Non-International Armed Conflict

Readings: Civilian Intelligence Agencies and the Use of Armed Drones by Ian Henderson

By Kenneth Anderson
Friday, June 27, 2014, 3:00 PM

Footnote 44 of the recently released and much-discussed OLC Awlaki memorandum is heavily redacted, but what's left reads, in part:

Nor would the fact that CIA personnel would be involved in the operation itself cause the operation to violate the laws of war. It is true that CIA personnel, by virtue of their not being part of the armed forces, would not enjoy the immunity from prosecution under the domestic law of the countries in which they act for their conduct in targeting and killing enemy forces in compliance with the laws of war-an immunity that the armed forces enjoy by virtue of their status. See Report of the Special Rapporteur sec 71, at 22; see also Dinstein, Conduct of Hostilities, at 31. Nevertheless, lethal activities conducted in accord with the laws of war, and undertaken in the course of lawfully authorized hostilities, do not violate the laws of war by virtue of the fact that they are carried out in part by government actors who are not entitled to the combatant's privilege. The contrary view "arises ... from a fundamental confusion between acts punishable under international law and acts with respect to which international law affords no protection." Richard R. Baxter, So-Called "Unprivileged Belligerency": Spies, Guerrillas, and Saboteurs, 28 Brit. Y.B. Int'l L. 323, 342 (1951) ("the law of nations has not ventured to require of states that they . . . refrain from the use of secret agents or that these activities upon the part of their military forces or civilian population be punished"). Accord Yoram Dinstein, The Distinction Between Unlawful Combatants and War Criminals, in International Law at a Time of Perplexity: Essays in Honour of Shabtai Rosenne 103-16 (Y. Dinstein ed., 1989) ...

The footnote is not incorrect nor is it central to the memorandum's overall conclusion (although it does indirectly run to the memo's question of "unlawful killings"). Its quotation from the eminent law of war scholar Richard Baxter's is well-taken of the legal difference between acts punishable under international law and acts with respect to which international law affords no protection.  We can add the third category mentioned in the footnote: acts which affirmatively benefit from immunity from domestic prosecution by reason of international law, which one core part of the combatant's privilege. I flag it in this Readings post in order to point out some important controversies in the status of CIA officers taking direct part in hostilities through armed drone strikes or other ways.

International law does not prohibit civilians from taking direct part in hostilities, but as a general rule it does not confer privileges or immunities of combatancy on them, either.  However, there are certain exceptions. International law of armed conflict recognizes circumstances in armed conflict under which even civilian agents of a national intelligence service such as the CIA might benefit from combatant immunity and POW status. Doctrinally, it is a complicated discussion, principally because it cannot avoid getting entangled in the differences in rules between international armed conflict (IAC) and non-international armed conflict (NIAC).

Customary law has long recognized that a state mobilized for war might incorporate into its war fighting forces paramilitary, civilian, or other government entities as part of its "armed forces" that, in peacetime, perform other duties. Many states have special or unique traditions in this regard--the US Coast Guard, for example, is by statute part of the armed forces of the United States but is also a federal law enforcement agency. Other countries have, for example, border patrol agencies or territorial guardias, etc., that can be mobilized as part of the armed forces in time of war.  Special rules have also evolved for the merchant marine. Some countries locate their foreign or domestic intelligence services within the uniformed military, whether in peacetime or wartime--the Soviet NKVD, for example--while others have created them as civilian agencies, such as the CIA.

This customary law rule of "incorporation" into the armed forces of a state is codified in Article 43 of Protocol I (and so far as I'm aware the US accepts its substance as customary). It defines the "armed forces" of a Party to a conflict as consisting of "all organized armed forces, groups and units which are under a command responsible to that Party for the conduct of its subordinates."  It goes on in Art. 43(3) to provide an additional rule which, in my view, was not customary law before Protocol I and likely not custom now, as an additional affirmative requirement, but rather a (rather sensible) innovation)--stating that whenever a Party "incorporates a paramilitary or armed law enforcement agency into its armed forces it shall notify the other Parties to the conflict." I'm not aware of the US government ever publicly asserting POW status for CIA operatives either in particular cases or as a general legal claim in a particular armed conflict--since 2001 or earlier. I've been unable to find an example for the Vietnam War, if there was one, or for the first days of the Afghanistan conflict. It's also unclear to me whether the US government's legal analysis of the status of civilian CIA is altered if the time period is characterized as part of the "preparation of the battlefield." I'm not sure there are any exactly parallel cases in earlier decades, because if one goes back very far historically, it's not clear to me that we frame the legal categories today in the same way that they were framed back in earlier conflicts such as Vietnam.  I don't know the answers to these questions, though my research has not gone to the ends of the earth; if anyone has any light to shed on these questions, by all means let me know.

Article 43 of Protocol I is applicable on its own terms only to an international armed conflict, because Protocol I as a treaty is applicable only to international armed conflict.  Yet, from the standpoint of custom, many of the core "conduct of hostilities" rules of Protocol I, such as targeting, are regarded today as applicable in either international or non-international armed conflict. Status of persons rules, addressing regarding various legal properties of combatancy, by contrast, are not so easy to translate across the IAC-NIAC divide.  Translation into the NIAC framework raises many serious questions about interpretation through relevant analogies or policy considerations, the relationship of any proposed IAC-NIAC rule to Protocol II, the subject of which is non-international armed conflict, and not to mention whether and to what extent any such extensions even enjoy genuine support as custom.  Things become even less clear in the case of the United States' claim--disputed by many states--of the existence of a non-international armed conflict against a transnational non-state actor (as noted in this earlier Readings post). Given these uncertainties, then, what can be said about the legal status of CIA operatives, civilian agents of a state intelligence service with some paramilitary functions, engaged today directly or indirectly in the conduct of hostilities, and particularly hostilities carried out using armed drones, in what the US calls a NIAC against Al Qaeda and associated forces?

I'm able to punt the hard questions, however, because I can instead point readers to an expert, scholarly discussion--the most provocative and useful I've read on this subject.  It's a 2011 article by Ian Henderson (well known and friend to many of us in the law of armed conflict community), "Civilian Intelligence Agencies and the Use of Armed Drones," available on SSRN, and published as Chapter 4 in Michael N. Schmitt, et al., eds., 13 Yearbook of International Humanitarian Law (2010), pp. 133-173.

Given that it's one of the very few direct parsings of this topic, it's very likely government national security lawyers have reviewed it and drawn their conclusions; I'm not sure how far beyond the specialized national security community it has traveled, and I think it a valuable read, whether one buys the analysis or not.  Royal Australian Air Force Wing Commander Henderson is not just a highly regarded LOAC scholar, he's a clear and graceful writer, and this article is certainly worth the attention of readers beyond the specialized LOAC community.

The article starts with armed drones and targeted killing, before turning to address the role of civilian intelligence agencies. In the national security legal community, today, in 2014, many of the issues raised about drones as a weapon system seem to have reached a rough common ground or at least a common ground on which to disagree--common ground reached mostly by shifting weapon platform debates into debates over targeted killing.  But in the actual practice of armed drone warfare today, it still leaves open the following question:  Does it legally make a difference whether the military carries out an attack or whether some civilian government entity carries it out instead? 'Civilian government entity' such as the CIA.   The article thus takes up, not the 'what, when, where, against whom' questions of drone warfare and targeted killing, but instead the 'by whom' (though the other questions can't be completely avoided).    Abstract:

It is hard to think of a more important legal issue than the legal authority for the use of lethal force. Concurrently with a State’s duty to protect its citizens from foreign threats, people must be protected from illegal acts; and government agents acting on behalf of their State should know their own legal position. The use of remotely piloted aircraft (drones) to conduct lethal strikes by the United States, and particularly the use by civilian intelligence agencies, against people associated with the Taliban and al Qaeda has been the subject of many recent publications. Three main questions on the use of armed drones are: (1) Who can be a legitimate target? (2) Where can that person be legally targeted? (3) Does it make a difference if the military carries out an attack, or whether other civilian government entities may legally conduct such attacks?

The focus of this article is on the third question. However, it is fair to say that the answer to any one of these questions might vary based on the answer to any other of the questions. While the discussion has tended to focus on US activities, and particularly those of the US Central Intelligence Agency (CIA) in Pakistan and other regions (eg, Yemen), the purpose of this article is to discuss the legal issues in a more general context. The fact that lethal force is being employed by armed drones is not a significant legal distinction. There is no unique law applying to armed drones. Rather, the employment of armed drones need to be analysed under the ordinary principles of international law applying to the resort to the use of force, the rules regulating the use of force and so forth. 

One of the crucial discussions in the article pertains to the general framework--confused and confusing, and unsettled on key matters--of the existence of a combatant's privilege in NIAC.  Under the surface, it nonetheless runs to the heart of footnote 44. The combatant's privilege is a powerful affirmative grant conferred by international law, and it overcomes domestic law in situations where it applies--including in states that regard themselves as neutral sovereigns but find themselves, for example, with belligerent parties to an IAC conducting operations on their territory.  This can trigger disputes even in an IAC, where the laws of war are quite clear. Whereas in a NIAC

there is no treaty articulation of the concept of the combatant’s privilege. There is no equivalent of that part of Article 1 Hague IV Regulations that implies the combatant’s privilege in international armed conflict; nor is there in Additional Protocol II an equivalent of Article 43(2) Additional Protocol I. Unless the conflict is an international armed conflict due to the application of Article 1(4) Additional Protocol I, then the current state of the jus in bello in a non-international armed conflict is that the non-state actors do not have lawful belligerent status nor enjoy the combatant’s privilege.

However, what about the government forces? Do they enjoy a combatant privilege? This question is more than just academic. While government forces who otherwise act lawfully are unlikely to be subject to prosecution by their own government for resorting to the use of armed force, there are at least three scenarios where the issue might arise. First, an opposing force may be successful in overthrowing a government and may then come into power itself. Second, the government forces may use force in the territory of another state and without that state’s consent.  Third, the government forces may use force in the territory of another state with that state’s consent, but the consent may be confidential with a result that there is no status of forces agreement or other like document granting the government forces immunity from prosecution under the domestic law of the territorial state.

The US has a serious interest in the status of both its military forces but also its civilian CIA officers insofar as they take direct part in hostilities in a NIAC--and particularly in whether whether there exists a combatant privilege.  Framed as a matter of Article 43 "incorporation" or its NIAC equivalent:

  • (i) If there is a case to be made that in an IAC, civilian intelligence officials can be incorporated into the "armed forces" of a party, per Article 43, with the combatant's privilege (and other legal conditions being met); and
  • (ii) if it accepted that the "armed forces" of a government side in a NIAC have the combatant's privilege (though the non-state "opposing forces," to use Henderson's terminology, do not); then
  • (iii) the "armed forces" of the government side can incorporate its paramilitary forces, such as civilian CIA special operators, with the same combatant immunities and status as uniformed military (other legal conditions met).

We can of course debate how good this argument is, and this is my framing at the deliberately simplest level of Article 43's language.  But Henderson draws upon the work of another eminent law of armed conflict scholar and practitioner, Kenneth Watkin, to make this basic argument in a deeper and more far-reaching way.  I quote it at some length below because if the argument hasn't already received attention in US government legal thinking, I'd hope it would receive discussion. Though I've framed the argument thus far on the basis of "incorporated" forces per Article 43's categories, Henderson offers it in a broader way, and explicitly concludes that, in a NIAC, there is no reason to presume that the

class of combatant is limited to members of the military or other armed forces (e.g., police or paramilitary) incorporated into the military. Noting the absence of positive state practice to the contrary, it does not seem likely that states would have chosen to limit themselves via international law as to what forces they can use to combat opposing forces in a non-international armed conflict. Based on the lack of a treaty provision or other international law imposing any limits or restrictions, the better conclusion is that the government may determine for itself who will operate on its behalf in a non-international armed conflict. (152)

I agree, and I also agree with Henderson (and Watkin's) overall arguments.  I would caution, as they do, that even if these arguments are correct legally, or at any rate the best view or a defensible view for a state, in order to avail itself of the status, a government still has an obligation to ensure in a reasonable and good faith way that any agency or agents acting on its behalf be subject to discipline and responsible command in a LOAC sense and under the LOAC rules applicable in a NIAC.  This is so whether a civilian intelligence agency is simply selected by the government to act as its agent or is regarded as legally incorporated per Article 43.  It is necessarily more than a pro forma requirement and arguably urges a policy of greater disclosure as to commitments, policies, and processes to ensure adherence to the rules of warfare than would otherwise be the case--or might otherwise be the preference of an agency that operates largely under secrecy and non-acknowledgment.
This is not an obligation owed to the non-state opposing forces, however.  A NIAC is legally asymmetrical.  Rather, its importance lies in ensuring the legitimacy of conferring the combatant's privilege on civilian actors, with respect to other states in which they might operate or into which they might direct, for example, remote aerial operations. Those states might reasonably question the basic commitment and adherence to LOAC in the conduct of clandestine or covert agents.  I would favor this as policy in any case, as a means of drawing even covert or clandestine action, whether in armed conflict or in any other use of force in national self-defense (including outside of a formal armed conflict), under acknowledged LOAC principles or rules.  From the article:

I am not aware of a positive assertion in publicly available military manuals or other official sources to the effect that government forces in a non-international armed conflict enjoy the combatant’s privilege. Nor does this point seem to be addressed in the learned commentary, with one exception. The highly-respected Kenneth Watkin writes that ‘it is logical to conclude that the concept of combatant immunity would apply to the armed forces of the state tasked with using force in a non-international armed conflict.’ No authority is cited. Rather, by his own words, the conclusion is based on logical deduction ... Notwithstanding the almost complete absence of writing on this point, I agree with Watkin’s conclusion and suggest there is a very persuasive argument in favour of there being the equivalent of a combatant’s privilege for government forces in a non-international armed conflict.

The argument starts from the premise that it is lawful for a government to engage in a non-international armed conflict with a non-state actor. And this is not a matter of international law being silent on the issue, but rather there is positive law recognizing this right. Where government forces combat opposing forces on the authorization of their government, they are acting as agents of the state. Surely, therefore, it must be legal for government forces to engage in acts pursuant to government orders (e.g., combat opposing forces) as long as the government forces otherwise comply with the law of armed conflict. Otherwise, a government’s right to combat a non-state actor would be somewhat illusory. It makes legal sense for the law to be that government forces enjoy a form of combatant privilege.

However, it also makes sense that non-state actors do not enjoy a combatant privilege, as there is no legal right under international law for a non-state actor to engage in a non-international armed conflict. And this line of reasoning is supported by the fact that on the limited occasions where the international community has recog- nised a right for a non-state actor to take up arms, those conflicts have been given the status of a an international armed conflict; and, therefore, only in such circumstances can the fighters on behalf of the non-state actor can obtain combatant status and the combatant’s privilege.

Presuming for current purposes that the equivalent of a combatant’s privilege does exist for the government forces, the next relevant question is who may lawfully form the government forces (and thereby lawfully participate in hostilities on behalf of the government)? In particular, is there any legal argument that limits, under international law, the use of force to the military or may the government also authorise other government agents to use lethal force on its behalf? There is nothing explicit in Common Article 3 of the Geneva Conventions that would prevent a government from authorising other government agents to use lethal force on its behalf. Of course, in doing so those government agents would not come within the limited protections provided by that article as they would be taking an active part in hostilities—which is no different from the fact that members of the armed forces who are still participating in the hostilities also do not come within the ambit of Common Article 3. Also, where the conflict is governed by Additional Protocol II, there is nothing in the Protocol that would prevent a government from authorizing other government agents to use lethal force on its behalf. The application of Additional Protocol II is predicated on the government side involving its ‘armed forces’, but there is no requirement in Additional Protocol II that only the armed forces operate on the government’s behalf. Also, the ICRC Commentary on Article 1(1) Additional Protocol II makes it clear that the government armed forces are not limited to military forces .... (150-151)

As a consequence of the above (assuming it is correct), Henderson (again drawing on Watkin), concludes it is not apparent that government forces

could not include civilian intelligence operatives. If, as argued above, an equivalent of the combatant’s privilege does exist for government forces in a non-international armed conflict ... it may be that CIA personnel might enjoy the equivalent of a combatant’s privilege in a non-international armed conflict when acting on behalf of the government.  (152)

The article's opening section was kind enough to quote me, testifying in a 2010 Congressional hearing on drone warfare. "Arguments against drones are really proxy for arguments against the very idea of the CIA using force," I said.  Today, in 2014, it's not the only thing arguments against drones are proxy for, but, yes.