014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 2/9/2012 3:56 PM Military-Intelligence Convergence and the Law of the Title 10/Title 50 Debate Robert Chesney* Leon Panetta appeared on PBS Newshour not long after the raid that 1 killed Osama bin Laden. He was the Director of the Central Intelligence Agency at that time, and during the course of the interview he took up the question of the CIA’s role in the attack. It had been “a ‘title 50’ operation,” he explained, invoking the section of the U.S. Code that authorizes the 2 activities of the CIA. As a result, Panetta added, he had exercised overall 3 “command.” This surely confused at least some observers. The mission had been executed by U.S. Navy SEALs from Joint Special Operations Command (JSOC) after all, and both operational and tactical command seemed to have 4 resided at all times with JSOC personnel. But for those who had been following the evolution of the CIA and JSOC during the post-9/11 period, Panetta’s account would not have been surprising. The bin Laden raid was, from this perspective, merely the latest example of an ongoing process of convergence among military and intelligence activities, institutions, and authorities. * Charles I. Francis Professor in Law, University of Texas School of Law. Special thanks to Suzanne Spaulding, and thanks as well to participants at workshops at Vanderbilt and New York University including Norman Abrams, Philip Alston, Diane Marie Amann, David Golove, Monica Hakimi, Peter Margulies, Michael Newton, Deborah Pearlstein, Richard Pildes, Harvey Rishikof, Christopher Slobogin, Stephen Vladeck, Matthew Waxman, Benjamin Wittes, and Ingrid Wuerth. I am also grateful to William Banks, David Barron, David Donatti, Chris Donesa, Louis Fisher, Jonathan Fredman, Martin Lederman, and A. John Radsan for their comments. 1. CIA Chief Panetta: Obama Made ‘Gutsy’ Decision on Bin Laden Raid, NEWSHOUR, May 3, 2011, available at http://www.pbs.org/newshour/bb/terrorism/jan-june11/panetta_0503.html. 2. Id. Title 50 is a section of the U.S. Code addressing a range of security topics, including the standing authorities of the CIA. Title 10, in contrast, is a section of the Code devoted exclusively to the armed forces. Reflecting this distinction, the argot of national security lawyers uses “Title 50 authority” and “Title 10 authority” as shorthands for the notion that there are distinct spheres of intelligence and military operations and that each is subject to a distinct set of standing statutory authorizations and constraints. 3. Id. 4. Nicholas Schmidle, Getting Bin Laden: What Happened That Night in Abbottabad, THE NEW YORKER, Aug. 8, 2011, http://www.newyorker.com/reporting/2011/08/08/110808 fa_fact_schmidle. Panetta was quick to add during his NewsHour interview that Admiral William H. McRaven, Commander of JSOC, had maintained actual command during the raid. See supra note 1. 539 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 540 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 The convergence trend is not a post-9/11 novelty. It has much deeper roots than that. The trend has accelerated considerably over the past decade, however, thanks to an array of policy, budgetary, institutional, and technological developments. And as the trend accelerates, it is becoming increasingly clear that it has profoundly important implications for the domestic law architecture governing military and intelligence activities. That architecture is a complex affair, including what might be described as “framework” statutes and executive branch directives generated in fits and starts over the past forty years. Ideally, it serves to mediate the tension between the desire for flexibility, speed, and secrecy in pursuit of national defense and foreign policy aims, on one hand, and the desire to preserve a meaningful degree of democratic accountability and adherence to the rule of law, on the other. Of course, the legal architecture has never been perfect on this score, or even particularly close to perfection. But the convergence trend has made the current architecture considerably less suited towards these ends. First, it reduces the capacity of the existing rules to promote accountability. The existing rules attempt to promote accountability in two ways. They promote it within the executive branch by requiring explicit presidential authorization for certain activities, and they promote accountability between the executive branch and Congress by requiring notification to the legislature in a broader set of circumstances. Convergence undermines these rules by exposing (and exacerbating) the incoherence of key categorical distinctions upon which the rules depend, including the notion that there are crisp delineations separating intelligence collection, covert action, and military activity. As a result, it is possible, if not probable, that a growing set of exceptionally sensitive operations – up to and including the use of lethal force on an unacknowledged basis on the territory of an unwitting and non-consenting state – may be beyond the reach of these rules. Second, the convergence trend undermines the existing legal architecture along the rule-of-law dimension by exposing latent confusion and disagreement regarding which substantive constraints apply to military and intelligence operations. Is international law equally applicable to all such operations? Is an agency operating under color of “Title 50” at liberty to act in locations or circumstances in which the armed forces ordinarily cannot? These questions are not in fact new, but thanks to convergence they are increasingly pressing. Government lawyers are well aware of these issues, and in fact have 5 been grappling with them for much of the past decade, if not longer. For 5. See John Rizzo, National Security Law Issues – A CIA Perspective, Address delivered at a conference of the American Bar Association Standing Committee on Law and National Security (May 5, 2010), available at http://www.abanet.org/natsecurity/ multimedia/WS_30274.mp3 (noting that “this discussion has been going on inside the 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] LAW OF THE TITLE 10/TITLE 50 DEBATE 2/9/2012 3:56 PM 541 many years, however, public reference to them was quite limited. The most important early post-9/11 example came in 2003, when The Washington Times reported that the Senate Select Committee on Intelligence was quietly attempting to expand its oversight authority in order to encompass certain clandestine military operations in response to concern about the 6 expanding role of special operations units in the war on terrorism. That effort failed in the face of fierce pushback from the Pentagon and the Senate 7 and House Armed Services Committees, but not before drawing at least some attention to the disruptive impact convergence even then was having 8 on the accountability system. In more recent years, the media has begun to pay more sustained attention, frequently noting that the complications associated with convergence impact question of substantive authority as well as accountability. In 2010, for example, The Washington Post reported that a fierce interagency debate was underway in connection with “which agency should be responsible for carrying out attacks” online, with the CIA categorizing certain attacks as covert actions which are “traditionally its turf” and the military taking the position that such operations are “part of its mission to counter terrorism, especially when, as one official put it, ‘al9 Qaeda is everywhere.’” And the same Washington Post story indicated that the Justice Department’s Office of Legal Counsel had produced a draft opinion in spring 2010 “that avoided a conclusive determination on whether computer network attacks outside battle zones were covert or not,” but that nonetheless concluded that “[o]perations outside a war zone would require the permission of countries whose servers or networks might be 10 implicated.” Subsequent stories about the use of lethal force in Yemen have also raised the issue of host-state permission, suggesting that JSOC but not the CIA would be obliged to act only with such permission, and that as a result JSOC units might at times prefer to operate under color of the 11 CIA’s authority (as happened in Pakistan with Osama bin Laden, and 12 again in Yemen with Anwar al-Awlaki). executive branch for many years . . . this is not a post-9/11 phenomenon”). See also Matthew Dahl, Event Summary: The bin Laden Operation – The Legal Framework (May 25, 2011), available at http://www.americanbar.org/content/dam/aba/administrative/law _national_security/covert_action_event.authcheckdam.pdf). 6. Bill Gertz, Congress To Restrict Use of Special Ops.: Presidential Finding Would Be Required, WASH. TIMES, Aug. 13, 2003, at A1. 7. See Jennifer Kibbe, The Rise of the Shadow Warriors, FOREIGN AFFAIRS 102, 107 (Mar./Apr. 2004). 8. See id. Kibbe deserves substantial credit for her early identification of the convergence issue and its disruptive impact on the accountability system. 9. See, e.g., Ellen Nakashima, Pentagon Is Debating Cyber-Attacks, WASH. POST, Nov. 6, 2010, at A1. 10. Id. 11. See, e.g., Julian Barnes & Adam Entous, Yemen Covert Role Pushed: Foiled Bomb 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 542 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 These accounts give a sense of the range of legal questions that convergence generates, as well as the debates that surround them within the government. And that in turn is enough to frame the investigation that follows. I proceed in two parts, beginning in Part I with a descriptive account of the convergence trend itself. Part I opens with a focus on events in the 1980s and 1990s that presaged the accelerated convergence of the post-9/11 period. Attempts by the military to develop within the special forces community capacities quite similar to those of the CIA are described in Part I.A, and CIA flirtations with the use of deadly force against terrorists are described in Part I.B. Against that backdrop, Part I.C. then explores how convergence has manifested over the past decade, with an emphasis on the CIA’s kinetic turn, JSOC’s parallel expansion, the development of hybrid CIA-JSOC operations, and the emergence of cyberspace as an operational domain. Readers already familiar with the convergence phenomenon may wish to skip ahead to Part II, which examines the impact of convergence on the 13 domestic legal architecture relevant to such activities. Part II.A. clarifies what I have in mind when I refer to a domestic legal architecture, as it traces the emergence and growth of standing rules relating to (i) the internal executive branch decisionmaking process, (ii) information-sharing between the executive branch and Congress, and (iii) substantive authorizations and prohibitions relating to certain types of activity. The remainder of Part II analyzes the impact of convergence on each of these rules, demonstrating the manner in which convergence creates new problems for (and exacerbates existing problems in) the existing legal architecture. The key issues include: the increasingly large and significant set of military operations that are not subject to either presidential authorization or legislative notification; lingering suspicion with respect to what law if any Plot Heightens Talk of Putting Elite U.S. Squads in CIA Hands, WALL ST. J., Nov. 1, 2010, at A1; Greg Miller, CIA will Direct Yemen Drones, WASH. POST, June 14, 2011, at A1; Siobhan Gorman & Adam Entous, CIA Plans Yemen Drone Strikes: Covert Program Would Be a Major Expansion of U.S. Efforts To Kill Members of al Qaeda Branch, WALL ST. J., June 14, 2011, at A8; Greg Miller & Julie Tate, CIA Shifts Focus to Killing Targets, WASH. POST, Sept. 1, 2011. 12. Jennifer Griffin & Justin Fishel, Two U.S.-Born Terrorists Killed in CIA-Led Drone Strike, FOXNEWS.COM (Sept. 30, 2011), http://www.foxnews.com/politics/2011/09/ 30/us-born-terror-boss-anwar-al-awlaki-killed/. 13. Many of the operations at issue in the convergence context, such as the use of drones to kill, raise a host of international law issues. See, e.g., Philip Alston, The CIA and Targeted Killings Beyond Borders (New York University Public Law and Legal Theory, Working Paper No. 303, 2011); Robert Chesney, Who May Be Killed? Anwar al-Awlaki as a Case Study in the International Legal Regulation of Lethal Force, 13 Y.B. OF INT’L HUMANITARIAN LAW 3 (2010). Those questions are beyond the scope of this paper. This paper does, however, address whether there is variation in domestic law with respect to whether and when U.S. government entities must comply with certain bodies of international law (though without regard to what those bodies of international law happen to require). 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] LAW OF THE TITLE 10/TITLE 50 DEBATE 2/9/2012 3:56 PM 543 restrains the CIA’s use of lethal force; confusion with respect to whether and why the CIA might be at greater liberty than JSOC to conduct operations without host-state consent; and the difficulty of mapping the existing architecture onto operations conducted in cyberspace. I embed my recommendations for reform within the analysis at each step along the way. To summarize, I offer four recommendations. Enhance Accountability within the Executive Branch. The current legal architecture requires presidential approval for “covert action” programs, but the situation is complicated with respect to unacknowledged military operations. An unacknowledged military operation must be authorized by the President or at least the Secretary of Defense if it is collateral to an anticipated overt military operation that is not yet imminent but for which operational planning has been authorized – a sweeping set of circumstances. But no such approval is required if the operation is collateral to ongoing hostilities. This makes sense if the unacknowledged operation occurs in the combat zone. If it occurs on the territory of another state outside the “hot” battlefield, however, the risks are sufficient to warrant extension of the requirement of presidential or at least secretarial authorization. Notably, press accounts indicate that former Secretary of Defense Robert M. Gates had insisted upon such an approach for lethal operations outside the hot battlefield, as a matter of policy. At a minimum, that policy should be codified. Better still to extend it to all unacknowledged military operations outside the combat zone. The degree of accountability involved should be commensurate with the risks, and in light of convergence there is little reason to calibrate that judgment differently for the military than for the CIA, at least not outside combat zones. Enhance Information-Sharing with Congress. Operations constituting “covert action” must be reported to the House and Senate Intelligence Committees; by contrast, the unacknowledged military operations discussed above are not subject to this requirement. A separate law requires notification to Congress when the armed forces are deployed in circumstances involving a likelihood of hostilities, but given the strict interpretation of “hostilities” adopted in relation to the conflict in Libya it seems clear that a considerable amount of unacknowledged military activity might escape notification to Congress under that regime as well. An effort was made in 2003 to close this gap by requiring unacknowledged military activity to be reported to the Intelligence Committees when activity occurs outside the geographic confines of a state where the United States has an overt combat presence. The effort failed in the face of resistance from the Pentagon and the House and Senate Armed Services Committees. It should be revived, but with notification being made to the Armed Services Committees, subject to an option for close-hold notifications, based on the Gang of Eight model. All such notification scenarios should be modified, however, to include participation by the chief majority and minority 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 544 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 counsels of the relevant committees (creating, in effect, a “Gang of Twelve” system). Clarify Substantive Constraints on Title 50 Operations. It should be made clear that all U.S. government agencies comply with the law of war in any operation to which the law of war applies, regardless of whether the operation is categorized as a Title 10 or a Title 50 activity and regardless of which particular agency carries it out. This is not necessarily a change from current policy, but it would help to address concerns that critics have raised with respect to whether the CIA conforms its drone operations to law of war standards. On the other hand, it would not be appropriate to adopt a similar express commitment vis-a-vis international law’s treatment of state sovereignty, given lingering uncertainty with respect to whether and when international law prohibits one state from conducting espionage, covert action, or other operations within another state’s territory in the first place. Clarify Authorization and Accountability for Cyberoperations. Operations in cyberspace tend to defy categorization by type (collection, covert action, or military activity) or geographic location. This causes problems on all the dimensions mentioned above, while also raising difficult questions regarding when an agency has the affirmative authority to conduct such operations in the first place. Legislation can resolve much of this uncertainty by (i) clarifying that the military has standing authority to conduct computer network attacks (unacknowledged or otherwise) when acting in a defensive capacity or under color of a statutory authorization for the use of military force, and (ii) providing timely notification to the House and Senate Armed Services Committees of such operations when they have or are likely to have significant consequences outside a theater of combat operations. I. THE CONVERGENCE TREND The notion of “convergence” between military and intelligence activities would likely have seemed strange prior to the second half of the twentieth century. The U.S. military was no stranger to the business of intelligence after all. On the contrary, it had engaged in the collection and analysis of intelligence throughout American history, at least during times of armed conflict, and during World War II had developed the Office of Strategic Services (OSS) as the very embodiment of a “military” organization devoted to the full spectrum of “intelligence” activities. But the center of gravity shifted in the late 1940s when the Truman administration and Congress began reorganizing the national security establishment to suit the imperatives of the Cold War and America’s newfound status as the predominant Western global power. To be sure, executive branch departments had often exercised intelligence functions in the past, but prior to 1946 there had never been a free-standing agency, let alone a civilian one, intended to be distinct from the military establishment 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] LAW OF THE TITLE 10/TITLE 50 DEBATE 2/9/2012 3:56 PM 545 and capable of relatively disinterested analysis, reporting directly to the President, devoted to collection, analysis, and covert action. Thus it was a significant novelty when President Harry S. Truman ordered the creation of the Central Intelligence Group in 1946 as a civilian successor to the OSS 14 and when Congress the next year transformed that body into the CIA. Over the following decades, the CIA became the predominant institution associated with human intelligence, or HUMINT, collection, 15 outside the context of open armed conflict. It also became the repository of America’s covert action capacity as that realm of activity became increasingly significant during the Cold War. As a result of both developments, the notion of a distinction between military and intelligence activities came to seem more meaningful than it had been in the past. Various legal developments described below in Part II.A., reinforced and embodied that notion as well. The distinction was always tenuous, however, and by the early 1980s the early signs of convergence already were apparent. A. Convergence and the Military in the 1980s Though it is true that the late 1970s and early 1980s saw a significant revival in Cold War tensions, there were some during that same period who were turning their attention at least in part to unconventional threats involving noncommunist, non-state adversaries. One such strategic thinker was General Edward Meyer, the Army’s Chief of Staff from 1979 to 1983. According to one of his subordinates, Meyer believed that America’s “adversaries were affecting us below the threshold of war,” and America 16 needed to build its capacity to respond in kind. From Meyer’s perspective, the situation did not call for a greater CIA role so much as for the military to expand its own capacity to fight in the shadows – above all through Special Operations Forces (SOF). Of course, the military already had some such capacity, as illustrated by the Army’s 1st Special Operations Detachment-Delta (Delta Force) and the Navy’s SEAL Team Six. These units were capable of executing small-scale kinetic operations such as hostage rescue, including in denied areas. Or at least they could do so when supplied with the necessary tactical intelligence to support such operations. And thus a question had arisen: Should SOF units rely on the CIA and the rest of the Intelligence Community (including the 14. See generally AMY ZEGART, FLAWED BY DESIGN: THE EVOLUTION OF THE CIA, JCS, AND NSC (1999). 15. Signals intelligence, or SIGINT, by contrast, has been the bailiwick of the DoD’s National Security Agency (NSA), even as to non-military surveillance. 16. Seymour Hersh, Who’s in Charge Here?, N.Y. TIMES, Nov. 22, 1987 (quoting Lt. Col. Michael Foster, who served in the Army’s Special Operations Division during Meyer’s tenure in the early 1980s). 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 546 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 Defense Intelligence Agency and the Service intelligence agencies) to provide them with intelligence and non-kinetic forms of covert support, or should the SOF community develop parallel, in-house capacities? The issue came to a head twice during the Tehran hostage crisis in 1979-1980. Famously, the Carter administration authorized a Delta Force 17 rescue operation (Operation Eagle Claw). But obtaining the tactical intelligence and covert logistical support necessary for that mission proved to be exceedingly difficult. The CIA failed to provide it, possibly reflecting contemporaneous personnel and morale problems CIA was experiencing in 18 the late 1970s and early 1980s and certainly reflecting the problems caused by the capture of CIA personnel and files during the embassy 19 takeover. In any event, the SOF community ultimately came to the conclusion that “there existed nowhere in the national capability an 20 organization to provide this vital support,” including within the military 21 itself. Operation Eagle Claw ultimately was aborted because of a deadly accident at a staging area within Iran. Planning for another rescue operation (Operation Snowbird) began soon after, however, and this time the military took steps to create the intelligence and logistical support services it 22 needed. Toward that end, the Joint Chiefs of Staff authorized creation of 17. See MARK BOWDEN, GUESTS OF THE AYATOLLAH: THE FIRST BATTLE IN AMERICA’S WAR WITH MILITANT ISLAM (2006). 18. Caryle Murphy & Charles R. Babcock, Army’s Covert Role Scrutinized: Financial Probe Raises Fear that Special Units “Got Carried Away,” WASH. POST, Nov. 29, 1985, at A1, A8-A9; see also James Bamford, Where Secret Armies Clash by Night, WASH. POST, July 3, 1988, at X11 (reviewing STEVEN EMERSON, SECRET WARRIORS: INSIDE THE COVERT MILITARY OPERATIONS OF THE REAGAN ERA (1988)); Hersh, supra note 16 (noting “widespread belief” that CIA “had been weakened” in this respect, and contending that this perception incentivized the Army to develop its SOF-based capacity for clandestine operations). 19. Richelson notes that some CIA officers had been taken hostage, and that some Iranian CIA assets were either missing or dead as a result of the revolution. See Jeffrey T. Richelson, “Truth Conquers All Chains”: The U.S. Army Intelligence Support Activity, 1981-1989, 12 INT’L J. INTEL. & COUNTERINTEL. 168, 169 (1999). Richelson also quotes an anonymous government official asserting that “the agency [CIA] people were preoccupied with keeping their cover and could not provide equipment or information for the [rescue] operation.” Id. 20. BRIEF HISTORY OF UNIT, available at http://www.gwu.edu/~nsarchiv/NSAEBB/ NSAEBB46/document11.pdf (partially-redacted government document describing the origins of the Intelligence Support Activity unit) [hereinafter BRIEF HISTORY]. 21. See Memorandum from Lt. Gen. Philip Gast, Director of Operations for the Joint Chiefs of Staff, to Director of the Defense Intelligence Agency (Dec. 10, 1980), available at http://www.gwu.edu/~nsarchiv/NSAEBB/NSAEBB46/document6.pdf. General Gast, notably, had served in Iran as chief of a military advisory contingent between 1977 and October 1979. See U.S. AIR FORCE, BIOGRAPHY OF LIEUTENANT GENERAL PHILIP C. GAST, http://www.af.mil/ information/bios/bio.asp?bioID=5502. 22. Richelson suggests that the military also took the initiative prior to the aborted Eagle Claw mission, dispatching “at least two teams of individuals” who entered Iran on false passports and “attempted to collect the required on-the-ground intelligence.” Richelson, 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] LAW OF THE TITLE 10/TITLE 50 DEBATE 2/9/2012 3:56 PM 547 the Foreign Operations Group (FOG), which “was an adhoc [sic] organization” tasked with providing “a combination of intelligence 23 collection and operational support to a striking force.” FOG’s work never came to fruition, as the Iranians released the hostages upon the inauguration of President Ronald Reagan in January 1981. But in the meantime the Joint Chiefs of Staff’s Director of Operations already had recommended institutionalizing FOG because the “current DOD/Service HUMINT structure is not organized to satisfy” the “need of military planners to have accurate and timely situation oriented operational and environmental 24 data.” Within two weeks of Reagan’s inauguration, Army Chief of Staff Meyer responded by “authoriz[ing] creation of the US Army Intelligence Support Activity, or ISA, to “institutionaliz[e] in a DoD special unit . . . a worldwide, immediately responsive capability similar to that developed 25 over a one year period in the Tehran crisis.” 26 ISA encountered rough waters almost immediately. It reportedly became involved in a private effort to rescue POWs allegedly still held in 27 Laos, and this and other allegations spurred the DoD Inspector General to conduct an investigation culminating in the conclusion that ISA “lacked 28 proper oversight mechanisms for its missions and its expenditures.” Secretary of Defense Caspar W. Weinberger was alarmed, as was Deputy Secretary of Defense Frank C. Carlucci, who had been the second highest ranking official at CIA previously. Carlucci wrote in May 1982 that he found the Inspector General’s report “disturbing in the extreme,” adding that “[w]e seem to have created our own CIA, but like Topsy, supra note 19, at 169. 23. BRIEF HISTORY, supra note 20. 24. Richelson, supra note 19, at 170 (quoting Memorandum from Lt. Gen. Philip C. Gast, Director of Operations, to Lt. Gen. Eugene Tigh, Director, Defense Intelligence Agency (Dec. 10, 1980)). 25. BRIEF HISTORY, supra note 20. 26. So too did the contemporaneous Yellow Fruit initiative, which may or may not have been related to ISA. Yellow Fruit operated under cover of a business based in Northern Virginia, apparently with the aim of ensuring operational security on the part of other special operations units (as well as possible involvement in securing logistical support for other military or CIA operations, such as the provision of transportation or communications equipment). See Murphy & Babcock, supra note 18. Whatever its origins, allegations of gross financial improprieties brought Yellow Fruit to an end in 1983. For an overview of Yellow Fruit and its demise, see Hersh, supra note 16. For more detail, including allegations that a Swiss bank account created for Yellow Fruit later was used to support arms shipments to the Contras, see Dan Morgan, Secret Army Account Linked to Contra Aid: North, Secord Possibly Involved, Official Says, WASH. POST, Apr. 22, 1987, at A1; Jeff Gerth, Pentagon Linking Secret Army Unit to Contra Money, N.Y. TIMES, Apr. 22, 1987, at A1. 27. See Tim Weiner, Covert Forces Multiply, Some Run Amok, PHILL.COM (Feb. 10, 1987), http://articles.philly.com/1987-02-10/news/26179065_1_covert-action-covert-operations-blackbudget. 28. Richelson, supra note 19, at 173. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 548 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 29 Invoking the “lesson of the 70s,” uncoordinated and uncontrolled.” Carlucci directed that all ISA operations be terminated in thirty days unless a more accountable structure could be devised, subject to approval from the 30 DoD General Counsel as well as the Director of Central Intelligence. ISA survived, but emerged a far more constrained entity. From 1983 31 on, ISA would operate under a formal DoD “charter.” The charter imposed more intra-Army transparency and accountability, specifying that certain senior Army officials would have tasking and oversight authority over ISA’s activities and requiring ISA to operate under the auspices of the Army’s Assistant Chief of Staff for Intelligence rather than its Assistant 32 Chief of Staff for Operations (as General Meyer originally had planned). The charter did not restrict ISA’s range of permitted activities in a substantive sense – on the contrary, the charter contemplated that ISA might engage in both collection and covert action – but it did make clear that the former could be conducted only with approval from both the CIA and Defense Intelligence Agency (DIA), and that the latter required 33 compliance with Executive Order 12,333 governing intelligence activities (discussed below in Part II.A.). Perhaps most significantly, the charter’s concluding provision specified that the Army’s General Counsel would be responsible for ensuring that “all congressional committees having pertinent legislative or appropriation oversight responsibilities are kept fully and currently informed of [ISA] activities in accordance with applicable 34 statutes, Executive Orders, and DOD directives and regulations.” It is unclear from the public record what became of ISA (and its 35 successor organizations) in the years that followed. The important point for now, however, is that more than thirty years ago, the military – and especially the SOF community within the military –was already reacting to a perceived trend in the direction of asymmetric threats by developing in29. Memorandum from Frank C. Carlucci, Deputy Secretary of Defense, on ISA Operations to Richard Stilwell, Deputy Under Secretary for Policy (May 26, 1982), available at http://www.gwu.edu/~nsarchiv/NSAEBB/NSAEBB46/document7.pdf. “Topsy” was a character in Harriet Beecher Stowe’s 1852 novel Uncle Tom’s Cabin who when asked how she had come into the world, could not explain how, saying only that she must have grown. This gave rise to the once-common saying “grew like Topsy,” which originally conveyed a sense of unexplained origins but later came to suggest unconstrained proliferation as well. 30. Id. 31. See Richelson, supra note 19, at 175-176. 32. On Meyer’s original approach, see Hersh, supra note 16. 33. See Richelson, supra note 19, at 175-76. See also UNIT CHARTER, CHARTER OF U.S. ARMY INTELLIGENCE SUPPORT ACTIVITY, available at http://www.gwu.edu/~nsarchiv/ NSAEBB/NSAEBB46/document8.pdf [hereinafter CHARTER]. 34. CHARTER, supra note 33, ¶10(d)(3). 35. See Richelson, supra note 19, at 192 (suggesting that ISA lived on past its formal disestablishment in 1989, under various names, as a subordinate unit of Special Operations Command). See also MARK BOWDEN, KILLING PABLO: THE HUNT FOR THE WORLD’S GREATEST OUTLAW 72-73 (2001) (providing a list of alternative names). 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] 2/9/2012 3:56 PM LAW OF THE TITLE 10/TITLE 50 DEBATE 549 house capacities that seemed to compete directly with functions associated with the CIA (including collection and covert action capacities). It was, in short, an early manifestation of convergence. B. Convergence and the CIA from the 1980s until 9/11 Convergence also manifested within the CIA in the pre-9/11 era in mirror-image fashion. Just as the military experimented with CIA-like activities outside the context of the existing military components of the Intelligence Community, the CIA at around the same time was experimenting with military-like activities. Specifically, the CIA was flirting with the use of lethal force in circumstances in which diplomatic and political constraints made overt military force unpalatable. To be sure, the use of lethal force by or at the direction of the CIA was not a 1990s novelty. Plots to kill Castro with exploding cigars and alleged plots to kill other Communist or Soviet-leaning political leaders in pursuit of foreign policy aims had, after all, played a central role in the embarrassing revelations and scrutiny of the 1970s (to which Carlucci alluded in his criticism of ISA, above), and had prompted a series of executive orders prohibiting “assassination” by executive branch employees 36 (as described below in Part II.A.). But the CIA’s use of lethal force in the 1980s and thereafter in relation to terrorism was not just a matter of pursuing mere foreign policy. It was, rather, a question of circumstances involving threats to American lives. That is, it was a matter implicating the use of force in national self-defense, a justification ordinarily associated 37 with military action. 1. Using Lethal Force Against Hezbollah in 1984 The CIA’s turn toward counterterrorism had its roots in a series of terrorist attacks resulting in American deaths in the early-to-mid-1980s, 38 including the bombing in Beirut in 1983. These events led to repeated 36. The CIA also has a long history of paramilitary activity. See, e.g., Richard A. Best, Jr. & Andrew Feickert, Special Operations Forces (SOF) and CIA Paramilitary Operations: Issues for Congress (Cong. Res. Service RS22017), Jan. 4, 2005, at 3. For a discussion of the many ways in which questions of lethal force may arise in relation to CIA activities, entirely apart from operations specifically intended to cause death, see Jonathan Fredman, Policy and Law: Covert Action, Loss of Life, and the Prohibition on Assassination, 40 CENTRAL INTELLIGENCE AGENCY: STUDIES IN INTELLIGENCE 15 (1997). 37. See, e.g., Abraham D. Sofaer, Terrorism, the Law and National Self-Defense 126 MIL. L. REV. 89 (1989). For a discussion of national self-defense concepts in relation to targeted killing and covert action after 9/11, see Kenneth Anderson, Targeted Killing in U.S. Counterterrorism Strategy and Law, in LEGISLATING THE WAR ON TERROR: AN AGENDA FOR REFORM 346-400 (Benjamin Wittes ed., 2009). 38. See STEVE COLL, GHOST WARS: THE SECRET HISTORY OF THE CIA, AFGHANISTAN, 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 550 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 debate within the Reagan administration over whether and when to use force to respond to or preempt terrorist attacks, including both the overt military option and the idea of instead using lethal force covertly (either 39 directly by the CIA or through CIA-trained proxy forces). The lethal covert action option appears to have arisen first in the spring of 1984. A proposal drafted by Lieutenant Colonel Oliver North (then serving on the staff of the National Security Council) included language that would authorize covert action to “neutralize” terrorists with lethal force, using CIA-trained proxies in cases where the terrorists either already 40 had attacked Americans or were planning to do so. This drew fierce condemnation from the Agency’s Deputy Director, John McMahon, who called North to berate him for forgetting the lessons of the 1970s relating to 41 CIA involvement in assassination. Director William Casey, already sympathetic to North’s proposal, at this point turned to his General Counsel, 42 Stanley Sporkin, to weigh in on McMahon’s objection. Sporkin concluded that there was a salient distinction between political assassination along the lines of the Castro plots of the 1970s and the exercise of force in national self-defense and that the task of preempting terrorist attacks fell on 43 the proper side of that line. Casey thereafter put his weight behind North’s proposal, and a version of it appears to have prevailed in the end. National Security Decision Directive 138 (NSDD 138), which President Reagan signed that April, remains a classified document. It has been reported, however, that NSDD 138 included language authorizing “the use of sabotage, killing (though not “neutralization” or assassination), [and] preemptive and retaliatory strikes” against terrorists, and that it also included authorization for CIA officers to cooperate with SOF personnel in 44 such missions. Noel Koch, a Pentagon official known as a major supporter of SOF, later lamented that NSDD 138 “was simply ignored. No part of it was ever 45 implemented.” What happened? Within a few months of NSDD 138, Director Casey and Secretary of State George P. Shultz urged President AND BIN LADEN, FROM THE SOVIET INVASION TO SEPTEMBER 10, 2001, at 137-138 (2004); see also DAVID C. WILLS, THE FIRST WAR ON TERRORISM: COUNTER-TERRORISM POLICY DURING THE REAGAN ADMINISTRATION (2003). 39. See TIMOTHY NAFTALI, BLIND SPOT: THE SECRET HISTORY OF AMERICAN COUNTERTERRORISM 145-147 (2005); see also id. at 148 (“The solution to the disagreement over using overt means was to choose covert action instead.”). 40. BOB WOODWARD, VEIL: THE SECRET WARS OF THE CIA, 1981-1987, at 361 (1987). 41. See id. at 361-362. Interestingly, it does not appear from the public record that this proposal, or others akin to it described below, generated comparisons to or criticisms involving the Agency’s involvement in the Phoenix program in Vietnam. 42. See id. at 362. 43. See id. at 362, 394. 44. WILLS, supra note 38, at 84. 45. Id. at 87. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] 2/9/2012 3:56 PM LAW OF THE TITLE 10/TITLE 50 DEBATE 551 Reagan to issue a finding that would seem to be in line with NSDD 138, as it would direct the CIA to train and support small units of foreign nationals in the Middle East which would conduct preemptive strikes against terrorists. When intelligence showed that someone was about to hit a U.S. facility, such as an embassy or a military base, the units would be able to move to disable or kill the 46 terrorists. The preemptive strike proposal was developed by the National Security Council (NSC) and called for the CIA to “train and equip Lebanese ‘hit men’ who would be responsible for tracking down the people responsible for the terrorist attacks on U.S. facilities and the abduction of three U.S. 47 citizens.” Not inconsistently, Woodward’s account specifies that there would be two stages of authorization. An initial approval would authorize the creation and training of the unit itself, while a second and more specific approval would be necessary in order to authorize any specific operation by 48 the unit. It appears that President Reagan approved this proposal in early November 1984, with a focus on as many as three separate Lebanese proxy 49 units. The CIA was thus put into the business of using lethal force via proxies as an alternative to an overt military response against terrorists including, at least, Hezbollah. Some at the time explicitly viewed this new 50 CIA paramilitary capacity as “in competition with the Pentagon.” Things did not proceed smoothly from there, however. Richard Helms, who had previously served as Director of Central Intelligence, got wind of the new program and, feeling that it smacked of the assassination programs that had gotten the CIA into such trouble in the past, reached out to Vice President George H. W. Bush (himself a former Director of the CIA) to 51 express his concerns. Internally, some senior CIA officials, particularly Deputy Director McMahon, felt much the same way, expressing concern 52 that the CIA would be blamed for instigating an assassination program. What ultimately derailed the program, however, were doubts about the proxies themselves. The CIA and the State Department worked together to make it possible for a team of SOF personnel to inspect the Lebanese proxies on two occasions, and the resulting negative evaluations simply 46. 47. 48. 49. 50. 51. 52. WOODWARD, supra note 40, at 393. NAFTALI, supra note 39, at 148. WOODWARD, supra note 40, at 393. See NAFTALI, supra note 39, at 148; WOODWARD, supra note 40, at 405. See WOODWARD, supra note 40, at 362. See NAFTALI, supra note 39, at 150-151. Id. at 151. See also WOODWARD, supra note 40, at 394. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 552 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 53 took the wind out of the program’s sails. Or at least it did so with respect to relying on Lebanese proxies. According to Woodward, Director Casey responded to the collapsing Lebanese effort by turning to the Saudi intelligence service with a request that they target Hezbollah’s leader, Mohammed Hussein Fadlallah; and in Woodward’s account, this occurred without the knowledge of McMahon or other internal CIA opponents, without any additional presidential authorization, and without notification 54 to Congress. Eventually, Casey’s alleged Saudi initiative may have culminated in a massive car bombing in early 1985 near Fadlallah’s Beirut residence. That attack, assuming it was indeed directed at Fadlallah, was a fiasco; dozens died, hundreds were wounded, Hezbollah made sure to 55 blame America, and Fadlallah was unharmed. In the aftermath, Casey told McMahon that he was going to “call the president . . . and tell him we have 56 to rescind the finding and shut down the operation.” These events underlined the level of aversion within the CIA and more broadly to anything that might be depicted as assassination, but also the strong desire to take whatever steps were possible in order to prevent further terrorist attacks. The latter force drove the CIA toward involvement in the use of lethal force in the name of self-defense, while the former acted 57 as a check on that impetus. The Fadlallah bombing for the time being tilted the balance in favor of caution, but so long as the underlying threat of terrorism remained – and so long as covert action through the CIA appeared to provide a politically-palatable alternative to the overt use of military force – the issue was bound to resurface. 2. Counterterrorism “Action Teams” in 1986 It did not take long. The terrorist attacks at El Al Airlines ticket counters in Rome and Athens in late 1985, which killed several Americans, prompted Casey to revisit the idea of a covert action capacity to use lethal 53. See NAFTALI, supra note 39, at 151-152. 54. See WOODWARD, supra note 40, at 395-397; NAFTALI, supra note 39, at 152. 55. See WOODWARD, supra note 40, at 397. 56. NAFTALI, supra note 39, at 152. Woodward asserts that Fadlallah was subsequently mollified by a $2 million Saudi bribe, in exchange for which he agreed to cease supporting attacks on U.S. and Saudi interests. See WOODWARD, supra note 40, at 397. 57. Another example of this tension arose in 1985 in connection with a covert action program directed at undermining the Muammar Gadhafi regime in Libya. Woodward reports that members of SSCI expressed concern to Director Casey that efforts to support Libyan dissidents and exiles could run afoul of the prohibition on assassination “since the exile movement wanted [Muammar Gadhafi] dead” and “support to potential murderers was murder, period.” Id. at 419. Director Casey did not object to the proposition that it would be a problem if Gadhafi’s death was sought, but rather responded that this was not the goal of the program. Id. The senators involved nonetheless protested in a letter to Reagan about the prospects of a CIA “assassination” program. Id. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] 2/9/2012 3:56 PM LAW OF THE TITLE 10/TITLE 50 DEBATE 553 58 force to preempt further attacks. Casey had in mind “action teams that could put the CIA on the offensive in a global campaign against terrorist groups,” and he tasked Duane R. "Dewey" Clarridge, a key figure in the CIA’s Directorate of Operations (DO), with developing a proposal to that 59 effect. Clarridge recommended “formation of two super-secret ‘action teams’ that would be . . . authorized to kill terrorists if doing so would preempt a terrorist event, or arrest them and bring them to justice if 60 possible.” Working with a new NSC-based interagency coordinating committee, Clarridge began developing a new covert action finding to 61 authorize the “action team” model. The NSC committee, notably, had a “founding directive” that grappled explicitly with the question of whether terrorism should be seen as an issue of law enforcement or national security, and whether, as paraphrased by Steve Coll, the CIA should “try to capture terrorists alive in order to try them on criminal charges in open 62 courts, or should the goal be to bring them back in body bags[.]” In a National Security Decision Directive (NSDD 207) issued in January 1986, the Reagan administration endorsed a nuanced position. Terrorism was a law enforcement issue in some contexts, but capture for trial would not always be possible, and in some situations, a military-style 63 response would be needed. And in a covert action finding issued that same day, President Reagan made clear that the U.S. military was not the 64 only instrument through which such force might be used. The finding reportedly authorized the use of “action teams” as Casey and Clarridge had 65 wished, including via foreign proxies or, apparently, CIA personnel. Once again, however, there was resistance. The “action team” concept “stirred nervous reaction on Capitol Hill,” with “[s]ome privately label[ing] 66 them ‘hit teams.’” Of particular concern were the boundaries of the authority to use lethal force. Apparently the finding authorized such action 58. See COLL, supra note 38, at 139. 59. See id. 60. Id. at 139-140. Clarridge also recommended establishing an integrated counterterrorism center to support these efforts, with integrated operational, analytical, and technical personnel. See id. at 140. Such a holistic approach ran counter to the traditional CIA model of strict separation between operations and analysis, not to mention its traditional emphasis on geographic regions. See id. at 139-140 The proposal encountered resistance because it violated the traditional model, and also because of a belief that counterterrorism was “‘police work’ best left to cops or the [FBI].” Id at 139-140. But Casey was persuaded, and this gave rise to CIA’s Counterterrorist Center. Id. at 140-141. 61. See id. at 140-141. 62. Id. at 140-141. 63. See id. at 141. 64. See id. at 141. 65. See id. at 141. 66. Id. at 141. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 554 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 only where an attack was imminent, for Robert Gates would later recall there was much debate about just where the line might lie in practical terms: [W]e got to the question of when you could kill a terrorist, and we had this almost theological argument. “Well, if the guy is driving toward the barracks with a truck full of explosives, can you kill him?” “Yeah.” “Well, what if he’s in his apartment putting the 67 explosives together?” “Well, I don’t know.” The debate over how imminent a threat must be in order to warrant lethal force remains a central question – perhaps the central question – 68 today. The important point for now, however, was that the CIA in 1986 again was being asked to embrace the use of lethal force as an instrument of national self-defense, albeit in a context of discomfort about the underlying principle and uncertainty about the metes and bounds of the authority in question. The public record sheds no light on whether and to what extent the action team concept was put into practice over the next decade. 3. Lethal Force Against al Qaeda from 1998 until 9/11 By the fall of 1997, the CIA was well aware that Osama bin Laden and 69 his al Qaeda organization constituted an increasingly important threat. The Agency had established a special unit within its Counterterrorist Center to focus specifically on bin Laden, and a plan to locate him was in the 70 works. But the goal at that time was not to kill bin Laden. The idea, instead, was to capture him and render him to the United States or elsewhere to face prosecution. Locating and capturing bin Laden was no simple task, however. The operation probably could not be executed by CIA officers. They would not be able to function effectively for this purpose in Afghanistan, at least not in comparison with Afghans. In any event, land-locked Afghanistan at the time was perceived as too remote and too hostile an environment to justify the risks, manpower, and resources that boots-on-the-ground American 67. Id. at 141. 68. See., e.g., John O. Brennan, Assistant to the President for Homeland Security and Counterterrorism, Strengthening Our Security by Adhering to Our Values and Laws, Address delivered at Harvard Law School’s Program on Law and Security (Sept. 16, 2011) (discussing the role of imminence in targeting decisions outside the “hot” battlefield), available at http://www.whitehouse.gov/the-press-office/2011/09/16/remarks-john-obrennan-strengthening-our-security-adhering-our-values-an; Robert Chesney, Malinowski on IHL away from the Battlefield and on the Meaning of Imminence, LAWFARE (Dec. 14, 2010), http://www.lawfareblog.com/2010/12/malinowski-on-ihl-away-from-the-battlefield-and-onthe-meaning-of-imminence/. 69. See COLL, supra note 38, at 367. 70. FINAL REPORT OF THE NATIONAL COMMISSION ON TERRORIST ATTACKS UPON THE UNITED STATES 110 (2004) [hereinafter 9/11 COMM. REP.]. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] LAW OF THE TITLE 10/TITLE 50 DEBATE 2/9/2012 3:56 PM 555 involvement would entail. The plan, therefore, was to use an Afghan proxy force in a non-lethal variant of the action team concept. Fortunately, a plausible proxy force was already in place. In connection with the hunt for Mir Aimal Kasi – a Pakistani man who had killed several CIA employees on the street outside CIA headquarters in Langley – other personnel at the Counterterrorist Center had been training and equipping a “family-based team of paid agents” to effectuate a similar 71 locate-and-capture operation. And after Kasi turned up in Pakistan, the team (known within the CIA as FD/TRODPINT, but referred to later in the 9/11 Commission Report simply as the “Afghan tribals”) was available for 72 a new project. The next question was whether a new presidential authorization would be necessary to undertake this operation or if instead it could be said to fall 73 within the scope of the 1986 action team program described above. Ultimately, the decision was made that this did fall within the scope of the 1986 program, and a Memorandum of Notification (MON) was duly issued to the congressional oversight committees explaining this new application of that existing authority. The process of drafting the MON, however, became the occasion for the reemergence of a familiar debate. Despite the fact that the goal of the proposed operation was to capture rather than kill bin Laden, the fact remained that it was quite possible, if not likely, that the attempt would produce a shootout in which bin Laden would be killed. That prospect – the unintended but nonetheless foreseeable killing of bin Laden – revived the debates of the 1980s regarding lethal covert action, despite the fact that the 1986 action team decision might seem to have resolved such disputes in favor of even the intentional use of deadly force. “Discussion of this memorandum brought to the surface an unease about paramilitary covert action that had become ingrained at least among some CIA senior managers,” according to the 9/11 Commission 74 Report. Echoing the position taken by Deputy Director McMahon and others in the 1980s, James L. Pavitt (who later became Deputy Director of 71. See Steve Coll, A Secret Hunt Unravels in Afghanistan: Mission To Capture or Kill al Qaeda Leader Frustrated by Near Misses, Political Disputes, WASH. POST, Feb. 22, 2004, at A1. 72. Id. 73. See 9/11 COMM. REP., supra note 70, at 113 (observing that a “1986 presidential finding had authorized worldwide covert action against terrorism and probably provided adequate authority,” though “senior CIA managers may have wanted something on paper to show that they were not acting on their own”); COLL, supra note 38, at 423 (“The agency already had legal authority to disrupt and arrest terrorists under the 1986 presidential finding that established its Counterterrorist Center. . . . It seemed wiser to use a MON to amend the legal authority the center already possessed, to make it more specific.”). 74. 9/11 COMM. REP., supra note 70, at 113. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 556 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 Operations) “expressed concern that people might get killed; it appears he 75 thought the operation had a least a slight flavor of a plan of assassination.” As a result of these concerns, case officers were at pains to convey to the Afghan tribals that they were not supposed to kill bin Laden: CIA officers met with their TRODPINT agents in Pakistan to emphasize that their plan to capture bin Laden and hold him in the Afghan cave could not turn into an assassination. “I want to reinforce this with you,” one officer told the Afghans, as he later described the meeting in cables to Langley and Washington. “You 76 are to capture him alive.” By this point, the project had come to focus on the prospects for capturing bin Laden while he stayed at the Tarnak Farms compound near Kandahar, a location where a large number of family members, including children, would be close by. There was considerable fear that the tribals would fire indiscriminately in the course of attempting the capture, causing collateral damage even if they succeeded in capturing bin Laden alive and 77 then somehow extricating him and themselves from the compound. The CIA’s top leadership ultimately determined not to support the operation, as 78 did key White House officials. By the summer of 1998 the project was put 79 on hold pending further developments. There things stood when two and a half months later al Qaeda struck the U.S. embassies in Nairobi, Kenya, and Dar es Salaam, Tanzania. Overnight, the propriety of using lethal force against al Qaeda looked dramatically different on all the relevant dimensions. Within a day of the bombings, an opportunity to attempt to kill bin Laden arose, and the Clinton administration seized it. The CIA informed the NSC principals that a large gathering of militants – including bin Laden – would take place in Afghanistan the following week. The principals accordingly recommended that the President order an overt strike – specifically, a cruise missile attack – with the “purpose to kill bin Laden 80 and his chief lieutenants” at that meeting. The President agreed, and ordered the attack – including notification to Congress “consistent with the 81 War Powers Resolution.” Congress, notably, had not in the interim passed 75. Id. at 113. It is interesting to note that the article by Jonathan Fredman cited above at note 36, which expressly addresses scenarios such as this, was published in 1997, and thus may well have been written while such questions were being debated in connection with the plan to conduct a similar capture targeting Mir Aimal Kasi. 76. COLL, supra note 71. 77. Id. 78. See id. 79. See id.; see also 9/11 COMM. REP., supra note 70, at 114. 80. 9/11 COMM. REP., supra note 70, at 116. 81. Letter from President William J. Clinton to Congressional Leaders Reporting on 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] LAW OF THE TITLE 10/TITLE 50 DEBATE 2/9/2012 3:56 PM 557 a statute to authorize the use of military force against al Qaeda, nor would it until after 9/11. But this did not stop the Clinton administration from resorting to such force, presumably (for domestic law purposes) pursuant to a claim of inherent authority under Article II of the Constitution to employ lethal force in national self-defense. Ultimately, the August 1998 cruise missile attack in Afghanistan was only a partial success. Some militants were killed, but bin Laden himself 82 left the scene hours before the missiles struck. When he was located again, would the U.S. government still be willing, at least in principle, to attempt to kill him overtly? Would it at least be willing to kill him via covert action? Or would the willingness to resort to lethal force prove to be temporary, soon to be replaced by an insistence upon capture instead? The Clinton administration’s response to these questions was complex, and remains the subject of disagreement. First, overt lethal force did remain on the table throughout the years that followed, even though it was 83 not actually employed. The fact that no further strikes actually were launched did not reflect a belated conclusion that overt lethal force was not, or was no longer, a lawful option. Rather, the failure to launch reflected a persistent lack of actionable intelligence regarding bin Laden’s location (though some, particularly within the CIA’s Counterterrorist Center’s unit focused on bin Laden, felt that the intelligence was good enough at various points), colored by grave concerns regarding the larger consequences of further strikes that if unsuccessful could make the United States appear 84 feckless or that could in any event incur too much collateral damage. With those constraints in mind, it stands to reason that the covert rather than overt use of lethal force might have been viewed at the time as an attractive alternative to further cruise missile strikes. Depending on the instruments involved, after all, a covert use of force might be brought to bear more quickly than cruise missiles, and in any event deniability would allow the United States to circumvent some of the costs of a failed attack or one that caused collateral damage. Those features, in turn, might have lowered the threshold of reliability policymakers otherwise would insist upon, as a matter of policy preference, with respect to the intelligence on Military Action Against Terrorists Sites in Afghanistan and Sudan (Aug. 21, 1998), available at http://www.gpo.gov/fdsys/pkg/PPP-1998-book2/html/PPP-1998-book2-docpg1464.htm. 82. See 9/11 COMM. REP., supra note 70, at 117. The attack in Afghanistan was accompanied by an attack against a suspected al Qaeda affiliated facility in the Sudan (the Al-Shifa pharmaceutical plant). There has been considerable controversy ever since regarding the accuracy of the intelligence depicting the latter as a chemical weapons facility. See id. at 118 n. 50. 83. See id. at 130-131 (discussing consideration given to a cruise missile strike in December 1998). 84. See id. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 558 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 bin Laden’s whereabouts. Yet the Clinton administration was less clear about its willingness to use lethal force covertly than about its continued willingness to launch cruise missiles. In the aftermath of the embassy bombings, the Clinton administration changed its posture regarding the use of Afghan tribals to capture bin Laden, first proposed in 1998. There was now consideration given to 85 allowing the tribals to kill in at least some circumstances. The issue that has proven controversial is pinning down what those circumstances were and whether they changed over the course of the next two years. It appears that from the aftermath of the East African embassy bombings until the end of his term, President Clinton issued at least four separate MONs relating to bin Laden and that these memoranda varied in significant ways with respect to the extent to which they authorized use of lethal force. The first MON issued after the embassy bombings in August 1998 revived the capture operation using Afghan tribals, and even though the Administration at that time was preparing a cruise missile strike to kill bin Laden, the tribals were, nevertheless, directed to use deadly force solely in self-defense, and in fact were warned that they would not be paid if bin 86 Laden were killed. This standard was changed, however, in the next MON in December 1998. The new MON still did not explicitly give the Afghan tribals open-ended authorization to kill bin Laden. But though they were to prioritize bin Laden’s capture, this second MON provided that they would be permitted to use “lethal force” not just in self-defense but also in the event that they determined that the attempted capture “seemed impossible to complete successfully” – which was a distinctly foreseeable, perhaps 87 even highly likely, eventuality. The December 1998 MON, in short, was close to a de facto authorization to kill bin Laden. For at least a brief period, then, the CIA’s efforts were brought closer into line with the aims of the overt military alternative. It was a moment of convergence; both the CIA and the military were seeking to kill bin Laden, even if the CIA approach differed from the military’s in that the CIA held out hope for a live capture. Indeed, the underlying legal rationale likely was the same regardless of which instrument was being used: echoing General Counsel Sporkin’s 1984 analysis, the Clinton administration’s lawyers apparently had concluded that “under the law of armed conflict killing a person who posed an imminent threat to the United States would 88 be an act of self-defense, not assassination.” The moment of convergence did not last, however. The December 1998 MON was relatively closely held, which the 9/11 Commission 85. See id. at 131. See also BENJAMIN WITTES, LAW AND THE LONG WAR: THE FUTURE 19-21 (2008). See 9/11 COMM. REP. supra note 70, at 126-127, 131-132. Id. at 131. Id. at 132. OF JUSTICE IN THE AGE OF TERROR 86. 87. 88. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] 2/9/2012 3:56 PM LAW OF THE TITLE 10/TITLE 50 DEBATE 559 concluded contributed to the continued perception among many at the CIA that use of lethal force remained a taboo, justified only in cases of national 89 Meanwhile, two additional MONs were issued over the self-defense. course of the next year, and both retreated from the relatively flexible 90 language of the December 1998 MON. The first was designed primarily to extend the proxy force concept beyond the existing Afghan tribals to 91 The CIA include Ahmed Shah Massoud’s Northern Alliance forces. proposed that Massoud’s men be given authorization to use force under the 92 same relatively flexible terms as provided in the December 1998 MON. For reasons that are not clear, however, “[o]n this occasion . . . President Clinton crossed out the key language he had approved in December and 93 inserted more ambiguous language.” And the second 1999 MON, which involved still another proxy force, went even further in constraining the lethal option, as it actually used the original capture language taken from the restrictive August 1998 MON, which predated the East African embassy attacks. The December 1998 MON, of course, remained the controlling document for the original Afghan tribal proxy force. But the declining practical significance of that force, combined with the closely held status of its uniquely flexible grant of authority and the subsequent promulgation of a series of more restrictive MONs, left the impression among CIA officials that purposefully killing bin Laden was not truly part of the covert action alternative. An incident involving a Northern Alliance attack on bin Laden in early 2000 underlines the gap between what the military was being asked to do overtly and what the CIA was permitted to do covertly through its proxies. The CIA had learned that bin Laden might be present at the Derunta training camp near Jalalabad and duly notified Massoud, whose men had 94 established observation posts nearby. But rather than merely confirm bin Laden’s location or attempt a capture, Massoud took the initiative of dispatching a team armed with Katyusha rockets to bombard the camp. After Massoud reported this, CIA’s lawyers convulsed in alarm. The White House legal authorities that provided guidance for the new liaison with Massoud had not authorized pure lethal operations against bin Laden. 89. 90. 91. 92. 93. 94. See id. at 133. Id. at 133. See id. See id. Id. See COLL, supra note 38, at 491-492. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 560 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 . . . The CIA was legally complicit in Massoud’s operation, the 95 lawyers feared, and the agency had no authority to be involved. The CIA directed Massoud to cancel the mission, but it was too late for him 96 to recall the attack party. “Langley’s officers waited nervously” to see 97 what would happen next. “Some of them muttered sarcastically about the absurd intersections of American law and a secret war they were expected 98 to manage.” In the end, weeks would pass without word as to what had happened before the CIA finally learned that the men claimed to have fired 99 off their rockets without any discernible effect. The situation grew still more complex after the October 2000 bombing of the USS Cole. There was much discussion of launching a fresh round of overt military strikes against whatever al Qaeda-related targets could be found. But the intelligence linking al Qaeda to the Cole bombing was uncertain at best at that early stage, and President Clinton did not want to strike until al Qaeda’s responsibility was established more clearly, notwithstanding the fact that al Qaeda already had been the authorized 100 object for such attacks over the past two years. And when the Bush administration took office in early 2001, it too was reluctant to launch missiles against al Qaeda-related targets in response to the Cole attack. For both administrations, this no doubt reflected at least in part the absence of substantial targets, above all bin Laden himself. Some Bush officials also emphasized, however, that “too much time had passed” and that the Cole 101 bombing had become “stale.” Yet even as the overt military option waned, the covert option for using lethal force unexpectedly waxed. By March 2001, the new Bush administration’s NSC had directed the CIA to begin drafting new authorizations, including a MON that would entail “more open-ended 102 language authorizing possible lethal action in a variety of situations.” A draft was in place by the end of that month, but things stalled at this point 103 One was that officials wished to embed the new for two reasons. authorities in the context of a broader regional policy review that was still 104 underway. The other was an ongoing debate about the way the expanded 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. Id. See id. at 492-493. Id. at 493. Id. See id See id. at 193-195. Id. at 202. Id. at 210. See id. See id. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] 2/9/2012 3:56 PM LAW OF THE TITLE 10/TITLE 50 DEBATE 561 authority would relate to a new technology: armed unmanned aerial 105 vehicles (UAVs), better known as “drones.” The Air Force already had made the MQ-1 Predator available for use in Afghanistan for reconnaissance purposes, and its capacity to loiter in place while providing real-time video was a remarkable step forward in overcoming the intelligence gaps that had hampered the ability to project force within Afghanistan. To arm a Predator could have a game-changing impact, however, as this might collapse the time horizon for a missile strike from multiple hours to mere seconds once the decision to attack was 106 made. And by the spring of 2001 it was clear that this would soon be a viable option. Operating armed Predators would remove the proxy element from the CIA’s lethal operations, forcing attention to the functional convergence with the military that already was underway but which previously had been obscured by the intervening role of the proxies. Or so it seemed to the CIA Director George Tenet, at any rate. According to the 9/11 Commission report, Tenet clearly perceived this development in convergence terms: Tenet in particular questioned whether he, as Director of Central Intelligence, should operate an armed Predator. “This was new ground,” he told us. Tenet ticked off key questions: What is the chain of command? Who takes the shot? Are America’s leaders comfortable with the CIA doing this, going outside of normal military command and control? Charlie Allen [of the CIA] told [the Commission] that when these questions were discussed at the CIA, he and the Agency’s executive director, A.B. “Buzzy’ Krongard, had said that either one of them would be happy to pull the trigger, but Tenet was appalled, telling them that they had no 107 authority to do it, nor did he. By August 2001, the NSC Deputies Committee had “concluded that it was legal for the CIA to kill Bin Ladin or one of his deputies with the 108 Predator.” Questions remained as to when the armed Predator could be fielded and who would pay for it. But the path forward had been cleared of legal obstacles, and so the NSC again directed CIA to prepare new 109 authorities including the use of lethal covert action. It was the day before 9/11. 105. See id. at 211. 106. Id. at 421 (describing a four-hour window “from a presidential order to missile impact in Afghanistan”). 107. 9/11 COMM. REP., supra note 70, at 211. 108. Id. at 212. 109. See COLL, supra note 38, at 212-214. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 562 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 C. The CIA as a Combatant Command After 9/11 After 9/11, the U.S. government publicly asserted that a state of armed 110 conflict existed between it and al Qaeda. In the months that followed, the resulting kinetic action was concentrated in Afghanistan, where the bulk of al Qaeda’s leadership and personnel happened to be and where the U.S. military acting through U.S. Central Command (CENTCOM) of course played the most visible role. But though CENTCOM remains deeply engaged in combat operations in Afghanistan to this day, the conflict is not (and from the beginning has not been) confined geographically to Afghanistan nor institutionally to CENTCOM. All along, there has been a “shadow” component to the conflict with al Qaeda, waged at times without formal acknowledgement by the U.S. government (though rarely without detection) in a variety of locations. The CIA has played a central role in this shadow war, serving not only as a source of HUMINT and covert logistical support for the actions of the military but also as a warfighter – a veritable combatant command – in its 111 own right. This development marks a sharp break from the hemming and hawing over the propriety of the CIA’s indirect involvement in the use of lethal force in the 1980s and 1990s and a substantial indicator of the post112 9/11 convergence of military and intelligence operations. The change has been described as a “fundamental transformation” of 113 The CIA’s Counterterrorism Center (CTC), the CIA as an institution. responsible for managing the CIA’s kinetic operations, has grown immensely in terms of budget and personnel since 9/11, and today some twenty percent of the CIA’s analysts function as “targeters,” whose primary task is to identify or locate specific individual targets, who may then be 114 attacked by a CIA-operated drone. Some lament that the “CIA now functions as a military force” that lacks the accountability structures 115 associated with the armed forces. In the words of former Director Michael Hayden, “CIA has never looked more like its direct ancestor, the 116 OSS, than it does right now.” 110. That determination has been the subject of considerable legal and policy controversy ever since. The merits of that debate are beyond the scope of this article. 111. See, e.g., Miller & Tate, supra note 11. 112. See id. “One former senior U.S. intelligence official described the agency’s paramilitary transformation as ‘nothing short of a wonderment.’ . . . ‘You’ve taken an agency that was chugging along and turned it into one hell of a killing machine,’ said the former official. . . .” Id. 113. Id. 114. Id. 115. Id. “‘We’re seeing the CIA turn into more of a paramilitary organization without the oversight and accountability that we traditionally expect of the military,’ said Hina Shamsi, the director of the National Security Project of the American Civil Liberties Union.” Id. 116. Siobhan Gorman, 9/11 A Decade After: Drones Evolve Into Weapon in Age of 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] LAW OF THE TITLE 10/TITLE 50 DEBATE 2/9/2012 3:56 PM 563 The transformation occurred quickly after 9/11. Tenet had presented what amounted to a CIA war plan to President Bush at a meeting at Camp David just four days after the attack. His proposal was sweeping, and included a request for “exceptional authorities” both to kill and to detain al Qaeda targets on a global basis. “It would give the CIA the broadest and most lethal authority in its history,” Woodward wrote, “a secret global war 117 on terror.” Significantly, Tenet requested that the authorization be broadly framed, providing programmatic approval rather than making it necessary to return to the President again and again to obtain specific 118 authorizations for particular actions. Bush agreed, reportedly signing an order on September 17th that formally modified Reagan’s 1986 counterterrorism finding and superseded the interim modifications of the 119 Clinton years discussed above. Going forward, CIA was authorized “to kill or capture Qaeda militants around the globe,” as paraphrased in media 120 reports. In contrast to the uncertainties associated with the MONs issued between 1998 and 2000, after 9/11 the use of lethal force was unambiguous. “My last meeting with [the head of the CIA’s Counterterrorism Center] before I left was interesting,” said Gary Schroen, who spearheaded the 121 initial CIA contingent to enter Afghanistan after 9/11. He basically said to me: “I want to make it clear what your real job is. All these other things – linking up with the Northern Alliance, preparing the battlefield, helping the special forces get in or whatever happens – is fine. But once the Taliban are broken, your job is to find bin Laden, kill him and 122 bring his head back on ice.” Terror – Intelligence Services Overcome Philosophical, Legal Misgivings Over Targeted Killings, WALL ST. J., Sept. 8, 2011, at A6. 117. BOB WOODWARD, BUSH AT WAR 76, 78 (2004). 118. See id. at 76. 119. See id. at 101. 120. Eric Schmitt & Mark Mazzetti, Secret Order Lets U.S. Raid al Qaeda in Many Countries, N.Y. TIMES, Nov. 10, 2008, at A1 (emphasis added). “A secret document known as a ‘presidential finding’ was signed by President George W. Bush that same month, granting the agency broad authority to use deadly force against bin Laden as well as other senior members of al-Qaeda and other terrorist groups.” Joby Warrick, CIA Assassin Program Was Nearing New Phase: Pannetta Pulled Plug After Training was Proposed, WASH. POST, July 16, 2009, at A1; see also Joby Warrick & Ben Pershing, CIA Had Program To Kill Al-Qaeda Leaders: Agency Didn’t Tell Congress About Bush-Era Plan To Use Assassins, WASH. POST, July 14, 2009, at A2. 121. Interview with Gary C. Schroen by PBS Frontline’s The Dark Side, Jan. 20, 2006, available at http://www.pbs.org/wgbh/pages/frontline/darkside/interviews/schroen.html, [hereinafter Schroen]. 122. Id. Schroen’s colleague Gary Berntsen, who later took command of the CIA’s paramilitary operations in Afghanistan, shared a similar story: “Gary Berntsen was working at the CIA’s counterterroris[m] center in October 2001 when his boss summoned him to the front office and told him, ‘Gary, I want you killing the enemy immediately.’ Berntsen left 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 564 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 Notwithstanding the emphasis on lethal force, the looming war initially unfolded in a relatively conventional manner in terms of the CIA’s role. CIA officers were the first Americans to enter Afghanistan after 9/11. The Agency had existing links to the Northern Alliance and had done extensive planning already with respect to a potential intervention on the Northern Alliance’s behalf. The CIA was also far more nimble than the Pentagon, which had not planned for this situation and apparently had trouble agreeing on which component of the SOF community ought to take the lead 123 initially. But the immediate task for the CIA officers was distinctly in the nature of a support mission, involving efforts to leverage the capacities of Afghan allies and to prepare the way for the impending arrival of the 124 military (though CIA officers on an episodic basis did take a direct hand in the fighting, alongside U.S. military and Afghan forces, and the CIA did 125 have armed Predator drones in theater). From this point of view, the broad authority in the September 17, 2001, order permitting the CIA to use lethal force against al Qaeda targets appeared more interesting on paper than in practice. Or at least that is how things looked from within Afghanistan. Beyond its borders, matters were different. From the beginning of the war with al Qaeda, the CIA has acted under color of the September 17, 2001 finding in a wide variety of locations, covertly exercising three sets of powers that were identical to those contemporaneously being exercised overtly against the same enemy by the U.S. military: detention without criminal charge, the use of proxy forces to conduct lethal operations (in an extension of the model debated throughout the next day for Afghanistan. . . . His primary target was bin Laden. . . .” Tora Bora Revisited: How We Failed to Get Bin Laden and Why It Matters Today: Report to Members of the S. Foreign Relations Comm. 111th Cong. 7 (2009), available at http://www.gpo.gov/ fdsys/pkg/CPRT-111SPRT53709/html/CPRT-111SPRT53709.htm. 123. See Schroen, supra note 121. “I said: ‘Reach out to these guys. Let’s talk to the SEALs. Let’s talk to Delta. Let’s talk to SOCOM [Special Operations] Command. Let’s talk to CENTCOM. Anybody you know, let’s invite. We need to have a military officer, a special operations guy, come along with us.’ Everybody that he talked to said: ‘God, I want to go. I’d go myself, but we can’t get the bosses to agree to even which special operations group is going to take the lead in this.’ It just seemed like total confusion there, and so we packed up and got ready to go. . . . It took several weeks before that sorted itself out.” Id. See generally GARY C. SCHROEN, FIRST IN: AN INSIDER’S ACCOUNT OF HOW THE CIA SPEARHEADED THE WAR ON TERROR IN AFGHANISTAN (2005) [hereinafter SCHROEN, FIRST IN]. 124. Schroen, supra note 121. 125. On the episodic involvement of CIA officers in combat, see SCHROEN, FIRST IN supra note 123, at 253, 292. On the presence of CIA-operated drones in the Afghan theater, see WOODWARD, supra note 117, at 289. Note that it is unclear from Woodward’s account whether or how often the CIA carried out drone strikes in Afghanistan, though it is clear that the strikes were used to collect intelligence in support of airstrikes carried out by the U.S. military. See id. at 211. Remarkably, the military was much slower to get Predators into the theater, and did not arm their Predators before sending them. See id. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] LAW OF THE TITLE 10/TITLE 50 DEBATE 2/9/2012 3:56 PM 565 the 1980s and 1990s), and the direct use of lethal force in the form of drone strikes. Consider first the CIA’s exercise of detention authority. At much the same time that the military was constructing a system for detention of “enemy combatants” (both within Afghanistan and at Guantanamo Bay), the CIA was constructing a parallel detention system at an array of undisclosed locations, a system that rested in an immediate sense upon the presidential finding described above but that ultimately relied upon the 126 same law of war arguments that were used to justify the military’s system. Indeed, over time there appears to have been some traffic between those two systems. Detention is not the most vivid or lasting example of militaryintelligence convergence, however. Many aspects of the CIA’s post-9/11 detention practices remain secret, after all. In any event, President Obama ordered the termination of the CIA’s detention program soon after taking 127 office in January 2009. (As of 2006, President Bush already had largely, 128 if not entirely, ceased to rely on it.) The CIA’s involvement in the use of lethal force against al Qaeda, in contrast, has scaled upward dramatically in recent years. As described above, the CIA’s potential involvement in the use of lethal force against terrorists in the 1980s and 1990s seemed consistently to involve the use of a proxy force of indigenous allies, capable of acting with greater freedom of action in denied areas and entailing a degree of plausible deniability, perhaps, should their actions prove problematic. These advantages to the proxy force concept did not disappear after 9/11, no matter how free the CIA became (thanks to technological and other developments) to use deadly force in its own right. And thus it should not have come as a surprise when Bob Woodward asserted in 2010 that the CIA had evolved well beyond the counterterrorism action team concept of 1986, 126. For several years, the CIA detention program – involving at least some detainees captured and held outside of Afghanistan – was kept secret. But in November 2005, the Washington Post reporter Dana Priest exposed the existence of the “black sites” in a dramatic article. Dana Priest, CIA Holds Terror Suspects in Secret Prisons: Debate Is Growing Within Agency About Legality and Morality of Overseas System Set Up After 9/11, WASH. POST, Nov. 2, 2005, at A1. In 2006, in an address to the nation, President Bush ultimately acknowledged the existence of the CIA detention program and announced his decision to transfer the remaining detainees from CIA to military custody. See Dana Priest, Officials Relieved Secret Is Shared, WASH. POST, Sep. 7, 2006, at A17; Dafna Linzer & Glenn Kessler, Decision To Move Detainees Resolved Two-Year Debate Among Bush Advisors, WASH. POST, Sept. 8, 2006, at A1. 127. See Exec. Order No. 13,491, Ensuring Lawful Interrogations, 74 Fed. Reg. 4893 (Jan. 22, 2009) (“The CIA shall close as expeditiously as possible any detention facilities that it currently operates and shall not operate any such detention facilities in the future.”). 128. See Priest, Officials Relieved Secret Is Shared, supra note 126 (“‘Although there is no one in CIA custody today, it’s our intent that the CIA detention program continue,’ said a senior intelligence official. ‘It’s simply been too valuable in the war on terrorism to not allow it to move forward.’”). 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 566 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 and had established a “3,000 man covert army in Afghanistan” consisting 129 “mostly of Afghans, the cream of the crop in the CIA’s opinion.” Called Counterterrorism Pursuit Teams (CTPT), according to Woodward, these forces “were a paid, trained and functioning tool of the CIA” that carried out “lethal and other operations” such as “kill[ing] or captur[ing] Taliban 130 insurgents” or going into “tribal areas to pacify and win support.” In Woodward’s account, the CTPTs amounted to a regiment-sized armed force operating under the ultimate command of the CIA Director, originally focused on combat and other operations in Afghanistan but later providing a rare capacity for projecting boots-on-the-ground force into Pakistan as 131 well. All that said, the CIA’s kinetic turn is best embodied by its creation of an extraordinary capacity to wage an air campaign using armed drones. The key development here was the timely maturation and proliferation of UAVs (first the MQ-1 Predator and then later the MQ-9 Reaper) equipped with increasingly reliable and discrete weapon systems (such as the AGM114 Hellfire air-to-surface missile). This enables the CIA to project force in locations where it would be far more difficult, if not impossible, to carry out commando-style raids using either CIA officers or proxy forces. ArmedUAVs can be maintained and launched from more accessible areas, can loiter over potential targets for an extended period (thus providing better intelligence as well as the ability to tailor the precise moment of an attack in a manner that might reduce collateral damage), pose no risk to American or allied personnel (and thus no need to establish and maintain a combat search-and-rescue capacity), and may be perceived as less intrusive than ground troops from the perspective of host governments or populations (though that is not to say that they would not also cause sovereignty concerns). Drones may lack plausible deniability – particularly in contrast 132 to the CTPTs – but these other qualities over time have proven to be more than adequate compensation. 129. WOODWARD, supra note 117, at 8. 130. Id. at 8, 52; see also id. at 355 (asserting that the CTPTs were conducting “multiple raids every night around Kandahar”); Miller & Tate, supra note 11 (“[T]he purpose of the Counterterror Pursuit Teams is a source of disagreement among senior officials in government. ‘They can fire in self-defense, but they don’t go out to try and kill a target,’ a U.S. official familiar with CIA operations in Afghanistan said. ‘They’re mostly arresting people and turning them over to’ the Afghan security services. But the former senior U.S. military official said the teams’ objectives were ‘more kill-capture’ than capturekill. . . . In some cases, the pursuit teams used more indiscriminate means, including land mines, to disrupt insurgent networks, the former official said.”). 131. See WOODWARD, supra note 117, at 367. 132. Miller & Tate, supra note 11 (noting, with reference to the CTPTs, that “[g]iven the scope of the CIA’s paramilitary activities, human rights groups say the death toll over the past decade from CIA-directed operations undoubtedly exceeds the casualty count associated with strikes from drones. U.S. intelligence and congressional officials insist that the number of people killed in CIA operations outside the drone campaign is negligible, but say they 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] LAW OF THE TITLE 10/TITLE 50 DEBATE 2/9/2012 3:56 PM 567 The first reported CIA drone strike occurred in November 2002, in Yemen. Al Qaeda had long had a substantial presence there, with its most 133 notable operation being the attack on the USS Cole just two years earlier. In this instance, the target was Qaed Salim Sinan al-Harethi (a.k.a. Abu Ali al-Harethi), the senior al Qaeda figure in Yemen and suspected mastermind of the attack on the Cole. Al-Harethi was traveling through a remote region of Yemen, packed into a vehicle with five colleagues including Kamal Derwish (a.k.a Ahmed Hijazi, an American citizen believed to have recruited the so-called Lackawanna Six to attend an al Qaeda training camp 134 in 2001). Unbeknownst to al-Harethi, he was entering a trap. A Predator drone circled overhead, transmitting a live feed both to CIA headquarters in Langley and to an operations center in the tiny east African nation of Djibouti, from which the Predator was controlled. Then-CIA Director George Tenet “gave a nod,” the command was transmitted to the controllers 135 in Djibouti, and the Predator fired. All in the vehicle were killed. This first drone strike made headline news, and U.S. officials were quick to offer legal justifications for the “covert” attack. It had taken place “with the approval and cooperation of Yemen’s government,” unnamed officials told reporters, and was appropriate both because al-Harethi and his colleagues were “‘combatants’ under international law” and the strike in any event could be viewed as an “act of self-defense . . . permitted under 136 the international laws of war.” The unspoken premise, of course, was that the CIA was fighting war against al Qaeda with the military, relying on the same ultimate justifications for using lethal force. The CIA and the military found themselves targeting not only the same enemy using the same legal rationale, but also using the same weapons platform. Nearly two years would pass before the CIA had the occasion to carry out another drone strike, at least insofar as we can tell from the public record. This might indicate that “assassination” concerns remained influential in the minds of the government’s lawyers and policymakers. It might indicate a lack of access to the intelligence necessary to launch a similar strike, as the laborious process of developing a supporting network 137 of on-the-ground HUMINT sources continued. It might simply indicate have never seen an agency-produced casualty count that includes other categories of operations”). 133. For a review of al Qaeda’s history in Yemen, see Robert Chesney, supra note 13, at 3. 134. See DINAH TEMPLE-RASTON, THE JIHAD NEXT DOOR: THE LACKAWANNA SIX AND ROUGH JUSTICE IN THE AGE OF TERROR 195 (2007). 135. See id. at 196. 136. See Dana Priest, CIA Killed U.S. Citizen in Yemen in Missile Strike: Action’s Legality, Effectiveness Questioned, WASH. POST, Nov. 8, 2002, at A1. 137. On the necessity of strong HUMINT sources to inform drone operations, see WOODWARD, supra note 117, at 106-107 (describing critical role played by local informants 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 568 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 lack of access to armed UAVs in the right places and at the right times, or a lack of targets in locations where a drone strike would be deemed diplomatically or legally appropriate from the point of view of sovereignty concerns. Some or all of the above might be true. But whatever the case, the logjam began to break in the summer of 2004 with the killing of Nek Muhammad, a Pakistani extremist and former Taliban fighter who was in 138 hiding in Pakistan’s South Waziristan region. According to an analysis published by the New America Foundation, two more drone strikes in Pakistan’s Federally Administered Tribal Areas (FATA) followed in 2005, with at least two more in 2006, four more in 139 2007, and four more in the first half of 2008. The pattern was halting at best. Yet that soon changed. U.S. policy up to that point had been to obtain 140 Pakistan’s consent for strikes, and toward that end to provide the 141 Pakistani government with advance notification. But intelligence suggested that on some occasions “the Pakistanis would delay planned strikes in order to warn al Qaeda and the Afghan Taliban, whose fighters 142 would then disperse.” A former official explained that in this environment, it was rare to get permission and not have the target slip away: “If you had to ask for permission, you got one of three answers: either ‘No,’ or ‘We’re thinking about it,’ or ‘Oops, where did the target 143 go?’” Declaring that he’d “had enough,” Bush in the summer of 2008 “ordered stepped-up Predator drone strikes on al Qaeda leaders and specific camps,” and specified that Pakistani officials going forward should receive only “‘concurrent notification’ . . . meaning they learned of a strike as it 144 was underway or, just to be sure, a few minutes after.” Pakistani 145 permission no longer was required. The results were dramatic. The CIA conducted dozens of strikes in in supporting drone operations in Pakistan). 138. See David Rohde & Mohammed Khan, Ex-Fighter for Taliban Dies in Strike in Pakistan, N.Y. TIMES, June 19, 2004, at A6; The Year of the Drone: An Analysis of U.S. Drone Strikes in Pakistan, 2004-2011, NEWAMERICA.NET, http://counterterrorism. newamerica.net/drones#2011chart (attributing the attack on Nek Muhammad to a U.S.operated drone) [hereinafter U.S. Drone Strikes in Pakistan]. 139. See U.S. Drone Strikes in Pakistan, supra note 138. 140. JOBY WARRICK, THE TRIPLE AGENT : THE AL-QAEDA MOLE WHO INFILTRATED THE CIA 13 (2011); see also ERIC SCHMITT & THOM SHANKER, COUNTERSTRIKE: THE UNTOLD STORY OF AMERICA’S SECRET CAMPAIGN AGAINST AL QAEDA 118-119 (2011) (describing Director Hayden and DNI McConnell’s efforts to persuade President Musharraf to permit an expanded U.S. combat presence in Pakistan, and Musharraf’s agreement to permit CIA drones to strike targets beyond specifically-identified leaders). 141. See WOODWARD, supra note 117, at 4. 142. See id. at 4. 143. WARRICK, supra note 140, at 13. 144. WOODWARD, supra note 117, at 5 (emphasis in original). 145. WARRICK, supra note 140, at 13. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE ) 2012] LAW OF THE TITLE 10/TITLE 50 DEBATE 2/9/2012 3:56 PM 569 Pakistan over the remainder of 2008, vastly exceeding the number of strikes 146 over the prior four years combined. That pace continued in 2009, which 147 eventually saw a total of 53 strikes. And then in 2010, the rate more than 148 doubled, with 188 attacks (followed by 56 more as of late August 2011). The further acceleration in 2010 appears to stem at least in part from a meeting in October 2009 when President Obama granted a CIA request both for more drones and for permission to extend drone operations into areas of Pakistan’s FATA that previously had been off limits or at least 149 discouraged. Whatever the cause, the fact is that the CIA has directed the use of lethal force from armed drones in Pakistan more than 300 times over the past three years, or nearly once every three days. Thus it was no surprise to hear Director Panetta make such martial claims as his 2009 statement that the CIA is “the point of the spear” in the hunt for al Qaeda’s top 150 leadership. Nor was it surprising, on this level at least, to find the director of the CIA’s Counterterrorism Center (CTC) asserting that “[w]e are killing 151 these sons of bitches faster than they can grow them now.” Such claims normally would be spoken by combatant commanders, if by anyone. But in light of the frontline role that the CIA and its CTC have come to play in the shadow war with al Qaeda, characterizing these officials as “combatant 152 commanders” in their own right might not be too far off the mark. With each additional sortie, the CIA’s functional similarity to a conventional military conducting an air campaign has grown. Like a military commander, the CIA’s Director now routinely decides whether to launch missiles to kill various targets, balancing the advantage to be gained with 153 the risks (including the risk of collateral damage). 146. See id. at 13. 147. See U.S. Drone Strikes in Pakistan, supra note 138. 148. U.S. Drone Strikes in Pakistan, supra note 138. The figure of 118 strikes in 2010 also appears in Miller & Tate, supra note 11. 149. See WOODWARD, supra note 120, at 208-209; WARRICK, supra note 140, at 91-92. It may also have mattered that Director Panetta in May 2010 urged the Chief of Staff of Pakistan’s Army, Ashfaq Kayani, to consent to drone operations in a southern region of the FATA in which Pakistani armed forces were present. According to Woodward, Kayani replied that “he would see that they had some access.” WOODWARD, supra note 117, at 366367. 150. WARRICK, supra note 140, at 91. 151. See Miller & Tate, supra note 11. The quote was in response to a question from a fellow CIA officer inquiring about the pace of drone strikes in 2010. See id. 152. See SCHMITT & SHANKER, supra note 140, at 102-103 (describing the CIA Director as “America’s combatant commander in the hottest covert war in the campaign against terror”). 153. Warrick’s account contains several vignettes in which former CIA Directors Michael Hayden and Leon Panetta exercise this authority. See, e.g., WARRICK, supra note 140, at 15-16, 87-88. 014_CHESNEY V14 1-19.DOCX (DO NOT DELETE) 570 2/9/2012 3:56 PM JOURNAL OF NATIONAL SECURITY LAW & POLICY [Vol. 5:539 To be fair, the CTC does far more than just drone strikes when it comes to counterterrorism, and its other operational activity is much less akin to 154 military activity. The drone program plainly has become a high-intensity operation, however, and is unlikely to abate much in the near term. Indeed, it is if anything likely to expand to other theaters, in some of which it will compliment or even replace existing military efforts. Expansion to Yemen, in fact, has already begun. As noted above, the CIA did conduct a drone strike in Yemen in 2002. But in contrast to Pakistan, this attack did not mark the beginning of a sustained air campaign, let alone one commanded by the CIA. It was not until 2009, so far as the public record indicates, that the U.S. government began again to use lethal force in Yemen, and though it has done so on many occasions in the years that followed, it appears that these latest strikes were until recently carried out exclusively by the U.S. military (using a combination of manned 155 aircraft, cruise missiles, and military-owned armed UAVs). This began to change in the summer of 2011. That August, The Washington Post reported that the CIA would soon resume drone strikes...
Purchase answer to see full attachment