Exporting Armed Drones – The United States Sets Policy

Barry Kellman
July 24, 2015

The U.S. State Department recently announced a new policy for exports of military drones (unmanned aerial vehicles).[1] Military drones are today’s most sophisticated tools for aerial surveillance, capable of persistent and distant overflight of any terrain. While not all military drones can fire weapons, armed drones have generated controversy because of their prominent role in targeted killings of foreign and American supporters of terrorist organizations.

The new policy raises two distinct sets of legal issues. First are issues pertaining to the State Department’s determination that the new policy satisfies the United States Conventional Arms Transfer (CAT) Policy[2] and guidelines of the Missile Technology Control Regime (MTCR).[3] Second are issues pertaining to whether using armed drones for targeted killings is legal in international law. This essay concludes that various reasons support the new policy, but a wider set of concerns about the legality of targeted killings using remotely operated weapons remain unaddressed.

Policy Consistency with Export Control Obligations

United States CAT Policy sets forth thirteen criteria for balancing support for weapons transfers that meet our allies’ legitimate security requirements with restraint on weapons transfers that would enhance hostile states’ military capabilities, facilitate human rights abuses, or otherwise undermine international security. Export of armed drones can certainly be seen to satisfy our U.S. allies’ security requirements. These weapons are designed to destroy ground targets with remarkable accuracy using the world’s best surveillance mechanisms. Operated remotely, no pilot is at risk. Armed drones have the additional advantage of being able to hover over a possible target for extended periods, waiting to fire when the target reveals itself or separates from non-targets.

Two especially pertinent CAT Policy criteria are the impact of approving or denying the transfer on the U.S. defense industrial base and the availability of comparable systems from foreign suppliers. Military drones represent the leading edge of the “most dynamic growth sector of the world aerospace industry this decade.”[4] The United States is the world’s leader in drone technology, featuring General Atomics’ Reaper and Northrup Grumman’s Global Hawk. Israel is a major competitor,[5] and the Chinese have developed a drone competitive with the Reaper.[6] A European consortium is considering building this kind of drone, and the Russians have shown a mock-up of a similar system.[7] India, Pakistan, and Turkey are reportedly developing armed drones, and Iran may have armed drones with a range of 2,000 km.[8] This list may indeed be longer as proliferation of drone technology renders their development and deployment less challenging.

Prior to the new policy, uncertainty about the Administration’s approach to exporting military drones provoked concerns that the United States would abandon its technological leadership.[9] Significantly, “military drones” refers not only to the vehicle itself but to the remote command system that receives and analyzes data and makes operational decisions. The new policy, by clarifying how export applications for the range of technologies that comprise such a system will be scrutinized, encourages American industry to compete in this lucrative global market.

Moreover, the prospect of expanding demand for military drones being satisfied by foreign suppliers would risk abandoning this market to states that might not uphold United States’ restraints on arms exports. For example, Israel, the chief competitor in this market, is not a member of the MTCR nor the Chemical and Biological Weapons Conventions and might therefore be less legally restrained from selling armed drones to questionable users. By contrast, United States export of military drones can foster military-to-military cooperation, thereby extending and entrenching influence on recipient states.

While dozens of nations may be eligible to buy military drones,[10] the new policy imposes significant restrictions. First, consistent with the MTCR, the policy establishes that exports of large (Category One) armed drones capable of carrying a 500 kg warhead at least 300 km, including the Reaper and the Global Hawk, should be met with “a strong presumption of denial.”[11] However, the MTCR affords States greater discretion in approving exports of smaller or shorter-range (Category Two) missile systems, such as cruise missiles, that have been viewed as less strategically threatening than long-range ballistic systems. MTCR members have recently debated whether some types of armed drones should be moved from Category One to Category Two, but no action has been taken.[12] Looking forward, tomorrow’s armed drones will likely carry lighter weight precision directed weapons, which means they may not qualify as Category One missiles, rendering inapplicable the strong presumption of denial.

The State Department also included within the new policy four criteria that apply to how exported military drones will be used. They must be:

(1) operated “in accordance with international law, including international humanitarian law (IHL) and human rights law,”

 (2) used only “when there is a lawful basis for use of force under international law,”

(3) not used “to conduct unlawful surveillance or use unlawful force against domestic populations,” and

(4) used only by technically and doctrinally trained operators so as “to reduce the risk of unintended injury or damage.”[13]

While these criteria clearly raise many questions for case-by-case application, they signal a sensitivity to the fact that drones, especially armed drones, are controversial in international law and that there is no firm legal doctrine to govern use of these new weapons. Indeed, the State Department, recognizing the ambiguity of legal standards, announced that its new policy is part of a broader review which “includes plans to work with other countries to shape international standards” for the use of military drones.[14] It remains to be seen whether international standards will be shaped so as to be relevant to military drone export applications.

Controlling The Use of Armed Drones for Targeted Killings

International law’s consideration of targeted killings was, not long ago, something of an arcane topic if only because of the sheer rarity of accomplishing the deed. But armed drones take targeted killings from the exception to the weekly (at least) reality. So viewed, armed drones are not just another type of missile delivery system. They are designed to search for and kill individuals, which they do effectively, cheaply, and without significant risk to the attacker. Moreover, armed drones augur an imminent future when technological advances will enable individuals to be killed with remarkable precision virtually anywhere on Earth,[15] literally without knowing what hit them and entirely without due process. The prospect of flotillas of aerial missile launchers selectively targeting individuals with lethal force—even those who might deserve such fate—represents a crossroads in the use of armed force and, accordingly, in international law.

The leading decision on the legality of targeted killing is the Israeli Supreme Court’s decision in Public Committee Against Torture in Israel v. The Government of Israeli, which held that while targeted killing of terrorists may be legal under the law of armed conflict, the discretionary use of force is not unlimited. It is within the judiciary’s purview to address questions as to whether a program of targeted killings satisfies international legal constraints on the use of force. President Beinisch’s concurrence elaborated three important criteria: (1) the information about the target “must be well based, strong, and convincing regarding the risk the terrorist poses to human life”; (2) the damage to innocent civilians must not be disproportionate to the military benefit; and (3) “‘targeted killing’ is not to be carried out when it is possible to arrest a terrorist.”[16]

The remainder of this essay considers the export of armed drones in light of these criteria.

Strong Informational Support and The Question of Authority

The potential for mistaken intelligence deserves attention. Even if an armed drone attack harms only the intended target, the target may have been undeservingly selected. The entire operation of drones is based on intelligence, and sometimes intelligence may be faulty. It may be a sincere misjudgment, faulty transmission of intelligence, or institutional bias having infected the decisional metrics—in any event, someone might be killed who did not do what his executioners believed, or at least did not deserve to die. Subsequent rectification of the loss is, of course, impossible.

Notably, the United States’ targeted killings outside of Afghanistan and Iraq have been operations of the intelligence community, not the military. It may be reasonable to expect that recipients of exported armed drones will imitate this allocation of authority over life-and-death to their intelligence communities. Looming is the prospect of a strategic environment in which multiple intelligence networks, the most secretive and unaccountable parts of government, share information about terrorists and, on the basis of that information, summarily kill them without the bother and expense of prosecution—arguably a major victory for security but a bit too Orwellian for comfort.

All this raises questions about how targeted killings are authorized and whether there should be juridical intervention in the authorization process. There have been proposals in the United States for a targeted killing warrant, perhaps issued by a specialized tribunal, but these proposals have not yet been accepted.[17] The new policy on export of armed drones signifies that the question of juridical intervention is no longer a matter exclusively for the United States. Unsurprisingly, nothing in the State Department’s new policy would require recipients of exported armed drones to implement a warrant for targeted killings.

Non-Combatant Casualties Under International Humanitarian Law

Under the IHL principle of distinction, an attacker must use every available means to avoid harm to non-combatants, targeting only legitimate enemy personnel consistent with a lawful military objective.[18] By this metric, armed drones’ surveillance and precision guidance capabilities enable an attack to be selectively executed far more precisely than with cruise missiles or aerial bombers.

 Moreover, armed drones will not linger after the conflict. Unlike landmines, they will not explode if later discovered by a child. Nor is likely that there will be an aftermarket where terrorists can acquire armed drones. Armed drones require an infrastructure of complex information management systems backed by timely intelligence. Disconnected from remote operators who receive and transmit necessary data, armed drones are harmless.

Critics respond that reliance on increasingly autonomous weapons is a slippery slope to dehumanization of warfare whereby IHL proscriptions against inflicting civilian casualties might be ignored by machines incapable of subtly assessing complicated battlefield conditions.[19] Also troubling is the potential for armed drones to spawn chemical-biological weapons (CBW) proliferation. A guidable hovering aircraft equipped for pinpoint warhead delivery would be ideal for disseminating chemical or biological agents.[20]

The strongest objection to armed drones focuses on how their persistence and range enables more attacks in remote areas, causing more—not fewer—civilian casualties. Heretofore, military leaders might have declined to fire cruise missiles at a target due to an unacceptable risk of ancillary casualties, but armed drones’ targeting capability undercuts such hesitation.

The available data on civilian casualties from armed drone attacks is troubling. Allied forces have flown over 1,000 armed drone attacks in Afghanistan, where, according to the United Nations Assistance Mission in Afghanistan (UNAMA), civilian deaths in 2013 totaled 45, representing a third of all civilian deaths due to air strikes.[21] But in Pakistan, Yemen, and Somalia, i.e., outside any explicit theater of armed conflict, drone attacks have reportedly killed at least 3,000 people, including 500–1000 civilian casualties including as many as 200 children.[22] And in Gaza in 2014, Israel conducted over 40 armed drone attacks; while Israel boasted of its “highly discriminating” use of advanced technology to minimize civilian losses, critics have alleged that many civilians were, in fact, killed or injured.[23]

Ultimately, the problem here may be one of accountability. It is difficult to assess the numbers of victims of attacks against targets outside combat zones in the world’s most secretive hiding places. [24] Moreover, there is no consensus on how to categorize victims: the United States, for example, counts as a combatant any adult male in the target’s immediate vicinity.[25] But most of the problem has to do with post-attack access to data. Armed drones generate a record of their use; barriers to such review are due to official secrecy, not technology. If access is opened, the record of how the armed drone was used can be available. However, no nation with armed drones has been willing to release data concerning civilian casualties from drone strikes.[26]

The Obligation To Arrest Terrorists When Possible

Law enforcers must consider and employ every means to capture a target before resorting to the use of lethal force. They may not, of course, summarily kill suspected criminals, no matter how heinous the crime. Only where the law enforcement officer’s life or others’ lives are in imminent jeopardy should lethal force be used. This constraint on the use of force does not apply, however, under the laws of armed conflict which permit use of lethal force unless the adversary surrenders. Lethality, international humanitarian law tells us, is inherent in armed conflict. The logic has been the near-impossibility of apprehending enemy combatants without significantly risking the lives of one’s own forces.

But armed drones may offer, in comparison with other missile delivery systems, wider opportunities to arrest rather than kill, especially if equipped with emerging incapacitants, so-called because they are designed to temporarily impede a target’s freedom of movement. There are projectile nettings for ensnaring a target, even a vehicle; electrical pulse weapons can be used to stun the target; directed energy beams can disable vehicles’ power systems. [27] Various foams can stick to a target or make traction impossible—in either case, the target cannot move. Darts can temporarily paralyze a target.[28]

Incapacitants are not without controversy,[29] and, in most cases, an incapacitated terrorist must still be extracted, perhaps putting troops at risk. A question in this context is whether the “virtues” of armed drones—precision, persistence, and range—offer opportunities to use incapacitants for capture instead of killing. If so, then a strike against terrorists should be more aptly governed by the rules on how law enforcers use force. Consideration could be usefully devoted, therefore, to specifying the circumstances where the use of nonlethal force should be legally mandated lest a targeted killing be deemed an extrajudicial execution.


The new policy on the export of military drones may be seen to advance United States’ military and trade policies, sustaining technological supremacy and offsetting the influence of competitors that may sell them more wantonly. More broadly viewed, however, American exports of armed drones will enter an anarchic global market lacking well-developed and broadly accepted legal constraints. The new policy means that longstanding debates over armed drones, previously concentrating almost entirely on how the United States uses lethal force, are now definitively international in scope. The new export rules reveal the need for international law to address the vast and nuanced challenges posed by emerging weapons.

About the Author: Barry Kellman, an ASIL member, is Professor of Law and Director of the International Weapons Control Center, at DePaul University College of Law


[1] Fact Sheet, U.S. Export Policy for Military Unmanned Aerial Systems, U.S. Dep’t of State (Feb. 17, 2015), http://www.state.gov/r/pa/prs/ps/2015/02/237541.htm, [hereinafter State Dep’t Fact Sheet].

[2] Presidential Policy Directive – United States Conventional Arms Transfer Policy, (PPD-27) (Jan. 15, 2015), available at https://www.whitehouse.gov/the-press-office/2014/01/15/presidential-policy-directive-united-states-conventional-arms-tra....

[3] MTCR Guidelines and the Equipment, Software and Technology Annex, http://www.mtcr.info/english/guidelines.html (last visited May 21, 2015).

[4] Teal Group Predicts Worldwide UAV Market Will Total $91 Billion in Its 2014 UAV Market Profile and Forecast, Teal Group Corporation (July 17, 2014, 11:01 AM), http://www.tealgroup.com/index.php/about-teal-group-corporation/press-releases/118-2014-uav-press-release.

[5] Mary Dobbing & Chris Cole. Israel and the Drone Wars: Examining Israel’s Production, Use and Proliferation of UAVs (2014), available at https://dronewarsuk.files.wordpress.com/2014/01/israel-and-the-drone-wars.pdf.

[6] Kimberly Hsu, et al., US-China Econ. and Sec. Review Comm’n, China’s Military Unmanned Aerial Vehicle Industry 9–10 (2013).

[7] Randy Huiss, Cong. Research Serv., R42539, Proliferation of Precision Strike: Issues for Congress (2012).

[8] Sarah Kreps & Micah Zenko, The Next Drone Wars – Preparing for Proliferation, Foreign Aff., (Mar./Apr. 2014).

[9] See Precision Strike Warfare and International Intervention: Strategic, Ethico-Legal and Decisional Implications 57 (Mike Aaronson, et al., eds., 2014).

[10] W.J. Hennigan, Obama Administration to Allow Allied Countries to buy Military Drones, L.A. Times, Feb. 17, 2015.

[11] See generally U.S. Gov't Accountability Office, GAO-12-536, Agencies Could Improve Information Sharing and End-Use Monitoring on Unmanned Aerial Vehicle Exports (2012), available at http://www.gao.gov/assets/600/593131.pdf.

[12] See Jefferson Morley, Drone Proliferation Tests Arms Control, Arms Control Assoc. (Apr. 1, 2014), https://www.armscontrol.org/act/2014_04/Drone-Proliferation-Tests-Arms-Control

[13] State Dep’t Fact Sheet, supra note 1.

[14] Id.

[15] See generally Derek Gregory, The Everywhere War, 177 Geographical J. 238, 238–250 (2011).

[16] Public Committee against Torture in Israel v. Government of Israel, HCJ 769/02, 46 I.L.M. 373 (2006). See generally Mayur Patel, Israel's Targeted Killings of Hamas Leaders, ASIL Insights (2004), http://www.asil.org/insights/volume/8/issue/9/israels-targeted-killings-hamas-leaders; Mary Ellen O’Connell, The International Law of Drones, ASIL Insights (2010), http://www.asil.org/insights/volume/14/issue/37/international-law-drones; David Kaye, International Law Issues in the Department of Justice White Paper on Targeted Killing, ASIL Insights (2013), http://www.asil.org/insights/volume/17/issue/8/international-law-issues-department-justice-white-paper-targeted-killing.

[17] The New York Times has offered that “[b]efore a decision is made to kill, particularly in areas away from recognized battlefields, the government needs to consider every other possibility for capturing the target short of lethal force.” Lethal Force Under Law, N.Y. Times, Oct. 9, 2010, available at http://www.nytimes.com/2010/10/10/opinion/10sun1.html?_r=0. The American Civil Liberties Union has also proposed a targeted killing warrant process. See generally Kevin H. Govern, Warrant Based Targeting: Prosecution-Oriented Capture and Detention as Legal and Moral Alternatives to Targeted Killing, 29 Ariz. J. Int’l & Comp. L. 477 (2013). See also Zack Beauchamp, What To Be Concerned About As Congress Mulls Targeted Killing Courts, Think Progress (Feb. 8, 2013, 5:00 PM), http://thinkprogress.org/security/2013/02/08/1562861/targeted-killing-courts-concerns/.

[18] Notably, the United States has endorsed a stricter policy of near-zero civilian casualties in areas outside battlefield theaters while tolerating more casualties (although still within IHL limitations) against ISIS in Iraq and Syria. See Ken Dilanian, Obama Administration Eases Policy on Preventing Civilian Casualties in Iraq, Syria, PBS Newshour, Oct. 1, 2014, available at http://www.pbs.org/newshour/rundown/civilian-casualty-standard-eased-iraq-syria/.

[19] See Markus Wagner, The Dehumanization of International Humanitarian Law: Legal, Ethical, and Political Implications of Autonomous Weapons Systems, 47 Vand. J. Transnat’l L. 1371 (2014); Armin Krishnan, UVs, Network-centric Operations, and the Challenge for Arms Control, 21 J. L. Inf. & Sci. 61 (2011).

[20] See Noel E. Sharkey, The Evitability of Autonomous Robot Warfare, 94 Int’l Rev. of the Red Cross 787 (2012).

[21] Alice K. Ross, Civilian Drone Deaths Triple in Afghanistan, UN Agency Finds, Bureau of Investigative Journalism, Feb. 8, 2014.

[22] These numbers are derived from Get the Data: Drone Wars, The Bureau of Investigative Journalism http://www.thebureauinvestigates.com/category/projects/drones/drones-graphs/ (last visited May 22, 2015).

[23] See Ann Rogers, Investigating the Relationship Between Drone Warfare and Civilian Casualties in Gaza, 7 J. Strat. Sec. 94 (2014), available at http://scholarcommons.usf.edu/jss/vol7/iss4/8.

[24] See Cor Oudes & Wim Zwijnenburg, IKV Pax Christi, Does Unmanned Make Unacceptable – Exploring the Debate on using Drones and Robots in Warfare (2011); Peter Asaro, On Banning Autonomous Weapon Systems: Human Rights, Automation, and the Dehumanization of Lethal Decision-Making, 94 Int’l Rev. of the Red Cross 687 (2013).

[25] Jo Becker & Scott Shane, Secret ‘Kill List’ Proves a Test of Obama’s Principles and Will, N.Y. TimesMay 29, 2012.

[26] Kreps & Zenko, supra note 8.

[27] See Mark Guzinger & Chris Dougherty, Changing The Game: The Promise of Directed-Energy Weapons, Center for Strategic and Budgetary Assessments (2012).

[28] See David Koplow, Tangled Up in Khaki and Blue: Lethal and Non-Lethal Weapons in Recent Confrontations, 36 Geo. J. Int’l L. 703 (2005).

[29] See Treasa Dunworth, The Silent Killer: Toxic Chemicals for Law Enforcement and The Chemical Weapons Convention, 10 N. Z. Yearbook Int’l L. 3 (2012).