By Kenneth Roth
Where are the proper boundaries of what the Bush administration calls its war on terrorism? The recent wars against the Afghan and Iraqi governments were classic armed conflicts, with organized military forces facing each other. But the administration says its war on terrorism is global, extending far beyond these typical battlefields. On September 29, 2001, U.S. President George W. Bush said, “Our war on terror will be much broader than the battlefields and beachheads of the past. The war will be fought wherever terrorists hide, or run, or plan.”
This language stretches the meaning of the word “war.” If Washington means “war” metaphorically, as when it speaks of the war on drugs, the rhetoric would be uncontroversial—a mere hortatory device designed to rally support to an important cause. But the administration seems to think of the war on terrorism quite literally—as a real war—and that has worrying implications.
The rules that bind governments are much looser during wartime than in times of peace. The Bush administration has used war rhetoric to give itself the extraordinary powers enjoyed by a wartime government to detain or even kill suspects without trial. Enticing as such enhanced power might be in the face of the unpredictable and often lethal threat posed by terrorism, it threatens basic due process rights and the essential liberty such rights protect.
War and peace rules
By literalizing its “war” on terror, the Bush administration has broken down the distinction between what is permissible in times of peace and what can be condoned during a war. In peacetime, governments are bound by strict rules of law enforcement. Police can use lethal force only if necessary to meet an imminent threat of death or serious bodily injury. Once a suspect is detained, he or she must be charged and tried. These requirements—what one can call “law enforcement rules”—are codified in international human rights law.
In times of war, law enforcement rules are supplemented by the more permissive rules of armed conflict, or international humanitarian law. Under these “war rules,” an enemy combatant can be shot without warning (unless he is incapacitated, in custody, or trying to surrender), regardless of any imminent threat. If a combatant is captured, he or she can be held in custody until the end of the conflict, without being charged or tried.
These two sets of rules have been well developed over the years, by both tradition and detailed international conventions. There is little law, however, to explain when one set of rules should apply instead of the other. Usually the existence of an armed conflict is obvious, especially when two governments are involved. But in other circumstances, such as the Bush administration’s announced war on terrorism as it extends beyond Afghanistan and Iraq, it is less clear.
For example, the Geneva Conventions—the principal codification of war rules—apply to “armed conflict” but do not define the term. However, the International Committee of the Red Cross (ICRC), the official custodian of the conventions, does provide some guidance in its commentary, in distinguishing between civil war and mere riots or disturbances.
One test suggested by the ICRC for determining whether wartime or peacetime rules apply is to examine the intensity of hostilities. The Bush administration, for example, claims that al-Qaeda is at war with the United States because of the magnitude of the September 11, 2001 attacks as well as the pattern of al-Qaeda’s alleged bombings including of the U.S. embassies in Kenya and Tanzania, the U.S.S. Cole in Yemen, and residential compounds in Saudi Arabia. Each of these attacks was certainly a serious crime warranting prosecution. But technically speaking, was the administration right to say that they add up to war? Is al-Qaeda a ruthless criminal enterprise or a military operation? The ICRC’s commentary does not provide a clear answer.
In addition to the intensity of hostilities, the ICRC suggests considering such factors as the regularity of armed clashes and the degree to which opposing forces are organized. Whether a conflict is politically motivated also seems to play an unacknowledged role in deciding whether it is “war” or not. Thus, organized crime or drug trafficking, though methodical and bloody, are generally understood to present problems of law enforcement, whereas armed rebellions, once sufficiently organized and violent, are usually seen as “wars.”
The problem with these guidelines, however, is that they were written to address domestic conflicts rather than global terrorism. Thus, they do not make clear whether al-Qaeda should be considered an organized criminal operation (which would trigger law-enforcement rules) or a rebellion (which would trigger war rules). The case is close enough that the debate of competing metaphors does not yield a conclusive answer. Clarification of the law would be useful.
Even in the case of war, another factor in deciding whether law-enforcement rules should apply is the nature of a given suspect’s involvement. War rules treat as combatants only those who are taking an active part in hostilities. Typically, that includes members of an armed force who have not laid down their arms as well as others who are directing an attack, fighting or approaching a battle, or defending a position. Under these rules, even civilians who pick up arms and start fighting can be considered combatants and treated accordingly. But this definition is difficult to apply to terrorism, where roles and activities are clandestine, and a person’s relationship to specific violent acts is often unclear.
Given this confusion, a more productive approach is to consider the policy consequences of applying wartime or law enforcement rules. Unfortunately, the Bush administration seems to have ignored such concerns.
Padilla and al-Marri
Consider, for example, the cases of Jose Padilla and Ali Saleh Kahlah al-Marri. Federal officials arrested Padilla, a U.S. citizen, in May 2002 when he arrived from Pakistan at Chicago’s O’Hare Airport, allegedly to scout out targets for a radiological or “dirty” bomb. As for al-Marri, a student from Qatar, he was arrested in December 2001 at his home in Peoria, Illinois, for allegedly being a “sleeper,” an inactive accomplice who could be activated to help others launch terrorist attacks. If these allegations are true, Padilla and al-Marri should certainly be prosecuted. Instead, after initially holding each man on other grounds, President Bush declared them both to be “enemy combatants” and claimed the right to hold them without charge or trial until the end of the war against terrorism—which, of course, may never come.
But should Padilla and al-Marri, even if they have actually done what the U.S. government claims, really be considered warriors? Aren’t they more like ordinary criminals? A simple thought experiment shows how dangerous are the implications of treating them as combatants. The Bush administration has asserted that the two men planned to wage war against the United States and therefore can be considered de facto soldiers. But if that is the case, then under war rules, the two men could have been shot on sight, regardless of any immediate danger they posed. Padilla could have been gunned down as he stepped off his plane at O’Hare, al-Marri as he left his home in Peoria. That, after all, is what it means to be a combatant in time of war.
Most people, I suspect, would be deeply troubled by that result. The Bush administration has not alleged that either suspect was anywhere near to carrying out his alleged terrorist plan. Neither man, therefore, posed an imminent threat of the sort that might justify the preventive use of lethal force under law enforcement rules. With a sophisticated legal system available to hear their cases, killing these men would have seemed gratuitous and wrong. Of course, the Bush administration has not proposed summarily killing them; it plans to detain them indefinitely. But if Padilla and al-Marri are not enemy combatants for the purpose of being shot, they should not be enemy combatants for the purpose of being detained, either. The one conclusion necessarily implies the other.
Even if they were appropriately treated as combatants, Padilla’s and al-Marri’s lives might still have been spared under the doctrine of military necessity, which precludes using lethal force when an enemy combatant can be neutralized through lesser means. But from the bombing of urban bridges in northern Serbia during the Kosovo war to the slaughter on the “Highway of Death” during the 1991 Gulf War, the U.S. government has been at best inconsistent in respecting the doctrine of military necessity. Other governments’ records are even worse. That terrorist suspects who pose no immediate danger might only sometimes be shot without warning should still trouble us and lead us to question the appropriateness of their classification as combatants in the first place.
A similar classification problem, though with an arguably different result, arose in the case of Qaed Salim Sinan al-Harethi. Al-Harethi, who Washington alleges was a senior al-Qaeda official, was killed by a drone-fired missile in November 2002 while driving in a remote tribal area of Yemen. Five of his companions also died in the attack, which was carried out by the CIA. The Bush administration apparently considered al-Harethi an enemy combatant for his alleged involvement in the October 2000 U.S.S. Cole bombing, in which seventeen sailors died.
In this instance, the case for applying war rules was stronger than with Padilla or al-Marri, although the Bush administration never bothered to spell it out. Al-Harethi’s mere participation in the 2000 attack on the Cole would not have made him a combatant in 2002, since in the interim he could have withdrawn from al-Qaeda; war rules permit attacking only current combatants, not past ones. And if al-Harethi were a civilian, not a member of an enemy armed force, he could not be attacked unless he were actively engaged in hostilities at the time. But the administration alleged that al-Harethi was a “top bin Laden operative in Yemen,” implying that he was in the process of preparing further attacks. If true, this would have made the use of war rules against him more appropriate. And unlike Padilla and al-Marri, arresting al-Harethi may not have been an option. The Yemeni government has little control over the tribal area where al-Harethi was killed; eighteen Yemeni soldiers had reportedly died in an earlier attempt to arrest him. However, even in this arguably appropriate use of war rules, the Bush administration offered no public justification, apparently unwilling to acknowledge even implicitly any legal constraints on its use of lethal force against alleged terrorists.
Bosnia and Malawi
In other cases outside the United States, the Bush administration’s use of war rules has had far less justification. For example, in October 2001, Washington sought the surrender of six Algerian men in Bosnia. At first, the U.S. government followed law enforcement rules and secured the men’s arrest. But then, after a three-month investigation, Bosnia's Supreme Court ordered the suspects released for lack of evidence. Instead of providing additional evidence, however, Washington switched to war rules. It pressured the Bosnian government to hand the men over anyway and whisked them out of the country—not to trial, but to indefinite detention at the U.S. naval base at Guantánamo Bay. If the men had indeed been enemy combatants, a trial would have been unnecessary, but there is something troubling about the administration’s resort to war rules simply because it did not like the result of following law enforcement rules.
The administration followed a similar pattern in June 2003, when five al-Qaeda suspects were detained in Malawi. Malawi’s high court ordered local authorities to follow criminal justice laws and either charge or release the five men, all of whom were foreigners. Ignoring local law, the Bush administration insisted that the men be handed over to U.S. security forces instead. The five men were spirited out of the country to an undisclosed location—not for trial, but for interrogation. The move sparked riots in Malawi. The men were released a month later in Sudan, after questioning by Americans failed to turn up incriminating evidence.
These cases are not anomalies. In the last two-and-a-half years, the U.S. government has taken custody of a series of al-Qaeda suspects in countries such as Pakistan, Thailand, and Indonesia. In many of these cases, the suspects were not captured on a traditional battlefield, and a local criminal justice system was available. Yet instead of allowing the men to be charged with a crime under local law-enforcement rules, Washington had them treated as combatants and delivered to a U.S. detention facility in an undisclosed location.
A Misuse of War Rules?
Is this method of fighting terrorism away from a traditional battlefield an appropriate use of war rules? At least insofar as the target can be shown to be actively involved in ongoing terrorist activity amounting to armed conflict, war rules might be acceptable when there is no reasonable criminal justice option, as in tribal areas of Yemen. But there is something troubling, even dangerous, about using war rules when law enforcement rules reasonably could have been followed.
Errors, common enough in ordinary criminal investigations, are all the more likely when the government relies on the murky intelligence that drives many terrorist investigations. The secrecy of terrorist investigations, with little opportunity for public scrutiny, only compounds the problem. If law enforcement rules are used, a mistaken arrest can be rectified at a public trial. But if war rules apply, the government is never obliged to prove a suspect’s guilt. Instead, a supposed terrorist can be held for however long it takes to win the “war” against terrorism—potentially for life—with relatively little public oversight. And the consequences of error are even graver if the supposed combatant is killed, as was al-Harethi. Such mistakes are an inevitable hazard of the traditional battlefield, where quick life-and-death decisions must be made. But when there is no such urgency, prudence and humanity dictate applying law enforcement rules.
Washington must also remember that its conduct sets an example, for better or worse, for many governments around the world. After all, many other states would be all too eager to find an excuse to eliminate their enemies through war rules. Israel, to name one, has used this rationale to justify its assassination of terrorist suspects in Gaza and the West Bank. It is not hard to imagine Russia doing the same to Chechen leaders in Europe, Turkey using a similar pretext against Kurds in Iraq, China against Uighurs in Central Asia, or Egypt against Islamists at home.
There is some indication that the Bush administration may be willing to abide by a preference for law enforcement rules when it comes to using lethal force. President Bush has reportedly signed a secret executive order authorizing the CIA to kill al-Qaeda suspects anywhere in the world but limiting that authority to situations in which other options are unavailable. But when it comes to detention, the administration has been quicker to invoke war rules.
Both the administration’s reluctance to kill terrorist suspects and its preference for detention over trial presumably stem in part from its desire to interrogate suspects to learn about potential attacks. Just as a dead suspect cannot talk, a suspect with an attorney may be less willing to cooperate. Moreover, trials risk disclosure of sensitive information, as the administration has discovered in prosecuting Zacarias Moussaoui. These are the costs of using a criminal justice system.
But international human rights law is not indifferent to the needs of a government facing a security crisis. Under a concept known as “derogation,” governments are permitted to suspend certain rights temporarily if they can show that it is necessary to meet a “public emergency threatening the life of the nation.” The International Covenant on Civil and Political Rights, which the United States has ratified, requires governments invoking derogation to file a declaration justifying the move with the U.N. secretary-general. Among the many governments to have done so are Algeria, Argentina, Chile, Colombia, Peru, Poland, Russia, Sri Lanka, and the United Kingdom. Yet instead of derogating from law enforcement rules, the Bush administration has opted to use war rules.
The difference is more than a technicality. Derogation is a tightly circumscribed exception to ordinary criminal justice guarantees, permitted only to the extent necessary to meet a public emergency and scrutinized by the U.N. Human Rights Committee. Moreover, certain rights—such as the prohibition of torture or arbitrary killing—can never be suspended. The Bush administration, however, has resisted justifying its suspension of law enforcement rules and opposed scrutiny of that decision, whether by international bodies or even by U.S. courts. Instead, it has unilaterally given itself the greater latitude of war rules.
The U.S. Justice Department has defended the Bush administration’s use of war rules for suspects apprehended in the United States by citing a U.S. Supreme Court decision from World War II, Ex Parte Quirin. In that case, the court ruled that German army saboteurs who landed in the United States could be tried as enemy combatants before military commissions. The court distinguished its ruling in an earlier, Civil War-era case, Ex Parte Milligan, which had held that a civilian resident of Indiana could not be tried in military court because local civil courts remained open and operational. Noting that the German saboteurs had entered the United States wearing at least parts of their uniforms, the court in Quirin held that the Milligan protections applied only to people who are not members of an enemy’s armed forces.
But there are several reasons why, even under U.S. law, Quirin does not justify the Bush administration’s broad use of war rules. First, the saboteurs in Quirin were agents of a government with which the United States was obviously at war. The case does not help determine whether, away from traditional battlefields, the United States should be understood as fighting a “war” with al-Qaeda or pursuing a criminal enterprise. Second, although the court in Quirin defined a combatant as anyone operating with hostile intent behind military lines, the case has arguably been superseded by the 1949 Geneva Conventions (ratified by the United States), which, as noted, treat as combatants only people who are either members of an enemy’s armed force or are taking active part in hostilities. Quirin thus does not help determine whether, under current law, people such as Padilla and al-Marri should be considered civilians (who, under Milligan, must be brought before civil courts) or combatants (who can face military treatment).
Moreover, Quirin establishes only who can be tried before a military tribunal. The Bush administration, however, has asserted that it has the right to hold Padilla, al-Marri, and other detained “combatants” without charge or trial of any kind—in effect, precluding serious independent assessment of the grounds for potentially lifetime detention. The difference is especially significant because in the case of terrorist suspects allegedly working for a shadowy group, error is more likely than it was for the uniformed German saboteurs in Quirin.
Finally, whereas the government in Quirin was operating under a specific grant of authority from Congress, the Bush administration, in treating suspects as enemy combatants, is operating largely on its own. This lack of congressional guidance means that the difficult judgment calls in drawing the line between war and law enforcement rules are being made behind closed doors, without the popular input that a legislative debate would provide.
A Policy Approach
So, when the “war” on terrorism is being fought away from a traditional battlefield, how should the line be drawn between war and law enforcement rules? No one should lightly give up due process rights, as the Bush administration has done with its “enemy combatants”—particularly when a mistake could result in death or lengthy detention without charge or trial. Rather, law enforcement rules should presumptively apply to all suspects, and the burden should fall on those who want to invoke war rules to demonstrate that they are necessary and appropriate.
The following three-part test would help assess whether a government has met its burden when it asserts that law enforcement rules do not apply. To invoke war rules, a government should have to prove, first, that an organized group is directing repeated acts of violence against it, its citizens or interests with sufficient intensity that it constitutes an armed conflict; second, that the suspect is an active member of the opposing armed force or an active participant in the violence; and, third, that law enforcement means are unavailable.
Within the United States, the third requirement would be nearly impossible to satisfy—as it should be. Given the ambiguities of investigating terrorism, it is better to be guided more by Milligan’s affirmation of the rule of law than by Quirin’s exception to it. Outside the United States, Washington should never resort to war rules away from a traditional battlefield if local authorities can and are willing to arrest and deliver a suspect to an independent tribunal—regardless of how the tribunal then rules. War rules should only be used in cases when no law enforcement system exists (and the other conditions of war are present), not when the rule of law happens to produce inconvenient results. Even if military forces are used to make an arrest in such cases, law enforcement rules might still apply; only when attempting an arrest is too dangerous should war rules be countenanced.
This approach would recognize that war rules may have their place in fighting terrorism, but given the way they inherently compromise fundamental rights, they should be used sparingly. Away from a traditional battlefield, they should be used, even against a warlike enemy, as a tool of last resort—when there is no reasonable alternative, not when a functioning criminal justice system is available. Until there are better guidelines on when to apply war and law enforcement rules, this three-part test, drawn from the policy consequences of the decision, offers the best way to balance security and rights. In the meantime, the Bush administration should abandon its excessive use of war rules. In attempting to make Americans safer, it has made all Americans, and everyone else, less free.
The Israeli-Palestinian conflict provides a useful context to apply this test. Since late 2000, the Israeli government has been deliberately assassinating Palestinians in the West Bank and Gaza Strip whom it claims are involved in attacks against Israelis, particularly Israeli civilians. In many cases, Palestinian civilians died in the course of these assassinations, sometimes because suspects were targeted while in residential buildings or on busy thoroughfares. Even if these attacks might otherwise have been justified, some would violate the international prohibition on attacks that are indiscriminate or cause disproportionate harm to civilians. In other cases, however, the assassinations have hit their mark with little or no harm to others. Can these well-targeted assassinations be justified?
Although the level of violence between Israeli and Palestinian forces has varied considerably over time, the violence in certain cases has been intense and sustained enough for the Israeli government reasonably to make the case that in those instances an armed conflict exists.
As for the second prong, the Israeli government would have to show, as noted, that the targeted individual was an active participant in these hostilities, such as by directing an attack, fighting or approaching a battle, or defending a position. The Israeli government used to claim that the Palestinians targeted for assassination were involved in plotting attacks against Israelis, although increasingly the government has not bothered to make that claim. Even when it does so, the summary nature of the claim means that there is nothing to stop Israel from declaring virtually any Palestinian an accomplice in the violent attacks and thus subject to assassination. Given that these assassinations are planned well in advance, Israel should provide evidence of direct involvement in plotting or directing violence before overcoming the legal presumption that all residents of occupied territories are protected civilians. Moreover, because unilateral allegations are so easy to make falsely or mistakenly, and in light of their lethal consequences, these claims should be tested before an independent review mechanism.
As for the third prong, Israel has made no effort to explain why these suspected participants in violent attacks on Israelis could not be arrested and prosecuted rather than summarily killed. Significantly, assassinations are taking place not on a traditional battlefield but in a situation of occupation in which the Fourth Geneva Convention imposes essentially law enforcement responsibilities on the occupier. These responsibilities do not preclude using war methods in the heat of battle, but the assassinations typically take place when there is no battle raging. In these circumstances, Israel has the burden of explaining why law enforcement means could not be used to arrest a suspect rather than war-like tools to kill him. Theoretically Israel might claim that its forces are unable to enter an area under occupation without triggering armed conflict, but in fact the Israeli military has shown itself capable of operating throughout the West Bank and Gaza with few impediments. In these circumstances, Israel would be hard-pressed to show that a law-enforcement enforcement option is unavailable. It would thus not be justified to resort to the war rules of assassination.