Warrick Page/Corbis

Israeli soldiers returning from the war in Gaza, near the Israeli town of Sderot, January 18, 2009

In 2005, Asa Kasher and Amos Yadlin published in an American academic journal “Assassination and Preventive Killing,”1 an essay that explores the issue of “assassination within the framework of fighting terror.” There are good reasons to believe that the political and practical significance of this essay goes far beyond its academic interest. Asa Kasher is professor of professional ethics and philosophy of practice at Tel Aviv University and an academic adviser to the Israel Defense Forces (IDF). Amos Yadlin is a major general who at the time the article appeared was the military attaché of the embassy of Israel in Washington; he is currently the head of Israeli army intelligence.

The writers are quick to point out that the “views expressed in the present paper are those of the authors and not necessarily those of the…IDF or the State of Israel.” But the issue is not whether their views are official, but whether they are in fact influential in the Israeli army. Soon after the recent Israeli intervention in Gaza, Amos Harel argued in Haaretz (February 6, 2009) that the guidelines suggested in the article are indeed the ones that govern the IDF’s conduct in battle. This claim has since been both affirmed and denied by Israeli soldiers. We will not join that dispute here, but given the intense interest in Israel’s rules of engagement in the Gaza fighting, it’s critically important to address Kasher and Yadlin’s argument.

We are not going to deal here with the issue of targeted assassination, which is the paper’s explicit subject. Instead we want to challenge what the authors say is their most “important and sensitive” claim. Kasher and Yadlin ask:

What priority should be given to the duty to minimize casualties among the combatants of the state when they are engaged in combat…against terror?

When they write of combatants of “the state,” the authors mean states in general, including the armed forces of the state of Israel. By “terror” they mean the intentional killing of civilians, as by members of Hamas in recent years. And this is their answer:

Usually, the duty to minimize casualties among combatants during combat is last on the list of priorities, or next to last, if terrorists are excluded from the category of noncombatants. We firmly reject such a conception because it is immoral. A combatant is a citizen in uniform. In Israel, quite often, he is a conscript or on reserve duty. His state ought to have a compelling reason for jeopardizing his life. The fact that persons involved in terror are depicted as noncombatants and that they reside and act in the vicinity of persons not involved in terror is not a reason for jeopardizing the combatant’s life in their pursuit…. The terrorists shoulder the responsibility for their encounter with the combatant and should therefore bear the consequences.

And they go on:

Where the state does not have effective control of the vicinity, it does not have to shoulder responsibility for the fact that persons who are involved in terror operate in the vicinity of persons who are not.

One quick remark is in order. There is nothing in these quotations that hinges on the word “terrorists.” Replace that word with “enemy combatants” and the argument remains the same. Kasher and Yadlin are simply assuming that the war against the enemy is a just war. Their claim, crudely put, is that in such a war the safety of “our” soldiers takes precedence over the safety of “their” civilians.

Our main contention is that this claim is wrong and dangerous. It erodes the distinction between combatants and noncombatants, which is critical to the theory of justice in war (jus in bello). No good reasons are given for the erosion.

The point of just war theory is to regulate warfare, to limit its occasions, and to regulate its conduct and legitimate scope. Wars between states should never be total wars between nations or peoples. Whatever happens to the two armies involved, whichever one wins or loses, whatever the nature of the battles or the extent of the casualties, the two nations, the two peoples, must be functioning communities at the war’s end. The war cannot be a war of extermination or ethnic cleansing. And what is true for states is also true for state-like political bodies such as Hamas and Hezbollah, whether they practice terrorism or not. The people they represent or claim to represent are a people like any other.

The main attribute of a state is its monopoly on the legitimate use of violence. Fighting against a state is fighting against the human instruments of that monopoly—and not against anyone else. It might be a morally nicer world if states would agree to limit their wars even more and to be represented by champions like David and Goliath. In settling disputes a gladiatorial duel would be better than a war. But duels like that take place in the Bible or in Homeric epics, not in the real world. In the real world, we watch with dismay a tendency to enlarge, rather than to reduce, the scope of wars. In World War I, only 15 percent of the casualties were civilians, whereas in World War II the civilian total reached 50 percent.


The crucial means for limiting the scope of warfare is to draw a sharp line between combatants and noncombatants. This is the only morally relevant distinction that all those involved in a war can agree on. We should think of terrorism as a concerted effort to blur this distinction so as to turn civilians into legitimate targets. When fighting against terrorism, we should not imitate it.

The contrast between combatants and noncombatants is not a contrast between innocent civilians on the one hand and guilty soldiers on the other. Civilians are not necessarily innocent, in the sense of being free from guilt for evildoing. German civilians who were enthusiastic supporters of the Nazis were certainly not innocent in that sense. Innocence is a term of art: noncombatants are innocent because they do not participate directly in the war effort; they lack the capacity to injure, whereas combatants qua combatants acquire this capacity. And it is the capacity to injure that makes combatants legitimate targets in the context of war. Men and women without that capacity are not legitimate targets. (Workers in weapons and munitions factories create the means to injure and are legitimate targets. As Elizabeth Anscombe argued long ago, workers making K-rations, food for the soldiers, do not create the means to injure and are not legitimate targets. But these issues are not our subject here.)2

Combatants are accountable only for their conduct in war. They do not become criminals because they are fighting in an aggressive war, and they don’t acquire immunity from attack because they are fighting a just war, on the side of the angels. The presumption of just war theory is that all the combatants believe that their country is fighting a just war. This is a necessary and also a reasonable presumption, given the way those who become combatants are brought up, educated, and indoctrinated. We can demand of soldiers that they react morally to concrete combat situations; we can’t demand that they judge correctly the moral merit of the reasons their political leaders give them for going to war.

The presumption that combatants have a subjective justification to fight can be rebutted. Mercenaries, or participants in a drug cartel’s battle with the government, or soldiers in a war of extermination are not presumed to have subjective moral justification. Mercenaries and mobsters believe, of course, that they join the war for good money, but they don’t believe that they do so for good moral reasons. The beliefs of genocidaires don’t matter in judging their behavior; we make no presumption in their favor. For the greater number of cases in modern history, however, the presumption holds.

When two sides to a war claim that justice is on their side, they usually make incompatible claims, but not contradictory claims. When claims are contradictory, the two sides cannot both be right, and they cannot both be wrong. If one side says, “10,000 civilians were killed in the war,” and the other side says, “No, it is not the case that 10,000 civilians were killed in the war,” one of them is right and the other is wrong. When claims are incompatible, by contrast, both sides can be wrong—as when one side says that UN vehicles are green and the other side says that they are yellow—but they cannot both be right.

Incompatibility rather than contradiction is the case in most (though certainly not all) wars: both sides fight a war that is objectively unjust, and both sides believe subjectively that justice is on their side. And indeed, they may each have just grievances against the other, and so their conflict may have elements of tragedy. But logically, they cannot both have just reasons for going to war. It may be that one of them is justified in fighting and the other isn’t. But it is also possible, and common in human history, that neither side has a good reason to fight. The fact that both sides may be wrong, and often are, is another reason to refuse to blame soldiers for participating in a war, any war. Even if their country is wrong, they have a right to fight. We blame them only for immoral conduct in the course of the war.


The position that we mean to oppose is the opposite of this view. It holds that only the side that is fighting for a just cause (our side) has a right to fight, and that soldiers on the other side have no rights at all. Anything they do is immoral, whether they attack our soldiers or our civilians. And since our soldiers and civilians are equally innocent, we cannot ask our soldiers to take risks to protect enemy civilians. Those civilians have been put at risk by the immoral conduct of their soldiers.

The two senses of just war, jus ad bellum, the justice of the decision to go to war, and jus in bello, the justice of the conduct of war, have to be kept separate. Heads of states should be mainly accountable for the first, soldiers and their officers for the second. Blurring this line of separation undermines the categorical distinction between combatants and noncombatants, and it puts noncombatants (on whichever is taken to be the wrong side) at risk in new and dangerous ways.

The presumption of subjective justification applies to the combatants of Hamas and Hezbollah. They should, of course, be accountable for their conduct in war, especially when they make civilians the primary targets of their attack—and also when they deliberately use civilians as human shields. But neither of these crimes allows their enemies to give up their own obligation to avoid or minimize civilian injuries and deaths.


Adel Hana/AP Images

Tents set up for Palestinians next to their destroyed houses, Jabaliya, northern Gaza, January 30, 2009

Each side in a war quite naturally views its soldiers not as helmeted warriors but as “our kids,” young, pure, and innocent, who have been trained and issued uniforms by the state and who find themselves endangered by a cruel enemy. On many occasions, the public cares more for the life of its soldiers than for the life of its civilians. It is this understandable but morally misguided sentiment that creeps into the Kasher-Yadlin paper when they write: “A combatant is a citizen in uniform”—so as to convince us that we should not ask our soldiers to take risks to save the lives of noncombatants on the other side. This isn’t the same as saying that a diplomat is a citizen in the uniform of a head waiter. A uniform in the case of combatants is not merely conventional; it is the crucial sign of the distinction between combatants and noncombatants—the distinction that guerrillas and terrorists try to obscure by not wearing uniforms.

This is what each side should say to its soldiers:

By wearing a uniform, you take on yourself a risk that is borne only by those who have been trained to injure others (and to protect themselves). You should not shift this risk onto those who haven’t been trained, who lack the capacity to injure; whether they are brothers or others. The moral justification for this requirement lies in the idea that violence is evil, and that we should limit the scope of violence as much as is realistically possible. As a soldier, you are asked to take an extra risk for the sake of limiting the scope of the war. Combatants are the Davids and Goliaths of their communities. You are our David.

How do Kasher and Yadlin blur the distinction between combatants and noncombatants? By enabling “our” combatants to jump the queue for their own safety—so that their safety comes before the safety of civilians (whoever they are). For Kasher and Yadlin, there no longer is a categorical distinction between combatants and noncombatants. But the distinction should be categorical, since its whole point is to limit wars to those—only those—who have the capacity to injure (or who provide the means to injure).

Here is a concrete example that will help us to think about whether “our” combatants or “their” civilians should be given priority. Before the 2006 war in Lebanon, there were rumors in the Israeli press that Hezbollah planned to capture kibbutz Manara in northern Israel adjacent to the Lebanese border. We don’t know how much credence to give to those reports, but the idea of capturing a kibbutz sounds plausible enough; we will use it as a thought experiment to test the rival priority claims.

Assume that Hezbollah carried out this plan and took effective control of the area of Manara. Now consider four possible scenarios:

  1. Hezbollah captured Manara and held all its members, Israeli citizens, as hostages. Hezbollah combatants mingle with the kibbutz members so as to be shielded by them from any counterattack.
  2. Hezbollah captured only the outskirts of Manara, and a group of pro-Israeli, noncombatant volunteers from outside Israel—not Israeli citizens—who worked in Ma-nara and lived near the border were seized and used as human shields.
  3. Instead of well-wishing volunteers as in scenario 2, we now have a group of protesters from abroad, who traveled to the northern border of Israel to raise their voices against Israel’s policy toward Lebanon. As it happened, Hezbollah did not pay much attention to their protest, but seized and used them as its human shields.
  4. Before Hezbollah captured Ma-nara, the kibbutz was evacuated, and now Hezbollah brings in civilian villagers from South Lebanon, in order to claim that the kibbutz land belongs to them, but also to use them as human shields.

In all four cases, Israel is about to launch a military operation to recapture Manara. Note that Hezbollah has effective control of the kibbutz and controls the fate of the different noncombatants held there. We claim that Israel is morally required to behave in all those cases the way it would behave in the first case, when its citizens are held by Hezbollah in “a mixed vicinity.”

Whatever Israel deems acceptable as “collateral damage” when its own captured citizens are at risk—that should be the moral limit in the other cases too. If, as an Israeli, you think that a military operation will cause excessive harm to Israeli civilians, you should have equal concern for the excessive harm done to other civilians, whether they are welcome guests, unwelcome guests, or enemy noncombatants. The rules of engagement for Israeli soldiers are the same in all the cases, no matter how they feel toward the different groups. And if they observe those rules, and take the morally necessary risks, responsibility for the deaths of Hezbollah’s human shields—in all the cases—falls only on Hezbollah.

What degree of risk should Israeli soldiers assume in the first scenario? We can’t answer that question with any precision. They don’t have to take suicidal risks, certainly; nor do they have to take risks that make the recapture of Manara impossibly difficult. They are fighting against enemies who try to kill Israeli civilians and intentionally put civilians at risk by using them as cover. Israel condemns those practices; at the same time, however, it kills far more civilians than its enemies do, though without intending the deaths as a matter of policy. (Thirteen Israelis died in the Gaza fighting, some of them from friendly fire; between 1,200 and 1,400 Gazans were killed, half or more of them civilians.) But merely “not intending” the civilian deaths, while knowing that they will occur, is not a position that can be vindicated by Israel’s condemnation of terrorism. So how can Israel prove its opposition to the practices of its enemies? Its soldiers must, by contrast with its enemies, intend not to kill civilians, and that active intention can be made manifest only through the risks the soldiers themselves accept in order to reduce the risks to civilians.

There is nothing unusual in this demand, and nothing unique to Israel. When soldiers in Afghanistan, or Sri Lanka, or Gaza take fire from the rooftop of a building, they should not pull back and call for artillery or air strikes that may destroy most or all of the people in or near the building; they should try to get close enough to the building to find out who is inside or to aim directly at the fighters on the roof. Without a willingness to fight in that way, Israel’s condemnation of terrorism and of the use of human shields by its enemies rings hollow; no one will believe it.

But shouldn’t there be a difference between noncombatants forced to intermingle with Hezbollah combatants and those who intermingle voluntarily—for example, the villagers who came to Manara perhaps hoping to acquire kibbutz land?

This is a tricky question, and Israelis should be sensitive to its implications. The military headquarters of the Israeli army—including its war situation room—is located in the middle of north Tel Aviv, in one of its most expensive residential areas. This is no secret, and the civilians living near the headquarters knowingly put themselves at risk. Should they be more at risk because they live there voluntarily? We don’t think so. They may be entitled to more protection from their own state. But whatever collateral damage is tolerable in a war justly conducted is tolerable for them, and no more. They are clearly noncombatants, and the rules that apply to the treatment of noncombatants apply to them.

Kasher and Yadlin claim that “jeopardizing combatants rather than bystanders during a military act against a terrorist would mean shouldering responsibility for the mixed nature of the vicinity for no reason at all.” We agree that the terrorists are often responsible for the “mixed nature of the vicinity”—they may mingle, for example, with people in a marketplace in order to hide or they may fire weapons from the houses of innocent civilians—but that does not alter the responsibility of soldiers to minimize as best they can the risks to noncombatant bystanders. If there is “no reason” for responsibility of this sort, if the lives of “our” soldiers really take priority over “their” civilians, then why couldn’t the soldiers use those civilians as shields? Since they have not created the “mixed vicinity,” why can’t they in turn take advantage of it? We don’t see how Kasher and Yadlin can avoid providing justification for a practice that Israel officially condemns and that we believe they believe is despicable: the use of noncombatants as human shields for combatants.

We hope that Kasher and Yadlin will agree that the degree of risk that Is-raeli soldiers must accept in the “mixed vicinity” of Manara is the same in all of our four cases, whoever is responsible for the presence of people with different allegiances and identities. It is not that responsibility for the mix is irrelevant, but that the side that creates the mix does not thereby free the other side from its own moral obligations. The claim, in this case, that Hezbollah’s actions conferred that kind of freedom would not be taken seriously in Israel if Israeli civilians, bystanders, were in the mix.

This is the guideline we advocate: Conduct your war in the presence of noncombatants on the other side with the same care as if your citizens were the noncombatants. A guideline like that should not seem strange to people who are guided by the counterfactual line from the Passover Haggadah, “In every generation, a man must regard himself as if he had come out of Egypt.”

This Issue

May 14, 2009