by John Yoo
In war, we cannot be 100 percent sure the target is in fact the enemy or that our information is utterly correct no matter how diligent we are. In war, we are always dealing with probabilities about the future. Terrorists' plans can change at the last minute. We may have identified the wrong man, or our intelligence services may have made a simple mistake (as with the erroneous bombing of the Chinese embassy in Belgrade during the Kosovo war). Using force to prevent future harms can never be done perfectly. No military can choose or hit the correct target every time. Noncombatants near enemy locations might be killed. Under the tougher standards of domestic law enforcement, we do not punish a police officer who fires his weapon believing that his attacker holds a gun, even if it turns out after the fact that it was a fake gun. We only ask that our soldiers and policy makers make reasonable decisions under the circumstances. Every such decision must balance estimated effects on allied governments, local populations, and nearby civilians against the benefit of possibly eliminating an al Qaeda leader and frustrating his plans.
All this brings us back to al-Harithi and Derwish. Apparently the United States had good intelligence on the two men--consider that our intelligence agencies were able to locate them in a single vehicle traveling at great speed across a desert. No one has stepped forward, neither the Yemeni government nor any relatives, to claim that al-Harithi or Derwish were not members of al Qaeda or were not in that car. Because they were traveling through the desert, the chances of collateral harm to civilians was extremely low. The odds of capturing al-Harithi were not good, as he had succeeded in evading American intelligence for several years. While Yemen is a U.S. ally, parts of it are like the Wild West, beyond the control of the national government. The United States knew that an attack on the two men would not lead to a breakdown in diplomatic relations. Capturing al-Harithi and Derwish would have been preferable, as they could provide useful intelligence, but if they remained outside the reach of our government and its allies, as was deemed likely, at least we could stop them from carrying out further plots to attack the United States.
A similar analysis applied to Saddam Hussein and his sons. At the outset of the Iraq invasion, an attack on Saddam, Qusay, and Uday Hussein promised a high likelihood of success, with low civilian casualties. The benefits of success were large. Without Saddam or his sons at the top, Iraqi resistance to an invasion might have unraveled and casualties and destruction might have been dramatically reduced. Both Congress and Presidents Bush and Clinton had pursued a policy of regime change in Iraq, in good part because of Saddam's history of seeking, possessing, and using weapons of mass destruction against both military and civilian targets. A surprise missile attack on Hussein's compound could have produced that regime change in hours rather than months. Those benefits outweighed the low probable cost in civilian casualties. The question is why anyone would treat the al-Harithi/ Derwish case any differently from the Hussein case.
Law
Killing an individual, of course, is not legal in all circumstances. Nor is it illegal in all circumstances. Killing an individual is legal as capital punishment imposed on a convicted first-degree murderer. It is legal when a police officer shoots an attacker armed with a weapon. It is illegal when it is murder, as are any of the hundreds of premeditated homicides that occur in the United States every year. It is illegal when it is assassination, except that killing the enemy in wartime is legal. The rules are fairly clear, and they leave room for the ordinary exercise of judgment in making policy.
Killing Martin Luther King Jr. was an assassination. Killing a foreign head of state in peacetime is an assassination. Firing a Hellfire missile to kill bin Laden is not an assassination. Until September 11, 2001, our government remained confused about this distinction. Now it is not an assassination. This has everything to do with the law of war.
In the 1990s, the United States passed up at least three chances to kill bin Laden. By 1997, the CIA and the National Security Council had become aware that he was more than just a financier of terrorists; he was the leader of the terrorist group that posed perhaps the most direct threat to American security.22 Before the 1998 bombings of the American embassies in Kenya and Tanzania, the CIA had developed a plan to work with Afghan tribes hostile to the Taliban to capture bin Laden at his Tarnak Farms compound. The CIA's counterterrorism head thought the proposal "the perfect operation," and the military found no "showstoppers."23 CIA planning went so far as to hold a rehearsal of the operation that stretched across three time zones.
Concerns about the legal and political implications of targeting bin Laden prevented execution of the plan. President Clinton's national security adviser, Sandy Berger, thought that the evidence against bin Laden was not enough to win a conviction in the U.S. court system.24 A senior CIA official worried that the operation would run counter to the ban on assassination.25 The cabinet-level principals pulled the plug out of concern that civilian casualties could prove too high and that "the purpose and nature of the operation would be subject to unavoidable misinterpretation and misrepresentation--and probably recriminations--in the event that bin Laden, despite our best intentions and efforts, did not survive."26
These concerns continued to paralyze the administration as other opportunities came along. Through 1998, the President's orders to the CIA continued to authorize the capture of bin Laden only. Lethal force could be used against him only in self-defense. These limitations, plus shaky intelligence, led the CIA to pass up a chance to take bin Laden out with a cruise missile when he was traveling to and from Kandahar. A few weeks after the embassy bombings in August 1998, the administration had no trouble launching cruise missiles against suspected terrorist camps in Afghanistan, in the hope that bin Laden would be there. While the cruise missiles hit their targets, bin Laden and his lieutenants were absent.
By Christmas Eve 1998, President Clinton had authorized the CIA to allow America's tribal allies to kill bin Laden if they could not capture him, overriding the objections of some that this order violated the assassination ban. Administration lawyers concluded that the assassination ban would not be broken if the United States acted in self-defense under international law against an imminent threat of attack.27 But Attorney General Janet Reno warned Clinton that relaxing the assassination ban on bin Laden could invite retaliation against U.S. officials.
These missed opportunities were caused by deep misconceptions about American law regarding assassination and killing in wartime, which many then voiced and some repeat today, even after 9/11. Their arguments are unpersuasive. Neither the Constitution nor federal statutes prevent the direct targeting of individual members of the enemy. Only Executive Order 12,333, issued by President Ronald Reagan in 1981, is explicit: "No person employed by or acting on behalf of the United States Government shall engage in, or conspire to engage in, assassination."28 It continues a similar ban first put into place by President Gerald Ford in 1976, reaffirmed by President Jimmy Carter, and followed by every President since.29
While it bans assassinations, Executive Order 12,333 does not define them. Since the 1980s, however, government specialists have borrowed standard dictionary definitions to explain assassination as "an act of murder for political reasons."30 Murder is a specific legal phrase that includes only intentional and illegal homicides. Deaths that occur by accident or are authorized by law, such as a police officer's use of force to protect the life of another, would not constitute murder, and hence cannot be assassination. Killing an enemy soldier in wartime would not be assassination, because the attack has a lawful military, rather than political, purpose. By contrast, the killings of Presidents Abraham Lincoln, James Garfield, William McKinley, and John F. Kennedy were assassinations. Their assailants killed these men during peacetime and with a political purpose. The attempt to kill Pope John Paul II, if it had succeeded, would have been assassination because the Vatican was not at war and he was targeted for political reasons. While popular conceptions of assassination probably include the idea that the
killing take place covertly, that connotation does not appear in Executive Order 12,333's text, the laws of war, or their interpretation.31
It is easier to understand the scope of the American ban on assassination by briefly examining its historical context. President Ford issued the prohibition in part to head off congressional efforts to ban assassination by statute. Congress was reacting to revelations during 1970s hearings on the intelligence community, known popularly as the Church Committee hearings, which publicized the CIA's role in assassination plots against Fidel Castro, Congo's Patrice Lumumba, Dominican Republic leader Rafael Trujillo, South Vietnam president Ngo Dinh Diem, and Chilean general Rene Schneider.32 Senate leaders were concerned that the CIA had pursued these plots without the approval or knowledge of the President. With the Ford order in place, congressional efforts to ban assassination, some with broader prohibitions, failed.33 Several observers have concluded from this history that presidents intended their executive orders to ban political killings in peacetime, but not lawful killing in wartime.34 Clearly, neither President Ford nor Congress thought at the time that the executive order would preclude the targeting of perpetrators of attacks upon the United States.
The United States has a national right to use force to defend itself. However, under the UN Charter, member states must refrain from the threat or use of force "against the territorial integrity or political independence of any state, or in any other manner inconsistent with the Purposes of the United Nations."35 No exceptions were granted, such as preventing humanitarian disasters or rooting out terrorist organizations, except for two: interventions authorized by the UN Security Council "as may be necessary to maintain or restore international peace and security,"36 or "the inherent right of individual or collective self-defense."37 According to long state practice, the right to self-defense applies not only after a nation has suffered an attack, but also in anticipation of an "imminent" attack.38 Despite the arguments of some well-known scholars to the contrary, every state has, in the words of onetime Secretary of State Elihu Root, "the right...to protect itself by preventing a condition of affairs in which it will be too late to protect itself."39 The United States need not wait until an al Qaeda attack has occurred before it can launch a missile against a terrorist camp or send a special operations team to take out a terrorist leader.
Imminence should not be understood as a purely temporal concept. The concept traces its origins to the 1841 Caroline incident, in which British forces in Canada crossed the U.S. border and destroyed the Caroline, a ship being used by rebels. British and American officials agreed that a preemptive attack was justified if the "necessity of self-defense [was] instant, overwhelming, leaving no choice of means, and no moment for deliberation."40 Imminence classically depended on timing. Only when an attack is soon to occur, and thus certain, can a nation use force in preemptive self-defense.
Imminence as a temporal concept fails to deal with covert activity. Terrorists deliberately disguise themselves as civilians, and they attack by surprise. This makes it virtually impossible to use force in self-defense once an attack is "imminent." There is no target to attack in the form of the army of a nation-state. The best defense will occur only in the small window of opportunity we have to take out terrorist leaders. This often becomes available, as in the case of bin Laden, well before a major terrorist attack occurs. Imminence as a concept does not address cases in which an attack is likely to happen, but we are not sure when.
In addition to imminence, we need to account for the degree of expected harm, a function of the probability of the attack and the estimated casualties and damage. According to conventional doctrine, a nation must wait until an attack is imminent before using force, whether the attack comes at the hands of a small band of cross-border rebels, as in the Caroline, or by way of a terrorist organization armed with biological or chemical weapons. Expected harm ought to be a factor, just as it ought to be a factor in ordinary acts of self-defense, as when one is attacked with a gun as opposed to a set of fists. At the time of the Caroline decision in the early nineteenth century, the main weapons of war were single-shot weapons and artillery, cavalry, and infantry. There was an inherent technological limit on the destructiveness of armed conflict.
The speed and severity of attack possible today means that the right to preempt now should be greater than in the past. Weapons of mass destruction have increased the potential casualties a single terrorist attack can wreak, from the hundreds or thousands of innocent lives into the hundreds of thousands or even millions. This does not even count the profound, long-term destruction to cities, contamination of the environment, and long-term death or disease for large segments of the civilian population. WMDs can today be delivered with ease--a suicide bomber could detonate a "dirty bomb" using a truck or spread a biological agent with a small airplane. Detection is difficult, as no broad mobilization and deployment of regular armed forces will be visible, and imminent attacks will be virtually impossible to prevent using conventional military force. Probability, magnitude, and timing must all be relevant factors in when to use force against the enemy.
This same logic explains why most applauded President John F. Kennedy's decision to blockade Cuba during the 1962 missile crisis with the Soviet Union.41 President Kennedy did not wait until the Soviet missiles in Cuba were on the launching pad, fueled for flight. Rather, he acted earlier, during a brief window of opportunity, to head off the threat before those missiles could become operational. In doing so he risked a U.S.-Soviet war but used military measures that fell short of an attack. To prevent the potential harm from a terrorist WMD attack, the use of force involved in an assault team or cruise missile to kill Osama bin Laden and his lieutenants may be relatively low, and would fall far short of imposing a naval blockade on Cuba.
Targeted attacks to kill the enemy are one of the primary tools of war. As Hugo Grotius, the father of international law, observed in 1646, "It is permissible to kill an enemy."42 There is no indication that the presidents intended the assassination ban to prevent traditional military operations. War would be made much more difficult to win, and certainly consume many more lives, if specifically targeting members of the enemy armed forces were prohibited.
In war, enemy forces include foot soldiers and command and control up to the commander in chief. Also included are personnel and assets not directly engaged in hostilities, such as combat support units, administration, communications, logistical personnel, and suppliers. Anyone who is a legitimate military target can be attacked with a variety of means, including aerial bombing, snipers, commando raids, and surprise assaults. "All are lawful means for attacking the enemy," Hays Parks, one of America's most respected authorities on the laws of war, has written. "The choice of one vis-a-vis another has no bearing on the legality of the attack. If the person attacked is a combatant, the use of a particular lawful means for attack (as opposed to another) cannot make an otherwise lawful attack either unlawful or an assassination."43 Those same rules govern our attacks upon al Qaeda leaders and planners today. If the United States may attack Qadhafi in response to Libya's terrorism, it should have the same legal right to attack bin Laden or Zawahiri. Of course, under this rule, the targeting of the White House and Pentagon on 9/11 was legal, although the method of the attack was not because of the hijacking of civilian airliners.
Limits
All of this is not to say that American agents have a hunting license on anyone they suspect is an al Qaeda operative. The rules of warfare, which give nations and their militaries the right to use deadly force to defeat an enemy, also impose guidelines. But they are not abstract, sterile, and restrictive. Contrary to the protests from professors and activists every time the United States launches a missile or drops a bomb, the rules of war are not broken when a missile goes astray or nearby civilians lose their lives. The laws of war take into account the harsh reality that war is not a precise science and that unanticipated harms or loss of life ancillary to a military attack will occur.
A corollary of the right to destroy enemy personnel and assets is the fact that the deaths of civilians that occur as a result of legitimate attacks against military targets are not illegal. This is the source of the idea of "collateral damage," which made its controversial appearance in the Vietnam War. But the rule is as old as war itself. The central principle of the laws of war is that innocent civilians should not be targeted. On the other hand, the rules of war accept the death of civilians at or near legitimate military targets. Law recognizes that war does not yet amount to antiseptic surgery where we can zap cancers with lasers that leave healthy tissue nearby unharmed.
Thus, the United States does not commit murder if it bombs a location that contains both bin Laden and his associates, on the one hand, and their family members on the other. It did not commit murder in Iraq when, in the course of a firefight with Uday and Qusay Hussein, who were holed up in a residential building in the middle of a densely populated city, civilians next door were harmed too. Rather, it is the terrorists who violate the rules of war by deliberately hiding themselves and their bases of operation within civilian populations, thereby drawing unwilling and unsuspecting innocents into the fighting. In another example of asymmetric tactics, terrorists multiply their strength by relying on the humanitarian morals of the West not to harm civilians. Terrorists know that American military and civilian leaders are reluctant to launch attacks that might cause civilian casualties.
Killing or disabling enemy personnel does not mean that anything goes. The United States cannot use poison on terrorist leaders, or refuse to accept surrender, or shoot the wounded. One of the early laws of war treaties, known as the Hague Regulations of 1907, prohibits "kill[ing] or wound[ing] treacherously individuals belonging to the hostile nation or army," as well as killing or wounding an enemy who is helpless or has surrendered, or declaring "that no quarter will be given." The U.S. military interprets this provision as "prohibiting assassination, proscription, or outlawry of an enemy, or putting a price upon an enemy's head, as well as offering reward for an enemy 'dead or alive.'"44 Assassination prohibits killing the enemy using treachery. The United States may offer a reward for information or capture, but it does not place bounties on al Qaeda leaders' heads.
