Congressional oversight serves a critical role in the American system of government; in the foreign policy arena in particular, the Founding Fathers established a system of checks and balances whereby Congress and the executive branch would perpetually “struggle for the privilege of directing American foreign policy.”1 With regard to activities of the armed forces, while the executive branch claims broad powers over war-making – principally through its reliance on the Commander in Chief clause of Article II of the Constitution – Congress too has compelling constitutional obligations as to the declaration and conduct of war, including its near plenary authority over appropriations of funds for use by the armed forces.2 What results is an area of overlapping and competing authorities, which makes the role of Congress critically important to any programmatic use of force by the U.S. government, no matter how sensitive – including the U.S. drone program. Some commentators go a step farther and ask whether Congress can do more given the “pitifully little oversight” of the drone program by the Executive branch,” and have therefore asked whether Congress can fill what they perceive as the void in oversight over the program. “Congress Perhaps,” Judge Janice Rogers Brown of the D.C. Circuit asks.3
Seeking to answer Judge Brown’s inquiry, this paper explores the role that Congress plays overseeing the drone program and the current limitations on its ability to conduct effective oversight. It also offers several proposals for increasing Congress’s oversight responsibilities in ways that would seek to minimize their operational impact.
Elements of Meaningful Oversight
Meaningful oversight enables the Congress to fulfill its objective of “hold[ing] executive officials accountable for the implementation of delegated authority.”4 It requires that the overseer has access to sufficient information about the activity being overseen and the means to influence the activity going forward. To deny the overseeing authority adequate access to information about the activity deprives it of the ability to evaluate the merits, implementation, and risks and rewards of the activity. And oversight where the overseer lacks the ability to bring about change to the activity reduces the overseer to a passive observer that lacks any direct means of influencing the activity.5
It is only when the American public is adequately informed about the activities of its government that it can fulfill its most vital responsibility – that of self-governance.
And there is another foundational point to congressional oversight in the American system – informing the public about the actions taken by its government and assisting the public in engaging in a national dialogue about government policy and practices.6 It is only when the American public is adequately informed about the activities of its government that it can fulfill its most vital responsibility – that of self-governance. But how can the public knowledgeably exercise that right when it lacks adequate information about government programs and activities? It is in this respect that Congress helps foster democratic accountability by ensuring that the American public is armed with information, the lifeblood of self-determination.
And Congress’s role is of paramount importance on matters of national security activities, which in large measure are classified and thus shielded from public scrutiny because of the need to protect sensitive sources and methods. In the national security context, therefore, Congress must play a dual role. It must on the one hand press the executive branch to provide as much information to the American public as it can and help the American public interpret the information that it is provided, and with regard to programs and activities that cannot be shared publicly, it must serve as a proxy – evaluating and making decisions about whether these activities are consistent with what the American people would want their government doing.7
Congressional Oversight of the U.S. Drone Program
As a general matter, the executive branch provides information about the drone program to the oversight committees, which for purposes of this paper refers to those committees that currently oversee implementation of the drone program both for intelligence, surveillance, and reconnaissance (ISR) missions and to carry out lethal direct action, including the armed services committees, intelligence committees, foreign affairs committees, judiciary committees, and appropriations committees.
In many ways the departments and agencies that have a role in the procurement and operation of drone platforms (both for ISR and strike missions) are eager to share information with their oversight committees to garner and ensure sustained support for the continued operation – and expansion – of drone use. The government provides briefings on drone strikes through “storyboards” (which provide narrative summaries of particular strikes, including the target of the strike, how it was carried out, the platform and munitions used, and the outcome), closed-door testimony, and even occasional congressional delegation visits for members and their staff to the “ops floors,” where drone operators pilot their aircraft. Because drone operations are in many ways inherently more “sexy” than many other programs undertaken by the U.S. government, the oversight committees and their staff are routinely captivated by these presentations and are always eager to learn more.
In some respects, however, current congressional oversight of the U.S. drone program suffers several defects, both in terms of the scope and breadth of information that the executive branch provides to Congress and the tools available to Congress to influence the program going forward.
Limitations on Congress’s Access to Information Regarding the Drone Program
Congressional access to relevant information about the drone program is in many respects limited. For one, the focus of congressional interest in the drone program centers on the cool stuff – who is being targeted, their importance as a counterterrorism target, and the drone platforms and weapons that are used. There tends to be less time spent on policy questions regarding the drone program’s role in the broader U.S. counterterrorism strategy. This focus on the tactical over the strategic creates blind spots in congressional oversight that render congressional review and scrutiny of the program more focused on individual strikes and their outcomes and less focused on broader questions of the overall effectiveness of the program. Congress tend to inquire as to why a particular strike was taken, why it took place in a certain location or at a certain time, why a particular drone platform or particular weapons were used, or how the removal of certain targets will impact the operations of the terrorist organization with which they affiliated. Congress less frequently asks about the effectiveness of the program – apart from killing individual terrorist targets – as a counterterrorism tool. For instance, to what degree do drone operations – both loitering for long periods of time over communities and the drone strikes themselves – create hostility against U.S. interests and serve as a recruitment tool for new recruits into terrorist organizations, and does the expanded use of drones create tension with the nation’s foreign partners?
But perhaps most critically with regard to information access, the oversight committees lack a comprehensive picture of the drone program. This occurs for several reasons. Primarily, the oversight committees often compete with one another for access to information and actively lobby the executive branch to deny the other oversight committees access to information that they deem within their sole jurisdiction. In addition, agencies are always seeking to appease their own oversight committee with information to garner support for their continued role in the program, and to continue to support generally the relationship between agency and overseer, which extends to all aspects of the agency’s activities and which, therefore, constitutes a matter of strategic (and in some cases existential) importance to the agency. Yet this eagerness to please comes at a heavy cost. Agencies will hand-feed the congressional committees that they care about, but they will do anything in their power to avoid cooperating with the “other” committees. By way of example, while the Department of Defense may go to extraordinary lengths to ensure that the armed services committee receives all the information it seeks, the DOD may completely ignore or even refuse information requests by the intelligence committees, due to the potential of angering DOD’s own overseers and its view that other committees do not have the strategic importance as its principal oversight body in Congress.
In addition to turf battles between oversight committees and the willingness of the agencies to indulge these battles, deficits in congressional access to information about the drone program can also occur because of resistance from within the executive branch about sharing certain information about the drone program and its execution. The reasons that the executive branch withholds information from Congress about the drone program include, but are not limited to: (1) concerns that the provision of certain operational details to Congress will have a chilling effect on the drone program itself, insofar as those who carry out strikes will alter their conduct given the foreknowledge that their actions will be subjected to congressional scrutiny; (2) arguments that the oversight committees do not need operational details for purposes of carrying out effective oversight; (3) fears about the added risks of unauthorized disclosures of classified information (both given the general principle that the likelihood of unauthorized disclosures increases as more people are given access to the information, and the view within the executive branch that unauthorized disclosures emanate with greater frequency from Congress); (4) interest in protecting certain information about drone strikes that, in the executive branch’s view, is protected by the deliberative process privilege; (5) concern that providing Congress with too many details about the drone program would risk unconstitutional interference by the Congress on what the executive branch views as a core exercise of the president’s powers under Article II of the U.S. Constitution – chiefly, his commander-in-chief powers; and (6) cultural and bureaucratic impulses that disfavor sharing with Congress any more information about its activities than is absolutely necessary.
In addition to turf battles between oversight committees and the willingness of the agencies to indulge these battles, deficits in congressional access to information about the drone program can also occur.
One well-known example that touched on several of these concerns was the repeated requests by Congress for access to the legal basis by which the U.S. government carried out a drone strike to kill U.S. citizen Anwar al-Awlaki in 2011 in Yemen. While the executive branch initially refused to share its legal justification for targeting al-Awlaki, claiming these legal opinions constitute privileged attorney-client advice provided by lawyers from the Department of Justice and to the drone operators, the White House ultimately directed the Department of Justice to provide Congress with access to its legal memoranda in the face of concerns that otherwise Congress would delay the confirmation John Brennan as director of the Central Intelligence Agency.8
As an additional example, aspects of the drone program are classified to various degrees: Some information is considered unclassified or has been declassified by the government in recent years, other information is classified Secret or Top Secret, and still other information is “compartmented,” meaning that a limited and deliberately controlled number of people within the United States government have access to that information. As it does with all access to highly sensitive information, the government will restrict the number of congressional staffers who are provided with the necessary clearances to gain access to this compartmented information. Some such information, for example, is compartmented pursuant to the classification framework managed by the director of national intelligence, thereby in effect limiting access to that information to members and staff on the intelligence committees. Still other information is classified under a separate classification regime managed by the secretary of defense, thereby in effect limiting access to that information to members and staff of the armed services committees, and, in the case of the U.S. House of Representatives, the intelligence committee.9
No single member of Congress, congressional staffer, or oversight committee has visibility over all drone platforms, all strikes, taken in all theaters both within and outside of areas of active hostilities.
The end result of these various forces is that numerous members of Congress, countless congressional staffers, and multiple committees all have access to some information about the drone program. But with perhaps the exception of congressional leadership and their national security advisors, who all lack the bandwidth to focus on any one government program, no single member of Congress, congressional staffer, or oversight committee has visibility over all drone platforms, all strikes, taken in all theaters both within and outside of areas of active hostilities. The appropriations committees, for example, will have access to information relating to the procurement of drone platforms, including their technical specifications, capabilities and limits, and cost. But they will not be privy to information about particular ISR or strike missions when these drone platforms are put into service. And the armed services committees will receive information about ISR and strike missions carried out by the U.S. military, but they are not routinely provided with information on missions carried out by the U.S. intelligence community (e.g., ISR-only missions). This means that there is no single person or entity within the Congress that has the ability to evaluate and contrast best practices used across the U.S. government to carry out drone strikes, to review the various legal frameworks under which those strikes are taken, and to evaluate the cost, precision, risks, and myriad other considerations necessary to conduct a comprehensive review of the U.S. drone program and develop an understanding for what is working and what, in Congress’s view, needs to change.
Limitations on Congress’s Ability to Influence the Program
Apart from limits on the oversight committees’ access to comprehensive information about the drone program, critical means by which Congress typically exerts influence over executive branch activities are lacking or in some instances nonexistent. One common means by which Congress seeks to influence government programs is through appeals to the American public, where it can endeavor to change public opinion about the activity or, more coarsely, exert political costs on the executive branch for continuing to engage in activities that members of Congress find objectionable. But where the U.S. drone program is concerned, Congress’s ability to use politics is similarly diminished.
Because significant aspects of the drone program are classified, members of Congress are limited in their ability to facilitate a public dialogue in which they can educate the American public about the program, share their concerns, and seek to galvanize public opinion. As a result, public discussion often begins and ends with generalized assertions of the need to continue to take the fight to terrorists worldwide, the effectiveness of drones in targeting terrorists overseas, and assertions about the lack of direct risk to U.S. personnel in the use of drones.10 By way of illustration, findings from a 2016 report from the Center for a New American Security suggest that “the American public is generally more likely to support unmanned than manned” uses of force even though the report found critical gaps in the American public’s understanding of the program, including all of its risks and its key distinguishing characteristics from conventional manned aircraft.11
This lack of meaningful public debate and understanding regarding the drone program also creates practical limits on Congress’s ability to exercise control over the program through its power of the purse, because doing so could prove politically perilous for members of Congress, again especially in light of their reduced ability to fully describe to the American public the basis for any such funding restrictions.
An absence of informed public insight into the drone program – and Congress’s oversight of it – creates another force that can degrade the quality of congressional oversight over time. With regard to other government programs, the degree to which Congress oversees any program is itself the subject of robust public debate. As a result, public and political pressures exist that help compel Congress to satisfy its oversight responsibilities, or at least to explain to the public why it is not doing so. But concerning the drone program, limited public and political pressures exist. While private advocacy organizations such as Amnesty International and Human Rights Watch seek to raise awareness about the drone program and voice their concerns, they have thus far had extremely modest success in convincing the general American public to adopt their views and questions and concerns.
An absence of informed public insight into the drone program – and Congress’s oversight of it – creates another force that can degrade the quality of congressional oversight over time.
In addition, discussion and briefings between the executive branch and Congress regarding the drone program occur behind closed doors and far away from any public scrutiny. This significantly increases the risk of “agency capture,” the phenomenon whereby the overseer, over time, becomes less independent and more beholden to the agency that it is overseeing. The referees, in essence, end up joining the fan club.
Lawyers and national security policymakers in the executive branch often push back against attempts by Congress to increase its level of oversight of the drone program; to these officials, the drone program should operate free of some of the traditional checks that exist against executive power. The fragility of intelligence used to identify targets, the swift operational tempo required to rapidly act on that intelligence to carry out strikes, and the delicate diplomacy required to secure overflight and basing privileges create challenges for traditional, ongoing, and real-time oversight. Moreover, to the extent that the executive branch has come to rely heavily on the use of drones in its counterterrorism mission, traditional oversight by Congress over the use of drones would be viewed by the executive branch as an unwelcome interference in the president’s exercise of his commander-in-chief powers. In other words, while the operating agencies are more than happy to “ooh and aah” their oversight committees, they will sour quickly to any attempts by those committees to “interfere” in the program’s execution.
Traditional oversight by Congress over the use of drones would be viewed by the executive branch as an unwelcome interference in the president’s exercise of his commander-in-chief powers.
But while these arguments might counsel caution in how the Congress is to oversee the drone program, they do not in any way address whether Congress should oversee the program. Drones have become the government’s weapon of choice against terrorist targets overseas. There were roughly 10 times more drone strikes carried out under the Obama administration than the George W. Bush administration,12 and all indications are that President Donald Trump intends to continue to rely heavily on their use in the ongoing fight against al Qaeda, the Islamic State of Iraq and Syria (ISIS), and their many affiliates worldwide.
Given the starring role that drones play in the U.S. government’s ongoing campaign to hunt down and kill suspected terrorists, the high absolute financial cost of the use of drones, the risks that they pose to noncombatants, and the foreign policy implications of their use, there is no question that Congress must provide meaningful and robust oversight of the program. The remainder of this paper offers several suggestions for restructuring or reinvigorating congressional oversight to ensure adequate democratic accountability over the program.
Potential Avenues for Enhancing Congressional Oversight of the U.S. Drone Program
Consolidation of Oversight
A number of congressional oversight committees assert a role in overseeing the drone program, including the intelligence committees, the armed services committees, the judiciary committees, the appropriations committees, and the foreign affairs committees. Shared oversight, in some respects, makes sense given that drones are deployed for both intelligence collection and strike missions. However, their use raises difficult and sensitive legal and sometimes constitutional questions, the procurement of drone platforms requires appropriations from Congress, and their operation implicates foreign policy, including to the extent that their use requires local basing and overflight rights. Nevertheless, as noted previously, shared oversight has become synonymous with fragmented oversight, which creates significant constraints on Congress’s ability to meaningfully oversee the program from a holistic perspective.
The arguments for expanded access by each of these committees are strong. How can the intelligence committees conduct effective oversight over the ISR capabilities and intelligence gathered from drones without understanding how that intelligence will ultimately be used? How can the armed services committees conduct effective oversight of the platforms themselves, and the strikes that are taken from them, without a strong understanding of the intelligence that is developed upstream, by these same platforms? And in assessing the foreign policy considerations relevant to the use of drones, don’t the foreign affairs committees need access to information about the drones themselves and their capabilities and level of precision?
And yet the two most obvious fixes – providing expanded access to all of the oversight committees or consolidating oversight of the drone program with a single committee – are unworkable at best.
Expanding access to all of the oversight committees would significantly complicate the executive branch’s efforts to keep the oversight committees fully informed, would risk creating confusion as to which committee within Congress the executive branch should take direction from, and could spark intra-committee disputes when differences of opinion arise. Moreover, enabling full oversight by all oversight committees would greatly expand the number of people who have access to information about the program, including some extremely sensitive and highly classified information, which would significantly increase the risk of unauthorized disclosures.
The two most obvious fixes – providing expanded access to all of the oversight committees or consolidating oversight of the drone program with a single committee – are unworkable at best.
Consolidating oversight with a single oversight committee is similarly unworkable. For one, those committees that would be stripped of their current oversight (and access to information about the program) would not go quietly. In many ways, having access to information about the drone program is a badge of honor in national security circles, given the sensitivity of much of the information about the program. Moreover, each agency within the executive branch involved in drone operations would jealously guard the continued role for its oversight committee, given the importance of the relationship between agencies and their overseers.
One potential middle ground would be for Congress to create a special committee with the sole responsibility of overseeing the drone program. Delegate members from each of the oversight committees could serve on the special committee, with the assistance of dedicated congressional staffers. Such a special committee would enable oversight of the drone program via a cadre of congressional staffers with full access to information about the program and the ability to evaluate its operation across platforms, operational agencies, theaters of operation, and missions. And with delegates from each of the oversight committees, the Drone Committee could serve as a conduit to the oversight committees with regard to those aspects of the drone program that fall within their traditional areas of responsibility, allowing them to preserve their role in overseeing the program, but through a single Drone Committee that could look across the entirety of the program.
Congressional Sponsored Advisory Board
One variant to the idea of creating a special committee is for Congress, or one or more of the oversight committees, to commission an independent advisory board to create a forum in which disparate voices with differing perspectives on the drone program can exchange ideas and develop well-informed and thoughtfully considered recommendations, with guidance and advice, for executive branch policymakers. The idea for an outside advisory board was offered by Luke Hartig of New America, who suggested that the board comprise administration officials, counterparts from allied nations, leading outside legal and policy experts, human rights groups, and representatives from Congress.13
But while Hartig did not articulate a leadership or convening role for Congress, but merely for it to constitute another of the advisory board’s members, there might be significant advantages to such a board operating under congressional direction. For one, were one or more oversight committees to establish such an advisory board, it would be more likely to be taken seriously by the executive branch than if it was a project spearheaded by, say, a national security think tank. This, in turn, would help it in gaining access to information about the program. Moreover, congressional sanction of such an advisory board would likely mean that the recommendations that flow from it would be directed, principally, to the congressional committees that established it, thus giving the advisory board a vested audience of decisionmakers inside Washington, with an ability to put pressure on the executive branch to implement the recommendations of the board or at least insist on the executive branch providing compelling reasons for not doing so. Again, were the board instead constituted by a think tank or other nongovernmental organization, it would lack the cache and attention of senior policymakers from inside both the executive branch and Congress.
If Congress were to establish an advisory board, it should establish protections to ensure the independence of the board’s review, deliberations, and policy recommendations from undue congressional influence.
Of course, there are also risks to too significant a role for Congress in such a board, particularly in establishing it. The current level of dysfunction in Congress could result in the politicization of the advisory board, which could pose risks to its agenda if, for example, those in Congress who established the board determined that it was in their political interest to slow or even halt its agenda, or to alter its mission or focus in an effort to influence or undermine its policy recommendations.
For these reasons, if Congress were to establish an advisory board, it should establish protections to ensure the independence of the board’s review, deliberations, and policy recommendations from undue congressional influence.
One potential model that Congress could follow would be the legislation that established the Privacy and Civil Liberties Oversight Board (PCLOB). Created by the Implementing Recommendations of the 9/11 Commission Act of 2007, the PCLOB provides oversight and advice to the executive branch (with semiannual reports to Congress) on ensuring that privacy and civil liberties are appropriately considered in the development and implementation of counterterrorism programs. Congress could create a similar legislative creature, either as an independent agency (as with the PCLOB) or as an advisory body of the legislative branch, to focus in particular on oversight and policy recommendations regarding the drone program. And in creating this board, Congress can establish processes to ensure that the board constitutes a cadre of representatives – perhaps seconded to the board for a period of time – from the administration, allied nations, outside legal and policy experts, human rights groups, and Congress itself. At the same time, however, Congress should carefully proscribe the board’s mandate – namely, to review the implementation of the drone program and provide policy recommendations regarding it. Absent clear lanes within which the board is instructed to operate, the board could seek to take on issues and questions beyond its intended focus, which could undermine the level of respect and cooperation that it gets from government as it sets about to provide recommendations.14
Expanded Role for Congress in Assessing Noncombatant Deaths
One common concern about the drone program is the risk posed to noncombatants. Yet a wide disparity exists between U.S. government and nongovernmental assessments of noncombatant deaths. For example, whereas the U.S. government in July 2016 assessed that, for the period January 20, 2009 through December 31, 2015, there were between 64 and 116 noncombatant deaths resulting from 473 drone strikes, nongovernmental organizations(NGOs) estimate 200 to more than 900 possible noncombatant deaths.15
In seeking to explain the basis for this disparity, the U.S. government has identified several factors that, in its view, lead to less accurate and higher estimates of noncombatant deaths by NGOs. First, the government “draws on all available information (including sensitive intelligence) to determine whether an individual is part of a belligerent party fighting against the United States in an armed conflict; taking a direct part in hostilities against the United States; or otherwise targetable in the exercise of national self-defense.” Thus, the government claims, it may have information indicating that certain individuals are combatants, whereas NGOs, which lack that additional information, count them as noncombatants. Second, unlike investigations undertaken by NGOs, the government has noted that its post-strike reviews involve “the collection and analysis of multiple sources of intelligence before, during, and after a strike, including video observations, human sources and assets, signals intelligence, geospatial intelligence, accounts from local officials on the ground, and open source reporting.” Finally, the government argues that assessments by nongovernmental organizations may be artificially elevated through the use of misinformation, which is deliberately spread by some actors, including terrorist organizations.
While these factors certainly explain inaccuracies that may pervade nongovernmental assessments of noncombatant deaths, such assessments by NGOs can and do yield valuable information about drone strikes. For one, unlike U.S. government assessments, the assessments prepared by NGOs are in part informed by witness interviews.
Moreover, the government’s criticisms of NGO assessments do not address the long-standing criticism of government assessments as being self-interested and thus biased. U.S. government after-action reviews and assessments of noncombatant deaths are made by analysts in the same organizations that carry out the strikes themselves. And while these analysts are consummate professionals, and notwithstanding procedures in place to ensure that the analysts who assess noncombatant deaths are walled off from those who carry out the strikes, the simple fact that they work under the same roof creates, at least, a significant problem of perception – that the U.S. government is grading its own work. Bureaucratic pressures and human nature can help subtly pressure analysts to resolve uncertainties about noncombatant deaths in favor of their colleagues, which can lead to an underestimation of the collateral damage, including harm to civilians, caused by drone strikes.
The government’s criticisms of NGO assessments do not address the long-standing criticism of government assessments as being self-interested and thus biased.
To address this concern, some NGOs have called for the designation of an outside body to conduct the after-action reviews and assess civilian casualties resulting from U.S. drone strikes. Such organizations could include a separate or new department within the executive branch, an outside entity constituted specifically for this purpose, or Congress.
As part of their ongoing oversight of the U.S. drone program, the congressional oversight committees are provided with after-action reports regarding each strike taken. These reports detail the strike itself, including the target of the strike, the location in which it was taken, the circumstances surrounding it, and whether the executive branch believes that any civilians were injured or killed in the strike. Moreover, in instances in which the government believes that civilians may have been injured or killed and therefore undertakes a more thorough after-action investigation, it will supply the results of that investigation to Congress.
Yet in this role, Congress is primarily a passive recipient of conclusions made by the executive branch as to the outcomes of its drone strikes. And of course the information that the Congress receives is that which the executive branch chooses to share, and in the form that it chooses to share it (not unlike an audit committee that receives from a corporate board of directors a final and filtered presentation regarding an issue within the organization). It is extremely difficult for Congress to independently evaluate the facts and circumstances surrounding the strike and render its own judgment about any civilian harm that may have resulted. Moreover, there does not currently exist a formal process for Congress to also review assessments of civilian harm made by NGOs, which can include eyewitness interviews and other information not evaluated by the government as part of its review.
Consequently, Congress’s current process for assessing civilian casualties from drone strikes is necessarily skewed to the government's conclusions; Congress does not fully interrogate the government’s assessments with potentially contradicting information provided by NGOs.
Congress’s current process for assessing civilian casualties from drone strikes is necessarily skewed to the government's conclusions.
To remedy this defect in process, Congress could create a formal process for reviewing assessments of drone strikes from both the government and NGOs, perhaps in tandem with consolidating oversight and creating an advisory board, as discussed previously. Doing so could arm the Congress with information that contradicts the government’s assessments and better enable Congress to question the government’s information and its conclusions. Ultimately, were Congress to create such a process, it would be able to render its own conclusions as to the results of strikes and whether any civilian casualties resulted. Congress could then provide its independent conclusions to the executive branch, which could help inform the government’s conclusions, and it would provide NGOs with a formal means for providing useful input. More fundamentally, such a role for Congress would allow it to serve as a more independent and credible voice to the American public and international community about the efficacy and risks to civilians posed by the U.S. drone program, which could ultimately lend support to continued use of the drone program as part of the U.S. government’s counterterrorism arsenal, given its fairly impressive record of low civilian harm in relation to its high success in meeting its counterterrorism objectives.16
Refreshed Congressional Authorization for Use of Military Force
The most direct way in which Congress can exercise its oversight responsibilities is through its legislative function – by authorizing, or not, activities of the executive branch. Shortly after September 11, 2001, Congress passed into law the Authorization for Use of Military Force (AUMF), to authorize the president to use “all necessary and appropriate force” against “those nations, organizations, or persons he determines planned, authorized, committed, or aided the terrorist attacks that occurred on September 11, 2001, or harbored such organizations or persons,” which was originally understood to constitute al Qaeda (for planning and committing the 9/11 attacks) and the Taliban (for harboring al Qaeda).17 The executive branch continues to rely on the 2001 AUMF as congressional authority to carry out military operations in Afghanistan, Iraq, and Syria, as well as to serve as the legal basis for drone strikes both inside of those theaters and against terrorist targets elsewhere in the world.
During the second term of his administration, President Obama urged Congress to pass a new authorization for the use of force. He did so based on his belief in the importance of congressional oversight of, and acquiescence to, the use of force. In his view, “without Congress showing much interest in restraining actions with authorizations that were written really broadly, you end up with a president who can carry on perpetual wars all over the world, and a lot of them covert, without any accountability or democratic debate.”18 But there was a more practical reason driving the executive branch to seek a new authorization: It has become increasingly difficult as a legal matter to rely on the 2001 AUMF given the substantial passage of time since the 2001 authorization, the evolution of the terrorist threat, the emergence of new terrorist organizations such as ISIS, and the expanded geographic use of military force, including armed drones, to include places such as Somalia and Libya. Notwithstanding these overtures, the Obama administration continued to rely on the 2001 AUMF as providing sufficient congressional authority to conduct military operations in new theaters and against ISIS and the al Qaeda affiliate in Syria, the Khorasan Group.19 But the legal arguments supporting continued reliance on the 2001 AUMF grow more tenuous over time, as the temporal gap between congressional authorization and force grows, and as the purpose for engaging in hostilities becomes harder to justify as a response to the attacks on September 11, 2001. Moreover, whereas combat operations in the immediate months following September 11, 2001, were primarily confined to Afghanistan, the geographic scope of the government’s use of drones to target terrorist operatives has expanded to numerous additional theaters, thus supporting calls for a renewed congressional authorization.
In October 2017, Senator Bob Corker, Chairman of the Foreign Relations Committee, convened a hearing of that committee to consider a new AUMF. In his opening, Corker noted that “President [Trump], just like President Obama, believes he has the legal authority he needs under the 2001 AUMF to fight ISIS, al Qaeda, and other terrorist groups. And I agree. I agreed with the Obama administration, and I agree with this administration that they currently have the authority.” Nevertheless, Corker noted his belief that Congress should “update the AUMF to reflect the current conflict and reassert Congress’s constitutional role.”20
For its part, the Trump administration has noted its belief that a new AUMF “is not legally required to address the continuing threat posed by al Qaeda, the Taliban, and ISIS.”21 Nevertheless, to the extent that Congress does pass an AUMF, the administration has signaled three requirements for it doing so. The first, that a new AUMF be in place prior to repeal of the 2001 AUMF, would enable ongoing military operations – including the use of drones – to continue under congressional authorization without any gaps.
The Trump administration has noted its belief that a new AUMF 'is not legally required to address the continuing threat posed by al Qaeda, the Taliban, and ISIS.'
The second, that the new AUMF not be time-constrained, would, according to the administration, help avoid “arbitrary” terminations of authority that could “unintentionally embolden our enemies with the goal of outlasting us.” And third, the administration would require that a new AUMF not be geographically restricted. In the administration’s view, geographic restrictions would hamstring the government in its ability to use force against ISIS in new theaters. Given that ISIS has made footholds in various regions outside of Iraq and Syria, congressional authorization that is not geographically bounded would allow drone operations (and other military activity) to follow ISIS wherever it lays root.
Viewed from the perspective of congressional oversight of the U.S. drone program, Congress should carefully evaluate the administration’s preferences and assertions and balance any valid operational arguments against the impact that they would have on Congress’s current and future ability to oversee U.S. policy on the use of armed drones and its implementation. Ideally, Congress should pass an AUMF that (1) meets the administration’s operational needs, but (2) provides for ongoing and robust congressional oversight of the program. The administration, to be sure, needs legal certainty and operational flexibility in carrying out strikes in an ever-changing global landscape, but a new authorization must also enable Congress to participate in the policymaking process and oversee the program’s implementation.
From an operational perspective, the administration’s objection to a time-bounded AUMF (oftentimes via a statutory mechanism commonly referred to as a “sunset provision”) is predicated on a belief that doing so would “arbitrarily” terminate authority that could “unintentionally embolden our enemies with the goal of outlasting us.” This concern presumes that a time-bounded AUMF necessarily precludes a future Congress from renewing the authority or passing a new authorization tailored to the circumstances as they then exist. To the contrary, a time-bounded AUMF authority would facilitate a periodic dialogue, between the executive branch and Congress and between the government and the American people, about military operations and the use of force, including with armed drones, and compel Congress to reassess its continued support for it. The debate as to whether an AUMF should contain a sunset provision, therefore, is less about the operational requirements for the use of drones but instead whether, following initial sign-off by Congress, the drone program as well as the use of force overseas more generally is thereafter a matter solely for the executive branch. A sunset provision, in that sense, serves a very meaningful role in ensuring a continuing voice for Congress in the future use of drones.
The administration’s third condition, that the AUMF not be geographically bounded, would similarly impact congressional oversight of the drone program insofar as the authority, granted when drones were in use in X number of theaters, could tomorrow serve as the legal basis to support operations in X+N theaters. While Congress may support an expanded use of drones to meet new and emergent threats in new theaters, given the significant foreign policy implications of engaging in hostilities in new countries, Congress no doubt has a responsibility to provide input on any such expansion. But, on the other hand, constraining the authority for the use of force is nearsighted and operationally problematic given the rapid expansion of ISIS worldwide and the time it takes for Congress to act to approve an expansion. In that sense, the executive branch is right to insist that a new AUMF not be tethered to the threat landscape today, but be flexible to meet the operational needs of tomorrow.
And yet, while formal statutory mechanisms of preserving for Congress the ability to object to particular choices made by the executive branch in the execution of existing congressional authority would likely be unconstitutional,22 a potential middle ground approach would be to require notifications prior to the start of operations in new theaters, coupled with limitations that Congress could place on the use of funds. In other words, whereas an AUMF may not limit the theaters in which force, including through drones, can be conducted, in practice Congress can place restrictions on the use of funds – which it reviews and re-approves on an annual basis – with regard to operations in theaters that it objects to.
The executive branch is right to insist that a new AUMF not be tethered to the threat landscape today, but be flexible to meet the operational needs of tomorrow.
Such notification requirements and appropriation riders are used with numerous other classified defense and intelligence programs and have the effect of forcing the executive branch to keep Congress informed, in a comprehensive and ongoing fashion, about executive action, thereby preserving for Congress a role – via its appropriations power – in guiding the future of these activities. So too here such provisions could preserve a role for Congress in reviewing and approving the use of drones in new theaters, even as it would retain for the executive branch the statutory authority to rapidly move into those theaters as required by the on-the-ground conditions and overall threat picture.
An AUMF that includes a sunset provision and that has some mechanism for ensuring that Congress is aware of and assents to the expanded use of drones into new theaters provides another critical benefit to the executive branch – shared responsibility and accountability. In instances in which the executive branch undertakes an activity on its own, or pursuant to a stale congressional authority and without the input of the current Congress, it will bear all of the political accountability of the activity. To the extent that the AUMF can provide a mechanism for ongoing congressional oversight and approval of the drone program, it instead provides the executive branch with, in bureaucratic speak, additional “cover.”
While “cover” helps diffuse responsibility between both branches rather than the executive branch bearing it alone, it also helps provide constitutional and international legitimacy to the activity. While public support for the drone program today is high, as with almost all other national security programs, there could very well come a day when public sentiment wanes and the public seeks to hold the government to account for its actions. When those days come, it is far better for the administration to be able to colorably tell the American public that it did not overreach, that it sought to include Congress to the maximum extent operationally feasible, and that Congress was aware of and approved the actions that it took. The administration’s requested AUMF would do just the opposite. It would set up an authorization that would allow the administration to carry the drone program forward without any requirement for periodic reevaluation by Congress.
The goal of this paper is to initiate dialogue between Congress, the executive branch, and the American public about the proper role for Congress in overseeing the U.S. drone program, which has become a critical and heavily used tool in the government counterterrorism arsenal, and to offer proposals for enhancing Congress’s role in overseeing the program in ways that are sensitive to and minimize impact on executive branch operational concerns.
- Edwin S. Corwin, The President: Office and Powers, 1787–1948, 4th ed. (New York: New York University Press, 1957), 208. ↩
- See, e.g., Charles A. Lofgren, “War-Making Under the Constitution: The Original Understanding,” Yale Law Journal 81 (March 1972), 672, 701 (discussing the quasi-war with France and Congress’s authority to declare limited hostilities). ↩
- Ahmed Salem bin Ali Jaber v. United States, No. 16-5093 (D.C. Cir. 2017). Janice Rogers Brown, concurring (“Our democracy is broken. We must, however, hope that it is not incurably so. . . . The executive and Congress must establish a clear policy for drone strikes and precise avenues of accountability”). ↩
- Walter Oleszek, “Congressional Oversight: An Overview,” Congressional Research Service (CRS) Report No. R41079, February 22, 2010, 1, https://fas.org/sgp/crs/misc/R41079.pdf. ↩
- Morris S. Ogul, Congress Oversees the Bureaucracy: Studies in Legislative Supervision (Pittsburgh, PA.: University of Pittsburgh Press, 1976), 11 (noting that oversight constitutes “behavior by legislators and their staffs, individually or collectively, which results in an impact, intended or not, on bureaucratic behavior”). Of course, while Congress does certainly have “broad” oversight responsibilities, its authority is “not unlimited.” Watkins v. United States, 354 U.S. 178, 187 (1957). ↩
- Oleszek, “Congressional Oversight: An Overview.” ↩
- Lack of public access to information is not unique to the U.S. drone program; any sensitive intelligence or military operation will be classified and thus shielded from public view and debate. But the public’s lack of insight into the drone program is perhaps more concerning given that the use of armed drones in conflict is still a fairly recent addition to the military arsenal and thus lacks a long track record of use and public awareness and concurrence, and that international consensus and the international legal framework governing drone use are highly unsettled. Adding to the concern is the ease with which drones can be expanded to new geographies without the natural visibility that otherwise follows the deployment of large-scale military resources to new theaters. ↩
- Michael Shear and Scott Shane, “Congress to See Memo Backing Drone Attacks on Americans,” The New York Times, February 7, 2013, http://www.nytimes.com/2013/02/07/us/politics/obama-orders-release-of-drone-memos-to-lawmakers.html. ↩
- The House Permanent Select Committee on Intelligence has long asserted its jurisdiction over both the national intelligence program and the military intelligence program. See https://intelligence.house.gov/about/history-and-jurisdiction.htm. ↩
- Two notable exceptions to this general lack of robust public debate relate to the use of a U.S. drone to kill American citizen Anwar al-Awlaki in 2011 in Yemen, and the national coverage following the death of American Warren Weinstein and Italian Giovanni Lo Porto, who were killed by a U.S. drone strike in 2015 targeting an al Qaeda compound in the Afghanistan-Pakistan border region. ↩
- Jacquelyn Schneider and Julia Macdonald, “U.S. Public Support for Drone Strikes,” Center for a New American Security, September 20, 2016, 4. ↩
- Jessica Purkiss and Jack Serle, “Obama’s Covert Drone War in Numbers: Ten Times More Strikes than Bush,” Bureau of Investigative Journalism, January 17, 2017, https://www.thebureauinvestigates.com/stories/2017-01-17/obamas-covert-drone-war-in-numbers-ten-times-more-strikes-than-bush. ↩
- Luke Hartig, “The Drone Playbook,” New America, August 2016. ↩
- An example of this type of mission creep is the role that the PCLOB played in evaluating the National Security Agency’s telephone metadata program in the months following the unauthorized disclosures by Edward Snowden in June 2013. The statutory mandate of the PCLOB is to “advise the President and the departments, agencies, and elements of the executive branch to ensure that privacy and civil liberties are appropriately considered in the development and implementation of such legislation, regulations, policies, and guidelines.” See 42 U.S.C. § 2000ee, “Privacy and Civil Liberties Oversight Board.” Nevertheless, in its publicly released report on the metadata program, the PCLOB undertook a detailed legal analysis of the program, on both statutory and constitutional grounds. The PCLOB’s legal analysis constituted a significant departure from its statutory mandate, inserted the PCLOB into a highly charged legal debate that was the subject of active litigation in the federal courts (a far more appropriate venue for the adjudication of difficult legal questions than the PCLOB), and largely overshadowed the important policy recommendations – appropriately within its mandate – raised by the PCLOB about the program. ↩
- Office of the Director of National Intelligence, “Summary of Information Regarding U.S. Counterterrorism Strikes Outside Areas of Active Hostilities,” July 2016, https://www.dni.gov/files/documents/Newsroom/Press%20Releases/DNI+Release+on+CT+Strikes+Outside+Areas+of+Active+Hostilities.PDF. The government’s release of aggregate statistics on casualties from drone strikes and the policy framework under which strikes were conducted from 2013 through 2015 under the Obama administration followed a lengthy internal review within the executive branch in which national security officials worked through numerous difficult issues relating to the release of such information, including the classification of details about strikes and their outcomes, legal concerns that the release of certain information can complicate the government’s ability to continue to protect from public disclosure other information that remains operationally sensitive, and policy questions regarding what information to release. The public disclosure of information about the drone program reflected President Obama’s “commitment to providing as much information as possible to the American people about U.S. counterterrorism activities.” And while one goal of this push for greater transparency was a belief that the provision of information about counterterrorism activities would help address concerns raised by critics of the program, the extent to which it had this effect are debatable. Shortly after the release of the policy framework, for example, the American Civil Liberties Union was quick to identify “key questions” that remain about the framework. And other nongovernmental organizations were similarly quick to focus on the information that remained classified and criticized the government for not providing more – without carefully evaluating, as government officials had hoped, how this new information contributed to their understanding of the program and whether any of their previously held views and public criticisms of the program were in need of reevaluation. ↩
- One potential complicating factor to this proposal is that while Congress would have little constraints in its ability to share its conclusions with the executive branch, the classified nature of much of the information will impose real barriers to Congress’s ability to share its conclusions, and the basis for them, with the American people. Nevertheless, it would be similar to the role that the intelligence committees play as proxies for the American public on matters that are classified, and the simple fact that Congress would play a role as an independent party in reviewing information from the government and NGOs can have a positive impact on public confidence. ↩
- Pub. L. 107-40, The Authorization for Use of Military Force (AUMF) (2001). ↩
- Jonathan Chait, “Five Days That Shaped a Presidency,” New York Magazine, October 2, 2016, http://nymag.com/daily/intelligencer/2016/10/barack-obama-on-5-days-that-shaped-his-presidency.html. ↩
- Matthew C. Weed, “2001 Authorization for Use of Military Force: Issues Concerning Its Continued Application,” Congressional Research Service (CRS) Report No. R43983, April 14, 2015, 4–5. ↩
- Senator Bob Corker, Statement to the Committee on Foreign Relations on Authorizations for the Use of Military Force, U.S. Senate Committee on Foreign Relations, October 30, 2017, https://www.foreign.senate.gov/press/chair/release/corker-statement-at-hearing-on-authorizations-for-the-use-of-military-force. ↩
- Secretary of Defense James Mattis, testimony before the Senate Foreign Relations Committee, October 30, 2017. ↩
- Immigration and Naturalization Service v. Chadha, 462 U.S. 919 (1983). ↩