Interpreting the Law of Self-Defense
The 13 June 2025 Israeli attack (Operation Rising Lion) on Iranian nuclear and military infrastructure, as well as senior military leaders and nuclear scientists, sparked a sophisticated debate over the parameters of the international law right of self-defense (see, e.g., Cohen and Shany, Gill, Haque, Heller, Kittrie and Corn, Milanovic, Schmitt, Tsagourias, and a forthcoming Articles of War piece by Biggerstaff). That debate only sharpened following Operation Midnight Hammer, the 22 June U.S. attacks on Iranian nuclear facilities.
Much of the discussion focused on whether Israel enjoyed a right of “anticipatory self-defense” against a potential Iranian “armed attack,” the condition precedent for self-defense under Article 51 of the UN Charter and customary international law. As explained in an earlier Articles of War post, I am of the view that Operation Rising Sun had to satisfy the so-called “last window of opportunity” test for compliance with the necessity criterion of self-defense, especially its imminency component. It is an approach the United States adopted well over a decade ago, wisely in my estimation (see my 2003 piece advocating its adoption).
If Israel enjoyed a right of anticipatory self-defense against a potential Iranian nuclear attack, the United States could likewise employ force against Iran pursuant to the right of “collective self-defense” found in Article 51 and customary law. However, if Operation Rising Lion did not fall within the four corners of the law of self-defense, both the Israeli and U.S. operations amounted to unlawful “preventive defense.” Simply put, States enjoy no right under international law, absent Security Council authorization under Chapter VII of the UN Charter, to use force solely because another State is about to acquire a threatening capability.
In this post, I will not relitigate the law or the facts as they apply to operations by and against Iran. In fact, I concede the assertion some of those cited above made that the “last window of opportunity” approach is a reinterpretation of the right of anticipatory self-defense. Instead, my purpose here is to take a step back and ask the broader question of when it is appropriate for States to reinterpret long-held international law rules, as the United States and other States supporting the last window of opportunity approach have done, and with what limitations. For reasons that will be explained, I label such reinterpretation “interpretive adaptation.”
The Nature of International Law Rules
Sometimes international law needs to be surgically precise. The paradigmatic example is the law governing the width of the territorial sea. Before the UN Convention on the Law of the Sea set the maximum width of a territorial sea at twelve nautical miles (art. 3), a standard now widely recognized as customary, coastal nations claimed territorial seas ranging from three to as many as 200 nautical miles. Constant disputes over navigational, fishing, and other rights resulted. Stability emerged only once States achieved consensus on a bright-line rule.
But when it comes to the use of force, States approach the relevant rules, their interpretation, and their application from widely differing perspectives. Some are weak, others are strong. Some are members of robust alliances, others stand alone. Some wield unique capabilities, such as nuclear weapons or cyber capabilities, while others will never do so. The rules (and their interpretation) carry different weights for different States.
Thus, treaty provisions tend to be a product of compromise, often expressed in a manner vague enough for States to read into them what they believe their national interests demand. Moreover, ambiguity allows them to hedge their bets against unforeseen future situations in which a particular interpretation might either tie their hands or be exploited by an adversary. And customary international law is usually even more obscure. By nature, customary law rules are “least common denominator” creatures, crystallizing only around points that muster sufficiently dense practice and opinio juris (on crystallization, see International Law Commission, Draft Articles on Identification of Customary International Law).
This reality affords States a substantial margin of appreciation in the interpretation and application of use of force-relevant rules, a margin they have taken advantage of repeatedly. For example, the UN Charter gives the Security Council “primary responsibility for the maintenance of international peace and security” (art. 24). Yet, in 1950, the General Assembly acted to enable collective action in Korea when the Security Council could not do so over Soviet objection (Uniting for Peace Resolution).
More to the point in the present context is the text of Article 51 itself, which provides in relevant part, “Nothing in the present Charter shall impair the inherent right of individual or collective self-defence 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.” As the late Professor Thomas Franck noted, “it is beyond dispute that the negotiators deliberately closed the door on any claim of anticipatory self-defense.” Indeed, this was the view of the American Delegation at San Francisco (Franck, Recourse to Force, p. 50).
However, over time, it has become relatively well-accepted that such a right exists, although there is substantial disagreement concerning when it matures. Indeed, the debates over the Israeli and U.S. operations generally accept a right of anticipatory self-defense; the disputes are over the normative content of that right and whether the facts on the ground triggered it. The one significant outlier with regard to anticipatory self-defense is the late Professor Yoram Dinstein, who claimed that the plain text of Article 51 made clear that there had to be an ongoing armed attack before forcible defensive measures were lawful (Dinstein, War, Aggression and Self-Defence, p. 222-28). Yet even he squared the circle by proposing so-called “interceptive self-defence,” which looks strikingly like anticipatory self-defense.
The point is that international law rules can be malleable, especially the prohibition on using force and its exceptions. And that is as it should be, for law must remain flexible enough to meet the needs of the diverse international community it serves. Should a rule no longer respond to those needs, States will inevitably begin to ignore it altogether; it will fall into desuetude. That may present a greater threat to the international community than interpretive adaptation to a changed context in which the rule will be applied. The question is, when is such adaptation appropriate?
Interpretive Adaptation of the Right of Self-Defense
Interpretive adaptation of international law rules can be disruptive, for predictability is a potent stabilizing factor in international relations, especially concerning the use of force. After all, if international law is to work, States must have a rough idea of where the red lines lie. This allows them to anticipate when other States are likely to believe themselves entitled to respond in ways that might not otherwise be permitted in the normal course of events, such as by taking countermeasures in response to an internationally wrongful act or using forcible defensive measures in the face of an armed attack (see discussion here). Moreover, being perceived as acting unlawfully almost always comes at some tangible or intangible cost to the State concerned.
So, States must proceed cautiously in asserting that a new interpretation of an existing customary or treaty-based rule is merited. In my view, “interpretive adaptation” of a use of force rule like the right of self-defense is appropriate when there is a significant change in circumstance that was not anticipated at the time the UN Charter was adopted or a customary variant like anticipatory self-defense crystallized.
The advent of cyber operations is a case in point. It merited adapting use of force rules to the new reality that a cyber attack could be highly disruptive, even disastrous, without causing any direct physical consequences. This has ignited a reconsideration by States of the meaning of “use of force” and “armed attack” that is still underway. For instance, some States have adopted the position that a cyber operation having nationwide economic consequences should be considered not only as a use of force, but also as an armed attack that triggers the right to respond forcibly in self-defense. This is so despite the fact that, following the adoption of the UN Charter, economic consequences had been expressly rejected as qualifying an action as a use of force (Tallinn Manual 2.0, p. 331)
In the context of self-defense, the most notable interpretive adaptation was the United States’ rethinking of the concept of anticipatory self-defense in the aftermath of the 2001 “9/11” attacks. In its 2002 National Security Strategy (NSS), the Bush Administration observed,
We must adapt the concept of imminent threat to the capabilities and objectives of today’s adversaries. Rogue states and terrorists do not seek to attack us using conventional means.
They know such attacks would fail. Instead, they rely on acts of terror and, potentially, the use of weapons of mass destruction—weapons that can be easily concealed, delivered covertly, and used without warning.
Given this changed context, which was undeniable in light of the previous year’s attacks, the NSS stated,
The greater the threat, the greater is the risk of inaction—and the more compelling the case for taking anticipatory action to defend ourselves, even if uncertainty remains as to the time and place of the enemy’s attack. To forestall or prevent such hostile acts by our adversaries, the United States will, if necessary, act preemptively.
The reference to preemption sparked quite a mêlée among international law scholars. I set forth my views on the matter in a 2003 Michigan Journal of International Law article. I argued the U.S. position, although unartfully styled as preemptive instead of anticipatory, was an interpretive adjustment, not a purported new rule, and one that was appropriate to the changed circumstances. I suggested that in light of the new reality, a State could act forcibly in anticipation of another State’s armed attack, but only when its putative attacker wields the capability of carrying out the anticipated armed attack, intends to mount it, and the “last window of opportunity” to defend itself effectively is about to close. If any of the criteria are not satisfied, the action is “preventive” instead of “anticipatory” (preemptive) and therefore unlawful.
The circumstances had changed, thereby opening the door to interpretive adaptation. And because the right of self-defense leaves a margin of appreciation to States, it is appropriate for States facing such a scenario to interpret the right in a manner consistent with their national interest in order to accommodate the changed or otherwise unforeseen circumstances.
I want to emphasize that I am not asserting that States can make new rules other than through the normal process of law formation (treaty and customary crystallization) or that changed circumstances entitle them to ignore existing law altogether. All I am saying is that when a rule admits of interpretive play either because it is vague on its face or because there are significant changed circumstances that were not in contemplation at the time the rule was adopted or crystallized, States may engage in interpretive adaptation within the four corners of the rule, indistinct though those corners may be.
The Limits of Interpretive Adaptation
However, there are limits because interpretive adaptation can be destabilizing. Order is essential in the international community, for if everyone interprets imprecise rules independently, they will quickly become meaningless. Accordingly, at least for treaties, there is a convention governing interpretation, the Vienna Convention on the Law of Treaties (the United States is not a party but generally accepts its provisions on interpretation).
It provides, inter alia, that text should be accorded its “ordinary meaning.” Consideration should also be given to any agreements or instruments that provide context and to subsequent practice that sheds light on the interpretation of provisions (art. 31). If application of these interpretive tools leaves the meaning ambiguous or generates results that are “absurd or unreasonable,” supplementary means of interpretation, such as looking at preparatory work or the circumstances under which the treaty was concluded, are permissible (art. 32).
But this process is limited to treaty interpretation, and even there, it can leave a degree of uncertainty as to how a provision should be interpreted, especially regarding circumstances for which the original rule was not designed. It is in such circumstances that interpretive adaptation is appropriate. The question is, when does a State’s interpretation go too far?
At a minimum, a State engaging in interpretive adaptation of a treaty or customary law rule like the right of anticipatory self-defense must satisfy three general conditions. First, the State proposing the interpretation must be acting in “good faith,” a fundamental norm of international law recognized as a general principle in Article 38(1)(c) of the Statute of the International Court of Justice and appearing in the Vienna Convention’s rules on interpretation (art. 31(1)). The significance of the principle varies depending on the context in which it is applied. However, with regard to use of force matters, its relevance is clear. Article 2(2) of the UN Charter expressly provides that “[a]ll Members, in order to ensure to all of them the rights and benefits resulting from membership, shall fulfil in good faith the obligations assumed by them in accordance with the present Charter.” The good faith principle has the same effect vis-à-vis the fulfilment of customary law obligations.
As a result, an interpretation that is mere subterfuge to get around the prohibition on the use of force or to deny other States their right to engage in self-defense is not within the State’s margin of appreciation. An uncontroversial example of such subterfuge is Russia’s position on using force against Ukraine (my analysis). To satisfy the condition, a State must be making a sincere effort to adapt its interpretation of self-defense to circumstances that the rule was not designed to address.
Second, and critically for our purposes, the interpretation must be consistent with the “object and purpose” of the rule in question, which typically reflects the overall object and purpose of the treaty in which it is found or the network of related customary rules in which it resides (on treaties, see Vienna Convention, art. 31(1)). In other words, international law rules should be construed to best achieve their object and purpose in light of the context in which they are to be applied.
Use of force rules have an object and purpose of balancing the international community’s desire to avoid uses of force with the legitimate need of States to be able to respond effectively when avoidance fails. This object and purpose is reflected in Article 1(1) of the UN Charter, which includes the maintenance of “international peace and security” as a purpose of the United Nations; indeed, it is the first purpose the Charter cites.
It is for this reason that purely preventative forcible actions remain out of bounds; they shift the balance too far in the direction of security. But interpreting anticipatory self-defense in a manner that requires a State to sit by idly while the window of opportunity to protect itself closes tilts the balance too far in the other direction, especially in an era in which the prospect of an attack with a weapon of mass destruction or some other catastrophic means or method is all too real. It is, therefore, unsurprising that those States most likely to face such an attack are engaging in interpretive adaptation to accommodate that increased risk.
Finally, interpretive adaptation must pass the straight face test. For treaties, “manifestly absurd or unreasonable” interpretations based on the general rules of interpretation under the Vienna Convention open the door to supplementary means of interpretation (art. 32(b)). But it is simpler than that. A manifestly absurd or unreasonable interpretation is dead in the water. To suggest, for example, that a State that has a right of self-defense may exercise that right by directly targeting its attackers’ enemy population would obviously fail on this basis. The State may be acting in good faith in wanting to defend itself. In fact, targeting the civilian population might influence the aggressor to desist. Still, it is unreasonable to suggest that it is lawful for a State to defend itself by resort to unambiguously unlawful means or methods.
Interpretive Adaptation and Context
A significant change in the circumstances may merit adaptive interpretation. However, as a practical matter, its actual application depends on the context in which it will be applied. When considering anticipatory self-defense, the most significant contextual challenge is uncertainty. The reality is that anticipatory self-defense, by its very nature, must often be exercised before certainty is attained. And modern threats, such as those involving missiles, cyber capabilities, and weapons of mass destruction, rarely lend themselves to clear-cut assessments. Decision makers must often act under conditions of factual uncertainty, where adversary capability, hostile intent, and the time available to prevent an attack are unclear. This reality begs the question of how a State should weigh legal and operational risk in light of such uncertainty.
In doing so, the attendant contextual background is paramount. For instance, past patterns of hostility, current adversary behavior, and strategic dynamics can heighten or lower the plausibility of inferences drawn by the State considering anticipatory action. To illustrate, a history of forcible exchanges might render what would otherwise be a moderate indicator of impending attack more provocative. When considering such indicia, cumulative indicators of adversary capability and intent matter more than isolated facts.
Additionally, the magnitude of the anticipated harm will inform how a State considering anticipatory action treats uncertainty. Where the projected consequences of failing to act are potentially catastrophic, as in the case of nuclear weapons or large-scale cyber attacks, a greater degree of uncertainty about adversary capabilities and intentions can be tolerated. Conversely, if the anticipated harm is modest or speculative, the requisite level of certainty before acting would be significantly higher.
Ultimately, a State’s assessment of both uncertainty and the need to act must be reasonable. The standard is one of a reasonable State making such decisions in the same or similar circumstances (on the legal effect of mistake, see Milanovic). The key is to calibrate the risk of inaction against the potential for harm.
Concluding Thoughts
There is no question that the law of self-defense, especially that governing anticipatory self-defense, leaves room for interpretation. We should not be surprised that this is the case, given the differing national interests of the States it is meant to serve. Nor should we be surprised when States engage in interpretive adaptation of the right of self-defense in order to safeguard their ability to respond in circumstances that were not considered when the rule was drafted in treaty form and crystallized as customary law.
But States need to proceed cautiously, for as Robert Jennings, then the Whewell Professor at the University of Cambridge, noted in 1938,
There is probably no branch of international law which is so calculated to encourage the skeptic as that mass of contradictory precedents, dogmatic assertions, and vague principles which are collected under the common head of “intervention,” and perhaps there is no more potentially dangerous ground of intervention than that which is variously described as “self-preservation” and “self-defence” (p. 82).
So, the treatment by States of anticipatory self-defense is necessarily marked by both normative caution and operational necessity. Accordingly, interpretive adaptation of our understanding of the right of self-defense is only appropriate when new circumstances upset the delicate balance between the international community’s desire to maintain peace and the need of States for security. Still, as Professor Michael Reisman wrote in 1992, in an observation as applicable to States as to scholars,
The difference between the legal scholar and the legal anthropologist is that the legal scholar must identify the conditioning factors in the past that shaped normative expectations; moreover, he or she must determine whether they continue to operate and are likely to be factors in the future. Most important, a legal scholar has an independent responsibility to examine all legal formulations in terms of their current and prospective consequences for (and contributions to) the basic goals of minimum and optimum world order. If they do not contribute to these goals, the legal scholars should suggest alternatives that are likely to approximate them.
***
Michael N. Schmitt is the G. Norman Lieber Distinguished Scholar at the United States Military Academy at West Point. He is also Professor of Public International Law at the University of Reading and Professor Emeritus and Charles H. Stockton Distinguished Scholar-in-Residence at the United States Naval War College.
The views expressed are those of the authors, and do not necessarily reflect the official position of the United States Military Academy, Department of the Army, or Department of Defense.
Articles of War is a forum for professionals to share opinions and cultivate ideas. Articles of War does not screen articles to fit a particular editorial agenda, nor endorse or advocate material that is published.
Photo credit: U.S. Air Force, Tech. Sgt. Emerson Nuñez
