On Monday, December 13, I attended a panel discussion on “Drones and Targeted Killings Abroad: What is Legal and Who Decides?” It was hosted by the Federalist Society, at the Cornell Club in Manhattan, and the panelists were Ben Wizner of the ACLU and Michael W. Lewis of Ohio Northern University Law School. The moderator was P. Kevin Castel, a U.S. district court judge.
I was led there by my having complained to my friend Wesley Yang that there has been alarmingly little debate in America about whether killing by remote-controlled drones is ethical—whether it’s philosophically consistent with just-war doctrine. Wesley has been researching and writing about the high-stakes legal debates associated with the war on terror—see his recent profile in New York Magazine of Evan Kohlmann, the government’s most prolific terrorism expert—and he invited me—challenged me?—to come along with him to the panel.
The debate took place in a mustard yellow conference room. The Federalist Society is well-known as a conservative, strict-constructionist legal group, so it wasn’t too surprising to find a fair number of older white men in dark suits, flag pins in their lapels. While I was raiding the cheese-and-crackers table, one such attendee, observing a pile of knives and an absence of forks, volunteered that “The caterer must be a lefty.” It transpired that the quiet man sitting beside me during the discussion was Michael Mukasey, George W. Bush’s second attorney general; his identity was revealed to me at the end of the evening when people of the left and of the right converged to have their pictures taken with him.
Introduction by Judge Castel
Jane Mayer outlined the facts about America’s drone program and raised important ethical questions about it in “The Predator War,” published in The New Yorker on October 26, 2009. In his introductory remarks, Judge Castel seemed to draw on Mayer’s article. He noted that the U.S. has two drone programs. The first, run by the military, augments the work of troops on the ground. Castel suggested that the drones were controlled remotely with joysticks and monitored on flat-screen televisions, as the New York Times has also reported. The second program, Castel continued, “is said—I don’t know—to be run out of Langley, Virginia, by the C.I.A.” (Castel was careful throughout the evening to specify that the information he was presenting was secondhand, acquired by him from journalistic reports only, and that any opinions he might voice were to be understood not as representations of his personal judgment but merely as provocations, potentially fictional, intended only to stir up debate. As it happens, in a recent Guardian article about the recall from Pakistan of the CIA’s station chief, whose cover was blown last week by a Pakistani man angry over drone strikes that killed his father and brother, Declan Walsh reports that the CIA now manages its drones out of its Islamabad office rather than Langley.) Castel added that “It’s been said that some of the personnel will wear flight suits” while operating the drones. (This may not be true, however; Charlie Savage has reported for the Times that CIA drone operators do not wear military uniforms, in apparent contravention of the Geneva conventions, an awkward state of affairs that not long ago obliged the Obama administration to rewrite its military commission rules at the last minute while negotiating a plea bargain with Omar Khadr, a former child soldier and Guantanamo Bay detainee, downgrading the charge against Khadr from war crime to a domestic law offense, in order “to avoid seeming to implicitly concede that the C.I.A. is committing war crimes.”) Castel repeated the vignette that begins Mayer’s New Yorker article—the story of the August 2009 remote killing of Baitullah Mehsud, leader of the Taliban in Pakistan, while he was reclining on his roof receiving an intravenous drip. The United States and Pakistan are happy to have got rid of Mehsud, but Castel noted that the killing is thought by some to have taken place on the CIA’s sixteenth attempt to strike Taliban leader. (“During this hunt,” Mayer wrote, “between two hundred and seven and three hundred and twenty-one additional people were killed, depending on which news accounts you rely upon.”) Given such a high rate of collateral damage, can the killing be considered a success?
Castel cited data provided to Mayer by the New America Foundation, according to which Obama in his first nine months ordered as many drone strikes as Bush had during his last three years. (More up-to-date statistics are available on the New America Foundation’s website and in its February 2010 report on drone warfare; New America now claims that 45 drones strikes were ordered by the Bush administration and that the Obama administration, by contrast, ordered 51 strikes in 2009 and another 113 so far in 2010.)
Castel raised a number of questions: Do drone bombings reduce casualties of innocents by allowing operators to wait for certain identification of their target and a minimum number of bystanders? Are they a necessary means of war? Do they change the nature of war in an undesirable way, such that war no longer requires the virtues of courage and honor? (A June 2010 United Nations report warned that the U.S. drone program licensed a “‘PlayStation’ mentality to killing.”) Are any laws broken if the individual targeted by a drone is a United States citizen? If judges have to sign warrants before the government can eavesdrop on U.S. citizens, why don’t they have to vet the government’s requests to assassinate them? Are any laws being broken when the drones kill people outside of Afghanistan—that is, outside the theater of war as conventionally understood? As a closing sally, Castel raised the prospect of “nanodrones”—remote-controlled killing devices small enough to slip into a window—which reminded me of Neil Stephenson’s sci-fi novel The Diamond Age.
Castel then introduced Ben Wizner of the ACLU’s National Security Project, who helped to represent Nasser al-Aulaqi in a recent lawsuit against the Obama administration, which in April 2010 listed as an approved target for killing al-Aulaqi’s son, Anwar al-Awlaki, a cleric with dual U.S. and Yemeni citizenship alleged to have incited terrorist attacks. The al-Aulaqi/al-Awlaki case was dismissed earlier this month, when federal district court judge John W. Bates declared that the father lacked sufficient legal standing to challenge the government’s targeting of his son. Castel then proceeded to introduce Michael W. Lewis, a law professor at Ohio Northern University, who before taking up law served as a pilot for the U.S. Navy during the First Gulf War and the run-up to it.
Ben Wizner’s remarks
“In what circumstances can Barack Obama order the death of U.S. citizens?” Wizner began. He noted that the ACLU chose to assist al-Awlaki’s father in his lawsuit because the Obama administration had provided a legal opportunity by leaking to the media earlier this year its decision to target al-Awlaki for killing—a moment of transparency that the administration later backpedaled from in court, when it invoked the privilege of state secrecy in efforts to have the case dismissed. Wizner asserted that the recent dismissal of the case had been on standing rather than merits—that is, the judge ruled only that al-Awlaki’s father did not have an interest in his son’s welfare strong enough to give him the right to bring the matter to court; the judge did not say anything about the conditions that might give the U.S. government the right to kill one of its citizens. In fact, Wizner continued, the judge noted that the case raised “stark, and perplexing, questions,” such as
Can the Executive order the assassination of a U.S. citizen without first affording him any form of judicial process whatsoever, based on the mere assertion that he is a dangerous member of a terrorist organization?
Torture, Wizner said, is always illegal, but targeted killing is sometimes legal. According to Wizner, the ACLU does not contend that the government needs to obtain warrants before conducting targeted killings, nor does the ACLU believe that courts should be involved in real-time decision-making. Their contention is merely that such killings are only lawful in a theater of war and against an imminent danger. The Obama administration hasn’t disclosed its definition of lawful killings, but since it has placed individuals on its target list for months at a time, its standards must be more permissive than those proposed by the ACLU. Wizner observed that unsupervised powers are almost always abused in the long run, and that a policy of taking the government’s word for the prudence and wisdom of its actions hasn’t always worked out: the Bush administration claimed that it was only detaining dangerous terrorists at Guantanamo Bay, but the majority, it turns out, are far less scary than advertised.
Through the al-Awlaki lawsuit, the ACLU was asking the government to disclose its standards for remote killing. “Due process,” Wizner said, “requires at a minimum that citizens be put on notice as to when they could be put to death.” He noted that the Israeli government’s targeted-killing program had been challenged in court by human rights groups there, and that the Israeli Supreme Court had ruled that lethal force could be used by the government against individuals, but not for retribution, not if arrest or another lesser intervention could solve the problem, and not merely on the basis of membership in a group. The Israeli Supreme Court required that every instance of targeted killing be followed by a post-hoc investigation.
Why, Wizner asked, can’t the U.S. also make explicit its legal standards? The Obama administration has set no limits so far to its power to kill specific individuals, he maintained. The administration has not said whether the killings are permissible only when a threat is imminent, and if so, how imminent. Wizner suggested that the lack of clarity about the U.S. policy might help to legitimize targeted killings by other nations, whose ethical standards and target lists might not be palatable to Americans. In March 2010, Harold Koh, legal adviser to the State Department, defended the Obama administration’s targeted-killing program as justifiable as a matter of national self-defense; Koh claimed that the killings by drone were being conducted in accordance with the just-war principles of distinction (that is, they target only military personnel and equipment, not civilian ones) and proportionality (that is, they don’t kill more civilians incidentally than the military target is worth). Wizner granted that Koh’s argument might hold in its broad outlines, but Wizner insisted that Koh had not revealed enough details to enable anyone outside the Obama administration’s inner circle to judge whether the policy was in fact justifiable.
What limits, if any, can be put to the use of military force? Wizner asked. Can a U.S. citizen be killed in Yemen, with which the United States is not at war, as well as in Afghanistan, where the United States has acknowledged it is waging war? Is an organization like Al Qaeda in the Arabic Peninsula (AQAP) covered by the Authorization for Use of Military Force Against Terrorists (AUMF), the declaration of war passed by Congress a week after 11 September 2001, even though AQAP did not exist at the time and had nothing to do with the September 11 attacks? In closing, Wizner quoted the Israeli Supreme Court: “It is when the cannons roar that we especially need the laws” (a flip of the more familiar, much-abused Ciceronian tag).
Michael W. Lewis’s remarks
Lewis began by differentiating what he called the “law enforcement standard” from the “law of armed conflict standard.” In law enforcement, you’re only allowed to kill someone who poses an imminent threat. In war, however, you’re allowed to kill your enemies whether or not they pose an immediate threat. In the al-Awlaki lawsuit, Lewis averred, the ACLU was attempting to apply a law-enforcement standard to an armed-conflict situation. Furthermore, Lewis continued, the laws of war don’t recognize geopolitical borders as limits to military endeavor, and the ACLU’s advocated position would create sanctuaries behind international borders where none had originally existed.
The laws of war, Lewis summarized, distinguish civilians and combatants. Combatants are only recognized as such if they belong to an organization that enforces the laws of war; in the American army, for example, a soldier who gratuitously kills an Iraqi civilian may be tried by an American court martial and imprisoned. Balancing this liability is something called the combatant’s privilege: so long as combatants do not violate the laws of war, they are immune from prosecution for arson, murder, or damage to private property.
Anyone not a combatant is a civilian, and according to the Geneva Conventions, civilians may never be targeted in a military operation. But civilians also may never take part in armed conflict; if they do, they become targetable. By directing the actions of Nidal Malik Hasan (the Fort Hood shooter) and Umar Farouk Abdulmutallab (the attempted underwear bomber of Christmas 2009), al-Awlaki crossed the line, Lewis said, and rendered himself a legitimate target of lethal force.
What’s more, Lewis continued, crossing borders in pursuit of an enemy has traditionally been an accepted practice during wartime. In pursuit of FARC, the Colombian military has crossed into Ecuador; in pursuit of Hezbollah, the Israeli military has crossed into Lebanon. Neither incursion is considered a violation of the laws of war. (Asked during the question-and-answer period later about the controversy over American forces moving into Cambodia during the Vietnam War, Lewis argued that the scandal was not on account of the border crossing, but because the facts of the war were being kept secret from the American public.)
If a neutral country finds itself the host of a person engaged in hostilities with the United States, Lewis maintained, the country need not grant the United States permission to enter its territories in pursuit of him. But if the country wishes to maintain its neutrality, it must deny that person harbor. Yemen, where al-Awlaki is thought to reside, has in fact chosen to act as an ally, Lewis noted—referring, with a smile, to a State Department cable recently released by Wikileaks, which reveals that Yemen’s president promised the head of U.S. Central Command in January 2010 that he would keep his people in the dark about the American bombing missions against AQAP. (“We’ll continue saying the bombs are ours, not yours,” said Yemen’s president.)
As a matter of strategy, Lewis argued, granting terrorists political sanctuary was tantamount to giving them the initiative—allowing them to choose the next battlefield. It was, he insisted, to turn international law upside-down. (Lewis has further elaborated his support of the Obama administration’s targeted-killing policy in a paper posted on the Federalist Society’s website.)
In the questions and answers that followed, the discussants made further interesting points, but this post is already rather long, and perhaps I should limit myself to relating a few of them in the course of sorting through some of my own thoughts.
When in the past I have read about the Geneva Convention’s definition of combatants, the emphasis has always seemed to be on the requirement that combatants wear an identifiable uniform—a requirement that has always seemed a little arbitrary to me, especially when used to justify the indefinite detention of people who were volunteer fighters from impoverished regions, where the purchase of a uniform may have been a relative luxury. I found more ethically convincing Lewis’s emphasis on the requirement that combatants belong to an organization that enforces distinction and proportionality in the waging of war. That said, unless I’m missing something, I’m not sure the difference between a lawful and an unlawful combatant is pertinent here. Lewis is arguing for the government’s right to kill combatants of either kind, and the legal point at issue seems to be whether the killings may take place in territory outside the theater of war. It doesn’t seem relevant that someone like al-Awlaki, by failing to belong to an armed force that obeys the laws of war, may have forfeited his right to the protections afforded by the Geneva Convention to prisoners of war.
Toward the end of the debate, Lewis said, almost as an aside, “I don’t see any difference, by the way, between a drone and a manned aircraft.” To my surprise, Wizner nodded. (For the record, I don’t know for certain whether he nodded in agreement or merely to express something like, ‘I recognize that argument,’ but it seemed to me that Wizner was signaling agreement.) I see now that Koh made a similar claim in his March 2010 remarks:
There is no prohibition under the laws of war on the use of technologically advanced weapons systems in armed conflict—such as pilotless aircraft or so-called smart bombs—so long as they are employed in conformity with applicable laws of war.
I see the force of the claim. It no doubt seemed unchivalric when swords were first superseded by gunpowder, but the point of war is to defeat the enemy while suffering as little injury oneself as possible. So the problem with drone killing isn’t the cocoon of safety around the person holding the joystick. That doesn’t mean there isn’t a problem with drone killing, but it does mean that some of my horror over the technology may be a little sentimental.
I was also surprised, during the question-and-answer period, to hear Lewis concede that he was comfortable with the proposal of “some form of post-action review” of targeted killings. It surprised me even further when he said that “the guys in Langley bother me, because they’re in Langley. Whether there’s law-of-war training in the CIA, I don’t know.” In other words, Lewis was worrying whether the CIA’s operators might themselves be unlawful combatants—a question that concerns me rather more than the question of whether al-Awlaki is one. When an American soldier goes rogue, he is court-martialed. What happens to a rogue CIA officer? No doubt it’s something unpleasant, or so one hopes, but the secrecy of the organization may make it all but impossible to verify whether CIA killings are in compliance with the laws of war. Given the CIA’s propensity in recent years for kidnapping and torture—of the innocent as well as the probably guilty—one isn’t disposed to take their good behavior on trust.
Lewis welcomed post-action review and was concerned by the secrecy shrouding the CIA’s involvement; Wizner did not object to targeted killings without prior judicial review. If the two of them were representative of informed political opinion, I found myself wondering, might it be possible for Congress to agree on some legislation? Surely the soldiers operating these devices would prefer to know for certain when they were and when they weren’t committing war crimes.
On further consideration, this fantasy began to seem a little overoptimistic on my part. The trouble is the chasm between the law-enforcement model and the armed-conflict model. Should terrorists be treated as criminals or as enemy soldiers? The great difference between these paradigms may be related to my niggling sense that there is something wrong with drone killing, even if it’s not the remoteness of the attack. Maybe killer drones upset the balance of power between individuals and governments by making it easier to kill a specific person—by making war microlocal. It is accepted that in war a certain number of innocent bystanders are killed. In law enforcement, however, any death of an innocent is an outrage. Why do we allow soldiers to kill the “wrong” people in relatively high numbers but not police officers? Perhaps it’s because we distinguish between peace and war, and we understand war to be a temporary state marked by conflict and chaos, in which a government uses force to accomplish broad aims such as taking control of a region. Under such conditions, we are willing to accept that the use of force may be imprecise. We understand that being at war is different from being at peace. Lewis’s concern about crossing borders is, I think, a red herring. If FARC’s soldiers move into Ecuador, then I think we understand that in their retreat, FARC’s soldiers trail the boundaries of the theater of war after them. The trouble is, what happens if a few of FARC’s leaders abruptly surface in Madrid? If, in such a case, the Colombian military were to start blowing up the FARC leaders’ apartments, it would be reasonable of Madrid’s citizens to object.
One of the benefits that people expect from government is personal security from attacks by government itself. (And protection from other governments; Madrileños expect that Spain will not attack them and that Spain will protect them from Colombia if Colombia tries to—and from FARC, for that matter.) The social contract is understood, at least in some strains of political philosophy, to be a haven from the state of war. If an individual may be killed by his government on its say-so, without judicial review before or after his killing, there is no haven from the state of war. In fact, security from such attack is so fundamental to well-being that a person vulnerable to it might be thought of as lacking a government at all. It is no remedy if a government merely promises that it will only execute dangerous enemies to social order. Citizens need to be able to feel confident about the government’s epistemology; they need to be able to challenge the government’s identification of enemies before the fact, or (less reassuringly) they need to be able to punish after the fact government officials who turn out to have been in error. A case as celebrated as al-Awlaki’s is not where the trouble is likely to lie; given the stir about him, the governments of the United States and Yemen have probably been obliged to think long and hard about the certitude of the evidence against him. (Still, to say a word for the criminal model: if Yemen’s president is willing to wink while the U.S. bombs terrorists within Yemen’s borders, can’t he bring himself to extradite a single terrorist for us? Does his tenure in power really depend to that extent on duplicity?) The possibility for injustice, however, will increase the further down one goes on the government’s target list. What if some individuals lower down are listed because the CIA has misidentified them, as it misidentified Khalid El-Masri, an innocent German citizen whom the CIA kidnapped and tortured in 2003 and 2004 because they mistook him for a terrorist with a similar name? What if the culpability of some individuals lower down on the list has been grossly exaggerated, as was the case with the Uighurs in Guantanamo Bay? Without the protections afforded to suspects under the criminal law model, such errors would never be redressed, and might not ever even be discovered.
I’m offering these thoughts with the caveat that I’m still thinking through these issues myself. I don’t feel that I’ve gotten to the bottom of them.