Is the CIA in the Drone Kill Chain? (Answer: Likely.)

by Kevin Jon Heller

Wells Bennett calls my attention to this statement by Marc Ambinder in a recent article in The Week entitled “Five Truths About the Drone War”:

The CIA does not “fly” drones. It “owns” drones, but the Air Force flies them. The Air Force coordinates (and deconflicts) their use through the CIA’s Office of Military Affairs, which is run by an Air Force general. The Air Force performs maintenance on them. The Air Force presses the button that releases the missile. There are no CIA civilians piloting remote controlled air vehicles. The Agency has about 40 unmanned aerial vehicles in its worldwide arsenal, about 30 of which are deployed in the Middle East and Africa. Most of these thingies are equipped with sophisticated surveillance gear. A few of them are modified to launch missiles. The Air Force owns many more “lethal” RPVs, but it uses them in the contiguous battlefield of Afghanistan.

Wells points out at Lawfare that “if Ambinder is correct, then it is military personnel who do the drone-flying and the button-pushing, and military personnel can invoke a public authority justification for strikes implicating 1119, in Kevin’s view.” In other words, Wells suggests that it might be irrelevant whether CIA officers are entitled to a public authority defence, because they may not actually be involved in lethal drone attacks, including the one that killed al-Awlaki.

I completely agree with Wells’ restatement and application of my position on the public authority defence. But I am less sure that Ambinder’s “truth” insulates CIA from potential criminal liability. Ken Dilanian, a leading national-security reporter, had a long article in the Los Angeles Times last month discussing the possibility of the military taking over much of the CIA drone program. Ambinder’s reporting seemed to contradict Dilanian’s article, so I tweeted Dilanian about it. Here was his reply:

Dilanian is right: the articles don’t necessarily contradict each other. Ambinder says that the military flies the drones and pushes the button that launches the weapon; he does not claim that the military chooses the targets and makes the decision to launch the attack. There are some interesting questions about what it means for the CIA to “give the order to fire,” but it seems clear that CIA officers are still involved in lethal drone attacks in a manner that gives rise to a potential violation of the foreign-murder statute — as conspirators or instigators or as aiders-and-abettors. So the fact that a CIA officer is not entitled to a public authority defence remains an important issue.

24 Responses

  1. Customary international law on whether an aircraft is a military aircraft draws a distinction between the ‘crew’ and the person in ‘command’ of the aircraft. This would appear to be a useful way of analysing a situation in which one person is operating the flight controls, the same person or another operating the target acquisition and weapon system, but either or both persons are taking tactical direction from another.

  2. As I noted here at OJ after the bin Laden raid, the legal affect of placing the military under the CIA in covert operations is unclear and under-analyzed.  Nevertheless, at the time, almost everyone agreed that the bin Laden operation complied with international humanitarian law.  (Some didn’t agree that IHL applied, and others thought there might be a jus ad bellum problem, but nobody questioned the authority of the SEAL Team to kill bin Laden.)
    If the military is working under the CIA’s covert action umbrella, as these articles suggest, then the CIA’s drone program is operating no differently than the bin Laden raid did. When other agencies are placed under the CIA’s control for covert action, they are subject to the rules (or lack thereof) governing the CIA unless otherwise provided.  In a comment to my post, Marty Lederman stated his understanding that the military adhered to its general policy to comply with IHL in such operations.
    All of this suggests, as Wells noted, that “unlawful” killing of citizens for purposes of the foreign murder statute does not include the killing of U.S. citizens in the context of armed conflict who are targetable under IHL.  Under such circumstances, Supreme Court decisions strongly suggest that such killings are constitutional.  Killing citizens outside the context of armed conflict, however, would be presumptively unconstitutional and therefore unlawful if the targeted individual did not pose an immediate threat of grave harm.  

  3. Foreign murder is a domestic crime. The White Paper suggests that IHL provides the requisite public authority for a public authority defence, yet IHL does not authorize the CIA to kill; indeed, IHL specifically leaves open the possibility of states criminalizing acts that would be perfectly legal if committed by a privileged combatant. The US was under no obligation to domestically criminalize an American killing an American abroad, but it has — and it has not exempted the CIA or anyone else from the ambit of that crime. IHL thus does not and cannot provide a CIA officer charged under the foreign-murder statute with a public authority defence.

  4. John,

    If Yemen managed to capture a CIA officer who ordered lethal drone strikes, do you think international law would prohibit Yemen from prosecuting him for murder? If so, why?

  5. Kevin,
    My view is that the AUMF provides authority for the President to actuate the war powers of the government.  This authority includes using the CIA in covert operations designed to engage enemy fighters/belligerents/combatants, including those who are citizens, through non-traditional military means.  The net effect of such congressional and presidential authorization is that those agents may lawfully aid in prosecuting the conflict under U.S. domestic law.  This means that they are acting on the public authority of the United States as articulated by Congress and the President.  
    I don’t think that the White Paper is suggesting that IHL provides authority for the CIA to participate in hostilities.  I think it is saying that American citizens may constitutionally be targeted by the U.S. government under circumstances permitted by IHL. Thus, it is both IHL and the AUMF that makes CIA action lawful for purposes of the foreign murder statute.
    This is an entirely separate question from whether agents of the U.S. are entitled to immunity from foreign jurisdictions/prosecutions.  Clearly, some of what the CIA does is entirely consistent with U.S. federal law while violating the domestic criminal law of the state(s) in which they operate.  No species of international immunity (combatant, diplomatic, consular, or other official) protects them from prosecution for their official acts in such circumstances.  But this does not mean that these agents do not act pursuant to U.S. public authority or that their actions are not lawful from the perspective of U.S. law.  Indeed, the logical implication of the Charming Betsy decision is that Congress may authorize the President to violate international law.

  6. John,

    Okay, then answer this question, which you’ve never done specifically:if the AUMF provides the authority for an American to kill an American overseas, despite the plain language of the foreign-murder statute, why does it not provide the authority for an American to commit torture, despite the plain language of the torture statute?

  7. To state it differently, and to combine my comments, Supreme Court decisions strongly support the proposition that the U.S. government may lawfully use war powers against its citizens when the use of those powers is consistent with the laws of war/international humanitarian law.  Targeting a citizen who is an enemy fighter/combatant/belligerent is consistent with IHL. Therefore, the White Paper relies on IHL, coupled with authority to engage in armed conflict (the AUMF), to establish the lawfulness of a use of lethal force against a enemy citizen fighter/belligerent/combatant.  The case law only addresses the power of the government to invade public and private rights, not the (international) status of the government agent invading those rights.  If the government agent is acting pursuant to domestic authority regardless of international status, it is lawful under domestic law.

  8. Kevin,
    The difference between what I am arguing and torture is that separate domestic law — the War Crimes Act — specifically makes it a crime to engage in torture during international or non-international armed conflict.  The act itself is prohibited in the specific context and regardless of the international status of the person engaging in it.  
    Regarding the killing of an American citizen, Congress has generally prohibited such conduct when “unlawful” without specifically addressing the armed conflict context.  In armed conflict, killing a citizen can be consistent with international and domestic law, and according to the Supreme Court is probably “lawful” when the government’s use of force is lawful under IHL.  
    Although I’ll go into this in later scholarship, this might mean that the government should be estopped from prosecuting foreign nationals who kill in the context of an armed conflict without meeting IHL requirements for prisoner of war status/combatant immunity (murder in violation of the law of war), at least in some circumstances.

  9. Okay, that’s helpful. So your position is that the AUMF trumps any federal statute that does not specifically apply to armed conflict, but does not trump a federal statute that does specifically apply to armed conflict? If so, what in the text or legislative history of the AUMF supports that interpretation?

  10. No, I am not saying that the AUMF trumps “any” federal statute.  I am saying that it supersedes or provides a relevant exception to clearly inconsistent federal law.  
    The very text of the AUMF authorizes the President to use military force, interpreted in Hamdi to mean war powers.  Those powers undoubtedly include targeting and detaining enemy fighters/belligerents/combatants.  Supreme Court precedent permits the use of war powers against citizens when they are part of an enemy armed force.  Federal law permits the use of the CIA to engage in a broad range of covert activity, which includes traditional war powers exercised in non-traditional ways.  Thus, the CIA may target and kill enemy fighters/belligerents/combatants who are also citizens.  Doing so is “lawful” for purposes of the foreign murder statute, just as the Court found with regard to detention despite a general federal criminal prohibition.
    Let me pose this question, Kevin.  If in the course of a covert activity unrelated to an extant armed conflict the CIA encountered an American citizen that posed an immediate threat of grave harm to other citizens, could they engage that threat with lethal force?  Does the “defense of others” justification apply in those circumstances?

  11. What federal law does not do is authorize the CIA to kill an American abroad. There is a federal law that prohibits precisely that — and the National Security Act prohibits the President from authorizing the CIA to engage in covert activities that violate federal law. There is nothing in the text or history of the AUMF that indicates the foreign-murder statute is “clearly inconsistent” with the AUMF;it is only “clearly inconsistent” because you think it is a good idea for the CIA to use lethal force abroad.

    As for your hypothetical, yes, the CIA officer could argue self-defense. That defense is available under the foreign-murder statute, just as the public authority defence is. The difference is that defence of others is made out under your facts, whereas the public authority defence is not.

  12. Reasonable minds may certainly differ, Kevin.  I am offering a plausible view of the legal effect of the AUMF and covert action statutes.  Your view is certainly plausible as well.  If it were an easy question, then nobody in government would have paused to consider it.
    My reading is not based on the text or history of the AUMF, however.  It is informed by the history of the CIA and its use in activities related to armed conflict.  You are fundamentally arguing that a general requirement to comply with a relatively context neutral criminal statute (let’s keep in mind that it does not specifically apply to covert action or the CIA by its terms) cannot be the subject of an implied exception.  I am arguing that the law can be reasonably interpreted to grant a context specific implied exception when such an exception is relatively clear and constitutional, as happened in Hamdi.  The exception need not comply with common law understandings of public authority.
    Let us hypothetically say that the CIA was used in the initial stages of the invasion of Afghanistan and worked side by side with special forces.  Under the Military Extraterritorial Jurisdiction Act, they would likely be subject to prosecution for any conduct that violates a federal felony statute applicable in the special maritime or territorial jurisdiction of the U.S., which includes murder.  Let us further say that an agent killed a foreign enemy fighter in circumstances not amounting to self defense or defense of others.  By your logic, the CIA agent would be guilty of murder.  In my view, so long as the agent was lawfully empowered to engage in hostilities under domestic law, there is no crime.  Another example that comes to mind are soldiers or agents who help form or guide irregular armies that do meet IHL requirements for immunity to engage and kill the enemy.  (Does this describe the Northern Alliance/Anti-Taliban Forces?)  Are they also criminals under MEJA’s extension of federal criminal law?  Or, if their actions were sanctioned by the appropriate government officials acting pursuant to lawful authority, are they acting lawfully?
    With all of that said, I am not arguing that any of this is a good idea.  Until I leave the government, I will not comment on the wisdom of involving the CIA in armed hostilities.  I should also note that I have no specific official information, classified or otherwise, informing my arguments.  These are my personal, private views expressed in my academic capacity.
    Finally, I should note that I may have overstated Marty Lederman’s comment to my earlier post on the bin Laden operation.  Marty expressed his understanding that the military always complies with IHL by policy.  He did not specifically say that this includes when they act, as I have coined the term, “under the covert action umbrella.”  I read his comment in the context of the larger discussion and gave it substance he likely didn’t intend.  Marty NEVER comments on such classified matters, to my knowledge, and is therefore unable to give us the benefits of his true insights.

  13. There is no “public authority” defense under international law — a la IMT at Nuremberg, just the opposite re: international criminal conduct (which begs the question of course). So the supposed defense is relevant merely to domestic law as such.
    In any event, control of drones and targetings by active duty military is more relevant to “combatant” status and “combatant immunity” for lawful acts of war under the laws of war — which should also inform the domestic constitutional issue.  The more interesting question involves implied immunity under the self-defense paradigm when laws of war are not applicable (as noted in my 19 J. Trans. L. & Pol’y article in 2009).  If targeting is permissible under the law of self-defense (in time of peace or in time of war), it appears that general patterns of practice and opinio juris regarding such practices support an implied immunity for lawful self-defense targetings (no known prosecutions, etc.).  This should also inform the domestic constitutional issue.

  14. John,

    The crux of our disagreement is statutory interpretation. You say that the foreign-murder statute does not specifically apply to the CIA. But I think it does — CIA officers are American, and the foreign-murder statute applies to all Americans. There is no indication that Congress intended to exempt the CIA from the statute — which would obviously have been very easy to do, had that been their intent. Moreover, although you are correct about the history of CIA involvement in armed conflict, I think the lesson to draw from that history is precisely the opposite of yours, given that the foreign-murder statute is only 20 years old: namely, that Congress would have specifically exempted the CIA from the statute if it believed that the the President should be able to authorize the CIA to kill Americans abroad. That use of force by the CIA, if not unprecedented, is extremely rare. So I don’t think it works to say that Congress must have intended to exempt the CIA from the foreign-murder statute because they didn’t specifically say that it applies.

  15. Also, my view does not imply that it is murder under US law for a CIA to kill a foreign fighter abroad. There is no domestic law that criminalizes such killing. The US could criminalize the CIA participating in hostilities or criminalize the commission of specific acts during such participation, because CIA officers do not have combatant’s privilege. But it has not done so — and I doubt it ever would.

  16. Kevin,
    Your view that CIA agents may not claim public authority means that, when operating with U.S. armed forces overseas, they might be prosecuted for killing a foreign fighter under the general federal murder statute (18 U.S.C. sec. 1111) applied through the Military Extraterritorial Jurisdiction Act (MEJA) or the Uniform Code of Military Justice (UCMJ) murder offense.  Either provision might be applied (whether it is constitutional to use the UCMJ we are not yet sure) to civilians serving with or accompanying armed forces overseas in armed conflict or other overseas contingency operations.  (Under their terms, neither MEJA nor UCMJ would apply to independent CIA operations.  Query whether CIA agents “accompany” the armed forces or vice versa in covert action operations.)
    When I say that the foreign murder statute does not “specifically” apply to covert action I mean that in the strictest sense of the term.  The statute criminalizes the unlawful killing of one U.S. national by another U.S. national in foreign territory generally.  It would apply to me traveling overseas as an academic.  It only applies to covert action only by a general requirement in the covert action statutes to comply with federal law, not by a specific provision of the foreign murder statute that applies it to covert action or CIA operations.  This certainly means that the statute generally applies to covert action operations.  But it also means that a more specific law might authorize conduct at odds with this generally applicable federal law.  My argument is that the AUMF provides such authority when a covert operation is an integral part of an armed conflict prosecuted pursuant to it, just as the AUMF provides the authority to kill foreign fighters overseas despite a generally applicable murder offense potentially operable under the MEJA or the UCMJ.

  17. And implied immunity under the international law of self-defense would inform the meaning of “unlawful” killing as “murder” under any relevant domestic law, especially given Charming Betsy and the many other Supreme Court cases that recognize that customary international law, at least, is a necessary background for interpretation of any relevant federal statute — and recall the Cook rule (expressed also in many other Supreme Court cases) that if a federal statute is unavoidably inconsistent with treaty law of the U.S., Congress must clearly and unequivocally express an intent to override the prior treaty law or the treaty law will prevail domestically as law of the United States. All in our IL casebook

  18. John,

    Your final comment crystallizes the difference between us. For my part, I’m with Philip Dore, who has explained in an excellent law-review article why it is difficult, if not impossible, to read the AUMF as implicitly superseding the foreign-murder statute. Here is the core of his argument:

    Thus, the Administration must rely on the AUMF, which results in a showdown between the AUMF and the foreign-murder statute. Both the legislative history of the AUMF and the long-recognized ban against implicitly repealing legislation weigh against interpreting the AUMF as authorizing a violation of the foreign-murder statute.

    In Morton v. Mancari, the Supreme Court held that Congress did not intend to repeal the 1934 Indian Reorganization Act when it passed the Equal Employment Opportunity Act of 1972. The Court’s analysis is instructive in examining whether the AUMF repealed the foreign-murder statute.

    In Mancari, the Court noted the “cardinal rule . . . that repeals by implication are not favored,” and that “[i]n the absence of some affirmative showing of an intention to repeal, the only permissible justification for a repeal by implication is when the earlier and later statutes are irreconcilable. And further, “[w]here there is no clear intention otherwise, a specific statute will not be controlled or nullified by a general one, regardless of the priority of enactment.”

    Applying these principles to the statutes at issue begins with a search for an “affirmative showing” in the AUMF that Congress intended to repeal the foreign-murder statute. There is certainly no such showing in the text of the AUMF. The AUMF’s ratification history is also devoid of any such “affirmative showing.” To the contrary, there is an abundance of evidence indicating that Congress intended for the President to act in accordance with the Constitution and existing federal statutes such as the foreign-murder statute.

    Even if it could be argued that the AUMF’s legislative history is ambiguous, the AUMF would still not impliedly repeal the foreign-murder statute unless the two statutes were “irreconcilable.” That is hardly the case here. The two statutes can “readily coexist” by construing the AUMF so as not to authorize the President to give orders that violate the foreign-murder statute.

    Yet another argument against implicit repeal concerns the nature of the two statutes. The foreign-murder statute is a specific statute while the AUMF is more broad. When there is no clear intention otherwise, a specific statute will not supersede a general statute. As explained above, there is no indication that Congress intended for the AUMF to control or nullify the foreign-murder statute. Therefore, because the foreign-murder statute is a specific statute and the AUMF is a broader statute, the latter should not be construed as repealing the former.

  19. It’s a good argument, Kevin, but it doesn’t persuade me.  We are not talking about an implicit “repeal” of a statute as that term is commonly used.  At issue is whether the AUMF legitimizes a narrow class of conduct within a criminal statute’s otherwise very broad scope.  Now, I would agree that one shouldn’t readily imply exceptions to criminal statutes. But in this context that implication seems unavoidable.  At any rate, I couldn’t disagree more that the AUMF is broader than the foreign murder statute—quite the contrary.

  20. Important update: according to a Lawfare post this morning by Jack Goldsmith, this debate might soon become moot.  Has Kevin’s line of argument persuaded the administration, or at least made them uncomfortable enough to change course?  

  21. John: and the AUMF’s authorization is necessarily limited by the word appropriate  — but, as recognized in Hamdi, what is appropriate under the laws of war (and presumbably under international law more generally, a la Charming Betsy, etc.) is relevant.

  22. I completely agree, Jordan!

  23. John,

    A sweeping authorization to use force is not broader than a prohibition of one and only one use of force — Americans killing Americans abroad? 

    And yes, we are talking about an implicit repeal, if only a partial one. CIA officers (at least those in question) are Americans, and the foreign-murder statute prohibits Americans from killing Americans abroad. So there is no way to get around the need to explain why and how the AUMF partially repeals the foreign-murder statute.

  24. Some final thoughts:
    “Implied repeal” was not the Court’s approach in Hamdi, nor do I think it would be its approach if called upon to address this matter.  Is it really a partial “repeal”—generally meaning “rescission” or “revocation”—for a statute to authorize and make lawful in a specific context that which would be generally unlawful under a broad, generally applicable statute?  Also, generally speaking, when courts are faced with laws that might conflict, they first attempt to harmoniously interpret them, finding a conflict only when those laws are clearly irreconcilable.  It seems to me that doing so in this case would likely yield the result I have suggested, especially in light of the Executive’s (apparent) interpretation.
    “Sweeping authorization?”  If it were so sweeping would we now be discussing extra-AUMF threats?  The AUMF has been stretched to the breaking point because it was not sweeping enough to encompass the numerous and varied threats that existed at the time it was enacted and have since then emerged.
    It surely depends upon one’s point of view, but I think a statute that categorically bans certain conduct between members of a rather large societal group almost anywhere in the world and regardless of context is broader than a statute that authorizes particular types of conduct (“military force”) against (somewhat) specific persons, organizations, etc.

Trackbacks and Pingbacks

  1. There are no trackbacks or pingbacks associated with this post at this time.