The Law of Worst-Case Scenarios

The present Gaza war, initiated by Hamas’s October 7 attack on Israel, poses important questions about both the morality and laws of war and the language of conflict. People everywhere are having heated conversations involving concepts such as targeting civilians, terrorism, proportionality, and genocide. They deploy these concepts when arguing about broad questions, like whether Hamas’s attacks on civilians are justified resistance, and about narrower ones, like whether a specific Israeli bombing of a refugee camp is justified to kill Hamas leaders and destroy underground tunnels.

To a trained international lawyer, these terms have concrete definitions. At the same time, the terms are also moral concepts, and most people who aren’t trained lawyers use them colloquially rather than legally. The trouble is that the general rhetorical meanings of many of the words don’t match the legal meaning. That leads to the confusion of people talking (or yelling) past one another. It also obscures the primary moral message that underlies the law of armed conflict.

The whole point of international humanitarian law is to establish a minimal morality that can apply even during the terrible and bloody business of war. The rules of conflict in war do not address the underlying question of which side is right. Instead they bracket that question, on the understanding that, in wartime, neither side is going to concede that the other’s cause is worthy. The terms, in other words, are designed to adjudicate what may and may not be done while fighting a war, not whether the war is just. The reason to do this is to protect human life and dignity under conditions of the deepest human disagreement. Using these terms in ways that either differ from or actually contradict their legal meaning erodes the power of international humanitarian law.

When used correctly, the terms of international humanitarian law thus apply universally, to all people fighting all wars, whether just or unjust. Their core value is to insist that some actions are so morally wrong that they should never be tolerated, even if the side that perpetrates them is morally right in a given conflict. Put another way, under international humanitarian law, the ends do not justify the means. You can fight for liberation and self-determination, and to protect hearth and home. But there are rules. Violating them means you have violated universal principles of humanity.

Let’s begin with the prohibition on targeting civilian noncombatants. This is Rule 1 in the pantheon of international humanitarian law as captured by the International Committee of the Red Cross. (The ICRC, created by treaty, has the job under the Geneva Conventions “to work for the faithful application of international humanitarian law applicable in armed conflicts.”) Sometimes called “the principle of distinction,” the prohibition in Rule 1 says that the participants in a conflict “must at all times distinguish between civilians and combatants.” Attacks on combatants and military targets are lawful. Attacks on civilians or civilian targets are not.

The power of the ban on targeting civilians, people who aren’t armed participants in a conflict, can hardly be overstated. It flatly rules out the notion that a weaker party or one that has justice on its side or that is engaged in a struggle for anti-colonial liberation can ever be justified in targeting people who aren’t fighting. That includes babies and old people, but not just them. It includes anyone who isn’t fighting, regardless of gender or age.

One of the most noteworthy features of some rhetoric validating Hamas’s October 7 attacks is its apparent rejection of this core principle. Hamas did attack military targets and kill some Israeli soldiers. Had it limited its attack to such objectives, it would not have violated Rule 1. Justifying the attacks as a whole means endorsing a basic and obvious violation of international humanitarian law.

The argument that liberation movements should be able to target civilians isn’t new. (Consider Jean-Paul Sartre’s provocation, “Killing a European is killing two birds with one stone,” in his introduction to Frantz Fanon’s anti-colonial classic, the 1961 Wretched of the Earth.) But it has been consistently rejected by international law, and for good reason: It undermines the very idea that some moral principles are universal. To repudiate the distinction between combatants and civilians is to make the rules of war (in lawyers’ Latin, jus in bello) into a subset of the rules of whether a given war is just in the first place (jus ad bellum). International humanitarian law understands and accepts that people fighting wars ordinarily think their side is morally justified and the other is not. If you make the legality—or morality, for that matter—of whom you can target depend on whether you are right, that will be the end of humanitarian law: Everyone will say they are in the right and can kill whomever they choose.

The principle of distinction also applies to Israel’s actions in Gaza. Israel may not target civilians, and if it did so, that would violate the law exactly as Hamas’s actions did. Collateral damage is a different matter. The principle of distinction bars attacks directed at civilians, not attacks directed at military targets that incidentally kill noncombatants. Such attacks are, however, governed by the law of proportionality, which I will discuss below.

One famous example of an attack that arguably violated the principle of distinction is the nuclear strike on Hiroshima. President Harry Truman claimed that the Hiroshima attack was aimed at a military base there. But observers continue to question that account, arguing that the intent was to kill as many civilians as possible to pressure Japan to surrender. If the United States intended to target civilians, then its actions violated international humanitarian law as it is understood today. It’s still common for states to defend the deaths of civilians by saying they weren’t targeted. The test, ultimately, is the truth of the assertion, measured by relevant evidence that can be mustered by reasonable observers.

Terrorism is also outlawed by international humanitarian law. The ICRC definition describes terrorism as violence or threats of violence with “the primary purpose” of spreading “terror among the civilian population.” The humanitarian-law definition applies equally to state and terrorist groups. No side can set out to terrify civilians by using violence, whether it’s Israel or Hamas.

Although individual instances of terrorism can certainly fall under the prohibition on directly targeting civilians, the terrorism ban has a specific goal, namely protecting vulnerable civilians from actions that may not kill them but nonetheless terrorize them. Indiscriminately bombing or shooting civilians can qualify. The International Criminal Tribunal for the Former Yugoslavia brought war-crimes charges for violating this principle, focusing on these sorts of attacks against civilian populations.

In this context, it’s worth noting that the category of “terrorist” doesn’t have a specific meaning under international humanitarian law. It doesn’t narrowly refer to nonstate actors who target civilians for ideological reasons. The crime of terror describes certain actions based on the nature of those actions themselves, regardless of who perpetrated them.

This, too, is a feature of humanitarian law, not a bug. The law seeks to universalize obligations and protections. And it is focused on protecting the vulnerable by restricting actions directly aimed at them, not on differentiating different types of war-fighters. Just as the law rejects the idea that freedom fighters are justified in targeting civilians, it denies that states would be any more justified than nonstate actors in terrorizing a civilian population. Once again, the whole point is to designate a basic moral minimum framework for situations where both sides are prepared to fight and die, and both sides believe they are in the right.

This brings us to proportionality, one of the most confusing and complicated principles of international humanitarian law. The principle applies to attacks that target a military objective and that are otherwise lawful, but for their “disproportionality.” (A military “objective” in this sense is an actual object that contributes to military action and whose removal would give the attacker a military advantage.) But what constitutes disproportionality?

The crucial thing to understand is that disproportionality is not an absolute measure of destruction, with certain levels of destruction off-limits, but a relative one: Under international humanitarian law, an attack is disproportionate if the side launching it knows or expects that the incidental damage to civilians and civilian objects will be “excessive” or “disproportionate” relative to the “concrete and direct military advantage” that is sought. The incidental collateral damage to civilians must be proportionate—not excessive—relative to the military advantage that the attacker is pursuing in that specific military act. For example, if the goal is to destroy a military base, and it is possible to target that base alone, destroying the entire surrounding neighborhood would be excessive or disproportionate. As one U.S.-military manual cited by the ICRC puts it, “In attacking a military target, the amount of suffering or destruction must be held to the minimum necessary to accomplish the mission.”

The idea is not that one side must only inflict damage proportional to the damage inflicted by the other side. That would be a recipe for keeping wars going forever by prohibiting any side from taking the kind of decisive action that can end a war. The laws of war accept that different actors have different capacities to inflict damage. Under these rules, the number of people killed by one side need not be proportional to the number killed by the other. Nor is the idea that the attacker may weigh collateral damage in relation to its overall goal of winning the war. The military advantage or objective is described as “concrete” and “direct” in order to focus the proportionality analysis on the gains from destroying a specific military target, not the gain from defeating the enemy and creating peace.

Importantly, this is different from the use of the term proportionality in another area of international law, the right of every state to defend itself against external attack. In that context, the principle of proportionality belongs to the topic of jus ad bellum, the justness of entering the war in the first place. There it does indeed say that the state acting in self-defense must use force proportional to repelling and defeating the attacker. Jus ad bellum analysis focuses on who may lawfully and justly go to war in the first place, and so deploys proportionality with respect to the damage inflicted by the entire military engagement in relation to its legitimate war aims.

But the question of proportionality in jus ad bellum, itself relevant to the legality of Israel’s overall war aims in Gaza in reaction to Hamas’s attacks, is legally and morally distinct from the question asked by international humanitarian law, which is how much collateral damage is permissible. International humanitarian law assumes that the two sides are already at war and imposes legal and moral limits on what they can do. That is why its consideration of proportionality relates to the damage and war objectives of each separate engagement in relation to the overall campaign.

Once you have clarity on which proportionality principle is relevant, the question of proportionality in war is nevertheless challenging. To say what is or is not “excessive” or “disproportionate” requires a judgment call based on a cost-benefit analysis. But how can anyone weigh the costs and benefits of incommensurable goods like lives and military advantage? In practice, militaries like those of the U.S. and Israel follow procedures that are supposed to anticipate damage in advance and try to keep it within reason. This effort captures Michael Walzer’s observation that it is not enough to not intend to target civilians; one must also intend not to target them—by making efforts to avoid their death.

What’s more, even when weighing lives against lives, there is no clear answer to the question of what calculus should be used. How many lives of an enemy’s civilians may be lost to save one civilian life of one’s own side? Is the right ratio 1:1? Is it 10:1? Higher?

Reasonable people could differ in almost any case about whether it is disproportionate or excessive to launch an action that everyone anticipates will have the effect of killing some approximate number of civilians. And under war conditions, hardly anyone can be reliably reasonable.

Given this challenging aspect of the proportionality principle, why does international humanitarian law rely on it at all? The answer, surely, is that proportionality, notwithstanding its ambiguity, provides a degree of guidance that covers extreme cases. We can say, or should be able to say, that killing hundreds of civilians to destroy a small military base—one that does not itself have the capacity to cause great harm—would be excessive and disproportionate. We can say with a high degree of confidence that it would be unlawful to carpet bomb a residential area to kill a few soldiers or achieve a tactical advantage. The law recognizes that proportionality is imprecise but relies on it so that we can avoid the most extreme deviations from its core principle.

Genocide is prohibited by a specific treaty: the Convention on the Prevention and Punishment of the Crime of Genocide of December 9, 1948. The treaty’s definition of genocide, famously proposed by the Jewish international lawyer Raphael Lemkin, includes “acts committed with intent to destroy, in whole or in part, a national, ethnical, racial or religious group.” (The broadly accepted norm against genocide extends to political groups as well, such as those killed by the Pol Pot regime in Cambodia.) Those acts in turn are exemplified by killing or otherwise harming members of the group, inflicting conditions calculated to bring about its physical destruction, imposing measures intended to prevent births within the group, and forcibly transferring children of the group to another group.

Determining whether a given act of violence constitutes genocide therefore turns on the intent of the actor: Was it meant to destroy a group, in whole or in part? When prosecutors have charged and convicted defendants for the crime of genocide, as they did in the International Criminal Tribunal for Rwanda and the International Criminal Tribunal for the former Yugoslavia, they have proved this intent in two different ways. Direct evidence of a genocidal intent can be drawn from public and private statements declaring, for example, the perpetrators’ aims of eliminating Tutsis or Bosnian Muslims. Circumstantial evidence, such as the scale of the atrocities and a pattern of targeting members of some groups but not others for killing, can also be used to prove intent.

In ordinary speech, the term genocide is often used expansively, to include ethnic cleansing, the practice of using force and fear to remove a group of people from some territory. Ethnic cleansing would probably implicate war crimes under international humanitarian law, like extermination, or deportation as part of widespread attacks against civilians. It could also involve acts of genocide, which would themselves be crimes. But ethnic cleansing on its own is not the same as genocide. And international law has not recognized ethnic cleansing as an independent crime of its own.

In the context of the Israel-Hamas war, accusations of genocide are being levied against both sides. Sometimes the charges take the form of legal analysis. Addressing those in detail would require much more space. My view, however, is that as things stand, it would be difficult to prove charges of genocide as a matter of law against either party, especially if the standard of proof were comparable to that used by the international tribunals for Rwanda and Yugoslavia.

Israel has declared the war objective of eliminating Hamas, which is a military-political organization, not a whole people. Statements by Israeli officials that describe Hamas terrorists as “human animals” could be proffered to suggest that Israel is dehumanizing Palestinians with the intent to destroy the Palestinian people. But the repugnant expression has, at least in statements of government officials, been directed at Hamas, rather than all Palestinians—a distinction that would make a great difference as a matter of law. Israel’s conduct toward the Palestinian population in Gaza and elsewhere does not, in my view, manifest an intent to destroy the Palestinian people, in whole or in part. Those Israelis who openly seek to “transfer”—that is, expel—Palestinians by force from Israel or the West Bank or Gaza are advocating a morally reprehensible policy that would violate international law, which prohibits such forced deportations. They are not necessarily advocating genocide as defined legally.

As for Hamas, its 1988 charter called for the liberation of Palestine and for Muslim sovereignty over the entire land. (As revised in 2017, the charter allows for a return to pre-1967 borders as a “national consensus” fallback option.) The fact that Hamas lacks the capacity to destroy the Israeli or Jewish people would not preclude a genocide charge for acts that killed only some. So it could be argued that Hamas’s actions on October 7 were crimes of genocide. The revised charter specifically disclaims a religious conflict with Jews, as opposed to Zionists and Zionism, and does not expressly specify the destruction of Israelis in whole or in part.

The upshot is that charges of genocide, made in either direction, likely do not satisfy the legal definition of genocide, certainly not as it would be adjudged by any international tribunal today.

Legal language has no claim to be the only or even the best way to talk about war. But international humanitarian law does strive to create a framework for a minimal morality that everyone can agree to accept. It makes a claim to universality. Rejecting it outright marks a kind of refusal to belong to the community of nations and people who believe that some things are always wrong, no matter how right they believe their side to be. The language of conflict can be a source of conflict. Used carefully, however, it can also be a guide to engagement in the process of trying to value every human life equally. That aspiration may not succeed. But its existence has value nevertheless.

The Atlantic