veto “unless three permanent members were to agree to veto”. The consolation is that not many real-world cases are likely to lend themselves to credible claims of this kind: it is difficult, for example, to see how the atrocity- related resolutions on Myanmar, Zimbabwe, and Syria vetoed in recent years by China and Russia could possibly have persuaded anyone else that they put at risk the “vital national interests” of either, even taking into account the realpolitik of Moscow’s political and military relationship with the Assad regime in Syria
  • Agreement to review the veto restraint agreement after experience of its operation. A further condition that might offer some element of comfort to reluctant P5 members would be to provide that their agreement is subject to review after five or ten years – perhaps through the application of a sunset clause, which would mean it lapsed unless renewed. Again this has no obvious ethical, but may have some political, appeal.

Next Steps

In taking its proposal forward – as I hope very much it will – I think the French Government needs to do four things, the first three of which at least are perfectly straightforward:

  • Draw up a detailed proposal fully spelling out its preferred position on what I have described as the clarifying and protective positions, and developing, circulating and advocating in all appropriate forums, full arguments in support.  There may be some temptation to avoid commitment to particular positions as long as possible, to avoid adverse reactions to particular language, but the general idea has been on the table for 15 years now, and if real progress is to be made in the context of the UN’s 70th anniversary in 2014, it would be desirable to sharpen the debate as soon as possible.
  • Develop and advocate detailed arguments in support of what I described as the scene-setting conditions, ie. that there is a strong ethical case for veto restraint, that the practical need for it is real, and that a voluntary agreement is worth having despite the risk of back-sliding. A systematic attempt should be made to get a very substantial majority of UN member states to make the same kind of supportive statement that 65 already have: it creates an environment of international pressure which reluctant P5 members will find hard to completely ignore.
  • Develop further specific arguments which may have some self-interest appeal to P5 members.  One such argument, made originally by ICISS in the context of atrocity crimes but also subsequently by others in the context of Security Council membership reform (admittedly in both cases so far to little effect), that if the P5 takes steps which undermine the credibility of the Council, and make it seem less relevant to the critical issues of the day “they can only expect that the Council will diminish in significance, stature and authority” [13]
  • Recognize what went wrong with the Security Council decision-making process in the case of Libya in 2011, when initial consensus disappeared in the face of what was seen as over-reach by the P3 [14] in translating, without appropriate further consideration in the Council, a narrow civilian-protection mandate into a broad ranging regime-change one, with this an important part of the explanation for the Council’s subsequent paralysis over Libya. And accept in that context the need to adopt some variation of the “Responsibility While Protecting” proposal originally made by Brazil, which would require at least informal agreement on the way use-of-force mandates are initially agreed and then monitored and reviewed by the Council during their lifetime. [15]

That last item on the list is rather less straightforward, and may be not quite so palatable for France, any more than it will be for the US or UK. But I think it will be crucial to the restoration of any kind of consensus on the Security Council as to how to handle in future the hardest mass atrocity crime cases. And I certainly think that some show of contrition on this front will be crucial if the two likely most reluctant members of the P5 on the veto restraint issue are to have any chance at all of embracing the cause – hugely attractive as that cause is to all of us here and, I believe, the overwhelming majority of the international community.  

Building political momentum for major change of this kind is never easy.  It requires a good,clear proposal; good, clear arguments in support of it; political sensitivity to opponents’ positions, but not to the point of capitulation on key issues; creativity, not least in meeting concerns without capitulation; a thick skin; and stamina. France is abundantly capable of delivering on all these fronts, and it is enormously to be hoped that this enterprise succeeds.

[1] International Commission on Intervention and State Sovereignty (ICISS), Gareth Evans & Mohamed Sahnoun Co-chairs, The Responsibility to Protect (Ottawa: IDRC, 2001) , paragraphs 6.19-21

[2] Ibid, p. XIII.

[3] Ariela Blatter and Paul D. Williams, The Responsibility Not To Veto, Global Responsibility to Protect 3 (2011) p.306

[4] Whose recommendation was that veto “use be limited to where vital interests are genuinely at stake” and that “the permanent members, in their individual capacities,…pledge themselves to refrain from the use of the veto in cases of large-scale human rights abuses”: UN Secretary-General’s High-Level Panel on Threats, Challenges and Change, A More Secure World: Our Shared Responsibility (New York: UN, 2004), paragraph 256.

[5]Whose recommendation was that “no permanent member should cast a [veto] in the event of genocide, crimes against humanity and serious violations of international humanitarian law”: Draft resolution A/60/L.49, 17 March 2006, Annex paragraph 14 (sponsored by Costa Rica, Jordan, Liechtenstein, Singapore and Switzerland); see also draft resolution A/66/L.42/Rev.2, 15 May 2012, Annex paragraph 20.

[6] Led by Liechtenstein and Switzerland, like S5, to focus on Security Council working methods: see http://www.centerforunreform.org/?q=node/541

[7] Which recommended in 2008 that a principal aim of US policy in the Security Council “should be informal, voluntary mutual restraint in the use or threat of a veto in cases involving ongoing or imminent mass atrocities. The P-5 should agree that unless three permanent members were to agree to veto a given resolution, all five would abstain or support it. This should apply, in particular, to resolutions instituting and/or authorizing peace operations in situations where mass atrocities or genocide are imminent or underway”: Genocide Prevention Task Force (GPTF), Preventing Genocide: A Blueprint for U.S. Policymakers (Washington DC: The US Holocaust Memorial Museum, The American Academy of Diplomacy, and the US Institute of Peace, 2008), p.106.

[8] Which urged the P5 “to refrain from employing or threatening to employ the veto in situations of manifest failure to meet obligations relating to the responsibility to protect, as defined in paragraph 139 of the [2005 World Summit] Outcome[Document]”: Report of the UN Secretary-General, Implementing the Responsibility to Protect (UN document A/63/677, 12 January 2009), paragraph 61.

[9] Who proposed that “a code of good conduct be defined by the permanent members of the Security Council, and that in the event of a mass crime they can decide to collectively renounce their veto powers”: http://gadebate.un.org/68/france .

[10] “The Charter would not be amended and the change would be implemented through a mutual commitment from the permanent members. In concrete terms, if the Security Council were required to make a decision with regard to a mass crime, the permanent members would agree to suspend their right to veto. The criteria for implementation would be simple: at the request of at least 50 member states, the United Nations secretary general would be called upon to determine the nature of the crime. Once he had delivered his opinion, the code of conduct would immediately apply. To be realistically applicable, this code would exclude cases where the vital national interests of a permanent member of the Council were at stake.”:  Laurent Fabius, “A Call for Self-Restraint at the U.N.” The New York Times, 4 October 2013.

[11] See References on the need for veto restraint by the UN Security Council in mass atrocity situations, Global Centre for the Responsibility to Protect at http://www.globalr2p.org/media/files/veto-restraint-references.pdf

[12] See UN draft resolutions S/2014/348, S/2012/538, S/2012/77, S/2011/612 (Middle East/Syria), S/2008/447 (Zimbabwe), and S/2007/14 (Myanmar).

[13] ICISS, paragraph 6.22

[14] France, the UK and US (since then widely referred to around UN corridors by the alternative acronym FUKUS)

[15] See for example Gareth Evans, Responsibility While Protecting, Project Syndicate January 2012,  http://www.gevans.org/opeds/oped118.html