Drones and the politics of international law

Drones (also known as unmanned aerial vehicles or remotely piloted aircraft systems among other monikers), though not yet ubiquitous, are increasingly permeating our daily lives. They are used by scientists, police forces, hobbyists and delivery companies (but only for burritos...). Only the other day, while my partner and I were walking in the isolated North York moors, a drone hummed determinedly across the sky. But it was not a quadcopter being flown by a member of the public, it was (most likely, having referred to the very handy Drone Survival Guide) a ‘Hermes’ military surveillance drone, operated by the Royal Air Force, providing a clear reminder that the growth of drones has arguably had its most palpable manifestation within the field of armed conflict.

Drones have had a place in this field since at least the First World War, and they have subsequently played a part in many conflicts, including Korea, Vietnam, the Yom Kippur War and Yugoslavia. However, though some of these drones were used lethally as guided bombs, most were purely for reconnaissance. It was not until the Kosovo war that a drone, the Predator, was armed[i] and later still, during the war in Afghanistan, that a missile was fired by a drone. Since then, the number of drones has grown at a staggering rate. In 2000 the US had fewer than 50 drones[ii] but by 2012 this had risen to over 19,000, set to rise as the Pentagon seeks to increase its daily drone flights by 50%. Drones therefore have the capacity to change the traditional picture of warfare from one of direct personal engagement to one that is indirect and impersonal, at least for those states with the infrastructure to fly them.

Nonetheless, this transformation does not have immediate implications for armed conflicts and the law that governs them—from a doctrinal perspective, drones are not per se legally problematic, they are ‘conventional weapons’ under international law and they are not prohibited under any of the treaties that restrict the use of certain weapons.[iii] However, this statement masks the fact that it is not the weapon itself that is controversial, but the uses to which it can be put. On 3 November 2002, the US fired a missile from a Predator drone operating in Yemen, which killed six people including a reputed high-ranking member of al-Qaeda, wanted by the FBI in connection with the attack on the USS Cole in 2000. The importance of this event is difficult to overstate as it represented the first drone strike undertaken outside of a combat zone, against a non-state actor with no link to the state in which they were present. It is therefore of a distinctly different character, demonstrating the ability of drones to transcend the traditional boundaries of ‘war’. Since that first extra-conflictual strike, the use of drones in this manner has increased dramatically and the US flies them constantly in Pakistan, Yemen and Somalia. The number of deaths from drone strikes is now well over 5,000, many of which have been civilian. What drones represent therefore is a dramatically increased ability of states to use extraterritorial force avoiding much in the way of public accountability.

Luckily, untrammelled force on the part of powerful states is circumscribed by the framework of law that governs the use of force by states, known as jus ad bellum.[iv] This area of law comes out of the historic practice of states (known as customary international law) and, most importantly, the Charter of the United Nations. Article 2(4) of the Charter creates a universal prohibition on the use of force by states, subject only to a handful of exceptions. Principal among these is force used in self-defence;[v] under Article 51 of the UN Charter and international customary law, a state may use force ‘if an armed attack occurs’ and if it is necessary and proportionate.[vi] The Charter has been acceded to by ostensibly every state in the world and therefore we may assume that when drone strikes are carried out, they are done so in line with these laws. Indeed the US has insisted (in the guise of assertions about its wider use of force) that its global drone programme is carried out in line with international law. Thus it is possible to aver that the resort to drone strikes can be lawful under the rubric of jus ad bellum and that as such their use is done so in accordance with the international law; those actions that would be illegal are not carried out. End of story.

But of course this is not the end of the story at all. To claim that the law restricts the use of force and that therefore any resort to drones must be legal (I mean, come on, states aren't going to breach international law are they? Right??) is to ignore part of the picture.  This is that the law is a political process, subject to competing interpretations vying to crystallise a particular view of the law, advancing specific interests. In the words of Martti Koskenniemi, the law is a ‘[p]olitical struggle … waged … on the meaning of legal symbols’.[vii]  And drones provide a catalytic point around which these divergent interpretations converge by virtue of the fact that they represent a (relatively) easy way of states to resort to forceful actions. To put it another way, when force is a practical difficultly and therefore less likely to be used, it makes no difference whether the legal system in which it is used is restrictive or permissive; however, when the resort to force is readily available, the practical barrier has been removed and only the legal one remains. This consequently rests a state’s foreign policy options on whatever is the dominant interpretation of the law, giving states a huge motivation to interpret the law in a way that suits them.

This interpretive struggle is evident throughout the jus ad bellum but it is at its most clear when we look at a state’s ability to resort to force not against another state, but against a non-state actor (NSA). Under Article 51 of the UN Charter, which preserves states’ ‘inherent right’ to self-defence, a state may respond forcefully only when they have suffered an ‘armed attack’, but there is no definition of what such an attack constitutes. Consequently, the notion of an ‘armed attack’ presents itself as a legal concept which cannot but exist through interpretation. The key site of interpretive struggle over its meaning is whether such an attack can only come from another state (this could be via a NSA but requires a link between that NSA at the state in which it resides), thus restricting a ‘victim’ state’s right to self-defence, or whether it can come from a NSA, therefore expanding a state’s right to self-defence.

The UN Charter regime has been traditionally understood to emphasise the limitation of the use of force and it has been said that during the foundation of the UN ‘the general tendency was towards a restrictionist interpretation of any permission in relation to the use of force.’[viii] The traditional teleology of the jus ad bellum is like the ‘perpetual peace’ of Kant’s cosmopolitanism.[ix]  This provides the first interpretation of the law—the ‘restrictionist’ approach. This interpretive underpinning has been borne out by judicial decisions considering the possibility of armed attacks originating from NSAs. In the case of Nicaragua v United States, the International Court of Justice (ICJ) held that for an armed attack to be carried out by a NSA, there must be some degree of connection between that NSA and the state in which they reside[x] (though without saying what this degree needed to be). Similarly, in the Wall advisory opinion of 2005, the Court referred to ‘the existence of an inherent right of self-defence in the case of an armed attack by one state against another state’.[xi]  Finally, in DRC v Uganda the Court again cited the need for a non-state actor to be linked with a state before self-defence can be invoked.[xii]  Under this interpretation of the international law of self-defence, drone strikes would be restricted to conflicts which are ostensibly between states. Attacks carried out by NSAs would necessarily be treated as crimes and accordingly be dealt with as a matter of law enforcement. It is thus that such an interpretation can be understood to focus on and promote peace rather than allowing states easily to use military force in the territory of others.

Despite this seemingly staunch support for the requirement of a link between NSA and territorial state, the traditional restrictionist approach has been challenged by an opposed interpretation that emphasises states’ security requirements over the pursuit of peace. This alternative interpretive paradigm entirely removes the requirement of a state-NSA nexus and consequently creates a regime in which the lawful resort to force is much more readily available, known as the ‘expansionist’ approach. Jordan Paust referred to 19^th Century state practice to interpret the armed attack concept as requiring no state-NSA link[xiii] while others have suggested that preventing self-defence against independent NSAs would undermine the object and purpose of the UC Charter.[xiv]  This expansionist interpretation appears to have found favour with states in their recent practice as well, an example of which being the use of force against ISIL in Syria, in which self-defence appears to have been raised.[xv]  In contradistinction to the restrictionist paradigm above, this formulation of self-defence allows states to undertake forceful actions in the territory of others much more readily, and makes conflict a much more likely prospect, to the point of normalisation.

The use of drone strikes in Pakistan, Yemen and Somalia has been justified by self-defence[xvi] and in these instances force has been used against NSAs which are not linked to those states. In this manner it is entirely clear how their use has been made possible through an interpretation of the law and that the expansionist interpretation is currently dominating. Likewise, it is easy to see how this understanding of self-defence could be employed in the future to justify their further use. Drones therefore represent a coming together of the concrete and the abstract; when an expansionist interpretation is adopted, drones are the point at which the practical and legal possibilities of the use of force become congruent. They allow those states with the capacity to operate them the ability to use military force to police an entire region, as is arguably the case in the Federally Administered Tribal Areas in Pakistan.

In conclusion, it is vital that those studying drones are cognisant of the inherent plasticity and interpretive flexibility of the international law governing the resort to force. It is submitted that drones have arisen in an epoch that is hallmarked by an increased desire on the part of states to use force and a consequent reduction of global stability. The law cannot be viewed as a fixed structure which checks state uses of force—to do so would be result in the loss of all critical legal faculty. Drones have pushed to the forefront many inherent tensions within international law and without examination and resistance these have the potential to resolve themselves into a system that is highly permissive of the use of drones. The risk is the creation of conditions that encourage powerful states to subject weaker ones to near perpetual militarised policing from the air. This rather bleak prediction may sound far-fetched but it is already the case in the tribal areas of Pakistan. The need for critique is immediate.

---

[i] M Benjamin Drone Warfare; Killing by Remote Control (Verso, 2013) 15.

[ii] M Benjamin Drone Warfare; Killing by Remote Control (Verso, 2013) 17.

[iii] For instance the United Nations Convention of Certain Conventional Weapons (1980), which regulates landmines, incendiary weapons and blinding lasers, amongst others.

[iv] It should be noted that this is distinct from international humanitarian law (or jus in bello) which governs the conduct of hostilities once they have begun, which is not considered here.

[v] The other exceptions to the prohibition are collective actions in response to threats to peace, under Chapter VII UNC, as well as the more controversial humanitarian intervention. A state is able to consent to foreign intervention within its territory as an inherent aspect of sovereignty, but this removes the act from the jus ad bellum framework, rather than rendering it lawful within it.

[vi] Necessity and proportionality do not originate with the UN Charter but from much earlier customary international law, as evidenced in the Carline affair of 1837 in which self-defence was deemed lawful only when the ‘necessity of that self-defense is instant, overwhelming, and leaving no choice of means, and no moment for deliberation’.

[vii] M Koskenniemi ‘What Should International Lawyers Learn from Karl Marx?’ (2004) 17 Leiden Journal of International Law 229, 236.

[viii] I Brownlie ‘Legal Regulation of the Use of Force’ (1959) 8 International and Comparative Law Quarterly 707, 718.

[ix] I Kant ‘Toward Perpetual Peace’ (DL Colclasure trnsl.) in P Kleingeld Toward Perpetual Peace and Other Writings on Politics, Peace, and History (Yale University Press, 2006) 73.

[x] Military and Paramilitary Activities in and against Nicaragua (Nicaragua v. United States of America) ICJ Reports (1986) [195].

[xi] Advisory Opinion Concerning the Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territories ICJ Reports (2004) [139].

[xii] Armed Activities on the Territory of the Congo (Democratic Republic of the Congo v. Uganda) ICJ Reports (2005) [146].

[xiii] JJ Paust ‘Self-Defence Targetings of Non-State Actors and Premissibility of US Use of Drones in Pakistan’ (2010) 19(2) J.Transnat'l L.& Pol'y 237, 244-7.

[xiv] AC Orr ‘Unmanned, Unprecedented, and Unresolved: The Status of American Drone Strikes in Pakistan Under International Law’ (2011) 44 Cornell Intl LJ 729, 739.

[xv] SJ Power ‘Letter dated 23 September 2014 from the Permanent Representative of the United States of America to the United Nations addressed to the Secretary-General’ S/2014/695

[xvi] See e.g. H Koh ‘The Obama Administration and International Law’ (March, 2010) Speech at the Annual Meeting of the American Society of International Law; and Eric Holder ‘Attorney General Eric Holder Speaks at Northwestern University School of Law’ (2012) <http://www.justice.gov/opa/speech/attorney-general-eric-holder-speaks-northwestern-university-school-law> (accessed 20 August 2015).

‘…otherwise targetable…’: the curious case of the US and its expansive targeting practices

Within the last few days, the US government has released figures detailing its estimate of the number of civilians and combatants killed by drone strikes ‘outside areas of active hostilities’ (presumably Pakistan, Yemen and Somalia). While much commentary has focused upon the veracity of the figures given (many outlets estimate far higher numbers of civilian fatalities, for instance the Bureau of Investigative Journalism) I would like to draw attention to a rather curious phrase which features not in the White House’s statement itself, but in the document produced by the Office of the Director of National Intelligence which contains the figures.  The passage in question goes like this:

‘The U.S. Government draws on all available information (including sensitive intelligence) to determine whether an individual is part of a belligerent party fighting against the United States in an armed conflict; taking a direct part in hostilities against the United States; or otherwise targetable in the exercise of national self-defense.’

This sentence seems to detail the US understanding of the way individuals may be targeted lawfully, according to international law.  It envisages three categories; the first two are unremarkable, but the third is, to my mind, pretty peculiar.  The US asserts that there exists a category of persons ‘otherwise targetable in the exercise of national self-defense’.  As will be shown, this does not fit with established law on who may and may not be targeted during an armed conflict.

The law which governs when a state may kill an individual during an armed conflict is known, prosaically, as… the law of armed conflict (LOAC) (or more poetically as the jus in bello; or more humanistically as international humanitarian law (IHL)).  Within the LOAC, there are clear rules detailing who may be targeted during an armed conflict (okay, they’re not always that clear, but they are if you ignore the vagueness... for our purposes they’re clear, alright?).  In the case of US drone strikes, the situation is a ‘non-international armed conflict’ (NIAC), in accordance with common Article 3 of the Geneva Conventions and relevant case law (see the Tadić case, Decision on Jurisdiction) as the fighting does not take place between two states.  In such a conflict, individuals may be targeted when they are: members of dissident armed forces (a category which does not apply in the areas in which drones are used); members of organised armed groups; or individuals directly participating in hostilities.  These categories arise out of treaty and customary law.  None of these categories resemble the notion of those ‘otherwise targetable’ posited in the DNI document.  So immediately it seems that the DNI document has revealed a method of targeting that operates outside of the legal framework.

The DNI document links those ‘otherwise targetable’ with ‘the exercise of national self-defense’, which suggests a cross-pollination from another area of law governing force.

It is axiomatic to those who study international law that there is a distinction between the LOAC and the law which governs the use of force in self-defence (jus ad bellum).  Indeed, they are two entirely separate areas of law.  The jus ad bellum provides criteria under which force may be resorted to—it governs the lawfulness of the resort rather than the specifics of the way force is used.  Generally, force is prohibited under Article 2(4) of the UN Charter but will be lawful if undertaken in self-defence, according to Article 51 of the Charter (or customary law).  This Article requires that, for self-defence to be lawful, there must have been an armed attack (itself a contentious concept) and that the use of force is necessary and proportionate to that attack.  It is the gateway which must be passed through before a state may use any force.  It is clear that the law of self-defence has no bearing whatsoever on who may be targeted; that is entirely within the purview of the LOAC.

So, is the US using a standard that governs the resort to force to determine a question of on-going force, to establish who may be targeted by armed drones?  It seems like it may be.  If so this would greatly expand the scope for targeting and allow the US to conduct strikes and call them legal if they were undertaken as part of an operation which was lawful under the jus ad bellum.  But, as we know, the jus ad bellum does not relate to the specifics of force—its remit ceases once force has been resorted to, at which point the LOAC takes over.  Taken to the extreme, the ‘otherwise targetable’ doctrine would mean that once force began in self-defence, any individual killed as part of the action would have been targeted lawfully.  To my mind, it recalls the concept of ‘naked self-defence’, proposed by Ken Anderson, in which lawful self-defence allows uses of military force but in the absence of an armed conflict (I reject this notion entirely: the use of military force outside of an armed conflict is almost certain to be illegal, regardless of whether the original resort was legal under jus ad bellum).  The idea that individuals (and let’s be honest here, we’re talking about civilians) are ‘otherwise targetable in the exercise of national self-defense’ appears to be an attempt by the US to create an extra-legal space in which actions may be justified as not unlawful by virtue of the fact that norms of international law have been bypassed.  It is hallmarked by broad categories (‘otherwise targetable’, how broad is that?!) and confusion.  It represents the fostering of uncertainty in a clear drive to influence the discourse surrounding extraterritorial drone strikes in a way that makes them appear generally legal. 

I am surprised that this phrase has not caused more of a storm, it’s incongruity on the legal landscape is stark.  It is no wonder that the US has managed to produce civilian casualty statistics so different from those proffered by NGOs: its categorisation of those who are lawfully targetable by drones is vast.

The implications of the inclusion of this phrase within the US targeting policy are very dire indeed.  Drones will continue to be used, as long as the enemies of the US reside in hard to reach areas.  Meanwhile the legal frameworks in place to protect civilians and mitigate the consequences of war are gradually eroded, to be replaced by a soup of interpretations in which (borderline) unlawful, indiscriminate and trigger-happy uses of force can be relabelled as precise, clinical and neatly legal.  I am alarmed; I wonder if others agree? 

A fairly accurate depiction of the present author's feelings on US targeting policy

International Law and The Eye in the Sky: a hasty critique

Gavin Hood’s The Eye in the Sky is a timely release for me.  I am just on the wrong side of halfway through my thesis examining the legality of armed drone strikes and all of a sudden a film arrives which attempts just that.  Well, sort of.  BEWARE of reading on if you’ve not yet seen the film, I’ve included some SERIOUS SPOILERS!  Go watch it now then come and read this later…

Much of the film focuses on the legality of a drone strike upon a meeting of identified members of al-Shabaab the Somali armed group, in a house within a compound in Nairobi, Kenya.  It concerns the debate surrounding the change of a mission from one aimed at capturing subjects to one in which they become targets, to be killed by a missile fired from a drone.  Prominent is the issue of collateral deaths, as a young girl named Fatima (played by Faisa Hassan), sets up a stall selling bread next to the targeted house and will likely be killed by the strike.  The purpose of this piece is to consider, in brief, the various legal elements of the film.  If you want to know where, in terms of cold hard doctrine, the film got it totally wrong, then skip ahead to paragraph five!

The first is that of jus ad bellum, the law governing when force can be used extraterritorially.  Normally, states are prohibited from using force under Article 2(4) of the UN Charter, except in cases of self-defence, as enshrined in Article 51 of the Charter and customary law.  At first glance, this might seem the most problematic aspect of the strike—there is reference in the film to Kenya being ‘a friendly country’, implying that the use of force would be in breach of the Article 2(4) prohibition—the UK is clearly not using force to defend itself against Kenya.  However, the strike is a joint operation, lead by the British but with clear support from the Kenyan military.  We can therefore infer that the Kenyan government has given its consent to the operation.  Consent—as I have written about in depth elsewhere—vitiates questions of jus ad bellum.  There is no infringement of a state’s sovereignty when that state has authorised the strike, an act which is inherently an exercise of sovereignty.  Kenya has consented so there is no infringement of the jus ad bellum.  So far, so legal.

Secondly is the question of whether the strike was lawful under international humanitarian law (IHL), the law that governs warfare (in peacetime human rights and domestic law applies) and it is this that occupies the majority of the film.  Much time is spent wrangling over the calculation to be made between the collateral damage likely to occur (Fatima being blown to pieces) and the ‘concrete and direct military advantage anticipated’.  This calculation found in Article 51(5)(b) of Additional Protocol I of the Geneva Conventions and customary international law.  The realities of this calculation are beyond my experience, but I was told soon after watching the film that, sadly, the death of one child balanced against those of three high-level members of a terrorist group, three additional low-ranking members and the prevention of two suicide attacks, is not one that would take long to make: the strike would definitely go ahead.  And doubtless, under IHL, it would be legal.  In this way the main legal questions of the film are ultimately uncontentious, and the main dilemma political and moral, rendered cold through the operation of strategic legal language (discussed excellently here by Ntina Tzouvala).

But there is a problem!  IHL and the collateral damage calculation are only applicable during an armed conflict.  The assumption throughout is that the situation in which the strike occurred was an armed conflict and that therefore IHL was applicable.  But this assumption is incorrect: the UK is not in an armed conflict with al-Shabaab.  For it to be, there must be violence which is protracted and an armed group which is organised (held by the ICTY in the Tadić jurisdiction decision, paragraph 70).  Al-Shabaab is certainly organised, demonstrating many indicative features identified in international jurisprudence (for instance, in the film itself we see the use of checkpoints, cited as evidence of organisation by the Tribunal in the Limaj judgement, paragraph 145).  But there is no protracted violence between the UK and al-Shabaab, nor has there been.  It is possible that there is an armed conflict with Kenya since the horrific attacks on the Westgate shopping mall and Garissa University College but these attacks are insufficiently intense to have produced a non-international armed conflict; though awful, they were terrorist acts which have been distinguished from armed conflicts (Tadić Trial Judgment, paragraph 562).  Thus the strike occurred outside of an armed conflict situation.

Therefore, the international law that should have been applied to the strike is not IHL, but human rights, rules which are far less permissive than IHL when it comes to killing: there is no balancing of collateral damage and military necessity within human rights.  Though Kenya consented, no state can consent to acts with contravene human rights.  I am no human rights expert (that comes in the final third of the PhD, I’m not there yet!) but the right to life under Article 2 of the European Convention on Human Rights is only derogable in, inter alia, instances of armed conflict, which is exactly what does not exist in The Eye in the Sky.  Article 6 of the International Covenant on Civil and Political Rights provides that ‘No one shall be arbitrarily deprived of [her or] his life’ but it seems that this is exactly what has occurred at the end, when the dust settles and Fatima lies dying on the ground.  

It seems, as such, that the strike was carried out contrary to international law, though this is not immediately evident from the film.  That the decision to carry out a drone strike is a political, rather than legal, choice is apparent by the fact that the lawyers in the film possess about as much backbone as a tridacna crocea.  Nonetheless, the likely initial and ongoing feeling elicited by the film is that the strike was unpleasant though legal, despite the fact that it was quite the opposite.  In this manner, the film makes a, presumably accidental, point about the nature of those drone strikes currently being carried out in the world, which are continually asserted to ‘comply with all applicable law, including the laws of war’.  In actual fact, the legal reality is infinitely less straightforward.