Introduction
“Are we witnessing the death of international law?” Thus began a recent newspaper article surveying the seemingly hopeless state of international rules. Indeed, recent events have triggered a flurry of epitaphs for international law spanning the ideological spectrum (Hindi, Posner). In particular, scholars like Oona Hathaway and Scott Shapiro have argued that recent events are triggering the “catastrophic collapse of norms against the use of force.” In other words, the most fundamental rules of the international system seemingly face an existential crisis.
It is perhaps helpful to recall that we have been here before. As early as 1915, a New York Times article considering U.S.-U.K. relations asked, “[I]s international law dead?” Thomas Franck famously sounded the death knell of the prohibition on the use of force in 1970, 1990, and 2003. Indeed, international law scholars periodically engage in cycles of pessimism regarding the durability of international law in general and the use of force prohibition in particular (Wippman, pp. 387-88). But what seems to be different now is the generalized assault on bedrock legal norms by the world’s major powers—Russia, China, and with increasing regularity, the United States. The U.S. attack on Iran in June 2025 and repeated strikes on boats in the Caribbean Sea and Pacific Ocean since September 2025 are just the latest manifestations of the current administration’s complete disregard of fundamental international rules (O’Connell). In a world thus threatened by competing hegemonic powers, it is worth revisiting the status of international law and its usefulness for less powerful states.
This Essay argues that the dynamic of gradual change built into the sources of international law can serve as a powerful tool for marginalized states to resist imperial hegemony. This rigid incrementalism has been criticized for preventing international law from reflecting the needs of post-colonial states (Chimni) and addressing contemporary problems like climate change (Brunnée). However, the gradualism of change in international law can act as a bulwark against powerful states’ abusive attempts to circumvent established rules. This Essay explores the prohibition on the use of force, in both its treaty and customary forms, as a case study to illustrate how this dynamic can work in practice.
Today’s diagnoses of the state of the jus ad bellum—much like yesterday’s—generally take one of two forms. The first, which one might call the legal erosion thesis, is the idea that powerful states have corroded the prohibition on the use of force by repeatedly advancing legal interpretations of Article 2(4) that water down its core protections (e.g., Hathaway, Hathaway & Patrick). These interpretations, repeated and endorsed by other states, ultimately become law, thus hollowing out the core of the jus ad bellum. The second, which one might call the normative erosion thesis, posits that repeated violations of the prohibition on the use of force are undoing the normative power of the jus ad bellum. Even if the “law in books” of Article 2(4) has not changed, the “law in action” rings increasingly hollow as states no longer believe that the prohibition has any meaningful bite (e.g., Hathaway & Shapiro, Falk). This Essay primarily responds to the first thesis, aiming to show that the dynamics of gradual change in international law have made the jus ad bellum resilient to abusive interpretations by powerful states. The second thesis remains troubling, yet the mechanisms underpinning the creation and change of international law highlighted here also provide tools to counter international law’s decay into desuetude.
The Essay proceeds in three main parts, discussing the dynamics of change in (I) treaties, (II) custom, and (III) jus cogens norms. These elements are discussed both in general and in the specific context of the prohibition on the use of force. The Essay concludes with a discussion of the ways marginalized states can use these rules of modification to protect their interests in the perilous current moment.
I. Treaties
Treaties form the cornerstone of an international law system that is rooted in state sovereignty and consent. Just as consent is the centerpiece of treaty formation, it is also essential to treaty modification (see, e.g., Hollis). Treaties can be modified in three main ways: (A) through the formal amendment, modification, and termination procedures set out in the Vienna Convention on the Law of Treaties (VCLT) or in individual treaties themselves; (B) through dynamic interpretation of treaty provisions as laid out in VCLT Articles 31 and 32; and, more controversially, (C) through modification (as opposed to interpretation) by subsequent practice (Crootof, pp. 247-264). This Part considers these three broad ways to alter treaty provisions and applies them to the treaty prohibition on the use of force.
By way of introduction, Article 2(4) of the UN Charter prohibits “the threat or use of force against the territorial integrity or political independence of any state.” This prohibition has only two explicit exceptions: self-defense if an armed attack occurs against a state (Article 51) and force authorized by the UN Security Council in response to threats to or breaches of the peace (Article 42). State consent to armed intervention is often cited as a third exception to the prohibition (Paddeu).
This Essay employs the terms “prohibition on the use of force” and Article 2(4) interchangeably to refer to the combined set of UN Charter provisions, including Articles 42 and 51, governing the use of force in international law (see Articles on State Responsibility Commentaries, Art. 21, ¶ 1). Although it does not take a firm position on the precise contours of the prohibition, the fundamental dynamic that the Essay aims to highlight—that international legal rules are inherently “sticky” in ways that can protect against abuse by major powers—applies regardless of the scope of the prohibition on the use of force.
A. Formal procedures: amendment, modification, termination
The Vienna Convention sets out the conditions for altering treaties, whether by amendment, modification, or termination. As with the rest of the “treaty of treaties,” the guiding principle is state consent. For example, a multilateral treaty may be amended by agreement of all of the parties or under the rules provided by the treaty itself (VCLT, Art. 40). Modifying obligations between a subset of the parties to the treaty is only permitted by agreement between those parties and if the modification does not affect the rights or obligations of the other parties (ibid., Art. 41). Withdrawal from or termination of a multilateral treaty is permitted under certain narrow conditions, such as the conclusion of a later treaty on the same subject matter, but otherwise also requires the consent of all the parties (ibid., Arts. 54-62).
In practice, these rules make it very difficult to change treaties, particularly the substantive provisions of multilateral treaties. The UN Charter is a good illustration. According to its terms, it may be amended by a vote and domestic ratification of at least two-thirds of its members (UN Charter, Arts. 108-109). All five of the Charter’s amendments—the last of which occurred in 1973—concerned largely procedural issues. These high bars for formal alteration of the UN Charter mean that Articles 2(4) and 51 are very likely to endure in their current textual form, regardless of the wishes of the most powerful states.
B. (Re)interpretation
Even if treaty provisions are not formally amended, their meanings may change over time. The Vienna Convention’s foundational rules on treaty interpretation, which are part of customary international law, set out the ways in which this can occur (VCLT, Arts. 31-32). The VCLT provides that a “treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose” (ibid., Art. 31(1)). A central question raised by this rule is whether the relevant meaning of the treaty is to be frozen at the time of ratification or assessed dynamically at the time of interpretation. The International Court of Justice (ICJ) has held that an evolutive interpretation of a treaty is appropriate in certain circumstances. In Dispute Regarding Navigational and Related Rights (Costa Rica v. Nicaragua), the Court recognized that the parties to a treaty should be presumed to have intended to give its terms an evolving meaning (1) where they used “generic terms” in the treaty that are likely to evolve over time, and (2) where the treaty is of “very long” or unlimited duration (Merits Judgment, para. 66).
These factors suggest that Articles 2(4) and 51 of the UN Charter were intended to have an evolving meaning. The UN Charter is a “living instrument” of unlimited duration (Rensmann, pp. 31-32; Pobjie, pp. 66-67). For example, Articles 2(4) and 51 both employ open-ended language like “force,” “territorial integrity” and “self-defense,” and Article 51 refers to an “inherent” right of self-defense which incorporates a dynamic customary rule (Akande & Johnston, pp. 687-88). This suggests that the meaning of the treaty prohibition on the use of force may evolve over time. Identifying whether and how it has done so is determined by the three methods contemplated by the Vienna Convention (ILC Draft Conclusions Commentary, p. 64).
Subsequent agreement by the parties
Rooted in explicit state consent, one method by which the interpretation of treaty provisions can evolve is through subsequent agreement by the parties. The most straightforward channel is a formal agreement or instrument made “in connection with the conclusion of the treaty,” either signed by all the parties or signed by some parties and accepted by all of them (VCLT, Art. 31(2)). In practice, for multilateral treaties like the UN Charter, this is an onerous process akin to formal amendment that essentially gives individual states veto power over interpretive changes.
Similarly, the VCLT also provides that the parties may subsequently agree on how a treaty or one of its provisions should be interpreted (ibid., Art 31(3)(a)). In its 2018 Draft Conclusions on the issue, the International Law Commission (ILC) clarified that, like Article 31(2), this rule requires the agreement of all the parties to a treaty (ILC Draft Conclusions Commentary, p. 28). In the case of a multilateral treaty with many parties like the UN Charter, these stringent conditions are difficult to meet. Nevertheless, multiple scholars have suggested—based on the language of the ICJ’s Nicaragua case and Kosovo Advisory Opinion—that the acts of plenary organs of international organizations, such as the UN General Assembly, can express subsequent agreement of the parties for the purposes of treaty interpretation (ibid., p. 99). Erin Pobjie has pointed to the 1970 Friendly Relations Declaration, the 1974 Definition of Aggression, the 1987 UNGA Resolution 42/22, and the 2005 World Summit Outcome Document, as four such decisions that expressed agreement of UN Member States on the content of Article 2(4) (Pobjie, pp. 109-115).
Yet there are two major safeguards built into this tool of evolutionary interpretation. First, an international organization’s decision must be unanimous; any dissenting votes will disqualify it as an authentic source of interpretation (Nolte, ¶ 79; cf. Nicaragua Merits Judgment, ¶ 203). Second, the ICJ has emphasized that mere adoption is insufficient and that the attitudes of the parties must demonstrate consent to the proposed interpretation (Nicaragua Merits Judgment, ¶ 188). This gives all UN Member States a powerful platform to shape the evolution of UN Charter provisions, including Articles 2(4) and 51. In practice, this consensus-based mode of evolutionary interpretation has generally resulted in UN resolutions that reaffirm the application of Article 2(4) in specific contexts, including boundary disputes, forcible reprisals, military occupation, annexation, bombardment, blockades, and sending armed bands into the territory of another state (Pobjie, pp. 109-115). In contrast, no significant reinterpretations of the prohibition of the use of force have occurred through this method of interpretation.
Subsequent practice by the parties
Subsequent practice by the parties provides an alternative method for the interpretation of treaties to evolve. The VCLT recognizes the interpretive value of “subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation” (VCLT, Art 31(3)(b)). But again, the ILC has stated that any conflicting positions regarding the interpretation preclude the existence of an agreement (ILC Draft Conclusions Commentary, p. 76). This means that any state, regardless of its size or influence, may block the emergence of a new interpretation by expressing its opposition.
Nevertheless, the ILC has recognized that the practice of a subset of all state parties can inform the interpretation of a treaty as a “supplementary means” under VCLT Article 32 (ibid., p. 30). But this should not be understood to override the stringent requirements under the other rules of treaty interpretation. The supplementary means of interpretation may only be used to confirm the interpretation reached through the ordinary techniques of treaty interpretation or in the limited situations where those techniques yield an ambiguous or absurd result. In any event, as the ILC has cautioned, this approach should be applied restrictively because “the view of one State does not make international law” (ibid., p. 36).
In the case of Article 2(4), the use of subsequent practice to establish agreement of all UN Member States is doubtful. Three leading areas of debate illustrate this: the legality of anticipatory self-defense, the right to respond to attacks by non-state groups, and the right of unilateral humanitarian intervention (discussed in more detail below). In all of these areas, there is evidence of dissenting state views, which demonstrates a lack of agreement and excludes the application of Article 31(3)(b) (Akande & Johnston, p. 684).
As for supplementary means of treaty interpretation, it is true that the provision’s core phrase, “the threat or use of force,” is textually ambiguous and the application of the interpretive techniques enumerated in VCLT Article 31 does not conclusively resolve that ambiguity (Pobjie, pp. 157-58). However, unilateral attempts to modify the interpretation of Article 2(4) have been routinely resisted by other states and rejected by scholars. One notorious example is the doctrine of unilateral humanitarian intervention, under which a state may use force in response to grave violations of human rights, such as acts of genocide or crimes against humanity. Despite its repeated invocation in the last two decades, the doctrine has been widely rejected as representing neither custom nor a practice that successfully modified the scope of Article 2(4) (see, e.g., Heller, Gill). For example, in 2000 the Group of 77 and China issued a declaration “reject[ing] the so-called ‘right’ of humanitarian intervention, which has no legal basis in the United Nations Charter or in the general principles of international law” (Declaration of the South Summit, ¶ 54). In contrast, only a handful of mostly Western states have asserted the right—hardly enough to establish the emergence of a subsequent practice informing the interpretation of the Charter prohibition (see Heller).
Any relevant rules of international law
Finally, VCLT Article 31(3)(c) provides that “any relevant rules of international law” are relevant to interpreting treaty provisions. This opens the door for customary international law rules to modify the interpretation of treaty rules. While this principle of “systemic integration” is undoubtedly significant in the way that it unifies different areas of international law (ILC Conclusions on Fragmentation, ¶¶ 17-19), the scope of application of this provision to the modification of treaties is limited by the standards for the evolution of custom, which are discussed in detail in the next section. In other words, the guardrails on the development of new custom also limit the extent to which treaty interpretation can evolve through this principle.
To summarize, each of the three avenues for evolutive treaty interpretation has built-in mechanisms slowing the pace of change. And beyond these individual features, the combined process contains a further brake. The ILC has stressed that the various ingredients in treaty interpretation must be applied in a “single combined operation” (ILC Draft Conclusions Commentary, p. 51). This approach further reduces the possibility of one or a few states unilaterally imposing their preferred interpretations of a multilateral treaty.
C. Modification by contrary practice
Although more controversial, another way in which treaties may change over time is by the emergence of contrary practice that modifies the application of a treaty provision. Here, it is important to distinguish between the interpretation and modification of a treaty. In theory, the former involves assigning one of multiple permissible meanings to a given provision while the latter entails substantially altering the provision (Crootof, pp. 259-60). Although the line between the two is hard to draw in practice, the difference between the two exercises explains why the possibility of modifying treaty provisions without the consent of the parties is controversial, unlike the interpretive changes identified in the section above.
Several scholars have argued that a rule of customary international law may modify or replace a prior conflicting treaty provision, such as through the application of the lex posterior (later in time) principle (ibid., pp. 264-288; Buga, pp. 195-235). It is worth noting that this remains a minority view that has not been recognized by the ICJ or ILC (see, e.g., Akande & Johnston, p. 685). In any event, this form of modification would be subject to the same conditions of customary international law formation discussed in the next section. In other words, this form of modification requires broad state acquiescence.
Even absent a fully crystallized customary rule, subsequent practice may work to modify a treaty through desuetude, a form of treaty termination. Desuetude occurs when repeated, long-term failure to enforce treaty rights serves as evidence of tacit consent by the parties to terminate a treaty or some of its provisions. (Kontou, pp. 73-74). Considering that the absence of state invocations of treaty-based rights serves as a proxy for consent of the parties to terminate the treaty, the bar for showing that a treaty has fallen into desuetude is high (Crootof, pp. 262-64). Tacit consent “must be evinced by all the Contracting Parties and not only by some of them” (Dinstein, p. 412). Evidence based on the silence of parties is only relevant when a reaction is to be expected (Crootof, pp. 262-63). Therefore, desuetude is rare, particularly in the case of important provisions of large multilateral treaties.
Nevertheless, some scholars have argued that Article 2(4) has fallen into desuetude, given the repeated state violations of the provision that have occurred since 1945 (Glennon, pp. 958-964). This is nevertheless a minority view and is undermined by the degree of state engagement with Article 2(4), both by states justifying uses of force as legal and by states expressing their views in response to uses of force (Wood). Moreover, empirical evidence on the reporting requirement under Article 51 suggests that states are cognizant of the framework on the prohibition of the use of force and usually compliant with this procedural element of justification for their uses of force (Green). In other words, the high bar for desuetude—a repeated failure to invoke a right demonstrating tacit consent by all parties—is not met in this case. Moreover, the frequent appeals by states to the prohibition on the use of force show not only that the rule has not fallen into desuetude, but also that it has enduring normative power. For the time being at least, states have signaled their belief that Article 2(4) continues to be binding law.
II. Customary international law
Customary international law, the other major source of international law alongside treaties, is similarly difficult to change. Unlike treaties, the processes of identification and modification of customary international law are essentially the same (Johnston). The traditional two-element test applies: for a rule to constitute customary international law, widespread and uniform state practice as well as opinio juris, or general practice accepted as law, must support it (ILC Draft Conclusions on CIL, Conclusion 2).
The process of identifying customary international law is notoriously imprecise (see, e.g., Kammerhofer). Be that as it may, the doctrine that has developed around the identification of customary rules builds in a criterion of consensus, thus offering a possible shield from abusive interpretations of existing rules by powerful states. This works in three ways.
First, state practice must be “both extensive and virtually uniform” to qualify in terms of Article 38 paragraph 1 of the ICJ Statute (North Sea Continental Shelf, ¶ 74). In other words, a critical mass of practice must conform to the rule and at most limited practice may breach it. Although the practice of “specially affected” states is generally given additional weight, the ILC has emphasized that this term “should not be taken to refer to the relative power of States” (ILC Draft Conclusions on CIL, pp. 136-37). In the context of customary law changes, this means that all states have a role to play, particularly when a generally applicable rule is involved. By conforming their behavior to the existing rule, states can demonstrate its vitality and defeat claims by deviant states that a new rule supersedes it.
Second, the ICJ has acknowledged that even breaches of customary rules can serve to reinforce the underlying rules in certain cases. In the Nicaragua case, the Court held that:
[I]nstances of State conduct inconsistent with a given rule should generally have been treated as breaches of that rule, not as indications of the recognition of a new rule. If a State acts in a way prima facie incompatible with a recognized rule, but defends its conduct by appealing to exceptions or justifications contained within the rule itself, then whether or not the State’s conduct is in fact justifiable on that basis, the significance of that attitude is to confirm rather than to weaken the rule. (Merits Judgment, ¶ 186).
This dynamic provides an additional safeguard against the erosion of customary rules, especially because states will often justify potential breaches of important customary rules by arguing that their conduct fits within the scope of those rules. For example, the Trump administration has justified the strikes on Iran’s nuclear facilities and Venezuelan boats in the Caribbean Sea on self-defense grounds (Article 51 letter; Congress letter). According to the Nicaragua judgment, these justifications, however thin, constitute opinio juris that strengthens the underlying legal rule. This bridges the gap between the legal and normative erosion theses identified earlier: the stability of international rules compels states to frame their conduct in terms of those rules, which implicitly strengthens their normative force. This point should not be taken too far, however. Tenuous justifications can weaken international norms by suggesting that they are malleable to the point of being meaningless (Hathaway). Moreover, there is a real risk that powerful states will stop justifying their conduct in international law terms altogether. Should that occur, the burden will fall on the rest of the international community to uphold the existing rules.
Third, the opinio juris requirement provides a parallel way in which states can block a change in a customary rule. As the ICJ stated in the Nicaragua case, opinio juris is furnished by both the states taking relevant action and those in a position to react to those actions (Merits Judgment, ¶ 207). In other words, states can participate in the process of forming (or blocking) customary law simply by reacting to other states’ behavior.
Matters are somewhat different in the case of termination, rather than modification, of a customary rule. In this case, it is unclear whether both or only one of the two elements of customary international law must be lost over time (Yap). In any event, terminating a customary rule requires more than occasional inconsistent practice, which can be considered a breach of the existing rule (Wouters & Verhoeven, ¶ 14). If a sufficient number of states continue to follow the existing rule and the deviant behavior is identified as a breach of that rule, then the rule will endure. And even if the critical mass of the two elements is lost such that the customary rule terminates, the persistent objector doctrine may offer states a way to resist the effects of the termination. Although the law in this area is unclear, a state that objects to the termination of the customary rule could theoretically maintain the application of that rule in its relations with other states (Yap). In this sense, the persistent objector doctrine is a double-edged sword: it allows powerful states to shield themselves from the obligations of customary international law (Chimni, pp. 23-24), but it also theoretically allows states to salvage valuable rules that would otherwise disappear.
The prohibition on the use of force provides a particularly complex case study of the dynamics of change in international law because it is simultaneously a treaty and customary rule (which is also likely jus cogens, as discussed below). This dual nature makes it theoretically important to distinguish between the effects of changing state practice on the customary prohibition, on the one hand, and on the treaty prohibition, on the other (see Van Steenberghe, pp. 59-62). In the case of the customary prohibition, a change in state practice will directly modify the scope of the rule as long as it is accompanied by the requisite opinio juris. But changes in the customary rule will only affect the treaty prohibition through the three narrower channels analyzed in the previous section: (1) as evidence establishing the agreement of (all of) the parties regarding the interpretation of Article 2(4) under VCLT Article 31(3)(b); (2) as “relevant rules of international law” pursuant to VCLT Article 31(3)(c); or (3) more controversially, as rules serving to modify the treaty prohibition. This complex process provides an additional check on the rapid erosion of the prohibition on the use of force because the vast majority of states are bound by both the treaty and customary prohibitions.
In contrast to Article 2(4) itself, the feedback process between changes to the customary right to self-defense and Article 51 is more direct. As the ICJ has recognized, the reference to the “inherent right of self-defense” in the provision incorporates the customary rule (Nicaragua Merits Judgment, ¶ 176). This means that the self-defense exception to the use of force—in both its treaty and customary law forms—evolves dynamically over time (Akande & Johnston, pp. 687-88).
So how has the customary law governing the use of force, including the right to self-defense, evolved over time? This has been one of the most hotly contested areas of international law over the last quarter century. As discussed above, powerful states’ attempts to evolve the customary prohibition to exclude so-called unilateral humanitarian interventions appear to have failed, at least so far (Dinstein, pp. 75-77). But in other areas, the literature suggests that the customary prohibition has softened to allow uses of force in more instances, even if the issues remain far from settled (Van Steenberghe, p. 44). With respect to the right to self-defense against non-state actors, the United States’s intervention in Afghanistan after the September 11th attacks and the generally supportive global response—exemplified by UN Security Council Resolutions 1368 and 1373—have been widely cited as evidence of an expansion in the customary right to self-defense (ibid., p. 47). Similarly, although the Bush doctrine’s view that states may use force to prevent non-imminent, future attacks has been rejected, in recent years some scholars have begun to accept the intermediate view that preemptive self-defense against attacks that are imminent, but have not yet occurred, is permissible (ibid., pp. 50-51).
Nevertheless, in both of these cases, scholars who endorse expansive views on rules regulating the use of force have relied heavily on policy-based arguments and selective cases of state practice (Van Steenberghe, pp. 45-54; see also Corten, p. 804). Notably, the ICJ has not endorsed any of these positions as having crystallized into customary international law. In fact, the Court endorsed the restrictive position on the use of force against non-state actors in the Wall Advisory Opinion (para. 139) and seemingly reiterated this position in the Armed Activities case (Merits Judgment, paras. 147-147). It has not revisited the issue since the 2005 decision.
In the absence of comprehensive evidence supporting the existence of the two elements of customary international law for each of these positions, the modification of these rules remains an open question. After all, scholarly commentaries, like international court judgments, are at most a subsidiary source of international law (ICJ Statute, Art. 38(1)(d)). This leaves an opening for states to resist the erosion of these norms by contributing their practice and opinio juris to solidify the customary norms on the use of force.
The June 2025 U.S. strikes on Iran’s nuclear facilities illustrate how states can block the evolution of a customary rule in the direction proposed by a powerful state. The United States justified its attacks by arguing that Iran was building a nuclear weapon that it would use, or threaten to use, against it or Israel in the future (Article 51 letter). It thus relied on the concept of preemptive self-defense that has been the subject of heated debate in recent decades. Although several mostly Western states supported the U.S. attack, numerous states condemned it as a breach of the prohibition of the use of force (e.g., Brazil, China, Oman, Pakistan, Russia). This broad-based opposition to the application of the anticipatory self-defense doctrine effectively serves as a veto on the United States’s proposed interpretation becoming customary international law—and reaffirms the ongoing vitality of the prohibition on the use of force.
III. Jus cogens rules of international law
Although not a separate source of international law, jus cogens (or peremptory norms of international law) are a central feature of the structure of international law and perhaps best exemplify the stability of international rules (Tladi, Kleinlein). The VCLT defines jus cogens as “a norm accepted and recognized by the international community of States as a whole as a norm from which no derogation is permitted and which can be modified only by a subsequent norm of general international law having the same character” (VCLT, Art. 53). Although jus cogens norms are usually based on customary rules, they can also be grounded in treaty provisions and general principles of law (ILC Draft Conclusions on Jus Cogens, Conclusion 5).
Although the mechanics of identification, modification, and termination of jus cogens are not entirely clear, it is generally agreed that “the threshold for change is incredibly high” (Tladi). The ILC posits that jus cogens may only be modified by the emergence of other jus cogens—either existing rules of international law that subsequently become recognized as having a peremptory character, or new jus cogens (ILC Draft Conclusions on Jus Cogens, p. 57). The two-step process for recognition of jus cogens—first, the acceptance of a norm as a rule of international law, and second, the acceptance of that rule as peremptory—makes modifying jus cogens especially difficult. Robert Kolb, for example, has posited that changing jus cogens requires triple opinio juris: acceptance of a modifying norm as legal, peremptory, and replacing the existing peremptory norm (Kolb, p. 101).
Moreover, jus cogens norms have safeguards that protect them from erosion by existing or emerging treaty or customary rules. The ILC has recognized, for example, that rules of customary international law that conflict with jus cogens do not come into existence (ILC Draft Conclusions on Jus Cogens, Conclusion 14). Similarly, treaty provisions that conflict with jus cogens are void (VCLT, Arts. 53, 64). This prevents an individual state or small group of states from watering down peremptory norms by developing conflicting treaty or customary rules.
The prohibition on the use of force illustrates the resilience of jus cogens. It is widely accepted that the prohibition on aggression is a peremptory norm of international law (see, e.g., ILC Draft Conclusions on Jus Cogens, Annex). Most scholars are of the view that peremptory status extends beyond aggression to the entirety of the Article 2(4) prohibition (see, e.g., Orakhelashvili, pp. 50-51), although not all agree (Green, pp. 225-240). State practice appears to firmly support this expansive view of the scope of the peremptory norm (Corten, pp. 201-207).
The thornier issue is whether the exceptions to the prohibition, such as self-defense, are also themselves jus cogens. This is significant because the legal status of the exceptions determines the ease with which they can be modified (Payandeh, p. 105). Conceptually, it is difficult to conceive of self-defense as peremptory because it is a right, not an obligation, to which the concept of derogation does not apply. Scholars have attempted to resolve this issue by arguing that the prohibition on the use of force as a whole, including its exceptions, is jus cogens (Tams, p. 95; Payandeh, pp. 108-109). This means that both the core use of force prohibition and its exceptions are subject to the stringent standards applicable to the modification of jus cogens norms. This has led Judge Tladi to conclude that “notwithstanding important geopolitical changes, the prohibition on the use of force, one of the most elementary norms of jus cogens, remains unchanged” (Tladi). As he points out, even if the self-defense exception had been enlarged under the process for modifying normal customary rules (which is unlikely, as previously discussed), this has almost certainly not occurred given the rule’s jus cogens status (ibid).
This discussion raises the question of whether the prohibition on the use of force can ever be modified, given its jus cogens status. The answer must be yes; otherwise, permanently frozen jus cogens norms would fade into irrelevance as they become divorced from the social reality of the international order. How this would actually occur is not entirely clear because no modification of jus cogens has been recognized to date. Nevertheless, it appears to require the vast majority of states to recognize a modified version of the prohibition as peremptory in nature.
These dynamics of gradual change of jus cogens mean that all states, including those marginalized economically or politically, can play an important role in shoring up fundamental principles of international law. This can occur on two levels. First, states can continue to affirm the jus cogens status of central norms like the prohibition on the use of force. This will help immunize them from a lower standard for their modification. Second, states can object to proposed modifications. Because the standard for recognition—and therefore, implicitly, modification—of jus cogens requires acceptance by a “very large and representative majority of states” (ILC Draft Conclusions on Jus Cogens, Conclusion 7(2)), dissenting states have a powerful voice in vetoing proposed modifications.
Conclusion
In 1970, Mohammed Bedjaoui characterized customary international law as “[b]ackward looking, conservative because [it is] static, iniquitous in its content, [and] ponderous in its formation” (p. 137). These characteristics, he asserted, mean that “custom as traditionally conceived cannot be of real use in the development of new rules, and could actually be an obstacle to any attempt at change” (ibid.). In many ways, this indictment still rings true today: custom seems too archaic and cumbersome to adapt to contemporary challenges ranging from climate change to cyberattacks (see Trachtman). Yet, it is precisely these features that make custom—together with its equally inflexible counterpart, treaties—such a powerful tool to resist the current foundational attacks on international law launched by the world’s hegemons.
What would this look like in practice? First, smaller states must continue to respect the prohibition on the use of force. Any attempt to take advantage of geopolitical uncertainty to wage local or regional wars would show to the world that the prohibition on the use of force lacks both legal and normative power, further undermining incentives to comply. Second, smaller states should be vocal in their opposition to abusive interpretations of Article 2(4). Such opposition can take several forms: unilateral statements, regional declarations, or UN resolutions. Condemnation of illegal conduct should be public, unambiguous, and explicitly invoke international law. To be sure, small states may be reluctant to speak out individually out of a fear of retaliation, as exemplified by the Trump administration’s sanctions on Colombian President Gustavo Petro after he denounced U.S. attacks on Caribbean boats as illegal. To mitigate this concern, states should consider joint action. For example, a group of twenty Arab and Muslim states led by Qatar condemned Israel’s June 2025 attacks on Iran as “ongoing military aggression” that “contravene[s] international law and the purposes and principles of the Charter of the United Nations” (Joint Statement). Finally, to further bridge the gap between the legal and normative strength of the prohibition, smaller states should consider developing alternative enforcement mechanisms to bypass the paralyzed UN Security Council. These might include collective countermeasures by coalitions of smaller and middle power states or enforcement actions by regional organizations like the African Union, European Union, or Organization of American States (e.g., Hathaway & Patrick).
This Essay has sought to shed light on the built-in resilience of the sources of international law. This characteristic provides an opportunity for smaller states to protect the core values of the international community in turbulent times. To be sure, the inertia of international rules is a double-edged sword which can also prevent states from effectively addressing pressing concerns like environmental issues (e.g., Brunnée). Yet the same coalition-building that shores up the prohibition on the use of force can also help the international community tackle these contemporary issues (Joyner). Despite the important criticisms leveled against international law over the years, broad-based engagement with the international system remains essential for smaller states, both to protect their sovereignty and advance their long-term interests.
* Associate, Foley Hoag LLP; J.D., Yale Law School; Diploma, Hague Academy of International Law. I am most grateful to Alaa Hachem, Alaa Hajyahia, and the editorial staff of the Yale Journal of International Law, especially Florian Kriener, Rick Da, Catey Vera, Madeline Babin, and Uma Menon, whose feedback greatly improved the piece. All errors are my own, and the views expressed in this Essay are solely mine and do not reflect the views of any institution with which I am affiliated.