An MQ-1 Predator is shown during post-flight inspection at dusk from Southern California Logistics Airport, formerly George Air Force Base, in Victorville, Calif., Jan. 7, 2012.

An MQ-1 Predator is shown during post-flight inspection at dusk from Southern California Logistics Airport, formerly George Air Force Base, in Victorville, Calif., Jan. 7, 2012. U.S. Air Force Photo by Tech. Sgt. Effrain Lopez

The United States Needs a Drone Board

If we’re going to keep killing citizens by drone, we need a better form of due process for it.

Today we learned that in January, a U.S. drone strike in Pakistan inadvertently killed an American and an Italian held as hostages by al Qaeda. The strike also killed a U.S. citizen who was a prominent member of al Qaeda. A separate operation in January killed an American-born al Qaeda spokesman. The deaths of hostages Warren Weinstein and Giovanni Lo Portois are tragic and the Obama administration has pledged to conduct an independent review to understand how to prevent this type of grievous mistake. The apparently unintentional killings of two American al Qaeda operatives raise an additional question that President Obama did not address in his televised statement today: under what circumstances may the United States intentionally use targeted lethal force against a U.S. citizen abroad?

This question is particularly salient in light of the growing risk that increasing numbers of American citizens will join ISIL’s ranks and become U.S. military targets. U.S. officials estimate that over 180 U.S. citizens have traveled or tried to travel to fight in Syria; some have died there. U.S. operations against ISIL have ramped up over recent months and Congress is now considering whether to expressly authorize the use of military force to stop ISIL’s advance.

The Obama Administration has previously considered the question; several years ago, Justice Department lawyers set out the legal rationale for targeting radical cleric Anwar al Awlaki, a dual Yemeni-U.S. citizen, in a now-public white paper and redacted memo. Satisfied that Awlaki was a permissible target, President Obama authorized a CIA-led operation that culminated in drone aircraft, armed with Hellfire missiles, striking and killing the cleric in Yemen in 2011.

So why bother rehashing the question? For one, lawyers across the ideological spectrum have challenged the Justice Department’s legal reasoning. In particular, critics argue that secret and internal executive branch review, however painstaking and careful, does not satisfy the Fifth Amendment, which provides that no person shall be deprived of life without due process of law.

Even if targeting Awlaki was lawful, President Obama himself recognized that the Administration’s targeted killing operations might warrant greater oversight and even more rigorous procedures. In a 2013 speech at the National Defense University, the president said he had asked his administration to review proposals that would establish oversight mechanisms beyond reporting to Congress. He also committed to engage with Congress to explore options for increased oversight. President Obama noted that authorizing the judiciary, through a special court, to approve lethal action would have “the benefit of bringing a third branch of government into the process, but raises serious constitutional issues about presidential and judicial authority.” He also considered the creation of an independent oversight board in the executive branch that would avoid the constitutional problems “but may introduce a layer of bureaucracy into national security decision-making, without inspiring additional public confidence in the process.”

Now is the time to address the questions President Obama has raised. We must seize the opportunity to institutionalize a more transparent and dispassionate process, defend the hallmarks of due process, and affirm that neither the executive branch nor U.S.-born terrorists are outside the law. We recommend a fine-tuned version of the second approach raised by President Obama: an independent, executive branch review panel designated to assess the evidence against proposed targets and make non-binding recommendations to the President as to whether the targeting is appropriate before efforts are made to kill the targets. One candidate for such a panel is the existing Privacy and Civil Liberties Oversight Board, an independent, bipartisan agency in the executive branch.

How it works now

We have some sense of the current decision-making process for targeted killing based on media reporting, public speeches, and available military documents. The current scheme is rigorous, but insulated from outside review and accountability. Indeed, the ACLU filed a lawsuit last month seeking Obama administration documents that establish the criteria for placement on the “kill list” for the use of lethal force. Military and executive branch departments nominate, vet, and validate candidates for the “kill list,” considering the legality of each strike and potential operational impact. The nominations trickle up to the National Counterterrorism Center, or NCTC, and to the National Security Council, or NSC, before the President signs off. The process can be contentious; government participants regularly press for more evidence and it can take multiple meetings before a strike is approved.

The Obama administration asserts that when the government targets U.S. citizens it engages in a more thorough review because, as President Obama acknowledged, targeting U.S. citizens raises “constitutional issues that are not present in other strikes.” CIA Director John Brennan, who used to coordinate the targeting decision-making as the Assistant to the President for Homeland Security and Counterterrorism, affirmed that “[w]hen that [targeted] person is a U.S. citizen, we ask ourselves additional questions.” Yet, the decision-making process appears to differ for U.S. citizens in only two ways. First, some congressional committees – but not all members of Congress – are given notice. The Administration briefed these committees before, rather than after, the strike against Awlaki. Second, the administration submitted information about the cleric to the Department of Justice months before he was killed, asking whether federal criminal law or the Constitution would preclude the strike. But these two additional layers of review may be less meaningful than they seem. We don’t know whether Congress was afforded an opportunity to respond to the notification because we don’t know when the administration briefed members of Congress, which members were briefed, and whether senior congressional staff were permitted to attend the briefing. Greater transparency regarding congressional oversight of targeting killing would shed more light as to whether it provides an effective check on this use of Presidential power.

According to the Awlaki memo, the Justice Department considers four factors when assessing whether it is constitutionally permissible to target a U.S. citizen:

  1. Is the person located outside the United States and a senior operational leader of al Qaeda or an associated force that is continually planning attacks against U.S. persons and interests?
  2. Has “an informed, high-level government official…determined that the targeted individual poses an imminent threat of violent attack against the United States”?
  3. Would “a capture option…be infeasible”? and
  4. Would “such an operation…be consistent with applicable law of war principles”?

But these four factors essentially mirror the requirements that Director Brennan laid out for all targets in a 2012 speech.

The only apparent discrepancy between the process afforded to U.S. citizen and non-citizen targets is the degree of threat: “imminent” instead of “significant.” But the Justice Department defines imminence so expansively that the meanings of “imminent” and “significant” converge. After the White House clarified in 2013 that all individuals must pose “a continuing, imminent threat” to be targeted, any daylight between the process afforded to U.S. citizens and non-citizens disappeared. Ultimately, the requirements for targeting U.S. citizens are not meaningfully different, or more stringent, than the factors that apply in the general targeting process.

Judicial Oversight

The U.S. government concedes that the Fifth Amendment, in particular the Due Process Clause, applies to U.S. citizens abroad. The Fifth Amendment establishes that “no person shall . . . be deprived of life, liberty, or property, without due process of law.” But due process does not always mean a trial in a federal court. As former Attorney General Eric Holder affirmed, “‘Due process’ and ‘judicial process’ are not one and the same, particularly when it comes to national security.” But if a person does not receive a federal court trial to determine whether the government can take away his life, how do we know whether the process he received is the process he is due?

The Supreme Court balances three factors to determine whether a particular process satisfies constitutional standards: the individual’s interest; the Government’s interest; and a weighing of the risk that the deprivation is a mistake and the expected value of adding procedural safeguards. The justices have applied this test in the national security context before. In the 2004 case Hamdi v. Rumsfeld, a plurality of the Court balanced these three factors and concluded that the existing executive branch process for justifying a person’s detention was insufficient. The Court held that a U.S. citizen detainee “seeking to challenge his classification as an enemy combatant must receive notice of the factual basis for his classification, and a fair opportunity to rebut the Government’s factual assertions before a neutral decision maker.” The plurality’s understanding of the Due Process Clause in Hamdi accords with the Court’s previous assertion in Goss v. Lopez that an individual ought to have “some kind of notice and some kind of hearing” before the government takes away his life, liberty, or property.

In his Hamdi dissent, Justice Thomas extended the plurality’s holding to its logical conclusion to suggest that even in the context of a targeted killing of a U.S. citizen, “due process would seem to require notice and opportunity to respond.” But Justice Thomas’s suggestion that it would be absurd for U.S. citizen targets to be granted notice and a hearing before a neutral decision-making body, or at least some process, may be far from ridiculous.

Critics of the Obama Administration’s targeted killing program began recommending greater oversight and judicial review of the decision-making process early in President Obama’s first term. Though Justice Thomas and others scoffed at the notion of a “drone court,” the idea gained traction in policy circles. The public debate around reforming existing procedure came to a head after the Justice Department, in February 2013, released a white paper evaluating the legality of targeting a U.S. citizen. During Brennan’s confirmation hearing to become CIA Director, Senator Dianne Feinstein declared her intent to review proposals for “legislation to ensure that drone strikes are carried out in a manner consistent with our values and the proposal to create an analogue of the Foreign Intelligence Surveillance Court to review the conduct of such strikes.” In March 2013, Jeh Johnson, formerly the General Counsel of the Defense Department and now the Secretary of Homeland Security, delivered a speech dedicated to evaluating the pros and cons of a drone court. Even the President said the value of a special court for targeting killing decisions should be considered. Some academics have also endorsed the idea of a “drone court,” in which federal judges would undertake a prior review of targeting decisions, in many cases using the Foreign Intelligence Surveillance Court as a model.

Notwithstanding the support for a drone court, prior review of the President’s targeting decisions by federal judges is not likely to succeed. Perhaps the most significant impediment is constitutional. Granting authority to the judiciary to regulate the president’s conduct of war would limit the President’s constitutionally afforded power as Commander in Chief of the military, thus raising serious separation of powers concerns. In addition, the executive branch enjoys a deliberative privilege, which means that a court could not obligate executive branch officials to disclose advice, opinions, and recommendations that are communicated during a deliberation that leads to the making of a decision within the executive branch. Critics of the drone court proposal contend that federal judges are not competent to make the real time assessments required for determining whether a target poses an imminent threat and whether capture is feasible. And they worry that judges would face irresistible pressure to sign off on strikes, even where they have doubts, thereby legitimizing targeted killing operations.

Recognizing the difficulties that prior review by federal judges poses, some scholars have called for judicial review after a targeting operation. While obviously not a deterrent to a completed targeting operation, a civil suit against the government might have a deterrent effect and offer accountability on future activity, such as erroneous targeting decisions or excessive collateral damage. But even after-the-fact review by federal judges would present practical and legal hurdles: targets (or those proximate to the target of an attack) would be unlikely to rely on U.S. courts for relief after nearly being killed by the U.S. government, especially those targets who are, in fact, enemies; non-targets likely lack the means to sue in a U.S. court; and legal rules that protect Executive Branch secrecy and officials—the state secrets doctrine, the political question doctrine, and qualified immunity—may preclude suit. More importantly, an after-the-fact hearing, whether by a court or some other body, would provide cold comfort to someone successfully but erroneously targeted.

Other proposals call for prior review by executive branch officials, or as Georgetown law professor Neal Katyal puts it, a “‘national security court’ housed within the executive branch itself.” For example, former Attorney General Alberto Gonzalez has suggested a review board modeled on the Combatant Status Review Tribunals (CSRTs) once used for detainees held at Guantanamo Bay. Gonzales’ CSRT-like proceeding would entail a tribunal of military officers whose decision would be binding on the executive and would include an advocate who could argue and call witnesses on the target’s behalf. Notably, the Supreme Court has criticized CSRTs. The Court in Boumediene v. Bush was concerned that the detainee’s representative did not serve as the detainee’s lawyer or even his advocate, the government’s evidence was accorded a presumption of validity, and the detainee had a limited ability to rebut the evidence against him because he lacked counsel – concerns that would also apply in the targeting context. Moreover, a military review board is unlikely to be a neutral body when reviewing military intelligence and decision-making. A binding decision would impermissibly constrain the President’s Commander in Chief power. A “national security court” would also be viewed as biased because the President’s own national security advisors, the same people who vet the kill list, would serve as adjudicators. Despite the drawbacks of these proposals, they may be on to something.

A New Model

As a way to provide due process to U.S. citizens, enable accountability for targeting decisions, increase public trust in the decision-making process, and avoid the use of excessive or erroneous force against targets, we propose an executive branch board (Drone Board) that would conduct a prior review of the use of targeted lethal force against U.S. citizens abroad.

It’s worth noting three points at the outset. We accept the government’s basic premise that there are circumstances in which it may use lethal force in a foreign country against a U.S. citizen who plays a significant operational role within an enemy group. U.S. citizenship does not immunize a person from the use of force abroad. We recommend that the Drone Board hear cases only when the government intends to target a U.S. person. Jurisdiction over all targeting decisions, regardless of citizenship, would unduly constrain the President’s Commander in Chief power. Such a significant constraint would not be justified because the Constitution does not protect non-citizens abroad who are not held in U.S. custody. Finally, we acknowledge that an exigent circumstances exception must allow the President to move promptly on his own when there is evidence that a target will engage in or help to orchestrate an attack on the United States in the immediate future. Hopefully a narrow definition of exigent circumstances, where danger is near certain and impending but for intervening action, would prevent this exception from swallowing the rule.

We envision a review process that would begin with notification to the Drone Board as soon as the relevant executive branch department has nominated a U.S. citizen to be the target of lethal force. Early notification and submission of relevant evidence to the Drone Board would enable it to make a timely decision if and when the President intends to authorize a strike against a U.S. citizen.

Second, in the very rare case where notice to the target is feasible, recognizing that notice unquestionably reduces the element of surprise and may make it more difficult to kill or capture the target, notice would be provided. The notice could either be given directly to the target or provided through a public announcement (which would constitute constructive notice). This is not as farfetched as it may seem. News media reported that Awlaki had been placed on the Joint Special Operations Command “kill list” nearly two years year before he was killed. Obviously, in that case, the absence of surprise did not preclude an effective operation. In fact, the notice to Awlaki gave his father time to institute legal action, albeit unsuccessfully, to enjoin the targeting of his son. Of course, the FBI has long published its top ten targets for apprehension, albeit not killing. In the rare case in which notice is feasible, the U.S. citizen target would be given the opportunity to appear before the Drone Board or designate counsel to appear on his behalf. In this context, information could be submitted by the target’s representative to challenge the accuracy of the target’s designation as an operational leader of an enemy group.

Third, the Drone Board would conduct an independent review to evaluate whether the named target is an operational leader of an enemy group and poses an “imminent” threat based on the government’s evidence and rationale for the targeting decision. Considerable deference would be given the military commander in control of the operation in determining the feasibility of capture. The Drone Board or the target’s representative could challenge the government’s evidence, and potentially submit additional evidence to rebut the government’s case. If the target is not given notice or does not designate a representative, the Drone Board should designate a staff attorney, who would be walled off from the Board’s deliberations but have full access to the case file, to argue the target’s position. The Drone Board would apply a high evidentiary standard—“near certainty”—when assessing the case against the target. This is the same standard the Administration currently applies when assessing whether civilians will be killed or injured as a result of a drone strike. The “beyond a reasonable doubt” standard used in criminal law would be impracticable here because the Drone Board would have limited time to deliberate on evidence often gathered from hostile territory half a world away.

Fourth, the Drone Board would make a non-binding recommendation to the President, based on the evidence before it, as to whether the named target has been correctly identified as operational leader of an enemy group who poses an imminent threat of violent attack against the United States. The President would retain final authority, consistent with his role as Commander in Chief of the armed forces, to authorize lethal efforts against the target. And, because the recommendation is non-binding, Board members would likely feel less pressure to approve of the strike than a judge issuing a final ruling.

Fifth, to enable greater transparency and promote public trust, we envision that the Drone Board would annually publicly release a report on its activities, disclosing the number of targeting matters considered, in how many of those a recommendation was made not to proceed with a strike, and how often the President followed the Board’s advice. The Board could also release its specific recommendations in redacted form if doing so would not pose national security risks.

Our proposed framework would afford the procedural protections to which U.S. citizens are due—notice where practicable, an opportunity to be heard, and decision-making by a neutral body—thereby bringing the targeting process more comfortably within the realm of constitutionality. In some if not most cases, this will be less process afforded than in the Hamdi detention context, which makes sense. It is more practical to provide process to a captured prisoner than to a target who is loose and potentially still trying to do harm. The Drone Board would serve as a more neutral and dispassionate arbiter because it has less of a vested interest in the outcome than current decision-makers. To enhance public confidence in the Board’s deliberations, its members would not be drawn from the military or intelligence communities. In some cases, there would be adversarial discussion before the Drone Board, which would encourage rigorous testing of arguments for and against targeting. In contrast, the existing environment is ripe for inherent biases to affect the outcome. Military officials and NCTC and NSC staff have operational responsibilities and report to the President. To our knowledge, there is no designated internal “red team” (the practice of viewing a problem from an adversary’s perspective) to challenge conclusions within the current scheme.

An executive branch review board would avoid separation of powers concerns, the deliberative privilege hurdle, and legal rules (like the state secrets doctrine) that preclude a federal court from hearing cases on targeted killing. And unlike damages suits, Drone Board review would occur before the deprivation of liberty, meaningfully addressing due process concerns.

In considering whether to create a new entity or have an existing one to carry out the duties of the Drone Board, we recommend that the Privacy and Civil Liberties Oversight Board (PCLOB) serve that role. The PCLOB is an independent agency established by the Implementing Recommendations of the 9/11 Commission Act of 2007. The bipartisan, five-member Board members serving staggered six-year terms, are appointed by the President and confirmed by the Senate. The PCLOB has several advantages over alternative executive branch entities to serve as the body that reviews the targeting of U.S. citizens. First, the PCLOB is already up and running. Neither Congress nor the President would have to create a new body. Second, the targeted killing of U.S. citizens who are affiliated with al Qaeda, ISIL, or other terrorist organizations falls precisely within the PCLOB’s counterterrorism mandate. Third, the PCLOB is uniquely positioned as an independent Executive Branch agency, not bound by the President’s policy direction, and has demonstrated its willingness to offer neutral and candid recommendations. Fourth, Board members and PCLOB staff already have Top Secret security clearances; the PCLOB regularly handles highly classified information and can review the relevant intelligence without additional bureaucratic hurdles. However, one reform would be necessary: in order to effectively address targeted killing cases, all five PCLOB members, not just the chairman, would need to serve in a full-time capacity.

Our proposal does not address every concern. The locus of decision-making would remain within the Executive Branch. Therefore, the PCLOB’s review would not balance executive authority in the same way that a judicial or congressional check would. The process cannot be wholly transparent. Much of the evidence would remain unavailable to the public in order to protect the government’s intelligence sources and methods. Notice to the target would rarely be practicable. And even an expedited review would take time, which—even in non-exigent circumstances—is a valuable resource. But the Drone Board framework need not be the only added layer of oversight. For example, this proposal does not preclude review after-the-fact by a court (if the legal hurdles could be surmounted), Congress, or another body.

The growing risk that U.S. citizens will join terrorist groups around the world and be targeted for killing overseas by their own government creates an imperative to resolve when and how the U.S. government may lawfully use lethal force against its own citizens abroad. A Drone Board would be an important step to provide additional process and greater public confidence in the method of targeting U.S. citizens overseas.

David Medine is the chairman of the Privacy and Civil Liberties Oversight Board. This article is based entirely on publicly available unclassified or declassified information. Eliza Sweren-Becker is an intern at PCLOB and a student at Harvard Law School and the Kennedy School of Government. The views expressed in this article are our own and do not represent the views of the PCLOB or any other board member. 

NEXT STORY: Make Manpower Matter