The Diversity of Rules on the Use of Force: Implications for the Evolution of the Law

Written by

Last month, I had the pleasure and honour to deliver one of the keynote lectures at the Canadian Council of International Law Annual Conference. The theme of the conference was “Diversity and International Law” and I chose to speak about the diversity of rules on the use of force and the implications of that diversity for the evolution of the law. I am pasting the text of my lecture here

In this lecture I wish to address the question whether the law relating to the use of force – as set out in the UN Charter – has been capable, and is capable, of adapting to meet new threats and challenges facing the international community. My focus is not on the substance of the rules but rather on how they change. In particular, I wish to show that we need to be attentive to the nature of diverse nature of the rules in this area as we think about the possibility of their evolution.

Yesterday was the 74th anniversary of the entry into force of the United Nations Charter. It was around the time of the 50th anniversary of the UN that literature began to emerge suggesting that we might think of the UN Charter as a constitution for international society. Whether one agrees with that characterisation or not the Charter shares at least a couple of features with constitutions – it aims to lay down an overarching framework for the community it applies to, and is intended to be an abiding document in terms of duration. This immediately raises questions about whether the document can continue to regulate new and unforeseen challenges. This is particularly true of the Charter rules relating to the use of force.

Areas Where Evolution of the Charter Rules on Use of Force Have Been Called For

One can think of at least four areas where it has been argued that rules of the UN Charter ought to be adapted (or have been adapted, depending on one’s point of view) to meet new challenges :

(1)  First, there are the debates around the permissibility of the use of force by states, acting individually or collectively, but without United Nations Security Council authorization, for the purpose of stopping or preventing a humanitarian catastrophe. These are the debates about whether there is a rule permitting humanitarian intervention. As we know, the UN Charter does not include such a rule but a few states now maintain that such a rule exists and scholars continue to debate whether such a rule should exist.

(2) Second, it may be argued, it has been argued, that the rules relating to use of force by the UN Security Council acting under Chapter VII have changed in the time between the adoption of the Charter and now. The Council is now seen to possess the competence to act both with regard to internal situations and to prevent or put an end to humanitarian crises. Arguably, the drafters of the Charter did not foresee this role for the Council and had their minds mainly on Council action in inter-state conflicts.

(3) Third, there have been arguments about the legality of the use of force by states in anticipation of an armed attack. This is the debate about anticipatory or pre-emptive self-defence. While Article 51 of the UN Charter provides that nothing shall impair the inherent right of individual or collective self-defence “if an armed attack occurs”, some have argued that the “victim state” does not need to have suffered the armed attack, before the right to self-defence can be exercised.

(4) Fourth, there is the debate about whether the right of self-defence can be exercised in response to armed attacks by non-state actors. While there is nothing in text of Article 51 of the UN Charter that specifies that the right of self-defence only exists in relation to an armed attack that comes from a state, in 1986, the International Court of Justice in the Nicaragua case held that the right of self-defence exists in response to armed attacks that are attributable to another state. On that view, self-defence can only be exercised with respect to an armed attack originating from a non-state actor where the non-state actor was sent by a state or a state had substantial involvement in the acts of that non-state actor. This view of the ICJ seems to have reflected state practice at the time, especially as was set out in the UN General Assembly’s 1974 Resolution on the Definition of Aggression. That resolution was adopted by consensus. However, since September 11, 2001 a number of states have taken the view that self-defence can be exercised in response to attacks by non-state actors even where the attack by that non-state actor is not attributable to a state. It was on this basis that the US and other states, took action in Afghanistan, against Al Qaeda in 2001, it was also on this basis that a coalition of states have taken action in Syria against ISIS. Most recently, we saw Turkey claiming to be acting in the exercise of this right when it invaded Kurdish areas of Syria about 10 days ago.

Evolution of Charter Rules Through State Practice: The Structural Issues

In each of these four areas that I have just outlined it may be argued that the rules as envisaged in 1945, or as articulated at some later point since then have required adaption to new challenges. The arguments as to whether these rules have evolved almost always go straight to analysing whether state practice today or over time is such that we should consider that a new rule has emerged. Or those arguments are more evaluative and consider whether the rule should change to reflect a perceived need or cure a perceived deficiency.

Less attention is usually paid to the question of how can these rules can change. This question is important, indeed critical, because the rules relating to the use of force are embedded in a treaty instrument with a difficult amendment procedure.  Article 108 of the UN Charter provides that amendments will come into force after a vote adopted by a 2/3rds majority in the General Assembly, and after ratification by 2/3rds of the membership, including by all the permanent members of the Security Council.

In the Nicaragua case, the International Court of Justice held that the prohibition of the use of force in Article 2(4) of the Charter and the right of self-defence under Article 51 are also to be found in customary international law. This might lead one to think that because these are rules of customary international law, they will change and adapt in the usual way that customary international law evolves over time and in accordance with evolving state practice. Indeed, in the debates about the four issues I mentioned earlier, it is typical to see extensive reference to state practice over time with the implicit suggestion (often implicit) being that as practice evolves so will the rules.

However, when one stands back to think about the structure and position of the rules in question – the diverse nature of those rules – it becomes clear that the issue of how they may or how they have changed is not that simple. Even if the rules relating to the use of force by states are contained in customary international law, they remain nonetheless treaty rules. Moreover, they are not just rules in any old treaty – these are rules in the UN Charter which claims superiority over other treaties.

What are the structural problems with simply analysing changes by reference to evolving state practice?

It should be acknowledged that state practice may be used in two different ways in order to support an argument that the rules have evolved.

(1) practice may be used to underpin an argument that customary international law has changed. Here of course we need a general state practice plus opinio juris.

(2) practice may be used in the interpretation of treaty rules where it establishes the agreement of the parties as to the interpretation of the treaty (Art. 31(3)(b) Vienna Convention on the Law of Treaties)

Three out of the four areas I highlighted earlier deal with the law relating to the use of force by states (as opposed to by the Security Council). In those three areas it is impossible to say that practice establishes the agreement of the parties as to the interpretation of the treaty. One need only think about the debates about anticipatory self-defence or self-defence in response to attacks by non-state groups. So, usually the claim is that practice has established a new customary rule or changed customary law.

But this leads to the question about the relationship between customary changes and the UN Charter.

There are three obstacles to accepting that customary law changes can affect the law on the use of force in the Charter.

(1) First, though it is conceivable that a customary rule can take precedence over a treaty, the ordinary presumption is that treaty rules prevail over custom – unless there is an indication that the parties intend to abandon the treaty provision. This means, that the treaty or the treaty provision has fallen into desuetude.

(2) The second obstacle derives from Art. 103 of the UN Charter, obligations under the Charter prevail over obligations under other treaties. This is not a rule that relates directly to the relationship between the Charter and customary law. Instead, it indicates that parties may not enter into a treaty which would derogate from or amend their Charter obligations except by way of amendment to the Charter itself. If the parties cannot change their Charter obligations expressly by treaty, it would be odd if they can amend the Charter, implicitly, by custom.

(3) The third obstacle is that the prohibition of the use of force – or at least aspects thereof – are considered to be rules of jus cogens – peremptory norms of international law from which no derogation is permitted. One consequence of the fact that some aspects of the prohibition are jus cogens is that it can only be changed by another peremptory norm of international law (Art. 53 VCLT).

So the bar seems to be set high – very high – for adaptation of the rules relating to the use of force. There are good reasons for this. These rules reflect some of the fundamental features of the post World War II legal system. Indeed the prohibition of the use of force is central to that system.

But adaptation is and has been possible to meet new challenges.

Subsequent Practice and Evolution of the Charter Rules on the Use of Force

(1) There are occasions when subsequent practice can legitimately be used to interpret the Charter (under Art. 31(3)b of the VCLT) – as opposed to forming new custom. This is particularly the case with regard to the UN collective security scheme under Chapter VII. There, we have seen practice of the Council that has been endorsed, implicitly or even explicitly, by the membership as a whole and which interprets the Charter in new ways. One can think of the interpretation of the concept of threat to the peace to include internal matters or humanitarian challenges. This has opened the door to the Council authorising force on numerous occasions now for the purpose of protection of civilians or in internal situations. That practice was explicitly endorsed by the membership of the UN in the provisions of the World Outcome Summit Document of 2005 dealing with responsibility to protect (paras. 138-139). That General Assembly resolution adopted by heads of states and governments, by consensus, expressed support for actions by the Security Council to help to protect populations from international crimes. Whatever the correct interpretation of Article 39 and of Chapter VII of the Charter in 1945, there is now subsequent practice expressing the agreement of the parties to the Charter, which must be taken into account, under the Vienna Convention on the Law of Treaties, in interpreting the Charter.

(2) Secondly, despite my points earlier – changes to customary law are not irrelevant to the law on the use of force. From what I have said earlier, it is difficult to see how changes to customary law can affect the prohibition of the use of force in Art. 2(4), that is, it is difficult to see how custom can create new exceptions to the prohibition.

However, changes to customary law are relevant to the evolution of the right of self-defence under Article 51 of the Charter. The text of Article 51 refers to the inherent right of self-defence. In the Nicaragua case, the ICJ has accepted that this reference to an inherent right is a reference to a customary international law right (para. 176, Merits Judgment 1986).  This is not a static reference to custom, by which I mean that this not simply a reference to custom as it was in 1945, but rather a dynamic reference to custom. In other words, it is a reference to the customary law of self-defence as it evolves over time. What Article 51 is saying in short is that despite the prohibition of the use of force, States may act in accordance with the customary right of self-defence, (while adding some procedural obligations that must be fulfilled). So customary self-defence is preserved by the Charter and that is why we have regard to the requirements of necessity and proportionality, though these are not to be found in the Charter.

What this means is that as the customary law of self-defence changes, the Charter accommodates those changes since the Charter rule on self-defence is only given content by the customary law on self-defence.  It is this dynamic reference to custom that allows the law of self-defence to adapt and that makes it legitimate for us to debate the implications of the practice around self defence against non-state actors. If customary international law allows such uses of force, then so does Article 51 allow.

However, we must be careful in taking this methodology of assessing custom outside the law of self-defence to other areas of the law relating to the use of force – particularly when one is thinking of somehow adapting the prohibition of the use of force or the UN collective security scheme.  There one would need a different methodology or concept of adaptation. Not customary evolution but practice that establishes the agreement of the parties as to the interpretation of the Charter.

Why does this methodology not apply to the prohibition of the use of force? Though the ICJ pointed to the identity of custom and the Charter rules on force, both in relation to the prohibition in Article 2(4) and the self-defence exception in Article 51, it is important to realise that the relationship between custom and treaty is not the same in both cases. As just shown, in the case of Article 51, the identity of custom and treaty arises because the treaty rules preserves customary rules of self-defence.

In the case of the prohibition of the use of force, the identity of the rules arises not because the treaty itself says anything about custom but because custom has come to mirror the treaty. There is no intrinsic link between the two. A change in the customary rule would not in and of itself change the treaty prohibition. That prohibition would remain unchanged unless it were to be interpreted differently using the rules of treaty interpretation.

The main significance of the point that any changes to the customary prohibition of the use of force does not change the treaty prohibition is to be seen with regard to the arguments around humanitarian intervention. Any rule permitting humanitarian intervention would have to create a new exception to the prohibition of the use of force. Recall that this prohibition has three dimensions:

  • It exists as a customary rule
  • It exists as a treaty rule in Article 2(4) of the UN Charter
  • It, or at least, the prohibition of aggression is a peremptory norm of international law, a jus cogens rule

There is much debate as to the implications of practice regarding the use of force for humanitarian ends for the doctrine of humanitarian intervention. Some, most notably, the UK government, argues that there is a rule permitting such, others have argued more recently, that the silence, the failure to condemn, by many states of the US/UK/French action taken in response to the use of chemical weapons in Syria means that even if the law has not changed, it is changing. Or at least that it is not as clear as it once was.

However, all of this assumes that the creation of an exception to the customary rule can change the treaty prohibition. Even if there was much more practice than we see now, this would not affect the treaty prohibition to be found in Article 2(4).

The argument that the creation of a customary law exception to this prohibition in the Charter would have dramatic implications for the primacy of the UN Charter. It would upend Article 103 of the Charter. In the context of the imposition of sanctions by the Security Council, there have been states that have refused to comply with sanctions for one reason or another. One might ask whether it is possible to develop a customary international law exception to the obligation imposed on States to comply with sanctions imposed under Charter VII of the Charter. I would imagine that some of the states that seem to argue that custom can change the UN prohibition of the use of force would take the view that such a customary rule could not do this.

We would also need to think about the implications of the prohibition of aggression being jus cogens. No rule of custom can come into existence if it comes into conflict with a peremptory norm of international law (see Conclusion 14, Draft Conclusions on Peremptory Norms of General International Law (Jus Cogens), adopted on first reading, 2019). Given that a rule permitting humanitarian intervention would not only modify the prohibition of aggression, but that it would also be in conflict with it, in that it would seem to permit things that the prohibition of aggression would otherwise prohibit, it is difficult to see how a rule of customary international law permitting humanitarian intervention could even come into existence in the first place. Norms of jus cogens can only be changed by other norms of jus cogens. Thus, for there to be a customary rule permitting humanitarian intervention it would have to be a rule of jus cogens.

Possibilities for the Emergence of a Rule Permitting Force for Humanitarian Purposes

Now does all of this mean that it is impossible for the law to be developed to permit states to use force for humanitarian purposes, without Security Council authorization. I see three possibilities.

The first would be to seek to argue that the there is a jus cogens rule that permits humanitarian intervention and which modifies the jus cogens prohibition of force or prohibition of aggression. For this to happen thought would have to be given to the possibility of a jus cogens rule containing permissions rather than prohibitions.

The second would be to argue that Article 2(4) is to be interpreted as not prohibiting the use of force for humanitarian purposes. For all sorts of reasons, which I won’t go into now, this would not be the interpretation of that provision that would be arrived at using the ordinary rules of interpretation contained in Article 31 of the Vienna Convention on the Law of Treaties. However, if there were to be subsequent practice which establishes the agreement of the parties as to this interpretation of the provision then matters may change

The third way of breaking the strangle hold of the UN Security Council on the force for humanitarian purposes is for that force to be authorised by the UN General Assembly under the “Uniting for Peace” Resolution of 1950. That resolution provides that

“if the Security Council, because of lack of unanimity of the permanent members, fails to exercise its primary responsibility for the maintenance of international peace and security in any case where there appears to be a threat to the peace, breach of the peace, or act of aggression, the General Assembly shall consider the matter immediately with a view to making appropriate recommendations to Members for collective measures, including in the case of a breach of the peace or act of aggression the use of armed force when necessary, to maintain or restore international peace and security.”

In that resolution the UN General Assembly arrogated to itself the power to the take collective measures, including the use of armed force, to maintain international peace and security. Whether the General Assembly does indeed have this power is ultimately to be traced back to the UN Charter rather than to this resolution. Article 11(2) of the UN Charter provides that the

“the General Assembly may discuss any questions relating to the maintenance of international peace and security” and that it may make recommendations with regard to any such questions to the state or states concerned or to the Security Council or to both. 

While that provision also says that its provisions are subject to Article 12 which provides that the Assembly may not make any recommendations while the Security is exercising its functions in relation to a dispute or situation. However, more recent UN practice, the legality and effect of which have been confirmed by decisions of the ICJ (in the Kosovo and Israeli Wall in Palestine Advisory Opinions), suggest that matters may be dealt with in parallel by the Security Council and the General Assembly, and that the Assembly may make recommendations.

Thus, the only question that will be left is whether an Assembly resolution recommending force would be deemed not to be a breach of the prohibition of force under Art. 2(4) in the same way that a Council resolution authorizing force would have that effect. This is an interpretation of the Charter that is open for UN member states to take. If there were to be agreement on this principle, that agreement would be an important point to be taken into account in the interpretation of the treaty.

This way of achieving the end would also not fall foul of the various structural problems I have alluded to above. It would avoid the suggestion that customary international law can change either the UN Charter or a rule of jus cogens.

Print Friendly, PDF & Email

Leave a Comment

Comments for this post are closed


Mary Ellen O'Connell says

November 11, 2019

Dear Dapo,

I very much appreciate your note of caution about methodology in interpreting UN Charter rules, in particular, Article 51. Christian Tams, Dire Tladi, and I stress exactly this in our approaches to treaty, customary law, and jus cogens interpretation of rules on the use of force in our new book, Self-Defence Against Non-State Actors (CUP 2019).

Let me mention just one significant difference I reach on the critical words "inherent right" in Article 51:

The ICJ found that the reference in Article 51 to ‘inherent right’ did bring forward pre-Charter law but only principles that further restrict self-defence, such as necessity and proportionality. These principles, not mentioned in the text, are nevertheless consistent with the express terms of Article 51, in particular, armed attack. ICJ, Military and Paramilitary Activities, 94.

Interpreting Article 51 to evolve away from its text restrictions would also also conflicts with the prohibition on derogating from jus cogens.

Thinking of veterans today and all who have experienced war,

Mary Ellen

Dapo Akande says

November 11, 2019

Dear Mary Ellen,
Thanks for your comment! I can see that reference to the inherent right was used by the Court to show that the right to self defence was further restricted by the principles of necessity and proportionality. However, I can't see that this means that those principles can only evolve in one direction. In my view, that they derive from custom simply means that they can evolve in line with custom.


Adil Haque says

November 11, 2019

Hi Dapo,

Terrific post. I do think this claim is too strong: “the Charter rule on self-defence is only given content by the customary law on self-defence.” This may be true with respect to internal constraints on the exercise of self-defence, like necessity and proportionality. But the Charter rule on self-defence has an explicit triggering condition: the occurrence of an armed attack.

So, I think, the permissibility of anticipatory self-defence under the Charter cannot turn on pre-Charter custom, or post-Charter custom (as distinct from subsequent practice in the application of the Charter). I suppose the parties could agree to interpret “if an armed attack occurs” as surplusage, even though it was originally understood as restrictive.

Personally, I think “armed attack” was understood to mean “armed attack by a State.” If I’m right, then the permissibility of (using armed force on the territory of a non-consenting State in) self-defence against non-State actors also cannot turn on pre-Charter custom, or post-Charter custom (again, as distinct from subsequent practice in the application of the Charter).

It’s also worth asking *which* customary right of self-defence Article 51 partly incorporates and partly limits. The customary right of self-defence against an aggressor State (arising roughly with the Paris Peace Pact) seems quite distinct from the natural law/customary right of self-defence/preservation that may have justified or excused infringing the territorial integrity of an innocent State under the pre-Pact law of peace. While it’s possible that the parties to the Charter conflated them, I would be surprised, since their contents and functions seem so very different.

See you at ELAC,


Dapo Akande says

November 11, 2019

Dear Adil,

I agree with your central point that custom does not define all the conditions for lawful self-defence under the Charter and that there are requirements which the Charter itself imposes. I also agree that since the requirement of an armed attack is one that is imposed by the text of the Charter it cannot be removed by custom. Therefore I share your view that the question of whether anticipatory self defence can evolve depends in the first place on treaty interpretation. I didn't address this point in the lecture but it did come up in the Q & A at the CCIL conference. I am less sure about the claim that armed attack was understood to mean armed attack by a state.

Darryl says

November 12, 2019

Hi Dapo,
I think that this is an excellent idea - to combine Uniting for Peace with R2P. It squares the circle. The understandable concern about humanitarian intervention is the obvious danger of unilateral, imperialistic, pretextual invocation. But the concern with having no such doctrine is that a state can kill its own people a massive scale, and concerned watchers around the world are forbidden to intervene to protect people if the UNSC is paralyzed by a P-5 veto.

The need for 2/3 of the GA would give a bright line procedural criteria, alleviating concerns about unilateral abuse. This proposal would update UN practice to 21st century conditions, and would reduce injustices and bad effects from a dysfunctional UNSC. There could still be problems with under- or over-permissiveness but it seems far better than the alternatives. I hope the idea gets attention.

John R Morss says

November 13, 2019

Very interesting and informative. Just wondering (at a tangent) about the terminology of 'evolution', (and relatedly, 'adaptation')... what do we seek to capture, or assert, with this 'naturalistic' (or is it 'apologetic'?) vocabulary..
Is it supposed to indicate (contra Darwin) that 'more evolved' is 'better'? What is adapting to what? And if Customary IL 'evolves', what does treaty law do?
Just food for thought!

JS says

November 15, 2019

Regarding humanitarian intervention, if our understanding of self-defence has evolved to permit response to armed attacks from non-state actors why can’t collective self-defence evolve to permit coming to the aid of non-state actors under attack from their own governments?

Simon Wood says

December 7, 2019

This is a world class discussion greatly appreciate all the contributions. Two issue immediately come to mind:
1) What if GA and SC take different views on a matter so that there is a possibility of not just disagreement, but a disagreement that some states might seek to resolve by force? There would need to be some agreeable way to resolve a disagreement between these two leadership organisations on matters of threat to international peace and security. eg. Israel-Palestine could easily lead to a quick flashpoint.
2) Intervention by SC can be disastrous: eg UNSC resolutions against Iraq in 1990s were deadly and have been roundly condemned eg2 resolutions to implement no fly zone and other measures in Libyan civil war ended up destabilising that country badly. Hence be careful of what you wish for as greater power to act may in fact result in greater harm caused.