Home EJIL Analysis Legal Bindingness of Security Council Resolutions Generally, and Resolution 2334 on the Israeli Settlements in Particular

Legal Bindingness of Security Council Resolutions Generally, and Resolution 2334 on the Israeli Settlements in Particular

Published on January 9, 2017        Author: 

As I have read commentary on the recently adopted resolution by the U.N. Security Council (Resolution 2334) addressing Israeli settlements in the occupied territories, I’ve noticed a number of commentators who appear to assume that, since this resolution was not explicitly adopted in exercise of the Council’s Chapter VII powers, therefore all of its operative provisions are per se legally non-binding. Orde Kittrie, writing over at Lawfare, seems to make this assumption clear when he writes:

“Resolution 2334 was not adopted under Chapter VII of the U.N. Charter and is not legally binding. The resolution does not create additional legal requirements on Israel, nor does the resolution require (or even explicitly encourage) U.N. member states to impose sanctions on Israel in response to Israeli settlement activity.”

I thought this would be a good opportunity to write briefly to clarify that the legal obligation for U.N. Charter states parties to comply with the decisions of the Security Council, contained in Article 24 and 25 of the Charter, is not contingent upon the Council’s acting in exercise of its Chapter VII powers. Any decision of the Security Council is legally binding upon all U.N. member states, whether or not the text of the resolution explicitly references Chapter VII.

Rather, the key question for determining whether a particular provision of a Security Council resolution is legally binding on member states (i.e. whether the provision is a “decision” of the Security Council), including the specific addressee of the resolution, is whether the Council has chosen to use words within the provision indicating its intent to create a legally binding obligation.

The International Court of Justice made these points clear in its 1971 Namibia advisory opinion, in Paragraphs 108-114. There, the Court was considering the juridical implications of provisions of Security Council Resolution 276, which had similarly been adopted with no textual indication that the Council was acting in exercise of its Chapter VII powers. The Court held that:

“It has been contended that Article 25 of the Charter applies only to enforcement measures adopted under Chapter VII of the Charter. It is not possible to find in the Charter any support for this view. . . It has also be contended that the relevant Security Council resolutions are couched in exhortatory rather than mandatory language and that, therefore, they do not purport to impose any legal duty on any State nor to affect any right of any State. The language of a resolution of the Security Council should be carefully analysed before a conclusion can be made as to its binding effect. In view of the nature of the powers under Article 25, the question whether they have been in fact exercised is to be determined in each case, having regard to the terms of the resolution to be interpreted, the discussions leading to it, the Charter provisions invoked and, in general, all circumstances that might assist in determining the legal consequences of the resolution of the Security Council. (Para 113-114)”

Applying this test for determining bindingness, the Court determined that the provisions in operative paragraphs 2 and 5 of Resolution 276 were legally binding on all U.N. member states. This included the determination by the Council in operative paragraph 2 that the presence of South African forces on the territory of Namibia was unlawful, and the Council’s call in operative paragraph 5 for all states to refrain from any dealings with South Africa that were inconsistent with this determination.

The question of which words will indicate the Council’s intent to create binding obligation is one that has been discussed in scholarly literature, though honestly not as much as the topic deserves. I recently engaged in an analysis of this subject in Chapter 6 (Pgs. 195-198) of my book on Iran’s nuclear program.

It is interesting to note in this context that in the Namibia advisory opinion, the Court found to be legally binding a provision (operative paragraph 5) which began with the words “Calls upon all States . . . ” Most scholarly commentary over the succeeding decades (including mine) has, however, categorized “calls upon” language as legally non-binding. So there would appear to be some room for disagreement over which words fall into which category.

Moving from general principles to the particular case of Security Council Resolution 2334, by far most of the resolution’s operative paragraphs do utilize words and phrases such as “calls upon,” “reaffirms,” “underlines,” and “stresses,” which are generally (with the above caveat from the Namibia case) understood not to indicate an intention on the part of the Council to create binding legal obligation for U.N. member states or for the specific addressee of the resolution. There is one exception, however. Operative paragraph 2 states that the Council:

“Reiterates its demand that Israel immediately and completely cease all settlement activities in the occupied Palestinian territory, including East Jerusalem, and that it fully respect all of its legal obligations in this regard.”

The word “demand” in this provision sticks out. “Demand” is a word that has been used by the Council to command an addressee to abide by its obligations existing independently in international law. An example of its use in this context is found in operative paragraph 17 of Resolution 687 (1991), in which the Council:

“demands that Iraq adhere scrupulously to all of its obligations concerning servicing and repayment of its foreign debt”

However, the word “demand” has also been used by the Council to create new legally binding obligations for an addressee, in contexts in which there does not exist a relevant obligation elsewhere in international law. The best example of its use in this way is in operative paragraph 2 of Resolution 1696 (2006), wherein the Council:

“Demands, in this context, that Iran shall suspend all enrichment-related and reprocessing activities, including research and development, to be verified by the IAEA;”

So how should we understand the use of the term “demand” in operative paragraph 2 of Resolution 2334? The substance of the demand that the Council makes in this provision is not new. The Security Council has enjoined Israel to discontinue settlement building in the occupied territories on a number of prior occasions, including in operative paragraph 3 of Resolution 446 (1979), and operative paragraph 6 of Resolution 465 (1980). However, in both of these prior examples the injunction was framed with the term “calls upon.” Resolution 2334 is, as far as I can tell, the first time in which the word “demand” has been used by the Council in association with the injunction to discontinue settlement building.

Is this escalation in wording by the Council legally significant? If so in what way? Again, when “demand” has been used by the Council in the past, it has either been to add the Security Council’s own command to an already existing international legal obligation by the addressee, or to create a new legal obligation for the addressee where one does not already exist.

So does Israel have an already existing international legal obligation to discontinue settlement building in the occupied territories, to which the Council is simply adding its own command here?  There would seem to be a strong basis for this argument in the determination by the International Court of Justice in its 2004 Wall advisory opinion that, under international humanitarian law, Israel’s settlements in the occupied territories are illegal, with the clear implication that Israel’s international legal responsibility requires their cessation and dismantlement.

So if the Council’s demand in Resolution 2334 is read, as seems most persuasive, as adding the Council’s own command to Israel to abide by its independently existing legal obligations under international humanitarian law, what juridical effect, if any, does the Council’s own demand have?

It could be argued that the Council’s demand acts to create a “topped up” legal obligation, i.e. one that is based on the already existing one (here grounded in international humanitarian law) but which, by operation of the Council’s demand, is additionally made part of the addressee’s U.N. Charter obligations which, pursuant to U.N. Charter Article 103, would prevail over obligations under any other international agreement.

Is that what the Council meant to do in this case? Is that why they changed the injunction from one called for to one demanded?

I’m not sure it’s clear what the Council meant by adding the word “demand” here. But we do know from the Namibia advisory opinion that the words used by the Council in its resolutions do matter for legal analysis. We also know that it is possible for provisions in a Council resolution not textually linked to Chapter VII to be legally binding.

I will refrain from reaching here a conclusion about operative paragraph 2 of Resolution 2334. But I definitely don’t think that one can conclusorily determine that Resolution 2234 does not create, or at least add to the juridical character of, legally binding obligations for Israel.

Print Friendly

12 Responses

  1. VJS

    Thank you for this insightful analysis!

  2. Marko

    Thank you, Dan, for your persuasive take on this.

  3. Roger O'Keefe

    Dear Dan
    Many thanks for your excellent post, which helpfully reminds readers that a decision of the Security Council need not be taken under chapter VII for it to be binding on UN member states by virtue of article 25 of the Charter. It simply happens to be as a matter of practice and logic that most decisions binding by virtue of article 25 are taken under chapter VII, by which the Council is empowered to take coercive measures – that is, to push member states into doing what they may not be minded otherwise to do – for the maintenance of international peace and security.
    As to what constitutes a binding decision, however, and in belated response too to Dapo and Marko’s post many moons ago on SC res 2249 (2015), has not the considered and plentiful modern practice of the Council been refined to the point whereby these days the Council signals that it is taking a decision within the meaning of article 25 of the Charter – which, being such, is ipso facto binding – solely by using the word ‘[d]ecides’? In other words, whatever it may or may not have done back in the day, surely today the Council does not use language like ‘[d]emands’, let alone ‘[c]alls upon’, to create de novo obligations binding on member states by virtue of article 25 of the Charter. No ‘[d]ecides’, no decision, whatever individual members of the Council may subsequently allege in favour of a position pushed during the drafting that did not meet in the event with the sufficient support of other Council members.
    I am interested to hear what you (or others) may think.
    All the best for 2017, despite the dark portents

  4. Kriangsak Kittichaisaree

    Dear Dan and Roger,

    Very interesting points. UNSC resolutions are products of realpolitik rather than IL; hence, there can be no ready-made IL answers to Roger’s question.

    At the UN, we have international lawyers and non-international lawyers working side by side on draft resolutions. And even international lawyers do not necessarily share the same opinions. I recall, not long ago, one prominent IL expert insisting at one UN mtg. that only UNSC resolutions with operative paras. coming after ‘DECIDES: –‘ bind UN Member States, because Art. 25 of the UN Charter says so. A lot of people there agreed with him (NB: This was long after the 1971 Namibia Adv. Opn. mentioned by Dan).

    However, the wording of UN resolutions also reflects the purely political, non-legal, views of the delegates who draft them (e.g., condemn, renounce, implore, deplore ..etc.) and the final outcomes are usually achieved after a long process of drafting changes to try to accommodate the political stances of every delegation in the room (i.e., toning down versus heating up the language; plus the divergent goals of the negotiators). International lawyers do not always have the final say.

    The UNSC resolution in question also followed the same journey. It would not have been acceptable to all the P5, who have the veto, if much stronger wording had been used (e.g., “DECIDES”: – …). But this does not necessarily mean that it is not binding — one has to look at paras. 113-114 of Namibia …… So?

  5. Jordan J Paust

    Dan: these are all important points and, as you recognize, the new SC res. is not in a vacuum. There have been several relevant SC resolutions, plus several GA resolutions, the ICJ Adv. Op., and many other evidences of opinio juris. See
    And as former Sec. Baker, Sec. Kerry, and others have noted, the recent vote was not the first abstention or approval by the US. See also ssrn article on US formal statements and reports.
    The ssrn article demonstrates why this matter could go before the ICC as well as cases involving a number of types of international crimes by Palestinian accused.

  6. Thanks to all for the comments so far, which I think are particularly useful and engaging.

    Roger, your argument makes a lot of sense to me, and I certainly agree that the practice of the Council has evolved since the Namibia opinion, such that at this point language such as “calls upon” should almost certainly be considered not to express the Council’s intention to create legally binding obligations. I also agree that in by far most cases, Council practice has evolved to focus binding intent on provisions beginning with the word “decides.” And the formalist in me wishes to god they would do us the favor of just making it that simple always – i.e. when and only when they use the word “decide,” a provision is legally binding. Honestly, I think that “demand” is the only other word that they have appeared to use in recent years in a legally-binding-intent sort of way. And Resolution 1696 that I cited to in the piece is the clearest example of it. Again, we’re all operating with imperfect information because the UNSC has never published its own lexicon (wouldn’t that be nice!), but all the commentary from states and commentators I’ve ever read about the “demand” on Iran to cease uranium enrichment, adopts the understanding that the Council intended this to be a legally binding provision. Now, I have no normative problem revisiting that conclusion if indeed we, collectively, want to hold the Council to having to say “decides.” Again, from a formalist, systemic perspective that would be ideal. But I suppose my main concern is consistency. What we can’t have is “demand” meaning legally binding in some contexts but not in others.

    Kriangsak’s comment also, though, usefully reminds us formalists that we cannot read UNSCRs as if they were drafted and agreed upon exclusively by lawyers who are mindful about systemic consistency, and I think that is an important point to remember.

    With that caveat fully taken into account, however, to the extent the Security Council and its member states desire the decisions of the Council to be understood to have legal implications – which I think they do – it behooves them to use language as consistently as possible when they wish to express their intent to create binding legal obligations.

  7. H.Melkman

    Dear Dan et al.,

    I feel like a dwarf amongst giants at this forum.
    May I try nevertheless to exert my audicity.

    1. UN Res.2334 is evidently contravening UN Res. 242 (1967).
    2. Occupied Palestinian territory. Never understood what it meant.
    3. References to Namibia case I suggest is more political rather than legally pertainable on this case.
    4. Two state “solution” is a contradictio in terms not to appearing on this level previously.

    Sincerely yours,


  8. Marcus

    I was wondering what could be said looking at the Kosovo AO, in paragraph 85. The ICJ puts a great deal of emphasis on the fact that the resolution being discussed was a chapter VII one to prove it is binding. This seems to contradict what it said in the Namibia AO. ie if the resolution would have been binding no matter how it was adopted, why did the ICJ need to emphasize it’s a chapter VII resolution to show it’s binding? Here’s the paragraph:

    Within the legal framework of the United Nations Charter, notably on the basis of Articles 24, 25 and Chapter VII thereof, the Security Council may adopt resolutions imposing obligations under international law. The Court has had the occasion to interpret and apply such Security Council resolutions on a number of occasions and has consistently treated them as part of the framework of obligations under international law (Legal Consequences for States of the Continued Presence of South Africa in Namibia (South West Africa) notwithstanding Security Council Resolution 276 (1970), Advisory Opinion, I.C.J. Reports 1971, p. 16; Questions of Interpretation and Application of the 1971 Montreal Convention arising from the Aerial Incident at Lockerbie (Libyan Arab Jamahiriya v. United Kingdom), Provisional Measures, Order of 14 April 1992, I.C.J. Reports 1992, p. 15, paras. 39-41; Questions of Interpretation and Application of the 1971 Montreal Convention arising from the Aerial Incident at Lockerbie (Libyan Arab Jamahiriya v. United States of America), Provisional Measures, Order of 14 April 1992, I.C.J. Reports 1992, pp. 126-127, paras. 42-44). Resolution 1244 (1999) was expressly adopted by the Security Council on the basis of Chapter VII of the United Nations Charter, and therefore clearly imposes international legal obligations. The Court notes that none of the participants has questioned the fact that resolution 1244 (1999), which specifically deals with the situation in Kosovo, is part of the law relevant in the present situation.

  9. Kriangsak Kittichaisaree

    Dear Marcus,

    UNSC resolutions adopted under Chap. VII of the UN Charter are definitely binding — and that was the gist of the para. in the Kosovo AO quoted by you. But, as Dan and Roger have argued, this does not necessarily mean that UNSC resolutions adopted outside Chap. VII are not binding…

  10. Roger O'Keefe

    Dear Kriangsak
    With all due respect, only ‘decisions’ of the UNSC are binding. A paragraph of a UNSC resolution may well be adopted under chapter VII, but if it does not constitute a ‘decision’ within the meaning of article 25 of the Charter it cannot possibly bind member states. Chapter VII does not render binding what article 25 does not.
    Best wishes

  11. Kriangsak Kittichaisaree

    Dear Roger,

    Sorry that I misunderstood your first message, and thanks for the correction.

    Dear All,

    The more I think of it, the more I am convinced that the question posed by Dan may not be the right one. The real issue is not “whether UNSC Res. 2334 is legally binding under IL”, but “what does UNSC Res. 2334 practically/really mean [under IL]?”

    We can look at the Namibia AO etc. to help us understand the practical implication of UNSC Res. 2334. One crucial circumstance in its adoption is the US position which did not veto the res. The Explanation of Vote by the US can be found at:

    As my first intervention tried to explain that not all the “pen-pushers” of UN resolutions are international lawyers or conscious of the need to be consistent in the choice of terminology. Normally, coordinators of resolutions pick up previous resolutions on the same or similar topic to work on, and then all sorts of changes are proposed by the others.

    Would a UNSC resolution without the word “decides” among its operative paras. be a “decision” in the sense Roger has explained? What of Dan’s argument re: “demand”, etc.?

    The practical meaning of UNSC res. 2334 will have to be judged by reading the explanations of vote by the UNSC members, esp. that of the US, without whose abstention the resolution would not have seen the light of day.

  12. […] various of those issues, and namely the question of its binding effect have already been dealt with here, there still remain quite a number of open issues that require further clarification, some of which […]