(Editor’s Note: This is Part II of a two-part series discussing issues of use of force, aggression, and armed attack in light of the 75th anniversary of the UN Charter. The series aims to revisit the time between the conception and birth of the Charter in order to find new insights relevant to ongoing debates. Part I provides a different understanding of the relationship between Article 2(4) and Article 51 of the Charter. This part explains Article 51 as an exception to the authority of the Security Council to enforce peace and security.)
The United Nations Charter opened for signature on June 26, 1945. Seventy-five years later, two of its provisions seem more important than ever. Article 2(4) prohibits the use of force, while Article 51 permits self-defense. On the standard view, a grave violation of Article 2(4) triggers Article 51, while Article 51 creates an exception to Article 2(4).
This standard view is not incorrect, but it is incomplete. A more complete picture emerges when we view the whole text of the Charter in light of its drafting and negotiation history.
Part one of this essay showed that those who drafted and negotiated Article 51 understood it to permit self-defense against armed aggression. Decades passed before States firmly established that grave uses of force constitute armed aggression. So, today, grave violations of Article 2(4) trigger Article 51 because they constitute armed aggression.
This part shows that Article 2(4) has an exception, but it is not Article 51. Article 51 is an exception, but not to Article 2(4). Between these two articles, between force and self-defense, lies the collective security structure established by Chapter VII of the Charter. Article 2(4) prohibits the use of force, but makes an exception for armed force permitted within that structure. That structure generally requires the Security Council to authorize armed force, but makes an exception for self-defense if an armed attack occurs before the Security Council takes necessary measures. So Article 51 is an exception to an exception to Article 2(4).
This more complete view explains why there are no unwritten exceptions to the prohibition of force. It explains why the Charter did not permit self-defense before an armed attack occurs, against minor uses of force, or targeting non-State actors. As always, subsequent practice in the application of the Charter may establish the agreement of the parties to a new interpretation. But to see where we are, and to imagine where we should go, we should first remember where we began.
The Text of the Charter
Article 2(4) of the Charter provides that
All Members shall refrain in their international relations from the threat or use of force against the territorial integrity or political independence of any state, or in any other manner inconsistent with the Purposes of the United Nations.
And Article 51 provides that
Nothing in the present Charter shall impair the inherent right of individual or collective self-defense if an armed attack occurs against a Member of the United Nations, until the Security Council has taken measures necessary to maintain international peace and security.
On the standard view, Article 2(4) lays down a general rule, while Article 51 carves out a specific exception. Crucially, there are no unwritten exceptions to the prohibition of force. Anticipatory self-defense, self-defense against non-State actors, and self-defense against low-gravity uses of armed force are either permitted under Article 51 or prohibited under Article 2(4).
The textual argument for the standard view is straightforward. Article 2(4) prohibits force. Self-defense against armed attack involves force. Yet self-defense is not prohibited. So the use of force in self-defense must fall within an exception to the general prohibition. As article 51 states, nothing in the Charter impairs the right of self-defense. And “nothing” includes Article 2(4).
The textual argument for the alternative view is also straightforward, though it has a few more steps. Article 51 appears at the end of Chapter VII of the Charter, which concerns action with respect to threats to the peace, breaches of the peace, and acts of aggression. Every article in Chapter VII involves the Security Council. Article 39 says that the Security Council shall determine the existence of any act of aggression and decide what measures shall be taken. Article 42 says that the Security Council may take measures involving the use of armed force. Of course, States acting in self-defense make their own determination that an act of aggression constituting an armed attack has occurred, and take their own measures involving the use of armed force. Yet self-defense is not precluded by the Charter. Article 51 permits States to make such determinations and take such measures until the Security Council takes its own measures to maintain international peace and security. On this view, Article 51 is an exception to Article 39, Article 42, and the collective security structure established by Chapter VII as a whole.
The alternative view makes better sense of Article 51’s placement at the end of Chapter VII. But let us turn to the Charter’s preparatory work to confirm its meaning, or determine its meaning to the extent that its text and purpose leaves its meaning ambiguous or obscure.
“An exception to the general rule”
The standard view finds little support in the negotiating history. States simply did not discuss article 51 in relation to the prohibition of force. They only discussed the article in relation to the authority of the Security Council and the permissibility of regional arrangements. This is understandable, since the article began as an effort to assure Latin American States that the authority of the Security Council would not displace the regional arrangement contemplated by the newly-adopted Act of Chapultepec (discussed at length here).
On May 12, 1945, the United States presented its first draft of Article 51 to the other so-called “Five Powers” (China, France, the Soviet Union, and the United Kingdom), providing that:
Should the Security Council not succeed in preventing aggression, and should aggression occur by any state against any member state, such member state possesses the inherent right to take necessary measures for self-defense. … The taking of such measures shall be immediately reported to the Security Council and shall not in any way affect the authority and responsibility of the Security Council under this Charter to take at any time such action as it may deem necessary in order to maintain or restore international peace and security.
This draft clearly situates self-defense as an exceptional measure permitted in an exceptional situation—the Security Council’s failure to prevent aggression before it occurs—which nevertheless preserves the general rule—the Security Council’s authority and responsibility to maintain or restore international peace and security. As one U.S. delegate explained, the draft article “relates to action in the event of the absence or failure of enforcement. . . . the right of self defense must be reserved to meet such a situation.”
Importantly, the terms “Nothing in the present Charter” were introduced in a subsequent draft, composed by a British delegate during a conversation with his U.S. counterparts. That draft was handed to technical experts of the British and U.S. delegations for refinement and then presented to the “Five Powers” around an hour later. During this brisk sequence of events, the prohibition of force was never mentioned, and there is no evidence that these terms were inserted with the prohibition of force in mind. In context, these terms were probably understood to refer to the draft provisions that would establish the authority of the Security Council.
The most revealing exchanges occurred later in the negotiations. A subcommittee recommended removing (what would become) Article 51 from (what would become) Chapter VII and making it a new section following (what would become) Chapter VIII. On May 23, 1945, the Soviet delegate strenuously objected, stating that “[t]he exclusion of this paragraph from [what would become Chapter VII] would be illogical and even artificial.” The Soviet delegate emphasized that the article “deals with the right of a member of the Organization to self-defense in case of an act of aggression, until the Security Council takes measures for the maintenance of international peace.” The Soviet delegate explained at length how the article, like the chapter as a whole, deals with both the powers and duties of the Security Council and the rights and duties of member States.
In response, the U.S. delegate did not address the merits of the question, but simply explained that “the object [of the recommendation] was simply to avoid a conflict of jurisdiction” between committees, and that the Coordination Committee would consider and resolve the matter. The Committee of Jurists later recommended placing the article at the end of Chapter VII, in part because of “its connection with and dependence upon the powers given to the Security Council in that Chapter.”
On June 20, 1945, the Coordination Committee agreed that Article 51 belongs in Chapter VII. The Chinese delegate argued that “this article amounts to an exception to the enforcement arrangements decided upon by the Security Council.” The Soviet delegate said that “the only reason for this Article was that it was an exception to the general rule provided in Chapter VII, and his view was that it should go at the end of that Chapter as the only place where it could be correctly understood.” The Chairman, from the United States, joined the Chinese and Soviet delegates and “insisted that [the article] was a general exception to the Council machinery and belonged where the powers of the Security Council were stated, in Chapter VII.”
While this view prevailed, other views were expressed. Notably, the Australian delegate “proposed putting the Article in Chapter II, if not as a separate chapter between Chapters VII and VIII.” But the Australian delegate seems not to have explained the rationale for his proposal, withdrew it rather than attempt to reword the article, and agreed to place the article at the end of Chapter VII.
The Purposes of the United Nations
The drafting history confirms that Article 51 was understood by its drafters and negotiators as an exception to the general rule that the Security Council shall determine the existence of any act of aggression and decide whether to take measures involving the use of armed force. But what about Article 2(4)? How, exactly, was the prohibition of force supposed to interact with the authority of the Security Council and the right of self-defense?
Let’s start with the text. Article 2(4) prohibits the use of force “in any … manner inconsistent with the Purposes of the United Nations.” Article 1(1) states the relevant purpose:
To maintain international peace and security, and to that end: to take effective collective measures for the prevention and removal of threats to the peace, and for the suppression of acts of aggression or other breaches of the peace . . . .
Such collective measures evidently include the use of armed force authorized by the Security Council under Chapter VII. It follows that such force is consistent with the purposes of the United Nations, and not prohibited by Article 2(4).
In contrast, self-defense is not in itself a collective measure for the suppression of acts of aggression. But self-defense is nevertheless consistent with the purposes of the United Nations to the extent that it is consistent with the authority of the Security Council to take such collective measures. Article 51 ensures such consistency, (i) by limiting self-defense to cases in which an act of aggression constituting an armed attack occurs before the Security Council takes effective collective measures, (ii) by precluding self-defense after the Security Council takes effective collective measures, and (iii) by requiring States to immediately report measures taken in self-defense to the Security Council so that it may take effective collective measures. Within the limits of article 51, the right of self-defense supports the collective security system established by Chapter VII as a whole. On this view, Article 51 does not operate directly on Article 2(4), but instead operates directly within the Chapter VII structure, which operates directly on Article 2(4).
The negotiating history confirms this interpretation.
First, Article 2(4) permits force authorized by the Security Council because the latter is not “inconsistent with the Purposes of the United Nations.” On June 4, 1945, the Brazilian delegate expressed concern that these terms “might well be interpreted as authorizing the use of force unilaterally by a state, claiming that such action was in accordance with the purposes of the Organization.” The Norwegian delegate agreed that “the Committee should reconsider the present language” or at least make it very clear in its report that the article “did not contemplate any use of force, outside of action by the Organization, going beyond individual or collective self-defense.”
In response, the U.K. delegate said “that the wording of the text had been carefully considered so as to preclude interference with the enforcement clauses” of what became Chapter VII. This confirms that Article 2(4) permits force authorized under these enforcement clauses because such force is consistent with the purposes of the United Nations.
Second, Article 2(4) permits the use of force in self-defense only to the extent that such force supports the collective security system established by the Charter. Norway, still concerned that the terms of the article were susceptible to abuse, submitted a proposed amendment, according to which
All members of the Organization shall refrain in their international relations from the threat of force and from any use of force not approved by the Security Council as a means of implementing the purposes of the Organization.
On June 9, 1945, the Committee’s Rapporteur responded as follows:
The Committee likes it to be stated in view of the Norwegian amendment to the same paragraph that the unilateral use of force or similar coercive measures is not authorized or admitted. The use of arms in legitimate self-defense remains admitted and unimpaired. The use of force, therefore, remains legitimate only to back up the decisions of the Organization at the start of a controversy or during its solution in the way that the Organization itself ordains. The intention of the Norwegian amendment is thus covered by the present text.
The first sentence confirms that the unilateral use of force is always inconsistent with the purposes of the United Nations. Under Article 1(1), a general purpose of the United Nations is “to maintain international peace and security.” But a specific purpose of the United Nations is “to take effective collective measures” to that general end, particularly “for the suppression of acts of aggression.” Unilateral force is always inconsistent with that specific purpose.
The second sentence confirms that the use of arms in legitimate self-defense is “admitted and unimpaired.” This indicates that the Committee did not consider the use of arms in legitimate self-defense as a ‘unilateral’ use of force. Why not?
The third sentence confirms that the use of force is legitimate only to “back up” the decisions of the Organization. It follows that the use of arms in self-defense is legitimate only to “back up” the decisions of the Organization. This explains why the Committee did not consider the use of arms in legitimate self-defense as a unilateral use of force. Legitimate self-defense supports the Organization rather than usurping, circumventing, or undermining it.
It is true that Article 51 permits self-defense if a “controversy” starts—that is, if an armed attack occurs—before the Organization makes a decision. Strictly speaking, such self-defense serves as a ‘back stop’ until the Organization makes a decision, not as ‘back up’ to a decision already made. But the basic point stands: self-defense is legitimate only if it supports effective collective measures for the suppression of acts of aggression, including by halting or repelling armed attacks until the Security Council takes such collective measures.
The Path Not Taken
To this point, we have examined discussions between States for evidence of their shared understanding of the text they ultimately adopted. Before closing, we should to revisit an early discussion within the U.S. delegation which reflects an alternative understanding that occasionally resurfaces today. The minutes of the U.S. delegation are extensive and reflect a range of views. Rather than attempt to summarize them all, let us focus on two discussions of particular interest.
On May 4, 1945, before Article 51 was drafted, debated, rewritten, and debated again, the U.S. delegation considered whether the right of self-defense was implicit in the proposed text or required express recognition. John Foster Dulles, then an Adviser to the U.S. delegation, argued that
the members of the Organization, under paragraph 4, Chapter II, Principles, pledged “to refrain in their international relations from the threat or use of force in any manner inconsistent with the purposes of the Organization.” Since the prevention of aggression would be consistent with the purposes of the Organization, action by a state in self-defense would be in accord with the principles of the organization.
Similarly, Durward Sandifer, Secretary General of the U.S. Delegation, argued that:
it was the general rule in international law that a state possessed the right of self-defense, and that this right remains unless there is an explicit limitation upon it. In the present proposals, instead of an explicit limitation, it is provided only that states shall not resort to force in violation of the purposes and principles of the Charter. Since any act of aggression is contrary to the purposes of the organization and since preventing aggression is consistent with the purposes of the organization, the right of self-defense against attack is not limited. Of course, if a state acted individually, its action would be subject to review by the Council.
At this early meeting, other members of the U.S. delegation seemed to agree that the right of self-defense did not require express recognition.
One notable feature of Dulles and Sandifer’s argument is that it does not extend to ‘self-defense’ against non-State actors. Under international law, violence by non-State actors, however grave, does not constitute aggression. It follows that the use of force targeting non-State actors cannot be reconciled with Article 2(4) on the grounds that “preventing aggression is consistent with the purposes of the organization.”
But the most striking feature of Dulles and Sandifer’s argument is that it does not reflect the actual text of Article 2(4), either as proposed or as adopted. As we have seen, it is a purpose of the organization “to take effective collective measures … for the suppression of acts of aggression.” The Dumbarton Oaks proposals, to which Dulles and Sandifer referred, used the same terms.
Unilateral measures for the suppression of acts of aggression are not consistent with the purposes of the organization. The use of arms in self-defense is legitimate only to “back up” decisions of the organization, including by halting or repelling armed attacks until the organization can take effective collective measures. Since the argument of Dulles and Sandifer does not reflect the text of Article 2(4), it is fortunate that the right of self-defense was explicitly recognized in Article 51.
Importantly, the discussions of the “Five Powers” indicate a shared understanding that the right of self-defense, though ‘inherent’ (that is, not created by the Charter), is nevertheless limited by the terms of Article 51. This view was expressed by Leo Pasvolsky, Adviser to the U.S. delegation, and went unchallenged:
Mr. Pasvolsky then called attention to the fact that the text originally proposed by the United States … arose out of the inherent right of self-defense and that it was this residual right of self-defense which was to be protected under the wording of the original United States amendment. Mr. Pasvolsky said that the original United States amendment would be limited to armed attack and would thus limit freedom of action which states could take, so that it was not as “open” an amendment as might appear at first glance.
He said that it would be impossible to permit complete freedom of action without smashing the entire international organization; if such an event were to be avoided, it would be better to limit the right of self-defense along the lines of the United States proposal.
John Foster Dulles suggested that one British delegate “disliked the United States proposal because of its limitations on the right of self-defense.” Notably, no one from the U.S. delegation (or any other) sought to assure this British delegate that the article would not limit the right of self-defense.
Finally, later discussions within the U.S. delegation itself indicate an intent to explicitly limit the right of self-defense by the terms of Article 51. Most notably, on May 20, 1945, Green Hackworth, the State Department Legal Adviser, and Harold Stassen, a former governor and an influential delegate, had the following exchange:
Mr. Hackworth expressed the view that the present draft greatly qualified the right of self-defense by limiting it to the occasion of an armed attack. Mr. Stassen stated that this was intentional and sound. We did not want exercised the right of self-defense before an armed attack had occurred.
The U.S. delegation accepted internally what it expressed externally: that Article 51 precluded anticipatory self-defense, explicitly limiting any right that may have previously or otherwise existed. It remains true that nothing in the Charter impairs the right of self-defense if an armed attack occurs, and continues, until the Security Council acts. But it is also true that the Charter limits the right of self-defense, which may not be exercised before an armed attack occurs, after it ends, or after the Security Council takes effective collective measures.
Article 2(4) prohibits the use of force in any manner inconsistent with the purposes of the United Nations. Collective measures to suppress acts of aggression are consistent with those purposes. Self-defense within the limits of Article 51 is consistent with such collective measures, and therefore consistent with those purposes. The unilateral use of force—that bypasses or subverts the authority of the Security Council—is always inconsistent with those purposes and always prohibited.
With this more complete picture, several details come into focus. Article 51 permits self-defense only if an armed attack occurs. Self-defense is not permitted before an armed attack occurs because only the Security Council may take forcible measures to prevent and remove a threat to the peace. Self-defense is not permitted in response to a use of force that is insufficiently grave to constitute an armed attack because only the Security Council may take forcible measures to suppress a breach of the peace. And self-defense is not permitted in response to violence by a non-State actor located in the territory of a non-consenting State (see here) because the States in question must settle their disputes by peaceful means, seeking the Security Council’s supervision where appropriate.
As always, States remain free to establish a new agreement regarding the interpretation of the Charter. Perhaps States have already established such a new agreement with respect to anticipatory self-defense if an armed attack is about to occur. Clearly some States would like to establish a new agreement with respect to self-defense against non-State actors (see here). States make international law. They can remake it as well. But until they do so, the law they made governs them still.