The war in Gaza has generated significant debate on myriad IHL issues, ranging from the antiquated practice of siege warfare to the modern use of artificial intelligence (AI) in targeting. Underlying many of these debates is disagreement about the overall proportionality of Israel’s response to the heinous acts committed by Hamas on October 7, 2023. Few deny that Israel had a right to act in self-defense, but there is heated disagreement on the scope of that right. For many, Israel’s campaign to destroy Hamas (and the more than 35,000 Palestinians killed to date) is clearly disproportionate to the attack it suffered. Others argue that proportionality is not so limited, and that Israel has the legal right to eliminate Hamas, an organization that is committed to the destruction of Israel.
This debate – which plays out at the United Nations, in cities, and on university campuses around the world – highlights the unsettled nature of one of the most important principles of international law: proportionality in jus ad bellum, the body of law that regulates States’ resort to force. The general contours of jus ad bellum are straight-forward. Article 51 of the UN Charter makes clear that a State can use force in self-defense in response to an “armed attack.” Customary international law further requires that the use of force in self-defense must be both necessary and proportionate.
The U.S. invasions of Afghanistan and Iraq generated significant debate about the definition of “armed attack” in Article 51, including whether the use of force in self-defense applies to armed attacks committed by non-State actors, whether States can use non-consensual force in the territory of States that are “unwilling or unable” to address the threat posed by such actors, and whether States can act in self-defense against “imminent” armed attacks. Those issues have not been fully resolved, although there is greater agreement today than when those conflicts began. The meaning of proportionality, however, is perhaps the most vexing and important jus ad bellum question, as it regulates how much force can be used in response to an attack. Despite the far-reaching implications of this deceptively complex principle, it remains “shrouded in uncertainty.” There is little consensus on what it requires or permits.
Here, I propose a test for jus ad bellum proportionality based on nine factors. These factors – which address the nature of the threat posed, the objectives of the use of force, and the expected good and harm that will result from such force – would assist States in assessing whether a resort to force is a lawful response to an armed attack or a pretext for an illegitimate end such as retribution. As discussed below, I also argue that expected harm to civilians must be assessed at the jus ad bellum stage, and not solely prior to each attack once hostilities have commenced as required by jus in bello. Accordingly, before resorting to force, the overall harm anticipated to civilian life and property must be evaluated against the aims to be achieved from the use of force.
Two Interpretations of Proportionality
It is accepted that the use of force in self-defense must be proportionate to the threat or harm that gave rise to the right of self-defense. There is little agreement, however, on the scope of the threat that can be considered in the proportionality analysis, or on what other factors should or must be considered. Are States confined to responding only to the immediate threat posed? Can States consider a pattern of prior attacks by the same actor in determining the scope of the threat to be addressed? Can States respond to attacks that have already concluded or act to deter future threats? If so, how imminent and likely must those threats be? Are there external limits that must be considered in a jus ad bellum proportionality analysis, such as the degree of civilian harm anticipated as a result of the forcible measures to be taken?
There are two general interpretations of the proportionality requirement. The narrow interpretation is that a victim of an armed attack can use force to halt and repel an ongoing attack. Once a State repels the immediate threat, the right to use force in self-defense is extinguished. This interpretation corresponds with how we understand the right of individual self-defense in criminal law. A person’s right of self-defense ends once she has repelled the immediate attack. One cannot use force to deter future threats (a role that is reserved to the government by means such as incarceration). Under this interpretation, Israel’s response must be limited to the objectives of securing the release of the hostages, and potentially capturing those responsible for the Oct. 7 attacks. But it does not extend to the total elimination of Hamas.
The broad interpretation (adopted by the United States among others) is that self-defense is not limited to repelling an immediate attack. The victim of an armed attack can also use force to prevent or deter likely future attacks. In 2004, Will Taft, a former State Department Legal Adviser, explained the U.S. position here:
“A proper assessment of the proportionality of defensive use of force would require looking not only at the immediately preceding armed attack, but also at whether it was part of an ongoing series of attacks, what steps were already taken to deter future attacks, and what force could reasonably be judged as needed to successfully deter future attacks.”
Under this theory, which remains the U.S. view, Israel is justified in its aim to destroy Hamas, as Hamas poses an ongoing and grave threat to Israel’s security (although some debate remains as to the level of gravity of that threat). Other commentators, including this author, fall somewhere between these two predominant camps.
Consequences of the Gulf Between Interpretive Views
There is an urgent need to develop a more nuanced understanding of the jus ad bellum proportionality requirement. The current gulf between the two interpretative camps generates profound legal uncertainty on a critical issue of international law and makes it difficult to ascertain the basis for objections to Israel’s campaign. States that criticize Israel as acting disproportionately often do not clarify whether they consider the overall campaign to be illegal per se (i.e., under jus ad bellum) or only that specific strikes are disproportionate (i.e., under jus in bello) due to the harm to civilians and civilian infrastructure (collateral damage) caused.
More problematically, the two competing theories of jus ad bellum proportionality are overly simplistic and ill-suited to regulate State behavior when confronted with different types of armed attacks. The narrow interpretation places unrealistic constraints on States’ ability to respond to ongoing threats and reduces the costs for aggressor States or non-State armed groups to commit unlawful uses of force. The narrow interpretation is also premised, like the criminal law conception of self-defense, on the existence of a collective security mechanism (i.e., the Security Council) that can maintain peace and security. The Security Council has consistently failed to perform this function. The broad interpretation, by contrast, increases the likelihood of unnecessary escalation, prolonged conflicts, and humanitarian crises – a lesson learned by the United States after decades of conflict in Iraq and Afghanistan.
A test for proportionality that does not account for the range of relevant factors in determining what force is appropriate is simply unrealistic in light of the diverse threats States confront. The test for determining the lawful scope of a forcible response to armed attacks must be flexible enough to accommodate State’s legitimate security interests while sufficiently robust to safeguard the fundamental purposes of the UN Charter, including the preservation of international peace and security and the protection of human rights.
Factors for Determining Proportionality
I propose nine factors to be considered when determining the appropriate amount of force that may be used in response to an armed attack. These factors overlap to some extent with the factors that Daniel Bethlehem developed in the context of clarifying when States can use force in response to an “imminent” armed attack. This overlap makes sense as part of the debate between the interpretive camps centers on the scope of the threat that can be addressed through the use of force.
The first four factors consider the nature of the threat posed by the aggressor State or armed group. The remaining five factors evaluate the force used in response to that threat. For the sake of brevity, I will only briefly summarize the purpose and implications of these proposed factors.
The Armed Attack and Threat Posed
(1) The severity of the harm caused by the armed attack.
The scale and gravity of the armed attack clearly is relevant to determining the scope of force that may be warranted in response. Attacks that are widespread or that cause substantial destruction will generally require more significant responses than discrete or minor attacks.
(2) The likelihood of future attacks
(3) The imminence of such attacks
(4) The severity of future attacks
There is general agreement that the victim of an armed attack can use force to halt and repel an ongoing attack. There is little consensus, however, as to whether a State can use force to address future threats by the same adversary. The stronger position is that States may use force to deter threats under exceptional circumstances, although States must be careful to tailor any such response to legitimate deterrence and not retribution. The use of force to deter or prevent a likely, imminent, and grave attack will weigh more heavily towards legality than uses of force to deter remote or minor threats. In assessing these factors, the prior actions of the aggressor, its stated intentions, and its capabilities will be relevant.
The Use of Force in Response
The following three factors are grounded in the UN Charter’s objective of maintaining peace and security. In short, forceful measures that are likely to escalate, expand, or prolong a conflict should be viewed with heightened scrutiny. Similarly, measures that are not tailored to addressing an identified threat, or that are unlikely to do so, are more likely to be acts of revenge or punishment than legitimate acts of self-defense.
(5) The risk that the force used will escalate or expand the conflict, including with other actors
(6) Defined criteria for success and the likelihood that forcible measures will accomplish the self-defense objectives
(7) The likelihood that the forceful measures will create conditions that would make restoration of peace more difficult
The final two factors relate to the protection of civilians, one of the harms that needs to be assessed along with the legitimate ends of self-defense.
(8) The foreseeable harm to civilians and civilian infrastructure
(9) Measures taken to avoid or minimize civilian casualties, such as commitments to ensure humanitarian access
There is no consensus (see here, here, and here) on whether harm to civilians or civilian objects should be considered at the jus ad bellum stage, although the International Law Association implicitly endorsed such an approach in 2018. It wrote that the proportionality analysis should “assess[] whether the harmful effects of the force taken in self-defence are outweighed by achieving the legitimate aims.” (ILA Report at 12).
It makes little sense to only consider civilian harm in jus in bello proportionality, the rule of international humanitarian law (IHL) that requires commanders to weigh the anticipated harm to civilians and civilian objects against the concrete and direct military advantage expected from an attack. This balancing of military advantage and civilian harm during conflict can only have normative value if the military campaign itself is expected to generate more good than harm. The use of force to prevent mass atrocities or the annihilation of a State can clearly justify more civilian casualties than the use of force to repel a minor incursion or a small-scale terrorist attack. Moreover, as is clear from States’ reactions to the current conflict in Gaza, the amount of civilian harm is a central focus in assessing the proportionality of Israel’s response.
The ninth factor is particularly novel, but it would be prudent for States to consider systemic civilian harm mitigation at the jus ad bellum stage. Measures taken to avoid or minimize civilian harm in attacks are generally understood to be a requirement of jus in bello, reflected most prominently in Article 57 of Additional Protocol 1. A State’s commitment to implement specific or systemic mitigation measures at the outset of a campaign to address foreseeable humanitarian consequences, however, can also be useful in discerning whether its use of force is a legitimate exercise of self-defense or whether it is driven (at least in part) by illegitimate motives.
Assessing Proportionality
Full consideration of these factors may not be feasible at the initiation of a State’s use of force in self-defense. This may raise objections from those who argue that jus ad bellum governs only the initial resort to force and need not be reconsidered after hostilities commence. But this interpretation of jus ad bellum is problematic, except in situations of full-scale invasion, and inconsistent with state practice.
First, it may be impossible to discern the scope or objectives of a State’s use of force at the beginning of its campaign, unless the response is discrete and time-limited. States may not immediately reveal their intended response, or they may adjust strategy in light of additional information about the threat posed.
Second, the proportionality of a State’s response remains relevant throughout the campaign. In the same way that a State relying on consent for its military presence in another State’s territory cannot exceed the scope of that consent, a State using force in self-defense cannot exceed the legitimate ends that justified the resort to force, at least absent a material change in circumstances. Consider the conflict between the UK and Argentina over the Falkland (or Malvinas) Islands. If the UK had bombed military objectives in Buenos Aires, countries would undoubtedly claim such escalation was disproportionate (in the jus ad bellum sense) to Argentina’s invasion of the islands even if the attacks were lawful under jus in bello. The ongoing relevance of jus ad bellum is particularly important in today’s context, where many States are engaged in prolonged armed conflicts, both with other States and terrorist or other non-State armed groups, whose intensity may ebb and flow over time.
Third, States already do in practice periodically assess the proportionality of a State’s use of force. Much of the criticism of Israel’s attacks on Gaza came weeks or months after October 7, as civilian casualties mounted. In addition to criticizing specific attacks, States increasingly argued (see also here, here, and here) that the scope and scale of the response, and the civilian casualties inflicted, were disproportionate to Hamas’s initial attack. Similarly, the fact that Israel and Hamas have arguably been engaged in a low-intensity armed conflict pre-dating October 7, 2023, has not dissuaded States from opining on the jus ad bellum proportionality of Israel’s response.
Conclusion
I do not claim that these nine factors represent the current state of the law; any such claim would be impossible given the unsettled nature of the proportionality requirement in jus ad bellum. Nor do I argue that any single factor should be dispositive in determining whether a response is proportionate. Rather, these factors, considered in their totality, would provide a more nuanced understanding of proportionality and assist States in identifying specific concerns with another State’s use of force in response to an armed attack. They may also assist in determining whether a State’s use of force is a legitimate exercise of self-defense or a pretext for some illegitimate purpose.
States that act in self-defense could also address these factors in letters submitted to the Security Council pursuant to Article 51 of the UN Charter. These letters are used to satisfy States’ obligation to report on the measures taken in self-defense, but in practice they are generally brief and conclusory. Providing the Security Council, UN Member States, and the international community with more detailed information regarding the proportionality of actions taken or planned would lead to more informed debate and likely bolster the legitimacy of States that are lawfully acting in self-defense.