Ethics of Drone Strikes: Restraining Remote-Control Killing 9781474483599

Citation preview

Ethics of Drone Strikes

6686_Enemark.indd i

02/03/21 12:22 PM

6686_Enemark.indd ii

02/03/21 12:22 PM

Ethics of Drone Strikes Restraining Remote-Control Killing

E d i t ed by C H R I S T I A N E NE MAR K

6686_Enemark.indd iii

02/03/21 12:22 PM

Edinburgh University Press is one of the leading university presses in the UK. We publish academic books and journals in our selected subject areas across the humanities and social sciences, combining cutting-edge scholarship with high editorial and production values to produce academic works of lasting importance. For more information visit our website: edinburghuniversitypress.com We are committed to making research available to a wide audience and are pleased to be publishing an Open Access version of chapters 4, ‘Drone Violence as Wild Justice: Administrative Executions on the Terror Frontier’ (Christian Enemark); 7, ‘Drone Warriors, Revealed Humanity and a Feminist Ethics of Care’ (Lindsay C. Clark and Christian Enemark); and 9, ‘Autonomous Armed Drones and the Challenges to Multilateral Consensus on Value-Based Regulation (Thompson Chengeta).

© editorial matter and organisation Christian Enemark, 2021 © the chapters their several authors, 2021 Chapters 4, ‘Drone Violence as Wild Justice: Administrative Executions on the Terror Frontier’; 7, ‘Drone Warriors, Revealed Humanity and a Feminist Ethics of Care’ (Lindsay C. Clark and Christian Enemark); and 9, ‘Autonomous Armed Drones and the Challenges to Multilateral Consensus on Value-Based Regulation (Thompson Chengeta) are published under a Creative Commons Attribution-NonCommercial-NonDerivative licence. Edinburgh University Press Ltd The Tun – Holyrood Road, 12(2f) Jackson’s Entry, Edinburgh EH8 8PJ Typeset in 10/13 Giovanni by IDSUK (DataConnection) Ltd, and printed and bound in Great Britain. A CIP record for this book is available from the British Library ISBN 978 1 4744 8357 5 (hardback) ISBN 978 1 4744 8359 9 (webready PDF) ISBN 978 1 4744 8360 5 (epub) The right of Christian Enemark to be identified as the editor of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988, and the Copyright and Related Rights Regulations 2003 (SI No. 2498).

6686_Enemark.indd iv

02/03/21 12:22 PM

C ONTENT S

Notes on the Contributors / vii Acknowledgements / x

INTRODU CTION / Thinking Ethically about Drone Violence / 1 Christian Enemark

ONE / Riskless Warfare Revisited: Drones, Asymmetry and the Just Use of Force / 10 Robert Sparrow

TWO / Jus ad Vim and Drone Warfare: A Classical Just War Perspective / 31 Christian Nikolaus Braun

THREE / The Complicated Reality of Drone Strikes for Law Enforcement / 50 Max Brookman-Byrne

FOUR / Drone Violence as Wild Justice: Administrative Executions on the Terror Frontier / 74 Christian Enemark

FIVE / ‘A New Departure’: Britain’s Lethal Drone Policy and the Range of Justice / 93 Christopher J. Fuller

SIX / Ethics for Drone Operators: Rules versus Virtues / 115 Peter Olsthoorn

6686_Enemark.indd v

02/03/21 12:22 PM

vi / Contents

SEVEN / Drone Warriors, Revealed Humanity and a Feminist Ethics of Care / 130 Lindsay C. Clark and Christian Enemark

EIGHT / Armed Drone Systems: The Ethical Challenge of Replacing Human Control with Increasingly Autonomous Elements / 149 Peter Lee

NINE / Autonomous Armed Drones and the Challenges to Multilateral Consensus on Value-Based Regulation / 170 Thompson Chengeta

CONCLUSION / 190 Christian Enemark Index / 195

6686_Enemark.indd vi

02/03/21 12:22 PM

NO TES ON THE C ONTRIBUT ORS

Christian Nikolaus Braun is a Senior Lecturer in Defence and International Affairs at the Royal Military Academy Sandhurst and a Visiting Fellow of the Institute of Political Science at the University of Luxembourg. Christian holds a PhD from Durham University. His thesis explored the morality of targeted killing, often carried out by armed drones, from a classical Just War perspective. His research agenda lies at the intersection of normative International Relations theory, the history of political thought and Just War theory. He has published in journals including International Relations, the Journal of Military Ethics, the Journal of International Political Theory and the Journal of Catholic Social Thought. Max Brookman-Byrne is Senior Lecturer in Law at the University of Lincoln. His research interests include the international law rules governing the resort to force, the conduct of hostilities, and uses of modern weapons technology. Max has published articles on the interaction between international legal doctrines and the use of armed drones by the United States in Pakistan, Yemen and Somalia, a topic that was also the focus of his PhD studentship funded by the UK Economic and Social Research Council. He has worked with non-governmental and civil society groups on the issue of armed drones and airstrikes generally. Thompson Chengeta is a European Research Council Fellow on Drone Violence and AI Ethics at the University of Southampton, where he undertakes research and project-related leadership on autonomous weapon systems (AWS). His PhD thesis (University of Pretoria) was on international law and ethics relevant to the governance of AWS, while his LLM thesis (Harvard Law School) was on elements that define human control over AWS. Thompson is an executive board member of the Foundation for Responsible Robotics and

6686_Enemark.indd vii

02/03/21 12:22 PM

viii / Notes on the Contributors

also serves as an expert member of the International Panel on the Regulation of AWS and the International Committee for Robot Arms Control. Lindsay C. Clark is a European Research Council fellow on the DRONETHICS project at the University of Southampton, focusing on drones and ethics from the perspective of drone operators. Her research interests include drone warfare, gender and feminist theory, and creative socialscience research methods. She completed her PhD at the Institute for Conflict, Cooperation and Security at the University of Birmingham, where she was also a research assistant for the Birmingham Policy Commission report on ‘ The Security Impact of Drones: Challenges and Opportunities for the UK’. She is the author of Gender and Drone Warfare: A Hauntological Approach (Routledge, 2019). Christian Enemark is Professor of International Relations at the University of Southampton and Principal Investigator for a European Research Council project on ‘Emergent Ethics of Drone Violence: Toward a Comprehensive Governance Framework’ (DRONETHICS). His research interests include global health politics, arms control, international security, and the ethics of armed conflict. Christian’s publications include three authored books: Disease and Security: Natural Plagues and Biological Weapons in East Asia (Routledge, 2007), Armed Drones and the Ethics of War: Military Virtue in a Post-Heroic Age (Routledge, 2014) and Biosecurity Dilemmas: Dreaded Diseases, Ethical Responses, and the Health of Nations (Georgetown University Press, 2017). Christopher J. Fuller is Associate Professor in Modern US History at the University of Southampton, specialising in American foreign policy from 1945 to the present day. His book See It / Shoot It: The Secret History of the CIA’s Lethal Drone Program (Yale University Press, 2017) focuses on US counterterrorism practices. Christopher’s research also explores the concept of the United States as a post-territorial empire. His current book project, Net Gain? The Internet and American National (In)Security, explores the role played by new technologies, in particular those in the cyber domain, in advancing and undermining US interests. Peter Lee is Professor of Applied Ethics and Director of Security and Risk Research and Innovation at the University of Portsmouth. He has been researching military drone operations since 2012, and in 2016 he was granted unprecedented research access to the two UK Royal Air Force (RAF) Reaper drone squadrons for his book Reaper Force: Inside Britain’s Drone Wars (John Blake, 2018). Peter currently serves as an Expert Adviser to the UK All Party Parliamentary Group on Drones. He holds a PhD in War Studies from Kings College London, and from 2001 to 2008 he served as an RAF chaplain.

6686_Enemark.indd viii

02/03/21 12:22 PM

Notes on the Contributors / ix

Peter Olsthoorn is Associate Professor in Military Leadership and Ethics at the Netherlands Defence Academy. Besides leadership and ethics, he teaches on armed forces and society, war and the media, and ethics and fundamental rights in the European Joint Master’s programme in Strategic Border Management. Peter’s research is mainly on topics such as military virtues, military medical ethics, armed drones and the ethics of border guarding. Among his publications are the books Military Ethics and Virtues: An Interdisciplinary Approach for the 21st Century (Routledge, 2010) and Honor in Political and Moral Philosophy (State University of New York Press, 2015). Robert Sparrow is a professor in the Philosophy Program at Monash University. He has been an ARC Future Fellow and has held prestigious visiting fellowships at universities throughout Asia. Robert has published extensively, in both academic journals and the popular press, on the ethics of military robotics, social robotics, videogames, and AI. He is co-chair of the IEEE Technical Committee on Robot Ethics and was a founding member of the International Committee for Robot Arms Control.

6686_Enemark.indd ix

02/03/21 12:22 PM

AC KNOWLEDGEMENT S

The editor and contributors gratefully acknowledge the support of the European Commission. Many of the research activities leading to this volume were supported by funding from the European Research Council, under the European Union’s Horizon 2020 research and innovation programme, for the project ‘Emergent Ethics of Drone Violence: Toward a Comprehensive Governance Framework’ (DRONETHICS) (ERC-CoG-2017 grant no. 771082). We also thank our fellow participants in a preparatory research workshop (‘Governing Drone Violence: Concepts, Moralities and Rules’) held in Southampton, UK, on 16–17 July 2019. For sharing their ideas and debating with us constructively, we are grateful to Daniel Brunstetter (University of California Irvine), Joseph Chapa (University of Oxford), Jessica Dorsey (University of Amsterdam), Alex Edney-Browne (University of Melbourne) and Jesse Kirkpatrick (George Mason University).

6686_Enemark.indd x

02/03/21 12:22 PM

INTRODUCTION

Thinking Ethically about Drone Violence Christian Enemark

This volume is a response to the continuing, worldwide increase in the acquisition and violent use of armed, uninhabited aerial vehicles (UAVs or ‘drones’). Over the last two decades, many different types of drone have been developed and deployed primarily for military surveillance purposes. However, political and popular attention has tended to focus on those drones that are equipped also to conduct missile strikes. These larger, ‘Class III’ aircraft (weighing more than 600 kilograms) typically have a payload capacity of several hundred kilograms, can operate at long range for twentyfour hours or more, and can fly at speeds of around 300 kilometres per hour (Gettinger 2019, iv). Two of the best-known examples of an armed drone, deployed first and most often by the United States, are the MQ-1 (Predator) and MQ-9 (Reaper) aircraft. The salient feature of such weapon systems – which combine aerospace, missile, satellite and video technologies – is that they enable the near-instantaneous killing of a person made visible at a precise location as far away as the other side of the world. This capacity to conduct remote-controlled airstrikes against closely observed targets is arguably an extraordinary development in world affairs. Users of armed drones can surmount physical limits of space and time to an unprecedented degree, and governments are increasingly equipping themselves to project state power in this way. According to recent assessments produced by non-government organisations, at least ten and as many as thirteen states have conducted drone strikes (Gettinger 2019, xiii; Drone Wars 2020). Another thirty states have the capacity to do so or are in the process of acquiring it (Gettinger 2019, xiii). Seven states run armed drone development programmes, five states (China, Iran, Israel, the United States and Turkey) export armed drones, and thirteen states have imported them. France and the United Kingdom, for example, have conducted drone strikes

6686_Enemark.indd 1

02/03/21 12:22 PM

2 / Christian Enemark

using imported, US-made Reaper aircraft. And Italy and Australia have placed orders for armed Reapers which are expected to become operational in the early 2020s (Drone Wars 2020). As lethal drone technology attracts higher levels of government interest and investment, remotely controlled aircraft are becoming more capable as well as more numerous, and technological momentum towards drones controlled by artificial intelligence (AI) is accelerating. Meanwhile, the notion of ‘risk-free’ engagement with remote enemies continues to be highly attractive to political leaders who are sensitive to their citizens’ aversion to casualties among national personnel deployed to distant and dangerous places. Thus, it is reasonable to expect that the amount of drone violence going on in the world could increase, with potentially many human lives at stake. This prospect raises important ethical concerns for scholars and policymakers alike. Yet uncertainty persists about the moral status of this mode of remote-control killing and why it should be restrained. We keep asking, for example: how, if at all, is drone violence morally distinguishable from (better or worse than) other forms of political violence? On what basis can and should the use of armed drones be morally justified or condemned? Who is justly or unjustly affected (directly and indirectly) by violent drone use, and how? And how might the risks of certain kinds of drone-related injustice be reduced? Currently, among scholars who address drone use from an ethical perspective, the debate over its moral advantages and disadvantages takes place predominantly on the terrain of Just War theory. This is the centuriesold tradition of thinking ethically about why we may go to war (jus ad bellum) and how we should fight it (jus in bello). Scholars interested in the ad bellum ethics of armed drones tend to ask whether the availability and capabilities of these aircraft increase the frequency of political decisions to use force (justly or unjustly) in world affairs. Armed drones have been commended, for example, as a means of facilitating humanitarian interventions which might otherwise be deemed far too risky (Beauchamp and Savulescu 2013). However, there is also some jus ad bellum concern that this risk-reducing technology could cause national leaders to feel less anxious than before about resorting to violence when seeking to solve political problems in world affairs (Sauer and Schörnig 2012; Kaag and Kreps 2014). When ethical attention turns to in bello matters, concern tends to focus upon whether, how or by how much armed drones increase or reduce the amount of unjust human harm caused by violence wielded for a political purpose. One perspective is that drone technology, incorporating powerful target-identification capabilities, has the capacity to enable a degree of discrimination and proportionality in the use of force that

6686_Enemark.indd 2

02/03/21 12:22 PM

Introduction: Thinking Ethically about Drone Violence / 3

is greater than what is achievable using other weapon platforms (Vogel 2010). Alternatively, from a jus in bello perspective, the accusation has been raised that armed drones might instead enable less discriminate and less proportionate uses of force because drone operators, being physically and morally disengaged from the killing process, are less likely to act carefully and humanely (Calhoun 2015). More generally, some Just War scholars have insisted that the use of armed drones is neither essentially new nor uniquely wrong, even though these aircraft might exacerbate existing ethical concerns about warfare in general. Rather, they argue, the advent of ‘drone warfare’ does not require or justify any significant modifications to the way we traditionally view the moral permissibility of killing in war (see Strawser 2010; Steinhoff 2013; Schulzke 2017). Against this general view, however, other authors have claimed that radical remoteness in the exercise of violence is distinctive enough to raise profound questions about its moral basis. Highlighting the absence of physical risk to the radically remote drone operator, these authors have challenged the presumption (underpinning ethical assessments that reference Just War principles) that drone violence always counts as war in the first place. Daniel Brunstetter and Megan Braun (2011) have argued, for example, that small-scale drone strikes need to be judged on non-war terms because they are essentially a form of vim (‘force short of war’). Hugh Gusterson (2016) has described the use of armed drones as ‘a new form of state violence’ that does not equate to ‘war’ as a legal, ethical or military concept. And for Grégoire Chamayou, the armed drone is an ‘unidentified violent object’ which gives rise to ‘intense confusion’ about the applicability of ethical categories (Chamayou 2015, 14). As a response to this uncertainty about how to conceptualise and make judgements about drone violence, this volume assembles a set of ethical assessments that includes but is not restricted to applications of traditional Just War principles. In presenting a series of original essays demonstrating a variety of ways of thinking ethically about current and future drone use, the aim is to expand the scope for discerning its potential to generate just and unjust effects. Accordingly, the content of the volume is structured around four broad themes: drone strikes as war; drone strikes as violent law enforcement; the conduct of drone strikes by drone operators; and drone strikes involving AI. When airpower is subjected to ethical assessment, it is commonplace to highlight 1945 as a key historical milestone. In August of that year, when the Japanese cities of Hiroshima and Nagasaki were struck by atomic bombs dropped from piloted US aircraft, an apotheosis of airpowered atrocity was achieved. The deliberate and near-instantaneous destruction of tens of

6686_Enemark.indd 3

02/03/21 12:22 PM

4 / Christian Enemark

thousands of civilian lives was a clear and colossal breach of in bello ethics. However, for the purposes of this volume, our moral interest is attracted not only to the endangerment of civilians but also to the protection of those who directly wield airpower. A more relevant milestone, therefore, is the 1999 campaign by NATO member states (Operation Allied Force) against Serbia in its southern province of Kosovo, undertaken to prevent ‘ethnic cleansing’ there. This campaign (conducted entirely from the air) exemplified a newly heightened political desire to use violence in ways that avoid exposing a state’s own personnel to physical risk. During bombing expeditions, Serbian anti-aircraft missiles posed a risk, but NATO commanders were so determined not to lose any pilots that they ordered bomber aircraft to remain above 15,000 feet (and thus well beyond the range of retaliation) (Creveld 2011, 327). Soon afterwards, with the advent of armed drones, it was possible to reduce the risk of losing pilots even further: by removing them from the cockpit. Kosovo is the starting point for the volume’s discussion on the theme of war. In Chapter 1, Robert Sparrow reflects upon the work of philosopher Paul Kahn as it relates to drone violence, focusing on two of Kahn’s articles: ‘ War and Sacrifice in Kosovo’ (1999) and ‘ The Paradox of Riskless Warfare’ (2002). Together, these articles represent a compelling expression of the intuition that there is something deeply troubling about the profound asymmetry of risk made possible by modern technologies of long-range killing. Kahn’s ideas have been influential, and much cited, in the debate that later emerged about the ethics of airstrikes using remote-controlled drones. However, as Sparrow argues in his chapter, most participants in that debate have struggled to motivate the intuition any more than Kahn did, let alone to justify it. Many have instead concluded that extreme asymmetry is something to be sought after, in order that states may better prosecute war (or defend themselves against terrorist threats) while imposing less risk on national military personnel. Taking stock of two decades of debate about the ethics of ‘risk-free’ warfare – and drone strikes in particular – Sparrow’s chapter assesses what we have learned from Kahn and his critics, and it returns afresh to the critical issue of whether the use of armed drones presents unique moral challenges. In Chapter 2, the ethical discussion of armed drones moves squarely into the realm of Just War theory. Christian Braun places drone use in the context of a contentious development in contemporary military affairs: the increase in instances of states resorting to so-called ‘force short of war’ (vim). On the one hand, he observes, a shift in modern warfare away from large-scale slaughter and towards more calibrated applications of force (such as drone strikes) may be a step in the right direction in so far as

6686_Enemark.indd 4

02/03/21 12:22 PM

Introduction: Thinking Ethically about Drone Violence / 5

it signals greater constraint. On the other hand, because vim can appear to be more compartmentalised and therefore containable, there is a concern that this concept of non-war violence might prompt states to care less about seeking non-violent methods of political problem-solving. Braun acknowledges that the idea and practice of ‘force short of war’ put traditional jus ad bellum principles under pressure. Even so, in drawing upon the classical Just War thinking of Thomas Aquinas, he concludes that the introduction of an alternative framework of jus ad vim is unnecessary. Braun argues instead that the jus ad bellum needs to be renegotiated in the light of novel technologies and circumstances. Providing such a renegotiation, he explains how the ad bellum principles of sovereign authority, just cause and right intention should be applied to address moral concerns surrounding drone-based targeted killings. Sometimes, it might make more sense to think ethically about drone violence as being neither vim nor bellum. In Chapter 3, Max BrookmanByrne begins to shift the volume’s focus away from war and towards another non-war concept of state violence: lethal law enforcement. From an international law perspective, analysing the extraterritorial use of armed drones for a law enforcement purpose can appear simple: if they are used during armed conflict, there is a requirement to abide by international humanitarian law (IHL); outside of armed conflict, the use of armed drones must adhere to international human rights law (IHRL). However, as BrookmanByrne argues, such binary thinking is not well matched to the ‘complicated reality’ of law enforcement drone strikes. In considering drone strikes conducted during an armed conflict, he challenges the presumption that these are necessarily part of that conflict and therefore governed by the permissive rules of IHL. The chapter’s examination of US drone strike campaigns in Yemen and Somalia suggests another possibility: that the use of armed drones emulates colonial-era techniques of territorial policing based on airpower. If such use is thus more accurately regarded as serving a law enforcement function, Brookman-Byrne reasons it should be regulated not by IHL but by the more restrictive rules of IHRL. In Chapter 4, Christian Enemark offers another perspective on the state use of armed drones as an object of governance manifesting outside the paradigm and morality of war. In some circumstances, he argues, the drone violence wielded remotely by the US government resembles more closely a non-war form of state violence – lethal, punitive law enforcement (capital punishment) – even though the lawfulness of that violence itself is questionable. In parts of the world that are relatively under-governed and outside zones of ongoing conflict, the remote user of armed drones appears sometimes to act in the manner of a lawman delivering ‘wild

6686_Enemark.indd 5

02/03/21 12:22 PM

6 / Christian Enemark

justice’ to outlaws. This wildness manifests in the US government’s recent practice of conducting ‘personality strikes’ against alleged terrorists. These strikes, Enemark argues, amount to a practice of administrative execution and thus fall short of ethical expectations for the state use of lethal force in non-war (law enforcement) circumstances. Instead, this punitive form of drone violence is ‘wild’ because it merely mimics the legalism of proper criminal justice practice. When drone-based killings in the guise of law enforcement are sanctioned and conducted secretly by non-judicial agents of government, the ethical problem is that there is too much potential for unjust (arbitrary) violation of a person’s right to life. As the chapter shows, drone violence as wild justice thus presents a governance challenge: either to prohibit personality strikes outside war zones or to tame such violence by arranging for it to be judicially authorised. In Chapter 5, Christopher Fuller continues problematising the assignment of drone violence to the war paradigm, shifting the discussion towards the UK experience. The chapter assesses Prime Minister David Cameron’s 2015 authorisation of the drone-based targeted killing of Reyaad Khan, a British citizen and member of Islamic State. This attack, carried out in Syria, was the first known instance of a British drone being used lethally inside a country where the UK is not involved in a war. Accordingly, Cameron described the strike as ‘a new departure’ for the UK when revealing his government’s new drone policy and its adoption of the US government’s controversial interpretation of international law for ‘war on terror’ purposes. Fuller provides an ethical assessment of the way in which the UK government has remained vague about the purpose of its post-2015 drone policy, the legal basis which underwrites it, the associated decision-making process and the accountability mechanisms which exist for strikes. He explores the evolution of the UK’s armed drone fleet, ministers’ statements and government policy documents in an effort to reveal the ethos which has driven the government’s adoption of lethal drones for counterterrorism. In doing so, Fuller advances an argument that Britain’s drone policy is the product of a set of morally problematic attitudes towards self-defence, justice and the imminence of terrorist threats. Beyond considerations of state policy and practice, individual agency is morally significant too, so it is useful also to conceptualise drone violence by reference to the perspective and personal experience of the drone operator who directly wields that violence. In Chapter 6, Peter Olsthoorn moves the ethical discussion away from traditional, rules-based, statecentric modes of thinking about military violence. Today, most militaries regard virtue ethics as an important complement to regulations and codes imposed from above. Established military virtues such as honour, courage

6686_Enemark.indd 6

02/03/21 12:22 PM

Introduction: Thinking Ethically about Drone Violence / 7

and loyalty dominate official lists of virtues, values and desired behaviours within those militaries, and there is a growing body of literature on military virtues. Yet, as Olsthoorn argues in this chapter, it is evident that some conventional soldierly virtues are not particularly helpful to military personnel operating armed drones. His chapter assesses the ethics of drone violence from the perspective of military virtues and military ethics education, and it addresses the question of what change might be needed: a new set of virtues; a different interpretation of the existing virtues; or an abandonment of efforts to teach virtues? This challenge arises, Olsthoorn argues, partly because armed drones raise risk asymmetry in war to a new level, so the chapter also addresses the extent to which, and how, the possibility of ‘riskless warfare’ make drone use ‘virtue-less’. In Chapter 7, Lindsay Clark and Christian Enemark follow Olsthoorn in exploring the necessity for refreshing our ethical thinking when it comes to armed drones. Drawing upon feminist theory, they explain the relevance and potential applicability of an ‘ethics of care’ to the life-and-death decisions to be made by individual drone operators. Although some scholars have applied care ethics to a range of peacetime contexts, this mode of morality has hitherto been largely absent from thinking about contemporary warfare. The chapter is premised on an expanded notion of what constitutes ‘just’ drone warfare, and the rationale for such expansion is that it promises to illuminate potential injustices which might otherwise go unnoticed. The authors argue that, in order for the operators of armed drones to care (ethically) for potential victims of drone strikes, two things must be recognised and respected: first, that in the exercise of wartime violence the potential unjust effects upon innocent bystanders are not limited to immediate physical injury and death; and second, that an armed drone equipped with a powerful video-camera has the capacity to reveal to its operator some of the intimate features of a targeted individual’s ordinary life. In combination, as the chapter shows, these factors can generate an ethical challenge for a conscientious drone operator: to avoid killing an individual in circumstances where he or she would thereby knowingly deprive dependent civilians (e.g. the individual’s family members) of a critical source of care. Currently, the process of conducting drone strikes requires intensive human input, but in Chapter 8 Peter Lee shifts the volume’s ethical discussion towards a future in which that process relies increasingly upon machine autonomy. Although some militaries have ambitions to develop and deploy a drone system (for example, the MQ-9 Reaper system) that incorporates autonomous functions, the expected emergence of AI-enabled drone strikes presents practical and ethical challenges. The relevant practical considerations are illustrated by the chapter’s explanation of the evolution in applications of

6686_Enemark.indd 7

02/03/21 12:22 PM

8 / Christian Enemark

military airpower involving piloted, remotely controlled and autonomous aircraft. As Lee then explains, the way we think ethically about and plan for airborne lethal autonomous weapon systems (LAWS) is made difficult by several factors. These include the ongoing contestation among philosophers and engineers about the meaning of non-human ‘autonomy’ and ‘degrees’ of autonomy. Also, there is persistent concern about human biases possibly being transferred into machine decision-making, and the issue of how fairly to distribute moral responsibility for LAWS-generated effects remains unresolved. To address these challenges, the chapter discusses some suggested principles that could guide and restrain the future use of armed and autonomous drones. In Chapter 9, Thompson Chengeta continues the exploration of ethical challenges potentially arising from AI-controlled drones, and he focuses on how their use might be restrained through international legal regulation. His starting point is the 2013 recommendation of a moratorium on the production of LAWS to the UN Human Rights Council by its Special Rapporteur on extrajudicial, summary or arbitrary executions. The response by UN member states to this recommendation was to resolve that relevant discussions should occur within the framework of the UN Convention on Certain Conventional Weapons (CCW). However, the critical problem here is that the introduction of CCW-based regulation requires consensus among all the treaty’s members. Thus, to achieve principled and legally binding restraints on the use of autonomous armed drones, scholars and policy practitioners need to confront a set of challenges to multilateral consensus. As Chengeta explains, these challenges include: threats to multilateralism in arms control generally; ongoing concerns about a military AI arms race; anti-activist sentiments and ‘banphobia’ among arms control diplomats; and differing international understandings of what moral values are applicable to the deployment of autonomous weapons systems.

References Beauchamp, Zack, and Julian Savulescu. 2013. ‘Robot Guardians: Teleoperated Combat Vehicles in Humanitarian Military Intervention’, in Killing by Remote Control: The Ethics of an Unmanned Military, edited by Bradley J. Strawser. Oxford and New York: Oxford University Press, 106–25. Brunstetter, Daniel, and Megan Braun. 2011. ‘ The Implications of Drones on the Just War Tradition’, Ethics & International Affairs 25 (3): 337–58. Calhoun, Laurie. 2015. We Kill Because We Can: From Soldiering to Assassination in the Drone Age. London: Zed Books. Chamayou, Grégoire. 2015. Drone Theory. London: Penguin.

6686_Enemark.indd 8

02/03/21 12:22 PM

Introduction: Thinking Ethically about Drone Violence / 9

Creveld, Martin van. 2011. The Age of Air Power. New York: PublicAffairs. Drone Wars. 2020. ‘Who Has Armed Drones?’, updated July 2020, https://dronewars. net/who-has-armed-drones Gettinger, Dan. 2019. The Drone Databook. Center for the Study of the Drone, Bard College, https://dronecenter.bard.edu/projects/drone-proliferation/databook Gusterson, Hugh. 2016. Drone: Remote Control Warfare. Cambridge, MA: MIT Press. Kaag, John, and Sarah Kreps. 2014. Drone Warfare. Cambridge: Polity. Sauer, Frank, and Niklas Schörnig. 2012. ‘Killer Drones: The “Silver Bullet” of Democratic Warfare?’, Security Dialogue 43 (4): 363–80. Schulzke, Marcus. 2017. The Morality of Drone Warfare and the Politics of Regulation. London: Palgrave Macmillan. Steinhoff, Uwe. 2013. ‘Killing Them Safely: Extreme Asymmetry and Its Discontents’, in Killing by Remote Control: The Ethics of an Unmanned Military, edited by Bradley J. Strawser. Oxford and New York: Oxford University Press, 179–208. Strawser, Bradley J. 2010. ‘Moral Predators: The Duty to Employ Uninhabited Aerial Vehicles’, Journal of Military Ethics 9 (4): 342–68. Vogel, Ryan J. 2010. ‘Drone Warfare and the Law of Armed Conflict’, Denver Journal of International Law and Policy 39 (1): 101–38.

6686_Enemark.indd 9

02/03/21 12:22 PM

ONE

Riskless Warfare Revisited: Drones, Asymmetry and the Just Use of Force Robert Sparrow

In 1999 Yale law professor Paul Kahn published some philosophical reflections on the then recent air campaign over Kosovo, under the title ‘ War and Sacrifice in Kosovo’. This paper and his subsequent ‘ The Paradox of Riskless Warfare’ (2002) contain some of the earliest, and still some of the most provocative, reflections on the emerging phenomenon of asymmetric wars conducted, by one party, almost entirely from the air.1 Unsurprisingly, both papers, but especially the latter, have become key sources in the literature on the ethics of drone warfare. The invitation to present at the ‘Governing Drone Violence’ workshop in 2019, some two decades after the first of these papers appeared, seemed a fitting opportunity to re-evaluate Kahn’s work in the light of the subsequent literature and experience gained from twenty years of ‘riskless warfare’. In this chapter, after setting out the key contributions of each paper in the first two sections, I note some limits of Kahn’s treatment of the ethics of asymmetric warfare, and his account of the structure of Just War theory, for debates about drones. In the fourth section, I discuss how Kahn’s claims have fared after two decades of discussion of his work, again highlighting the implications of his (mis)understanding of Just War theory, and also of his relative neglect of the normative significance of sovereignty, for the plausibility of his account of the ethics of riskless warfare. I then turn to the history of ‘riskless’ warfare over the last two decades and argue that, for the most part, it suggests that Kahn was prescient in his concerns and (thus) that the attention paid to his work has been warranted. I conclude by suggesting that, even though Kahn’s most famous claim, about the role played by risk in justifying the moral privileges of combatants, is false, his papers are likely to remain an important source of insights into the ethics of drone warfare for many years to come.

6686_Enemark.indd 10

02/03/21 12:22 PM

Riskless Warfare Revisited / 11

The Paradox of Riskless Warfare Although it was preceded by ‘ War and Sacrifice in Kosovo’, Kahn’s ‘ The Paradox of Riskless Warfare’ is more often cited, and for that reason I have chosen to begin with it. A central claim in the paper, which has been highly influential in the subsequent debate about drones, is that war is constituted by the mutual imposition of risk amongst combatants. According to Kahn, combatants have a moral permission to direct lethal force against enemy combatants because – and in so far as – enemy combatants pose a risk to their lives (Kahn 2002, 2–3). On this account, then, the right of self-defence plays a central part in explaining the ethics of war. Another key claim, which follows from the first, is that the removal of risk from ‘war’ fundamentally alters the ethics of the use of lethal force. In the absence of risk, according to Kahn, ‘war’ becomes policing (Kahn 2002, 4). Where, on Kahn’s account, the ethics of the use of force in war is structured by the difference between combatants – who pose a military threat – and noncombatants – who do not – the ethics of the use of force in policing is structured by the distinction between guilt and innocence. Combatants in war are justified in directing lethal force against other individuals who have, in an important sense, ‘done no wrong’. Moreover, combatants are also permitted to kill non-combatants as an unintended but foreseeable consequence of attacks targeted at combatants. However, neither of these permissions exists in the context of policing. Police are only justified in directing force in extremis at those who are individually morally guilty and they are under a much more demanding obligation to avoid ‘collateral damage’, if they are permitted to do so at all. Kahn’s arguments have been widely taken up in the debate about targeted killing and drone warfare (see, for example: Kaag and Kreps 2014, 107–17; Enemark 2013; Galliott 2017; Schulzke 2017). However, it is worth noting at this point that Kahn himself (2002, 5) suggests that his arguments imply that those who are themselves not at risk as a result of enemy action have no moral permission to attack enemy combatants in conventional wars – a much more radical conclusion. I will return to these central claims below. For the moment, I want to note that the ‘Riskless Warfare’ paper also includes a number of other arguments that have played important roles in subsequent debate about drones. First, Kahn suggests that, even as it allows powerful states to adhere more strictly to the Just War principles of distinction and proportionality, the pursuit of riskless warfare may lead to a circumstance where the enemy has no choice but to violate jus in bello (Kahn 2002, 6).2 For instance, weak states may only be capable of targeting enemy civilians through terrorist attacks on foreign soil if they want to strike back at an aggressor that is conducting

6686_Enemark.indd 11

02/03/21 12:22 PM

12 / Robert Sparrow

high-altitude bombing beyond the range of their anti-aircraft weapons. This argument was developed at greater length as a critique of drone warfare by Suzy Killmister (2008). Second, Kahn acknowledges that armed forces have strong moral reasons to try to reduce the risk to the lives of their own personnel and therefore to pursue riskless warfare. Famously, Bradley Strawser (2010) has endorsed this claim in order to argue that states are morally obligated to use drones in some circumstances. Third, Kahn (2002, 7) highlights the danger that an ability to fight riskless wars may undercut the democratic legitimacy of those wars that governments do choose to fight. Where there is no risk of their sons and daughters coming home in body bags, democratic publics may be disinclined to engage with debates about the wisdom of going to war, and governments may be able to initiate wars without public consultation. As Kahn notes, and Kaag and Kreps (2014) and Enemark (2013), amongst others, have argued further, the capacity of drones to facilitate the use of force without the risk of (friendly) casualties may ultimately work to the detriment of the social contract. Finally, ‘Riskless Warfare’ includes another argument about the implications of asymmetry for the ethics of wars justified as humanitarian interventions, which has not received so much attention. In so far as war requires risk, Kahn suggests, conducting riskless warfare on behalf of another requires at the very least that they are at risk (2002, 7–8). Otherwise, ‘our’ forces could not be justified in firing on targets who themselves pose no risk to our troops. It is worth noting that this argument is clearly in tension with Kahn’s claim that combatants themselves must be at risk before they have a moral permission to use deadly force. Nevertheless, he suggests that, if our use of force is ultimately going to be justified by the fact that our targets have made themselves ‘dangerous men’, those targets must genuinely be dangerous to those with whom we have chosen to identify ourselves, at least. This argument implies that the use of force to support humanitarian interventions could only be justified when massacres are taking place or there is at least a widespread threat to human life.

War and Sacrifice in Kosovo This last argument is a less ambitious version of an argument that is made at greater length in the paper ‘ War and Sacrifice in Kosovo’. This paper contains some interesting ideas that do not receive as much emphasis in ‘ The Paradox of Riskless Warfare’, including, I believe, some key insights that have been neglected in subsequent discussions. In particular, in ‘ War and Sacrifice’ Kahn argues that wars of humanitarian intervention require one to form a moral community with the population on whose behalf

6686_Enemark.indd 12

02/03/21 12:22 PM

Riskless Warfare Revisited / 13

one is intervening (1999, 6). Otherwise one could not be acting in ‘selfdefence’ when one deploys lethal force to protect them. A refusal to accept risk, according to Kahn, jeopardises this. Willingness to sacrifice is the cost of moral community (Kahn 1999, 6). Relatedly, Kahn argues that there is a rhetorical and pragmatic contradiction involved in calling something a moral emergency of the sort that might justify going to war yet at the same time being unwilling to put the lives of our troops at risk in the course of that war (1999, 4). Here, and in a number of other places, Kahn is emphasising the ‘expressive’ content of the actions of states. What does it say about us when we are unwilling to accept risks in the course of the pursuit of some goal (Kahn 1999, 4)? In particular, what does it say about us when we are unwilling to risk the lives of ‘our’ troops? How can we say that what’s happening in some other nation is so terrible that it justifies the resort to war and then imply that it’s not worth the cost of American lives? As Kahn notes, this seems especially problematic when the decision to pursue ‘riskless’ warfare may actually place the lives of those on whose behalf we are intervening at more risk; for instance, because bombs dropped from beyond the range of enemy air defences may be more likely to go astray and cause ‘collateral’ damage. An unwillingness to tolerate US casualties therefore risks implying that the lives of US citizens are of more value than the lives of the citizens of the nations in which the US goes to war (Kahn 1999, 4). ‘ War and Sacrifice’ also observes that an ability to conduct riskless wars might place powerful states under more obligations to intervene to prevent wrongs around the world (Kahn 1999, 3). This thought has been further developed by Beauchamp and Savulescu (2013) to suggest that those concerned for human rights should actually welcome the development of drones, precisely because these aircraft do lower the threshold of war and thus make it more plausible for powerful liberal democratic states to conduct wars of humanitarian intervention. Conversely, Kahn notes that the capacity to conduct riskless warfare might make it too easy for powerful states to go to war. Governments that can conduct riskless wars do not need to worry about winning over their civilian population in the face of the prospect of significant casualties in their armed forces (Kahn 1999, 5). As I shall discuss further below, this has also been a central concern in the debate about the ethics of drone warfare. Finally, Kahn observes that the willingness to intervene militarily to defend human rights anywhere in the world implies a significant transformation – and, arguably, an erosion – of sovereignty (1999, 5–6). Most obviously, the argument that large-scale human rights violations anywhere in the world justify any nation in the world taking military action to try to halt or prevent them grants sovereign immunity

6686_Enemark.indd 13

02/03/21 12:22 PM

14 / Robert Sparrow

only to states that guarantee the basic human rights of those within their territories. More subtly, Kahn suggests, in so far as we are required to postulate a moral community in order to explain how risk to others can grant us a right to use force, riskless warfare implicitly denies, or at least discounts, the normative significance of borders.3

Some Limits of Kahn’s Treatment Although Kahn’s arguments have been widely taken up in the debate about the ethics of drone warfare, it is worth emphasising that both papers predate the first reported uses of armed drones. Kahn’s meditations were prompted by the use of airpower, especially at high altitudes, and precision munitions, in the NATO campaign against Serbian forces in Kosovo, rather than by the use of armed drones. Indeed, neither paper even mentions drones. Thus, not only do Kahn’s arguments provide little support for the idea that drones raise distinctive moral issues (‘drone exceptionalism’), they actually undercut it. It also has to be said that the intellectual fecundity of Kahn’s papers is achieved at the cost of philosophical precision. In particular, Kahn tends to use the terms ‘riskless warfare’ (or ‘risk-free warfare’) and ‘asymmetry’ interchangeably. Yet a moment’s thought reveals that these are not equivalent. One might, for instance, be involved in (putatively) riskless warfare that was completely symmetric, as, for instance when two states engage in aerial warfare over the territory of another using only remotely piloted aircraft (Strawser 2013, 13–14).4 Perhaps more importantly, asymmetric warfare (or ‘asymmetry of risk’) is a much more widespread phenomenon than riskless warfare and is enabled by a much wider range of technologies. As we shall see in the discussion below, the distinction between riskless and asymmetric warfare is sometimes quite important to the discussion of the ethics of the use of force. Kahn’s discussion also suffers from two further problems that have been to the detriment of the subsequent literature on the ethics of ‘riskless warfare’. First, while Kahn’s account of the distinction between the ethics of ‘war’ and ‘policing’ purports to be structured by the distinction between ‘classic’ (Walzerian) Just War theory and revisionist Just War theory, his account of both – and of the former in particular – is inaccurate.5 As a result, he fails to properly distinguish between these two different ways of thinking about the ethics of war. As I will discuss further below, his lack of accuracy on these matters has also afflicted many of the subsequent discussions of riskless warfare. Second, and relatedly, although Kahn pays some attention to the role played by the concept of sovereignty in our thinking about war, he fails to take it seriously enough. Again, the later literature on the use of drones has often repeated this error under his influence.

6686_Enemark.indd 14

02/03/21 12:22 PM

Riskless Warfare Revisited / 15

Lessons from the Literature Let us turn now to this literature in order to evaluate Kahn’s discussion in the light of the now two decades of arguments about riskless warfare. Much of the literature drawing on Kahn has been concerned with the ethics of the use of armed drones. However, the first lesson of the literature on drones is that drones need not be armed in order to change the calculus of the use of force in particular circumstances. An unarmed drone with a laser target designator, in combination with a platform capable of delivering a precision munition, will often facilitate striking targets of opportunity with the same degree of accuracy as an armed drone. Moreover, as I observed above, riskless war can be fought without drones by means of aerial bombardment, cruise missiles, and long-range artillery fires directed at targets who lack weapons of the same range. It is not the absence of risk but the capacity to loiter that makes drones morally distinctive – if indeed there is anything morally distinctive about them. While these are important lessons of the literature post-Kahn, they do not contradict his claims so much as the claims of some of those who have relied upon him: Kahn himself was not writing about drones. Second, the last two decades have seen a vast flourishing of what has become known as ‘revisionist’ Just War theory (Lazar 2017). In so far as revisionist Just War theory has been concerned to distinguish itself from – and to criticise – classic Just War theory, it has also greatly improved our understanding of the structure and merits (or lack thereof) of the latter. As a consequence, it is now clear that the way Kahn attempts to connect the moral permission to kill with the right to individual self-defence in war is mistaken. There is not necessarily anything morally problematic about the existence of asymmetry between combatants or even the total absence of risk to particular combatants in war. It is permissible to fire at someone who is not, and has no intention of, pointing a gun at you. Indeed, in wartime it is permissible to kill enemy combatants while they are asleep or from such a range that they are not even aware of one’s existence. ‘Classic’ Just War theory explains this with reference to the collective nature of the enterprise of war. By enrolling in the armed forces, each combatant has made themselves, as Walzer puts it, a ‘dangerous man’ (sic) – but the danger combatants pose is to enemy combatants collectively, not individually (Walzer 2015, 145). Sleeping enemy combatants may be targeted because they might kill one of ‘us’ at some stage in the future (Sparrow 2011). Only if there was no threat to any of the combatants on one side of the conflict might the intuitions that Kahn trades on have some force. Even then, as long as the conflict is properly characterised as war, combatants

6686_Enemark.indd 15

02/03/21 12:22 PM

16 / Robert Sparrow

on both sides will arguably retain the moral privilege to direct lethal force at other combatants. On classic accounts, war is fundamentally a relation between states rather than between individuals, and combatants are legitimate targets because they are the means by which states attack each other and defend themselves (Sparrow 2005; Zohar 1993).6 Revisionist Just War theory is a broader church than ‘classic’ Just War theory – especially in so far as Walzer has come to represent, almost singlehandedly, the latter tradition. What unites revisionists is their commitment to individualism and consequent willingness to abandon the idea that jus in bello and jus ad bellum are independent (Lazar 2017; McMahan 1994, 2004). That is to say, according to revisionist accounts, the ethical constraints on those engaged in combat will vary depending on whether or not they are fighting in a just cause. At first sight, it might seem as though this establishes more room for the idea that enemy combatants must pose a risk to me personally before I am permitted to kill them. However, in fact, the emphasis on the different moral permissions granted to just versus unjust combatants by revisionist Just War theory has the result that risk is less important for the ethics of war. Even if just combatants do pose a risk to the lives of unjust combatants, the latter have no right to target the former because one has no right to self-defence against someone who is defending themselves against an unjust threat one has imposed oneself. Because they fight in an unjust cause, unjust combatants have no moral permission to kill (McMahan 1994; Rodin 2002). Conversely, it is the unjust threat to the basic human rights of just combatants that grant the latter the moral permission to kill unjust combatants. While this threat might involve a threat to life, it need not. For instance, just combatants would be justified in directing lethal force against unjust combatants who were armed only with (hypothetical) non-lethal weapons if this was the only way the former could prevent the latter from successfully invading and taking over their homeland (McMahan 1994). Even according to (most) revisionist accounts, then, the absence of risk need not render it illegitimate for (just) combatants to use force against enemy combatants. There is an ethical quandary associated with the existence of asymmetry in military resources available to states rather than individuals, which is already discussed in Walzer (2015, 68–72). The jus ad bellum criterion of ‘reasonable prospects of success’ suggests that weak states that are threatened by strong states may not be justified in going to war, as there is little prospect that they would win. On the one hand, this seems reasonable, as futile wars of self-defence may lead to high numbers of casualties on both sides to no good end. On the other hand, this is clearly a perverse consequence of the structure of Just War theory, in so far as it seems both

6686_Enemark.indd 16

02/03/21 12:22 PM

Riskless Warfare Revisited / 17

to encourage states to pursue military might and to deny the right of weak states to defend themselves in practice. While the use of drones may serve to dramatise such asymmetry, drones are no more or less likely to establish it than any other military technology. Thus, while there is indeed a profound dilemma here when it comes to the presence, interpretation and justification of the ‘reasonable prospects of success’ criterion in the doctrine of jus ad bellum, this is a question for the broader literature on Just War theory rather than for the literature on riskless warfare. There is, though, a related tension within jus in bello, which Kahn and, later, Killmister (2008) have highlighted. Where warfare is highly asymmetric, especially where it is conducted mostly by drone strikes, it may effectively be impossible for combatants from the weaker state to attack combatants of the more powerful state, especially while taking reasonable precautions in attack (Enemark 2013, 60; Killmister 2008). The only options available to those being targeted by drones, to place the attacking state under political pressure to desist, may be to adopt means and methods of warfare that are forbidden by jus in bello, for instance, attacking enemy combatants with improvised explosive devices (IEDs), without taking reasonable precautions to avoid civilian casualties, or directly targeting civilians in violation of the principle of distinction. This does generate a pragmatic dilemma for the attacking state. Should they continue to employ drones and risk elevating the likelihood of terrorism? Or should they reduce the chance that their enemy will resort to illegal means by allowing the enemy some opportunity to direct their attacks at legitimate targets, thereby placing the lives of their own troops at risk? In so far as the choice involves trading off different sorts of risks to different parties this is also an ethical dilemma (Enemark 2013). However, the suggestion that, in such circumstances, the prescriptions of jus in bello are waived or weakened for the weaker state is a dangerous one, with radical implications for Just War theory, at least as classically conceived of – and probably also for revisionist accounts. Traditionally, the prescriptions of jus in bello are supposed to be absolute; if a state, or even a combatant, is unable to fight without violating them, then they are not permitted to fight at all. Allowing that these prescriptions may sometimes be ignored for the sake of the pursuit of a morally important victory risks rendering them irrelevant, as no party to a conflict ever believes that it is not important that they win the battle – let alone the war. At the very least, then, there are strong rule consequentialist reasons for continuing to hold that asymmetry provides no excuse for violations of jus in bello. What asymmetry may do is place the combatants of the more powerful military under stronger obligations when it comes to the operationalisation of the requirements of jus in bello. In particular, what is required of the operators of

6686_Enemark.indd 17

02/03/21 12:22 PM

18 / Robert Sparrow

drones when it comes to ‘feasible precautions in attack’ (see ICRC 1977, Article 57) may in practice be more demanding than what is required of warfighters who are physically present in the battle space (Rosén 2014). In order to comply with the principles of distinction and proportionality, combatants must take all feasible precautions to identify that they are not inadvertently attacking noncombatants or placing a disproportionate number of non-combatants at risk when attacking combatants (Henderson 2009, ch. 7). What counts as feasible will be context-dependent, but a significant consideration will be the extent to which any given action to confirm the nature and identity of the target (and the presence or absence of non-combatants within the area likely to be affected by the weapon used to attack the target) places the combatants tasked with this action at risk (Schmitt 2005). Because the operators of remotely piloted aircraft do not subject themselves to more risk by taking more care to identify targets and determine the location of non-combatants, it might seem that it is appropriate to demand more of them when it comes to satisfying the requirement to take precautions in attack. However, while personal risk is one consideration in determining what it is reasonable to expect of warfighters by way of precautions in attack, it is not the sole or even the main consideration. The price of the moral permission to kill that is granted to combatants includes the requirement to take on some personal risk where necessary to ensure that this permission does not come at too great a cost to non-combatants (Walzer 2015, 156). The military capacity represented by the attacking (or reconnoitring) units is also relevant. For this reason, risk to the remotely operated platform itself will be relevant to the calculation of what is required by way of precautions in attack (Rosén 2014). Another important lesson from the literature is that the advent of riskless war has been even more corrosive of intuitions about sovereignty than Kahn foresaw. After the September 11 attacks, the United States reserved for itself the right to attack those it identified as its enemies wherever they were located around the globe (Boyle 2015; O’Connell 2015). Perhaps as a result of the apparent ease with which the US was able to assert this ‘right’ in the international arena, both as a matter of fact and at the level of rhetoric, subsequent discussions of drones have tended to soft-pedal the extent to which drone strikes constitute an affront to the sovereignty of the states in whose territories they occur (O’Connell 2018). However, because – and in so far as – sovereignty is established by the existence of a monopoly on the legitimate use of armed force within a territory, drone strikes in the territory of another state without the permission of its government violate sovereignty and are themselves acts of war (Boyle 2015, 120).7 If Cuba launched drone strikes in Florida there would be no equivocation about this. It is the infringement on the right of another sovereign government, rather than

6686_Enemark.indd 18

02/03/21 12:22 PM

Riskless Warfare Revisited / 19

the scale of the destruction that results, which makes the use of force an act of war. As long as we maintain this intuition, then the ethics – if not the legality – of the use of lethal drones is more straightforward than is often appreciated. If they are used in a war, then they must only be used to target combatants as per any other weapon. Drone strikes in the territory of another state without that state’s permission will be an act of war and – unless they are conducted in self-defence or (perhaps) with permission from the UN Security Council – will almost always be a violation of jus ad bellum for that reason (O’Connell 2015). If they are used with the permission of a state then the use of force should be governed by the domestic law of that state, with some constraints from international humanitarian law (Schmitt 2010, 315). In these circumstances the ethics of the use of force is indeed governed by the logic of policing rather than of war. The exception to this clear dichotomy is where strikes are used in territories over which there is no clear sovereignty, either because the state has failed or because there is a civil war. In such cases, there may well be space for arguments about jus ad vim of the sort that have been developed by Brunstetter and Braun (2013). If there is no violation of sovereignty, then the use of force per se may not constitute an act of war. If the scale of the destruction that will result from the use of drone strikes is small enough, then the conflict may not qualify as war on that basis. If both of these conditions are met, then it may be reasonable to hold that the expectation that the highly demanding requirements of jus ad bellum should be met before the use of force is legitimate is no longer justified. That is to say, it may be that we should be instead thinking about jus ad vim instead of jus ad bellum. However, it is, I think, an open question as to whether Kahn’s arguments about the logic of policing have application here. It is plausible to think that, where the burden of justification for the use of force is lower, the obligation to avoid non-combatant casualties is higher (Brunstetter and Braun 2013). If states are not fighting for survival or to defend the international order, then there can be less justification for risking the deaths of non-combatants. And if states are going to be resorting to the use of force more often, we have good reason to demand that they take more care when it comes to the selection of targets. That is to say, whatever replaces jus in bello when it comes to regulating the means used to pursue political goals through the use of ‘force short of war’ (vim) should probably be more demanding. Yet it remains hard to believe that risk to individuals – especially risk to combatants – should play a central justificatory role here. The justification of the use of force short of war still proceeds by means of reference to reasons of state, and the state’s justification for killing particular people, or attacking particular sorts of targets, is not a function of whether those

6686_Enemark.indd 19

02/03/21 12:22 PM

20 / Robert Sparrow

people or targets have the capacity to threaten the lives of those tasked with attacking them. Given that the logic of the extraterritorial use of force is already so alien to the social contract regulating the relation between the sovereign and its citizens (Kahn 2013, 203), it does not seem implausible to me that, in claiming for itself the right to take the lives of those it identifies as its enemies, the state also assumes the right to risk the lives of those who it acknowledges have done nothing wrong. That is to say, while those engaged in the just exercise of force short of war might be required to take more care to avoid innocent casualties than combatants in war, unlike those agents of the state involved in policing they may nonetheless retain, contra Kahn, the right to risk innocent casualties in situations where their own lives are not at risk and/or where there is not an urgent necessity to prevent an immediate threat to the lives of others. None of this is to imply that there are not profound ethical problems with the use of force short of war. It may well be that we should insist that all extraterritorial use of force outside of war be conceptualised as extrajudicial killing and condemned on that basis (O’Connell 2018). My point here is solely that, even if we do endorse the idea of jus ad vim, it does not follow that the absence of risk to combatants therein would dictate that such use of force should be governed by the logic of policing rather than some distinctive conception of jus ad vim. Moreover, in any case, we should resist absolutely the idea that the ability of powerful states to get away with attacks conducted in the territory of other states without the outbreak of widespread hostilities means that these are cases of vim rather than bellum. The use of force across territorial borders without permission will always be an act of war and, as such, a threat to the stability of the international order. According to my reading of the literature, then, one of the central claims of Kahn’s work on riskless warfare – that war is constituted by the mutual imposition of risk amongst combatants considered as individuals – has fared poorly in the light of subsequent reflection. However, the discussion that it provoked has contributed much to our understanding of the ethics of asymmetric and ‘riskless’ war, including, especially, drone strikes.

Lessons from History The various lessons enumerated above are only the lessons that I have drawn from the literature on riskless warfare: there may be others. Moreover, not only do we have the benefit of two decades of philosophical discussion in assessing Kahn’s arguments today, we also have the advantage of another two decades’ experience of ‘riskless’ wars waged by the United States.8 While the US invasion and occupation of Iraq and Afghanistan involved large

6686_Enemark.indd 20

02/03/21 12:22 PM

Riskless Warfare Revisited / 21

numbers of ‘boots on the ground’, all of the United States’ post-2000 military operations have been radically asymmetric. In Pakistan, Yemen, Somalia and Libya they have arguably been ‘riskless’.9 The US’s operations in Syria have involved some ground troops but have also relied heavily on the use of precision airpower. Even in Iraq and Afghanistan a significant portion of US military operations have been conducted according to the logic of riskless warfare; in particular areas and/or at particular times this logic has arguably been the dominant one. The lessons of this history are harder to draw than one might have hoped due to the controversy about the justice of the cause of the US in these wars and also about key empirical claims about the consequences of the use of drones. Nevertheless, I believe it is possible to draw a number of lessons about the practice of riskless warfare, highly relevant to Kahn’s original two papers and to subsequent discussions of the ethics of drone warfare. Again, one clear lesson is that asymmetry is not solely the product of drones. For instance, casualties amongst US ground forces in combat in Iraq and Afghanistan have been a fraction of the casualties amongst the insurgent forces against whom they have fought. This profound asymmetry is generated by use of indirect fires, superior small arms, night vision equipment, body armour, and rapid medical evacuation by helicopter, amongst other technologies. Any conclusions we reach about asymmetry will therefore have wider implications than Kahn, with his emphasis on the use of airpower, intended. The evidence from subsequent conflicts also suggests that Kahn was prescient in his worries about the implications of the pursuit of ‘riskless’ warfare for the safety of those in whose names we claim to fight. While the figures on the number of civilian casualties due to drone strikes are notoriously controversial, there is no doubt that drones have caused a significant number of civilian casualties and there is some evidence that drones cause more civilian casualties per strike than do manned aircraft (Lewis and Holewinski 2013). This may well be because the capacities of drones encourage their use to strike individuals, and targets of opportunity more generally, in circumstances where a manned aircraft would not be employed. A confounding factor here, though, is the identification of the relevant counterfactual. Arguments about the ‘precision’ of drones often proceed by drawing a contrast with the use of airpower in previous generations of conflict. If the alternative to a drone strike is saturation bombing from a B-52, drones will obviously be preferable. However, in some cases, as suggested above, were it not for the availability of drones, particular targets might not have been attacked, especially where the struggle to win ‘hearts and minds’ and/or domestic political opinion weighs against

6686_Enemark.indd 21

02/03/21 12:22 PM

22 / Robert Sparrow

causing civilian casualties. Assessing the extent to which the availability of drones contributed to civilian casualties in any given attack therefore requires assumptions about the existence and impact of alternative means for pursuing military/political goals. A similar dialectic is at play in discussions about the implications of the desire to pursue riskless warfare for the number of civilian casualties incurred in conflicts in which drones – or the use of precision airpower more generally – have played a large role. Airstrikes may start wars, but they seldom end them; conflicts that begin with airstrikes on military targets may continue via the use of small arms against civilians. The ongoing civil war in Libya is the obvious case in point. More generally, as I discuss further below, a reluctance to commit ground troops may result in conflicts continuing where a willingness to accept friendly casualties might have facilitated a rapid and decisive settlement. That being said, at least when it comes to triggering wars, the opposite worry is also sometimes apposite: in some cases, the use of ground forces might well trigger a broader conflict, causing more civilian casualties, than the use of drones. One lesson from the history, then, is that we need a better understanding of the implications of the affordances of military technologies – and of the ways these relate to political considerations – for the uses to which they are put. Important normative arguments in the debate about the ethics of riskless war turn on counterfactual claims that implicitly rely on assumptions about the affordances of drones and about the public’s tolerance for casualties, both friendly and civilian, in the course of war. This intersection of Science and Technology Studies and Political Science/Peace Studies is a productive site for further investigation. The historical evidence also seems to bear out concerns about the capacity to wage riskless war lowering the threshold of war and undercutting the democratic legitimacy of war. The two decades since the publication of Kahn’s papers have been distinguished by the number of countries in which the United States has conducted kinetic strikes. Drones have become the ‘go to’ tool for foreign policy whenever the US government has felt compelled to ‘do something’ about undesired events unfolding elsewhere in the world. While there is, admittedly, some controversy about the impact of friendly casualty rates on domestic support for participation in war (see, for instance, Gelpi et al. 2006, which also includes a useful overview of the broader controversy), it is difficult to believe that the American public would have tolerated such extensive military engagements abroad over such long periods, especially in so far as these have all been ‘wars of choice’, except for the fact that casualty rates have been dramatically reduced as a result of the US preference for riskless war.10

6686_Enemark.indd 22

02/03/21 12:22 PM

Riskless Warfare Revisited / 23

Contra Beauchamp and Savulescu (2013), though, the increased willingness of the US to use military force has not led to better outcomes when humanitarian emergencies have occurred. While there are some well-publicised incidences of drone strikes being used to avert particular abuses (Vale 2017), one would be hard pressed to nominate an example where the use of airstrikes has been associated with a reduction of the number of human rights abuses in some territory or conflict. The case of the 2011 NATO intervention in Libya, which Beauchamp and Savulescu cite with some approval, has subsequently provided evidence for precisely the opposite conclusion. There, as in Syria and Yemen, in the absence of the willingness of the international community to commit substantial number of ground troops, civil wars and insurgencies have continued – and continued to generate human rights abuses – despite significant numbers of drone strikes. In fairness, Beauchamp and Savulescu’s argument is explicitly couched as a promissory note about what might happen in years to come as drone technology improves. To date, however, evidence of the contribution that drones make to responses to humanitarian emergencies remains thin. Indeed, if anything, the ability to respond with drone strikes seems to reduce the political pressure on governments to take more effective action to respond to humanitarian emergencies. There is, however, at least one interesting example where the involvement of drones arguably increased the threshold of conflict. In June 2019 Iranian military forces shot down a US Global Hawk drone, which (they claimed) had violated Iranian airspace while conducting surveillance operations (BBC 2019). Although the US initially responded by planning retaliatory airstrikes, these were quickly vetoed by US President Trump and a campaign of offensive cyber operations was announced instead (Nakashima and Sonne 2019). It is hard to imagine that there would not have been significantly more pressure on the US president to respond with kinetic attacks had Iranian forces shot down a manned aircraft. Whether the surveillance flights would have been flown – or flown with the same mission priorities – had an unmanned aircraft not been available is another question. However, presuming that the answer to this question is yes, then the use of drones instead of manned aircraft in this instance appears actually to have reduced the risk of war. Another lesson of the last two decades is that Kahn was right to draw attention to the implications of the capacity to wage riskless war for the institution of – and for our thinking about – sovereignty. As I noted above, subsequent to the September 11 attacks the US claimed for itself the right to use military force to disrupt the plans of those it identified as threatening its interests no matter where they were located. While subsequent

6686_Enemark.indd 23

02/03/21 12:22 PM

24 / Robert Sparrow

administrations have backed away from defending this presumptive ‘right’ so aggressively in public, the US continues to conceptualise the use of drones as a legitimate tool of foreign policy. Critics of the ‘global war on terror’ have pointed out that the assertion of this right denies, or at least discounts, the sovereignty of other states. However, it is now clear that the implications of the capacity to wage riskless warfare for sovereignty are radically uneven. A select group of states, primarily in Western Europe and North America, have ‘real’ sovereignty. Everybody knows that the use of force on their territories without their permission would be acknowledged globally as an act of war, which would be responded to with a level of force easily recognisable as war. But the rest of the world has no presumptive immunity to the operation of drones. The capacity of powerful states to wield force across the entire face of the globe has therefore arguably led to the evolution of a two-tier international order. In so far as sovereignty is a normative as well as an empirical notion, it is difficult to clearly distinguish ‘theoretical’ from practical threats to states’ sovereignty. Nevertheless, there is a sense in which the willingness of powerful states to use force across national borders represents an external and conceptual threat to sovereignty. If they do choose to use force, of course, this constitutes a practical, albeit still external, threat to sovereignty. However, in some cases, drone strikes also pose a practical internal threat to the sovereignty of the states in which they take place. The political consequences of a policy of drone strikes may include the collapse of civil order and the onset of civil war. This happened in Libya and worries about the possibility it might occur again have played a significant role in foreign policy deliberations about the use of drones in Pakistan and Syria. Note that if State A conducts military operations, including airstrikes, within the territory of State B without State B’s permission, this is itself sufficient to show that State B no longer possesses a monopoly on the use of force within its territory. Should the public come to believe that the state can no longer protect it from the predations of foreign actors, this perception is likely to undercut support for the government and increase the risk of civil unrest. Kahn’s arguments about the role played by imagined communities and differential weighting of human lives in the conduct of riskless war have also, for the most part, been borne out by the history of conflicts since he published them. As I noted above, Kahn suggested that the advent of riskless warfare establishes – or perhaps just intensifies – two ethical and rhetorical challenges when it comes to the justification of wars of humanitarian intervention. According to Kahn, in the absence of risk to members of one’s own political community, one could only be justified in the use of force against those who are carrying out atrocities abroad if one was willing to expand

6686_Enemark.indd 24

02/03/21 12:22 PM

Riskless Warfare Revisited / 25

one’s sense of moral community to include those one intends to defend. When one does this, a risk to them then becomes a risk to ‘us’ and so the use of force may be justified. However, at the same time, because the use of airpower will often elevate the risk to the lives of those on whose behalf we intervene, an unwillingness to place ‘our’ own troops at risk implies that the lives of those we claim to be defending are worth less than the lives of our own citizens. This in turn undercuts the claims of intervening states to be genuinely motivated by humanitarian concerns. Both of these dynamics have been present to some degree at least in arguments about the justification of military interventions in Libya, Syria and Iraq. The outrages being committed against civilians in these conflicts were widely publicised at key points in the campaign to rally political support for military interventions. Yet there has been little political will for anything more than ‘riskless’ warfare, with the consequence that the situation of the civilian populations in these countries has remained dire. The failure of the international community to do more does indeed call into question whether these interventions have genuinely been motivated by humanitarian concerns rather than by realpolitik. Moreover, not only have civilians sometimes been exposed to more risk as a result of the US’s preference for remote targeting on the basis of aerial surveillance over more demanding military operations conducted on the ground, but also the US’s decisions about where it chooses to employ drones suggest that not all human lives count the same in their strategic and ethical calculations. For instance, while the US has been willing to attack ‘terrorists’, with concomitant risk to civilians, in Yemen and Somalia, it is hard to imagine them doing so in France or Germany. The availability of alternative means to arrest suspected terrorists or disrupt terrorist plots in the latter states is, one suspects, only part of the explanation. As Uwe Steinhoff (2013) has observed, the practice of riskless warfare seems essentially continuous with a racist history of imperialism in which white people assumed the right to settle the affairs of people with brown skin and tried to do so by bombing them. It is, for instance, striking how few people with white skin have been killed by drones in the last two decades. One complexity here is that, as I observed above, the use of airpower alone is not capable of protecting civilians from atrocities. Perhaps because this is becoming increasingly obvious, the argument for the use of drones has, to some extent, shifted from protecting the innocent to punishing the guilty. It is the ‘evil’ of ISIS that has justified drone strikes rather than the suffering of the civilian populations where ISIS has operated. Again, though, one wonders whether there would be governmental or public support in the US for these strikes except for the fact that they take place in the territories of (what are, according to an essentially racist narrative) ‘savage’ peoples.

6686_Enemark.indd 25

02/03/21 12:22 PM

26 / Robert Sparrow

The Future of Riskless War? War is in part a technological endeavour. New technologies have the potential to reshape the face of war. However, the timeline over which such changes occur is uncertain and often longer than we expect. The modern use of drones and the development of increasingly autonomous weapons seem likely to prompt such a revolution in military affairs. While it is still, in a sense, early days for the practice of riskless warfare, one thing its brief history to date shows is that its impact on military culture and organisation is likely to be profound. What does it mean to be a warrior when one no longer needs physical courage to participate in war (Sparrow 2013; 2015a; 2015b)? What sorts of people, with what sorts of capabilities and strengths, should militaries be looking to recruit to operate their drones? The amount of philosophical and anthropological attention being paid to the idea of ‘moral injury’ at the moment (Enemark 2019; Kirkpatrick 2015; Lee 2018; Meagher 2014) is one indication that various armed forces are already seeking to recast courage as something that still has a central role to play in an age of riskless war. The challenge of reimagining the warrior is only likely to become more profound as drones and autonomous weapons become more capable. Kahn was one of the first people to draw attention to the challenges posed by these technologies for the ethical framework we use to think through the justification of war. With the benefits of two decades of hindsight we can now see just how prescient these early papers were. While, as I have argued, the central claim about the role of risk in justifying the moral privileges of combatants can now be seen to be mistaken, the observations and provocations that Kahn made alongside of them have proved to be incredibly productive in terms of the subsequent debate about the ethics of drone warfare. When it comes to the future of the debate about riskless war, then, we may still find things to learn from its origin.

Notes I would like to thank Christian Enemark for the invitation to participate in the ‘Governing Drone Violence’ workshop in Southampton in 2019. I would also like to thank the other participants in the workshop for discussion of the paper that became this chapter: conversations with Daniel Brunstetter and Jesse Kirkpatrick were especially helpful. Thanks are also due to Tony Coady, Larry Lewis and Mary Ellen O’Connell, for correspondence and assistance over the course of the drafting of this chapter. 1. Michael Walzer also published an essay ‘Kosovo’ in 1999 (Walzer 1999). The only other competitor in this regard, Michael Ignatieff’s Virtual War: Kosovo and Beyond, was published in 2000 (Ignatieff 2000).

6686_Enemark.indd 26

02/03/21 12:22 PM

Riskless Warfare Revisited / 27

2. Roughly speaking, the principle of distinction requires that combatants only directly target other combatants or objects that have a clear military role, while the principal of proportionality requires them to use only the amount of force required to achieve their military objective. 3. Kahn later expanded on this idea in a 2013 paper, ‘Imagining Warfare’ (Kahn 2013): I regret that space restrictions prevent me from treating that discussion in the current context. 4. As we will see below, even in this case war would involve risk to the interests of the warring states. 5. It is worth observing that there is significant controversy about the extent to which Walzer’s account itself revises rather than accurately reflects the history of the Just War tradition. 6. Kahn himself acknowledges this in later work (Kahn 2013). 7. It is worth observing that it is possible, perhaps even likely, in some cases that the United States did in fact have permission from the government of the relevant territory to conduct drone strikes, although those governments denied this publicly in order to avoid domestic political backlash. 8. British forces have also been involved in ‘riskless warfare’ in Afghanistan and Syria, as have French forces in Libya and Mali. There is also, of course, a long history of the use of drones by Israel in Palestine, Syria, Egypt, Lebanon and Iran. For reasons of space I shall concentrate here on those lessons that might be drawn from the history of the US’s engagements. 9. In so far as these campaigns have also involved covert operations by Special Forces on the ground, ‘risk’ here is a matter of degree. Nevertheless, it is plausible to hold that these campaigns were conducted in the expectation that it would be possible to dramatically reduce friendly casualties by prioritising attacks that could be conducted from the air. 10. In this context, the distinction between risk-free and asymmetric warfare is important – at least at the level of the imaginary of war. As the US experience of its invasions of Vietnam, Iraq and Afghanistan attest, asymmetric wars still generate friendly casualties. ‘Risk-free’ warfare, by definition, does not. When it comes to the initial level of public support for going to war, it is not the actual numbers of casualties but rather the imagined casualties that are at issue. As conflicts proceed, the actual casualty rates become increasingly important for the level of public support for war.

References BBC. 2019. ‘Strait of Hormuz: US Confirms Drone Shot Down by Iran’, BBC News, 20 June, https://www.bbc.com/news/world-middle-east-48700965 Beauchamp, Zack, and Julian Savulescu. 2013. ‘Robot Guardians’, in Killing by Remote Control: The Ethics of an Unmanned Military, edited by Bradley J. Strawser. Oxford and New York: Oxford University Press, 106–25. Boyle, Michael J. 2015. ‘ The Legal and Ethical Implications of Drone Warfare’, International Journal of Human Rights 19 (2): 105–26.

6686_Enemark.indd 27

02/03/21 12:22 PM

28 / Robert Sparrow

Brunstetter, Daniel, and Megan Braun. 2013. ‘From Jus ad Bellum to Jus ad Vim: Recalibrating Our Understanding of the Moral Use of Force’, Ethics & International Affairs 27 (1): 87–106. Enemark, Christian. 2013. Armed Drones and the Ethics of War: Military Virtue in a Post-Heroic Age. London: Routledge. Enemark, Christian. 2019. ‘Drones, Risk, and Moral Injury’, Critical Military Studies 5 (2): 150–67. Galliott, Jai C. 2017. Military Robots: Mapping the Moral Landscape. London: Routledge. Gelpi, Christopher, Peter D. Feaver and Jason Reifler. 2006. ‘Success Matters: Casualty Sensitivity and the War in Iraq’, International Security 30 (3): 7–46. Henderson, Ian. 2009. Contemporary Law of Targeting: Military Objectives, Proportionality and Precautions in Attack under Additional Protocol I. Leiden: Martinus Nijhoff. ICRC. 1977. Protocol Additional to the Geneva Conventions of 12 August 1949, and Relating to the Protection of Victims of International Armed Conflicts (Protocol I), 8 June, International Committee of the Red Cross, https://ihl-databases.icrc.org/ihl/ INTRO/470 Ignatieff, Michael. 2000. Virtual War: Kosovo and Beyond. London: Chatto & Windus. Kaag, John, and Sarah Kreps. 2014. Drone Warfare. Cambridge and Malden, MA: Polity. Kahn, Paul W. 1999. ‘ War and Sacrifice in Kosovo’, Philosophy and Public Policy Quarterly 19 (2/3): 1–6. Kahn, Paul W. 2002. ‘ The Paradox of Riskless Warfare’, Philosophy and Public Policy Quarterly 22 (3): 2–8. Kahn, Paul W. 2013. ‘Imagining Warfare’, European Journal of International Law 24 (1): 199–226. Killmister, Suzy. 2008. ‘Remote Weaponry: The Ethical Implications’, Journal of Applied Philosophy 25 (2): 121–33. Kirkpatrick, Jesse. 2015. ‘Drones and the Martial Virtue Courage’, Journal of Military Ethics 14 (3–4): 202–19. Lazar, Seth. 2017. ‘Just War Theory: Revisionists Versus Traditionalists’, Annual Review of Political Science 20: 37–54. Lee, Peter. 2018. Reaper Force: Inside Britain’s Drone Wars. London: John Blake. Lewis, Larry, and Sarah Holewinski. 2013. ‘Changing of the Guard’, Prism 4 (2): 56–65. McMahan, Jeff. 1994. ‘Innocence, Self-Defense and Killing in War’, Journal of Political Philosophy 2 (3): 193–221. McMahan, Jeff. 2004. ‘ The Ethics of Killing in War’, Ethics 114 (1): 693–732. Meagher, Robert E. 2014. Killing from the Inside Out: Moral Injury and Just War. Eugene: Wipf and Stock. Nakashima, Ellen, and Paul Sonne. 2019. ‘U.S. Military Carried Out Secret Cyberstrike on Iran to Prevent It From Interfering with Shipping’, Washington Post, 29 August, https:// www.washingtonpost.com/national-security/us-military-carried-out-secret-cyberstrike-on-iran-to-prevent-it-from-interfering-with-shipping/2019/08/28/36202a4ec9db-11e9-a1fe-ca46e8d573c0_story.html

6686_Enemark.indd 28

02/03/21 12:22 PM

Riskless Warfare Revisited / 29

O’Connell, Mary E. 2015. ‘International Law and Drone Attacks Beyond Armed Conflict Zones’, in Drones and the Future of Armed Conflict, edited by David Cortright, Rachel Fairhurst and Kristen Wall. Chicago: University of Chicago Press, 63–73. O’Connell, Mary E. 2018. ‘ The Presumption of Peace, Illegal War, Human Rights, and Humanitarian Law’, in Seeking Accountability for the Unlawful Use of Force, edited by Leila N. Sadat. Cambridge: Cambridge University Press, 526–47. Rodin, David. 2002. War and Self-Defense. Oxford: Clarendon Press. Rosén, Frederik. 2014. ‘Extremely Stealthy and Incredibly Close: Drones, Control and Legal Responsibility’, Journal of Conflict and Security Law 19 (1): 113–31. Schmitt, Michael N. 2005. ‘Precision Attack and International Humanitarian Law’, International Review of the Red Cross 87 (859): 445–66. Schmitt, Michael N. 2010. ‘Drone Attacks Under the Jus ad Bellum and Jus in Bello: Clearing the “Fog of Law”’, Yearbook of International Humanitarian Law 13: 311–26. Schulzke, Marcus. 2017. The Morality of Drone Warfare and the Politics of Regulation. London: Palgrave Macmillan. Sparrow, Robert. 2005. ‘“Hands Up Who Wants to Die?”: Primoratz on Responsibility and Civilian Immunity in Wartime’, Ethical Theory and Moral Practice 8 (3): 299–319. Sparrow, Robert. 2011. ‘Robotic Weapons and the Future of War’, in New Wars and New Soldiers: Military Ethics in the Contemporary World, edited by Jessica Wolfendale and Paolo Tripodi. Burlington, VT: Ashgate, 117–33. Sparrow, Robert. 2013. ‘ War Without Virtue?’, in Killing by Remote Control: The Ethics of an Unmanned Military, edited by Bradley J. Strawser. Oxford and New York: Oxford University Press, 84–105. Sparrow, Robert. 2015a. ‘Drones, Courage, and Military Culture’, in Routledge Handbook of Military Ethics, edited by George R. Lucas, Jr. Oxford and New York: Routledge, 380–94. Sparrow, Robert. 2015b. ‘Martial and Moral Courage in Tele-Operated Warfare: A Commentary on Kirkpatrick’, Journal of Military Ethics 14 (3–4): 220–7. Steinhoff, Uwe B. 2013. ‘Killing Them Safely: Extreme Asymmetry and its Discontents’, in Killing by Remote Control: The Ethics of an Unmanned Military, edited by Bradley J. Strawser. Oxford and New York: Oxford University Press, 179–207. Strawser, Bradley J. 2010. ‘Moral Predators: The Duty to Employ Uninhabited Aerial Vehicles’, Journal of Military Ethics 9 (4): 342–68. Strawser, Bradley J. 2013. ‘Introduction: The Moral Landscape of Unmanned Weapons’, in Killing by Remote Control: The Ethics of an Unmanned Military, edited by Bradley J. Strawser. Oxford and New York: Oxford University Press, 3–24. Vale, Jon. 2017. ‘RAF Drone Strike Stops Isis Carrying Out Public Execution in Syria’, The Independent, 16 May, https://www.independent.co.uk/news/world/middleeast/raf-drone-strike-isis-public-execution-syria-stop-ministry-defence-daesh-isilislamic-state-a7738836.html Walzer, Michael. 1999. ‘Kosovo’, Dissent 46 (Summer), https://www.dissentmagazine .org/article/kosovo

6686_Enemark.indd 29

02/03/21 12:22 PM

30 / Robert Sparrow

Walzer, Michael. 2015. Just and Unjust Wars: A Moral Argument with Historical Illustrations, 5th edn. New York: Basic Books. Zohar, Noam J. 1993. ‘Collective War and Individualistic Ethics: Against the Conscription of “Self-Defense”’, Political Theory 21 (4): 606–22.

6686_Enemark.indd 30

02/03/21 12:22 PM

TWO

Jus ad Vim and Drone Warfare: A Classical Just War Perspective Christian Nikolaus Braun

Introduction One of the most contentious developments in contemporary international relations has been the increase in uses of ‘force short of war’, such as targeted killings, limited retaliatory drone strikes or the imposition of no-fly zones (see, for example, Ford 2013). On the one hand, in so far as it signals greater constraint, the shift away from the mechanised slaughter of modern warfare towards more calibrated applications of force may be hailed as a step in the right direction. On the other, because force short of war appears more compartmentalised and therefore containable, it may encourage greater profligacy on the part of states in respect of the recourse to arms. Just War theory, the predominant moral framework for addressing the ethical questions that warfare provokes, has struggled to make sense of force short of war. Not only has one possible response, Michael Walzer’s suggestion of jus ad vim (the just use of force short of war), been hotly debated, but this suggestion has also become a discursive ‘battlefield’ in the ‘war of ethics within the ethics of war’ (Vaha 2013, 183) between Walzerian and revisionist Just War thinkers.1 For example, Helen Frowe (2016) and Daniel Brunstetter (2016) have disagreed about the moral purchase of jus ad vim, with the former critiquing the concept from a revisionist perspective as being redundant, and the latter, arguing in Walzer’s terms, continuing to support the viability of the suggestion. In this chapter, as a contribution to ethical thought about armed drones, I argue that the debate about the use of force short of war would benefit from a third way approach to Just War that sides with neither Walzerians nor revisionists all of the time. On the basis of the classical Just War thinking of

6686_Enemark.indd 31

02/03/21 12:22 PM

32 / Christian Nikolaus Braun

St Thomas Aquinas, I support revisionists in their critique that jus ad vim as a distinct moral framework is redundant. At the same time, however, I side with Walzerians who argue that force short of war puts the principles of Just War, in the way they are commonly applied today, under pressure. What is needed, I assert, is not a distinct new moral framework, but a renegotiation of Just War principles in the light of novel circumstances. I illustrate my argument by providing a renegotiation of the three deontological jus ad bellum criteria – sovereign authority, just cause and right intention – with regard to one of the most controversial contemporary military practices: drone-based targeted killings.

Jus ad Vim as a Distinct Moral Framework In the preface to the fourth edition of Just and Unjust Wars, Walzer (2006a, xv–xvi) introduced a novel moral framework in response to the measures taken by the US and its allies against the regime of Saddam Hussein prior to the 2003 invasion of Iraq. In what he called jus ad vim, Walzer pondered over a theory governing the just use of force short of war. In his treatment of counterterrorism policies, on which most of the jus ad vim debate subsequently concentrated, Walzer (2007, 484) observes that there is a ‘different “feel”’ to such operations as they are neither outright acts of war nor acts of peacetime policing. Consequently, Walzer (2006b, 12) argues that ‘we can’t stop with just war theory’, identifying a need to ‘maneuver between our conception of combat and our conception of police work, between international conflict and domestic crime, between the zones of war and peace’. He thus abandons the strict distinction between two moralities which he continues to uphold for what he sees as a clear dichotomy between war and peace, and his ‘maneuver’ between these two moralities gives birth to jus ad vim as a distinct third moral framework. Only ten days after the 9/11 attacks, Walzer (2001) distinguished between a ‘metaphorical war’ against terrorism and the ‘real thing’, expecting that the line between policing and war paradigms would become blurred. In order to demonstrate the difficulty of applying either paradigm, Walzer provides an example which illustrates his idea about jus ad vim. Referring to a US Hellfire missile strike against five alleged al-Qaeda militants driving in a van in the Yemeni desert in 2001, Walzer (2006b, 10) hypothetically transfers the strike to Afghanistan, a zone of war where the United States was engaged in armed conflict, and to Philadelphia, a zone of peace within the US itself.2 Had the attack happened in Afghanistan, Walzer reasons, it would have been justified to directly target terrorists as enemy combatants, as ‘It is part of the awfulness of war that people actively engaged on the

6686_Enemark.indd 32

02/03/21 12:22 PM

Jus ad Vim and Drone Warfare: A Classical Just War Perspective / 33

other side can legitimately be killed without warning’ (2006b, 10). Had the attack happened in Philadelphia, however, Walzer asserts that the policing paradigm would have had to be applied: ‘In Philadelphia, the (suspected) terrorists would have to be arrested, arraigned, provided with lawyers, and brought to trial. They could not be killed unless they were convicted – and many Americans, opposed to capital punishment, would say: Not even then’ (2006b, 10). With regard to the Yemen attack, Walzer (2006b, 11) provides the moral rationale behind jus ad vim as a hybrid between war and policing paradigms: Yemen is somewhere between Afghanistan and Philadelphia. It isn’t a war zone, but it also isn’t a zone of peace – and this description will fit many, not all, of the ‘battlefields’ of the ‘war’ against terrorism. [. . .] The Yemeni desert is a lawless land, and lawlessness provides a refuge for the political criminals called terrorists. The best way to deal with the refuge would be to help the Yemeni government extend its authority over the whole of its territory. But that is a long process, and the urgencies of the ‘war’ against terrorism may require more immediate action. When that is true, if it is true, it doesn’t seem morally wrong to target Al Qaeda militants directly – for capture, if that’s possible, but also for death. Yemen in this regard is closer to Afghanistan than to Philadelphia.

Under circumstances as found in Yemen, the concept of jus ad vim might function as justification for counterterrorism operations; although legally acts of war, jus ad vim actions should not be judged in the same way as actual war because the lethal force employed resembles neither the quantum nor the duration of traditional warfare (Walzer 2006a: xiv). Unfortunately, Walzer never elaborated in any greater detail on his idea of jus ad vim. This task has been taken on by a new generation of Just War thinkers, of whom Daniel Brunstetter (see, for example: Brunstetter and Braun 2013; Brunstetter 2019) has been the most prolific. I believe that the importance of this task is increasing as the use of armed drones (arguably a type of vim) increases worldwide.

Jus ad Vim and the ‘ War of Ethics’ Walzer’s interpretation of the Just War has been challenged ever since he first put forward his argument in 1977. Most of his contemporary critics work within the analytical tradition of moral philosophy and have thoroughly investigated his argument (see, for example: McMahan 2009; Frowe 2014). Employing Rawls’s method of reflective equilibrium, revisionists start from Walzer’s Just War idea as the ruling theory that must be scrutinised for

6686_Enemark.indd 33

02/03/21 12:22 PM

34 / Christian Nikolaus Braun

logical incoherence and pursue the goal of constructing a better theory. The primary target of these thinkers is the prominent role Walzerians award to the state. Revisionists reject the domestic analogy, which presumes that the relations between individual human beings and states are alike, and allocate moral responsibility for killing in war to individuals, not states. Their hope is to discredit Walzer’s Just War theory, which Frowe (2014, 13) describes as ‘a very state-based, collectivist approach to war’, and to argue for a ‘reductive individualism’ which is reductivist because revisionists hold that the rules which regulate killing in war are the same as those regulating interpersonal killing outside of war. The core argument of reductivism is that there exists only one set of moral principles which applies all the time, rather than distinct principles for different moral domains such as war and peace. At the same time, revisionists are individualists as they argue that moral theory must concentrate on individuals rather than collectives such as states. Given this conviction that there is only one morality, Walzer’s call for jus ad vim as a distinct third moral framework besides those of war and peace inevitably triggered a negative response from revisionists. That critique was provided by Helen Frowe, who assessed Walzer’s idea of jus ad vim as elaborated by Brunstetter and his co-author Megan Braun. Not surprisingly, as for revisionists there is no moral difference between the conditions of war and peace, Frowe (2016, 122) detects an unnecessary concentration on the question of whether a specific use of force counts as war or not. Moreover, she (2016, 119–20) criticises Brunstetter for considering the jus ad bellum principles as a ‘one-off judgment’ which only needs to be met at the onset of war, while, if interpreted correctly, they must continuously be reassessed as long as the war lasts. Most importantly, Frowe (2016, 123–6) argues that jus ad vim, as a distinct moral framework besides jus ad bellum, is ‘redundant’; the existing jus ad bellum framework can appropriately judge uses of ‘force short of war’. In his rebuttal, Brunstetter situates the conversation about jus ad vim within the wider debate between Walzerians and revisionists. Brunstetter (2016, 131) asserts that Frowe’s understanding about the use of force derives from a worldview that is ‘fundamentally’ different from his. In particular, he takes issue with the revisionist claim that there is no moral difference between the state of war and the state of peace. Brunstetter (2016, 131) further rejects the rights-based liability account which revisionists propose instead. Overall, despite their exchange about substantive aspects of jus ad vim, a reader who does not associate themselves with either camp cannot fail to notice that the conversation between the two authors seems rather narrow as a consequence of the ‘war of ethics’. Moral debate about the worrisome expansion in the use of force short of war would thus benefit from a third

6686_Enemark.indd 34

02/03/21 12:22 PM

Jus ad Vim and Drone Warfare: A Classical Just War Perspective / 35

perspective that, although not neutral in the ‘war of ethics’, can provide a distinct and valuable perspective on jus ad vim. At the same time, a third way perspective can make a valuable and distinct contribution towards ethically assessing the use of armed drones.

The Classical Contribution to the Debate about Jus ad Vim The classical approach to Just War thinking, exemplified by the ideas of medieval theologian Thomas Aquinas, can make an important contribution with regard to jus ad vim because it sides with neither Walzerians nor revisionists all of the time. While the Thomistic idea of Just War supports the Walzerian argument that there is in fact a moral difference between war and peace, it at the same time embraces the revisionist view that jus ad vim, as a distinct moral framework, is redundant. Aquinas distinguished between two moralities in the sense that only ‘general war’, or bellum generale, as a confrontation between two or more polities could constitute a Just War ‘in the most proper sense of the term’ (Reichberg 2013, 188). That is why Aquinas (ST, II-II, q. 40, a. 1),3 in his seminal definition of a Just War, gave a prominent role to the authority criterion, which limited the right to wage bellum generale to the sovereign who has been entrusted with the responsibility for the common good of the political community. Acts of proportionate self-defence, the only morally justifiable use of force by private individuals, were acts of war in a limited sense only. For ‘force used by or directed against private individuals’ (Reichberg 2013, 188) Aquinas employed the term ‘particular war’, or bellum particular (see, for example, ST, II-II, q. 123, a. 5). Consequently, bellum generale and bellum particular would not be subject to the same normative principles and the former type of war would not be reducible to the rules that apply to private self-defence (Reichberg 2013, 188). As a result, the revisionist argument that any individual has the authority to wage war in the broader sense is, from a classical Thomistic perspective, morally problematic. Aquinas, who systematised the Just War thinking of his day, carried on his immediate predecessors’ concern to employ the authority criterion as a means of restraint. The immediate problem canonists (theological lawyers) had been facing was a worrisome multiplicity of actors who all claimed to possess the right to wage war and which had resulted in ‘widespread banditry and warlordism’ (Johnson 2007, 7). By arguing that only sovereign authorities without a superior authority possessed the right to wage war in the actual sense, these thinkers delegitimised any use of force by actors who had a superior authority that went beyond the necessity of immediate self-defence. Distinguishing between force employed for public

6686_Enemark.indd 35

02/03/21 12:22 PM

36 / Christian Nikolaus Braun

and private interests, then, the Thomistic account rejects the revisionist argument that any individual can potentially wage a Just War. At the same time, however, Aquinas can be read as being supportive of Frowe’s critique that jus ad vim as a distinct moral framework besides jus ad bellum is redundant. As I will argue below, Aquinas would probably have conceptualised the targeted killing (drone-based or otherwise) of culpable unjust individuals on foreign soil by a public authority as bellum generale. Consequently, for Aquinas, Just War reasoning would have applied to such uses of force, and a distinct moral framework of jus ad vim, imagined as blending the ethics of military conduct and policing, would have seemed redundant indeed. Having said that, while targeted killings abroad would have been acts of bellum generale, Aquinas would also have used the term ‘war’ for today’s domestic police uses of force. However, such a ‘war’ would be war in the limited sense of bellum particular and would not have been subject to the same proportionality calculus as bellum generale (Reichberg 2013, 188). The reason for this is that normally domestic employments of force by the state do not rise to the magnitude or duration of war between political communities. While there may be instances of bellum particular, such as sedition, which may call for a more permissive interpretation of the amount of force that can be employed, such cases constitute exceptional circumstances. More common, during the days of Aquinas, was the domestic employment of lethal force through the imposition of the death penalty. For Aquinas, the death penalty constituted an act of bellum particular carried out by a sovereign authority against a culpable wrongdoer. In fact, one particular reading of Aquinas which concentrates on a punitive reading (see Murphy 2012) argues that he imagined the death penalty as the domestic parallel to war between political communities. The death penalty, however, if executed after a trial that fairly establishes the culpability of the wrongdoer, is the most discriminate of employments of lethal force, as only the wrongdoer is targeted and there is no risk that innocent people will be harmed. As a trial to determine the right punishment takes place outside the heat of battle, Aquinas accepted that different limitations would apply to bellum generale. In other words, while both forms of force, domestic and external, constitute ‘war’, they should not necessarily be subject to the same rules. That is why, in my following argument, I will engage with the asymmetry rationale of policing, which holds that a criminal may not resist a justified police action against her or him, and I will investigate what lessons it can teach us with regard to drone-based targeted killings. Arguing that a moral and legal equality of combatants as advocated by Walzerian Just War theorists was alien to classical thinkers like Aquinas, I suggest employing

6686_Enemark.indd 36

02/03/21 12:22 PM

Jus ad Vim and Drone Warfare: A Classical Just War Perspective / 37

the so-called asymmetry objection to drone strikes against culpable unjust individuals by a public authority. Seen from a Thomistic perspective, while the concept of jus ad vim as an independent moral framework is redundant, it is still a useful category in the sense that we might want to answer the question of how to regulate uses of force short of war by one state within the territory of another. Jus ad vim foregrounds the limitations of contemporary Just War thinking regarding the use of force short of war. As classical political thought did not distinguish between the modern paradigms of war and policing, there are interesting parallels with the recent argument for jus ad vim as a hybrid between the two and, consequently, it can help delimit which rules should govern contemporary uses of force short of war such as drone strikes.

The Ethics of Drone Warfare In this chapter, I cannot address all the moral issues raised by drone warfare. Instead, in line with the classical Just War perspective I adopt, I will concentrate on the deontological Just War criteria of sovereign authority, just cause and right intention, and I will not explicitly discuss the prudential criteria of proportionality, last resort and probability of success, which have been at the centre of much contemporary Just War thinking. Moreover, I will not provide a separate treatment of jus in bello concerns. Importantly, this does not at all mean that I seek to downplay the importance of these considerations. Critics of armed drones, like Kreps and Kaag (2012), who specifically point to the jus in bello questions of discrimination and proportionality, raise important moral questions. However, while I pay attention to these questions in my discussion of right intention, my main focus in this chapter lies elsewhere. In particular, I seek to refocus the debate about the morality of drones on the Just War tradition’s concern for the ‘ends of good politics’, namely, order, justice and peace, which are, respectively, reflected in the criteria of sovereign authority, just cause and right intention (Johnson 2013, 19–20).

Sovereign Authority The authority criterion was arguably the most important consideration for classical Just War thinkers. The reason for this is that Just War is part of an understanding of politics that aims for the common good of the political community. The sovereign at the top of the political community was considered to have a special responsibility for the common good of the political community entrusted to him, her or them, and was thus the only

6686_Enemark.indd 37

02/03/21 12:22 PM

38 / Christian Nikolaus Braun

authority who might wage war justifiably. Given the contemporary spread of non-state actors such as Islamist terrorist groups who claim the authority to wage war, and against whom armed drones have frequently been employed, one feels reminded of the multitude of medieval actors who claimed the right to war and influenced classical thinkers to develop their restrictive attitude towards the authority criterion. As indicated above, the classical conceptualisation held that only sovereign authorities could wage bellum generale. Private individuals, like today’s terrorist actors, could not. Their private employments of force would be acts of rixa, an unjust bellum particular. Rixa constituted ‘a use of force to effect private vengeance or to secure other unjustifiable ends, in the absence of authorization from legitimate authority, or in a manner disproportionate to the initial offense’ (Reichberg 2017, 270–1). Consequently, seen from a classical Just War perspective, there is no way that terrorist groups like al-Qaeda, as groups which do not take responsibility for a political community, can wage war in the sense of bellum generale because they lack sovereign authority. Admittedly, the case of the so-called Islamic State (IS) is more complex. In contrast to al-Qaeda, IS, during periods of its expansion in Syria and Iraq, could claim to rule over the people living in the territory it had conquered and, in some respects, the so-called caliphate carried out tasks commonly associated with functioning states. However, given the brutality of its reign, there can be no credible claim that IS was aiming for the common good of the people living under its rule. Consequently, from a classical Just War perspective, IS could not claim the right to wage bellum generale. This argument has important implications for the use of armed drones. One of the most common moral issues related to drone technology has been the concern that armed drones will proliferate and be employed by so-called rogue states and non-state actors. And, in fact, it has been reported that non-state actors such as IS, Hezbollah and Hamas have already used this technology (see Singer 2019). While, of course, these groups have not been able to employ the sophisticated models possessed by advanced military powers, the concern is that their efforts to master drone technology will become more advanced in the future. The contribution that classical, Thomistic Just War thinking can make in this regard is to emphasise that terrorist groups, because they lack the authority to use force that goes beyond proportionate self-defence, cannot be justified in employing armed drones to achieve political ends. Armed drones can justifiably be used in external territories only by sovereign authorities which work towards maintaining and establishing the common good: either the good of the people entrusted to them directly or, by extension, the good of all political communities.

6686_Enemark.indd 38

02/03/21 12:22 PM

Jus ad Vim and Drone Warfare: A Classical Just War Perspective / 39

Furthermore, related to the concern about the authority to wage war, the classical Just War can make a contribution with regard to the moral issue of drones and the asymmetric operator–target relationship that is characteristic of drone warfare. As the employment of armed drones creates a situation in which the operator can kill a target without having to fear a counter-attack, critics hold that such fighting is unjust (drawing, for example, upon the reasoning in Kahn 2002). In response to this critique, Bradley Strawser (2010), a leading revisionist Just War thinker, argues there is a ‘duty to employ uninhabited aerial vehicles’ in cases when the use of lethal force is justified and the just combatant, if other military means were to be employed, would face a risk of being harmed by the unjust threat. Exposing a just combatant to the risk of being harmed by an unjust combatant who has made themselves liable to lethal attack while, alternatively, an armed drone could be employed is, in Strawser’s view, morally indefensible. A crucial argument underpinning Strawser’s position is his rejection of the moral equality of combatants, which holds that when a state of war has been established combatants on both sides enjoy the same moral right to harm each other, independently of the consideration of just cause. The argument for the moral equality of combatants, most prominently defended by Walzer (2006a, 34–47), builds on contemporary international humanitarian law which grants the same legal rights to the parties opposing each other in armed conflict. Revisionists object to the moral equality of combatants thesis as they reject Walzer’s logical separation of jus ad bellum and jus in bello principles. For them, only the side that is justified in going to war can morally resort to the use of force. Consequently, the combatant fighting for an unjust cause cannot be the just combatant’s moral equal. Jai Galliott, in response to Strawser, provided a defence of the moral equality of combatants with regard to drone warfare. Building on the work of Paul Kahn, Galliott (2012, 63) argues that in order for war to be justifiable, there must be ‘some degree of mutual risk involved’ between combatants: ‘ When one side’s armed forces are incapable of posing an adequate threat to the other side, the more powerful side’s moral justification for targeting the weaker side’s armed forces is removed.’ In other words, if the operator– target relationship reaches a certain degree of asymmetry, the advantaged ‘state may be in violation of the jus ad bellum convention’ (Galliott 2012, 63). In such circumstances, Galliott reasons, the use of force should be governed by a framework that draws on the policing paradigm, which is built around the assumption that there is a moral asymmetry between the just police and unjust criminals. While he does not argue for the strict rules on the use of

6686_Enemark.indd 39

02/03/21 12:22 PM

40 / Christian Nikolaus Braun

force that apply to the police to be applied to drones, he holds that ‘less war and more successful international policing is surely an admirable goal’ (Galliott 2012, 64). As I have argued elsewhere, the revisionist rejection of the moral equality of combatants is at best old wine in new skins (Braun 2018). Classical thinkers like Aquinas had already emphasised that a combatant fighting for a just cause would not be the moral equal of his opponent fighting for an unjust cause. However, these thinkers were concerned that often it would be difficult to objectively determine who had just cause and, therefore, it could be morally justifiable to grant combatants on each side equal rights. Put differently, while unjust combatants were fighting for an unjust cause, they could be exonerated and granted equal rights because it was their sovereign who was to blame for waging an unjust war. This move, subsequently, became enshrined in the laws of war and Walzer, as he takes international law as a ‘frame’ (Johnson 2014, 5) through which he sees the moral world, accepts this compromise of sorts in his Just War argument (see Walzer 2006a, xix–xxv). In contrast, most revisionists are interested in the absolute, objective moral truth only and they show little interest in the dilemmas raised by actual warfare and the compromises it may require. In other words, regarding the equality of combatants imagined as a practical compromise arising from the realities of war, these thinkers put aside the subjective reasons that often undergird the decision to fight wars, arguing as if all parties had access to the same objective knowledge about just cause. Consequently, their rejection of the moral equality of combatants by no means reinvents the wheel of Just War thinking. Rather, what revisionists have achieved is to refocus attention on the uneasy compromise between the legality and morality of war, which has resulted in the argument for the moral equality of combatants. I argue that the classical Just War can provide a qualified defence of the revisionist argument for a ‘duty’ to employ drones. Before I begin, I need to emphasise that Galliott’s article considers war as a confrontation between states. This understanding of war as a practice between political communities also underpinned classical thinkers’ reasoning for granting combatants on each side equal rights. In what follows, however, I concentrate on the confrontation between states and terrorist groups. Interestingly, in this regard, Walzer has extended the moral equality thesis to a war between a collective and individuals. While terrorists necessarily fail the Just War criterion of sovereign authority and thus fight an unjust war, for Walzer, the confrontation with Islamist terrorists can nonetheless rise to the level of war. Consequently, once in a state of war, following the

6686_Enemark.indd 40

02/03/21 12:22 PM

Jus ad Vim and Drone Warfare: A Classical Just War Perspective / 41

moral equality thesis, individual terrorists would be the legal and moral equals of soldiers representing their collective. In particular, he argues that if the war paradigm was applied in the confrontation with groups like alQaeda, drone operators on US soil would be legitimate targets for these terrorist groups (see Walzer 2013). In contrast, for the classical Just War thinker, terrorists cannot be considered to be the moral equals of the soldiers who, representing a sovereign authority, fight them. In addition, they cannot be exonerated of their personal culpability, as might be the case for legitimate soldiers fighting on the unjust side, and should thus not be granted equal rights. The reason for this is simply that terrorist violence always resembles an illicit private use of force, not bellum generale. This understanding, crucially, is highly relevant to the moral issue of the asymmetric operator–target relationship that arises in drone warfare. Galliott’s argument that, in order for war to be justified, belligerents on both sides must face the risk of being harmed is alien to a classical view of Just War. Strawser’s revisionist argument that there is a duty to employ drones against culpable unjust threats that have made themselves liable to lethal attack, when alternative action would expose just combatants to the risk of being harmed, is thus vindicated by the classical Just War. However, there is a curious parallel between Galliott’s argument for employing a variation of the policing paradigm in asymmetric warfare and the classical understanding of Just War. I argue that drones used for targeted killings of culpable unjust individuals can be reconciled with the classical understanding of belligerency in which, as Reichberg (2013, 182) puts it: the legal effects of a just war were viewed as benefiting the just belligerent only. The unjust adversary, by contrast, was not even properly speaking a belligerent; rather he was deemed the rebellious object of armed enforcement action and in this respect was likened to a criminal resisting arrest.

Thus, in being comparable to a bank robber who may not claim a right to self-defence when surrounded by the police, a culpable unjust terrorist may not claim the belligerent rights the drone operator who targets her or him enjoys. Not unlike Galliott, who imagines an adapted policing paradigm to govern drone warfare, I suggest a renegotiation of the remaining classical jus ad bellum principles of just cause and right intention in order to make these principles applicable to the new circumstances introduced by armed drones. Importantly, while this particular renegotiation may be given the name of jus ad vim, it does not amount to a distinct third morality as it is firmly situated within the war paradigm.

6686_Enemark.indd 41

02/03/21 12:22 PM

42 / Christian Nikolaus Braun

Just Cause Both Walzerian and revisionist approaches to Just War consider selfdefence as the prototypical just cause for war. This modern consensus differs noticeably from classical Just War thought, which accepted the rationale of self-defence but focused on the just cause of retribution. The classical thinkers considered the use of force as a means to re-establish a state of justice which had been disrupted by prior wrongdoing. That is why Aquinas (ST, II-II, q. 40, a. 1) defined just cause as follows: ‘a just cause is required, namely that those who are attacked, should be attacked because they deserve it on account of some fault’. In other words, the classical Christian idea of war as punishment is built around a ‘judicial analogy’ (Luban 2011, 309) which ‘likens warfare to a judicial proceeding that metes out punishment’ (Luban 2011, 310). Seen from a historical perspective, today’s limitation of just cause to self-defence has been the result of a changed understanding of political authority as well as of prudential considerations. Enshrined in the Westphalian settlement, this new understanding no longer followed the conceptualisation of the sovereign as a divinely instituted avenger of injustice who held responsibility for the common good. Instead, the sovereign was commonly imagined as the representative of the people only, who had been entrusted with the people’s fundamental right of self-defence. In addition, the new understanding of sovereignty limited the ruler’s responsibility for the common good to the people entrusted to her within a defined territory, whereas the earlier understanding had also emphasised the common good of all humankind (Johnson 2014, 1–2). In consequence, the Westphalian principles of political sovereignty and territorial integrity effectively replaced the earlier responsibility for the common good of all humankind which, in principle, allowed for intervening in other sovereigns’ affairs in response to grave injustice. Furthermore, the objective of stopping sovereigns from intervening in each other’s affairs over the issue of religion contributed to the limitation of just cause to self-defence. Centuries later, the dreadful experience of modern warfare, as well as the destructiveness of modern weaponry, added additional prudential concerns which contributed to the establishment of the 1945 UN Charter framework and Just War theory’s contemporary consensus about self-defence as the only just cause for war between states. While force employed for the defence of others, such as humanitarian interventions, has recently received considerable attention, it seems fair to argue that the morality of such uses of force remains contested.

6686_Enemark.indd 42

02/03/21 12:22 PM

Jus ad Vim and Drone Warfare: A Classical Just War Perspective / 43

In the context of targeted killings of culpable unjust individuals, for which drones are often the weapon of choice, there is something to be learned from Aquinas’ thinking about just cause, and I have argued that it can be morally defensible to kill in this way not only in self-defence but, in rare circumstances, for retribution, too (Braun 2019). I base my argument on the classical understanding that culpable wrongdoing causes an imbalance in the equilibrium of justice which needs to be restored. Restoring the equilibrium of justice, as pointed out above, falls within the responsibility of the sovereign who has been wronged and who takes on a function similar to that of a judge. Under normal circumstances, this restoration of justice would mean that an alleged terrorist would be given a fair trial in which they would have the chance to defend themselves and which, if their guilt could be proven, would result in the punishment they deserve. Critics of the death penalty, like myself, would argue that the maximum sentence a terrorist should be subjected to is life imprisonment. Unfortunately, the reality of the so-called war on terror does not allow for such an ideal way of meting out a deserved punishment. Under the circumstances of what has been called a ‘regime of non-state responsibility’ (Heinze 2011, 1080) in which states are either unable or unwilling to prosecute terrorists operating within their territory, I argue that the wronged state may be justified in disregarding the Westphalian principles of political sovereignty and territorial integrity. I argue that there can be cases, resulting from the respective state’s failure to meet its responsibility to establish a state of justice within its territory by holding the culpable unjust individuals accountable, when this responsibility transfers to the wronged state, if the alternative would be to let the terrorist go unpunished. It is in this sense that I suggest that Aquinas can be read as supporting the use of targeted killing by a public authority against culpable unjust individuals of or within another polity as acts of bellum generale. While a targeted killing, if it was carried out within the respective polity’s own territory, would be an act of bellum particular, force used by or directed against private individuals, under the circumstances of a regime of non-state responsibility, becomes an act of bellum generale, a war directed against another polity. Importantly, arguing that a wronged state can have just cause to punish a culpable unjust individual hiding on foreign soil does not yet say anything about what kind of coercive action may be morally justifiable. Likewise, while a wronged state may be justified in acting when a third country fails to meet its obligations, there may be considerations that caution against punitive action. In order to make these determinations, the criterion of right intention takes on a crucial role.

6686_Enemark.indd 43

02/03/21 12:22 PM

44 / Christian Nikolaus Braun

Right Intention Having established that, in principle, the punishment of culpable unjust individuals on foreign soil can be morally defensible, the question remains whether such retributive action should be carried out. For Aquinas, the criterion of right intention was to ensure that wars were not fought for illicit motivations, and sought to guarantee that peace was the goal of any Just War. Importantly, Aquinas did not imagine earthly peace to be a condition where violence would be absent. Rather, he imagined it as a tranquillitas ordinis (a tranquillity of order) in which an equilibrium of justice was established or restored, which, at times, could require the use of armed force (ST, II-II, q. 40, a. 1). Consequently, following Aquinas, it may well be that grave wrongdoing should better go unpunished when not licit retribution but illicit vengeance is the actual rationale, and if such action would be an obstacle to the goal of peace. With regard to potentially problematic motivations, the distinction between retribution and vengeance is instrumental. As Luban (2011, 318–19) explains, the former ‘is undertaken for moral reasons as a practice of justice’ while the latter ‘is undertaken out of rage and hatred’. Of these two rationales, Luban (2011, 319) is correct in pointing out that only retribution ‘is a genuine moral basis for punishment’. In order to avoid giving in to the temptation of vengeance, retributive punishment should thus be measured according to a terrorist’s individual culpability. In order to make that determination, based on the conceptualisation of the sovereign as judge, a trial in absentia should be held in cases where alleged terrorists who are accused of great wrongdoing actively seek to flee from prosecution. If the wrongdoers, then, after having been sentenced and asked to turn them themselves in, continue to hide they should be considered as unwilling to take responsibility for their misdeeds. While seeking to capture such individuals and taking them to court should be the option of first resort, I argue that if there is credible reason to believe that soldiers, as just combatants who make a capture attempt, may be harmed, there is no moral obligation for them to take this risk. In other words, assuming the sovereign authority has a reasonable assumption that these individuals will resist capture and if it is possible to kill them with a drone strike, such lethal action would be justifiable on moral grounds. Crucially, however, flowing from the moral culpability account, the culpable unjust terrorists alone would be targetable. From a retributive Just War perspective, as it is only these individuals whose culpability has been determined, retributive targeted killing would prohibit strikes which may foreseeably kill or injure innocent individuals nearby. In cases where it is impossible to strike at the culpable unjust individual without harming

6686_Enemark.indd 44

02/03/21 12:22 PM

Jus ad Vim and Drone Warfare: A Classical Just War Perspective / 45

others, the use of force cannot be justified morally as a means of just punishment. Even the grave state of injustice which the terrorists caused through their wrongdoing cannot justify the shedding of innocent blood on retributive grounds, and such action would taste of illicit vengeance. Consequently, my renegotiation of the right intention principle is more restrictive than today’s war paradigm and closer to the rules of policing. While, under the war paradigm, the doctrine of double effect allows for foreseen but unintended harm caused by military operations, the police are not permitted to kill innocent bystanders. Moreover, the war paradigm holds that, once a state of war has been established, targetability can be based on group membership, turning most members of the opponent’s forces into legitimate targets. My argument, in contrast, similar to the policing paradigm, permits the use of force against only those individuals who have committed wrongdoing and who resist the re-establishment of a state of justice. Given these very strict limitations for retributive targeted killings, occasions on which such action is morally justifiable will be rare. However, drones, due to their particular technological capabilities, especially their loitering, surveillance and precision-strike capabilities, can function as a tool of justice by striking only at those individuals who have made themselves liable to attack. Drones can thus make a contribution in order to make sure that force is employed for retribution only and vengeance can be avoided. Consider, for example, that the US has employed the R9X, a modified version of the Hellfire missile commonly carried by drones. Also known as the ‘flying Ginsu’, this missile deploys six long blades moments before impact and is supposedly capable of striking so discriminately that it can kill a target in the passenger seat of a driving car while leaving the driver physically unharmed. Reportedly, the Obama administration considered the employment of this missile to kill Osama bin Laden, the prototypical example of a culpable unjust individual hiding in a third country, but instead it decided to send in a special operations team (Lubold and Strobel 2019). In retrospect, this seems to have been a prudent decision, given that bin Laden’s identity could be confirmed and no just combatants were harmed. However, following my argument, had it been possible to kill bin Laden and bin Laden alone by employing the R9X, the US government would have been justified in doing so based on the just cause of retribution (for the 9/11 terrorist attacks) and in accordance with the principle of right intention. Of course, drones, like any weapon, can be used justly and unjustly and the specific technological features just mentioned can be abused. It is important that critics point to these issues, but the point I want to make here is

6686_Enemark.indd 45

02/03/21 12:22 PM

46 / Christian Nikolaus Braun

that this weapon system can make a contribution to the goal of peace imagined as a tranquillitas ordinis. New weapons capable of highly accurate, discriminating and low-yield effect should, in my opinion, be welcomed by Just War thinkers as they put the moral decision-making where it rightly belongs, namely with those ‘in authority’. Yet, interestingly, critics of such weapons concentrate on the argument that they become easier to use than more destructive, less discriminating weaponry. This tendency underscores the seemingly underlying purpose of many critics of armed drones, namely, to seek to end the resort to military force altogether (see Johnson 2006, 190). Importantly, while I have just spelled out circumstances where dronebased lethal force may be morally justifiable against certain culpable unjust individuals, there are prudential considerations that caution against the actual employment of such force. In particular, the likelihood of an outbreak of war between the wronged party and the country on whose territory the military operation would take place seems to conflict with the goal of peace. While abstaining from the operation would allow for a situation of injustice to continue – the culpable unjust individual would not be punished – the killing and destruction likely to result from war between two sovereign authorities cautions against military action. As far as my argument for retributive military action is concerned, whether or not to abstain from a morally justifiable retributive operation would be a judgement call that those ‘in authority’ would have to make. And if it indeed led to an escalation to full-scale war, they would be responsible for that.

Conclusion In conclusion, seen from a classical Just War perspective, instances of socalled ‘force short of war’ such as targeted killings, limited retaliatory airstrikes or the imposition of no-fly zones are acts of war. The reason for this is that they are uses of force carried out by a sovereign authority. However, because such employments of force are small-scale and look very different from a major ground war between states, a reassessment of Just War principles in the light of new circumstances seems warranted. In other words, seen from a classical Just War perspective, the established principles are sufficient to regulate force short of war, and no distinct new moral framework (jus ad vim) is needed. However, the spread of force short of war poses distinct new challenges which require an adaptation of how the Just War principles are commonly interpreted today. Drawing on classical Just War thinking, I have provided an argument as to how the use of force short of war can be regulated in a morally justifiable way. The particular renegotiation I propose sits in between what today are the distinct paradigms of

6686_Enemark.indd 46

02/03/21 12:22 PM

Jus ad Vim and Drone Warfare: A Classical Just War Perspective / 47

policing and war. While a culpable unjust individual would not be liable to lethal force if they turned themselves in and stood trial, they can, by actively seeking to escape from justice, make the use of lethal force against them morally justifiable if an attempt to capture them would expose just combatants to an unacceptable risk of being harmed. In this sense, my argument has parallels with the policing paradigm, which is grounded in an asymmetric moral relationship between the police and the criminal, although, of course, the use of armed drones exceeds the firepower normally associated with the police. At the same time, my renegotiation is more restrictive than the war paradigm, which allows both for the doctrine of double effect and for the making of targeting decisions based on group membership.

Notes I would like to thank the participants in two workshops for very helpful comments on earlier versions of this chapter. The first was held at the University of Southampton in July 2019; the second was held at Radboud University in November 2019. The arguments in this chapter are those of the author alone and they are not necessarily those of the UK Ministry of Defence. 1. I accept that the term ‘ Walzerian’ is considered to be an oversimplification by some scholars. It is not my intention to question the distinctiveness of any scholar. However, I believe that the category can be helpful to grasp the fundamental differences between scholars whose argument is close to Walzer’s, on the one hand, and the revisionists, on the other. In this sense, I would consider scholars such as Brian Orend (see, for example, Orend 2019) and Daniel Brunstetter as Walzerians. 2. It seems that Walzer is mistaken when he refers to the strike as having taken place in 2001. The details he provides match a drone strike that occurred in November 2002 (see Grossman 2018, 39). 3. In this traditional mode of citing the work of Aquinas, ‘ST’ is Summa Theologica, a 1948 translation of which is listed in the references for this chapter.

References Aquinas, Thomas. 1948. Summa Theologica. Trans. Fathers of the English Dominican Province. Allen, TX: Christian Classics. Braun, Christian N. 2018. ‘ The Historical Approach and the “War of Ethics within the Ethics of War”’, Journal of International Political Theory 14 (3): 349–66. Braun, Christian N. 2019. ‘ The Morality of Retributive Targeted Killing’, Journal of Military Ethics 18 (3): 170–88. Brunstetter, Daniel. 2016. ‘Jus ad Vim: A Rejoinder to Helen Frowe’, Ethics & International Affairs 30 (1): 131–6.

6686_Enemark.indd 47

02/03/21 12:22 PM

48 / Christian Nikolaus Braun

Brunstetter, Daniel. 2019. ‘In Defence of Jus ad Vim: Why We Need a Moral Framework for the Use of Limited Force’, in Force Short of War in Modern Conflict: Jus ad Vim, edited by Jai Galliott. Edinburgh: Edinburgh University Press, 284–307. Brunstetter, Daniel, and Megan Braun. 2013. ‘From Jus ad Bellum to Jus ad Vim: Recalibrating Our Understanding of the Moral Use of Force’, Ethics & International Affairs 27 (1): 87–106. Ford, S. B. 2013. ‘Jus ad Vim and the Just Use of Lethal Force-Short-of-War’, in Routledge Handbook of Ethics and War: Just War Theory in the 21st Century, edited by Fritz Allhoff, Nicholas G. Evans and Adam Henschke. New York: Routledge, 63–75. Frowe, Helen. 2014. Defensive Killing. Oxford: Oxford University Press. Frowe, Helen. 2016. ‘On the Redundancy of Jus ad Vim: A Response to Daniel Brunstetter and Megan Braun’, Ethics & International Affairs 30 (1): 117–29. Galliott, Jai C. 2012. ‘Uninhabited Aerial Vehicles and the Asymmetry Objection: A Response to Strawser’, Journal of Military Ethics 11 (1): 58–66. Grossman, Nicholas. 2018. Drones and Terrorism: Asymmetric Warfare and the Threat to Global Security. London: I. B. Tauris. Heinze, Eric A. 2011. ‘ The Evolution of International Law in Light of the “Global War on Terror”’, Review of International Studies 37 (3): 1069–94. Johnson, James T. 2006. ‘ The Just War Idea: The State of the Question’, Social Philosophy and Policy 23 (1): 167–95. Johnson, James T. 2007. ‘ Thinking Morally about War in the Middle Ages and Today’, in Ethics, Nationalism, and Just War: Medieval and Contemporary Perspectives, edited by Henrik Syse and Gregory M. Reichberg. Washington, DC: Catholic University of America Press, 5–10. Johnson, James T. 2013. ‘ The Right to Use Armed Force: Sovereignty, Responsibility, and the Common Good’, in Just War: Authority, Tradition, and Practice, edited by Anthony F. Lang, Cian O’Driscoll and John Williams. Washington, DC: Georgetown University Press, 19–34. Johnson, James T. 2014. Sovereignty: Moral and Historical Perspectives. Washington, DC: Georgetown University Press. Kahn, Paul W. 2002. ‘ The Paradox of Riskless Warfare’, Philosophy & Public Policy Quarterly 22 (3): 2–8. Kreps, Sarah, and John Kaag. 2012. ‘ The Use of Unmanned Aerial Vehicles in Asymmetric Conflict: Legal and Moral Implications’, Polity 44 (2): 260–85. Luban, David. 2011. ‘ War as Punishment’, Philosophy & Public Affairs 39 (4): 299–330. Lubold, Gordon, and Warren P. Strobel. 2019. ‘Secret U.S. Missile Aims to Kill Only Terrorists, Not Nearby Civilians’, Wall Street Journal, 9 May, https://www. wsj.com/articles/secret-u-s-missile-aims-to-kill-only-terrorists-not-nearbycivilians-11557403411 McMahan, Jeff. 2009. Killing in War. Oxford: Oxford University Press. Murphy, James B. 2012. ‘Suárez, Aquinas, and the Just War: Self Defense or Punishment?’, in From Just War to Modern Peace Ethics, edited by Heinz-Gerhard Justenhoven and William A. Barbieri. Berlin: De Gruyter, 175–96.

6686_Enemark.indd 48

02/03/21 12:22 PM

Jus ad Vim and Drone Warfare: A Classical Just War Perspective / 49

Orend, Brian. 2019. War and Political Theory. Cambridge: Polity. Reichberg, Gregory M. 2013. ‘ The Moral Equality of Combatants: A Doctrine in Classical Just War Theory? A Response to Graham Parsons’, Journal of Military Ethics 12 (2): 181–94. Reichberg, Gregory M. 2017. Thomas Aquinas on War and Peace. Cambridge: Cambridge University Press. Singer, P. W. 2019. ‘ The Future of War is Already Here’, New York Times, 18 September, https://www.nytimes.com/2019/09/18/opinion/drone-attack-saudi-arabia. html Strawser, Bradley J. 2010. ‘Moral Predators: The Duty to Employ Uninhabited Aerial Vehicles’, Journal of Military Ethics 9 (4): 342–68. Vaha, Milla E. 2013. ‘ The Ethics of War, Innocence, and Hard Cases: A Call for the Middle Ground’, in Ethical Reasoning in International Affairs: Arguments from the Middle Ground, edited by Cornelia Navari. Basingstoke: Palgrave Macmillan, 182–202. Walzer, Michael. 2001. ‘First, Define the Battlefield’, New York Times, 21 September, http://www.nytimes.com/2001/09/21/opinion/first-define-the-battlefield.html Walzer, Michael. 2006a. Just and Unjust Wars: A Moral Argument with Historical Illustrations. New York: Basic Books. Walzer, Michael. 2006b. ‘ Terrorism and Just War’, Philosophia 34 (1): 3–12. Walzer Michael. 2007. ‘On Fighting Terrorism Justly’, International Relations 21 (4): 480–4. Walzer, Michael. 2013. ‘Is the Use of Drones Ethically Defensible?’, lecture given at the Berkley Center for Religion, Peace and World Affairs, Georgetown University, 13 March, https://www.youtube.com/watch?v=Pc2kOMJQJoQ

6686_Enemark.indd 49

02/03/21 12:22 PM

THREE

The Complicated Reality of Drone Strikes for Law Enforcement Max Brookman-Byrne

Introduction It is not unusual for the various discourses surrounding armed drones to draw upon the language of law enforcement and criminal justice. In excoriating their use outside of armed conflict, some have presented drone strikes as ‘extrajudicial execution[s]’ (Gordon 2016; Amnesty International 2013; 2017), or ‘death sentences’, carried out ‘without indictment or trial’ (Scahill 2015). When framed this way, the moment at which a drone-launched missile impacts and kills a targeted individual is the culmination of a quasi-trial in absentia, in which the person has been prosecuted and convicted, the sentence carried out remotely. The drone operation thus becomes a radically truncated version of the user state’s criminal justice system, doling out lethal punishment for terrorist offences. An alternative narrative that also gestures towards law enforcement is the notion that the proliferation of drone use coincided with the abandonment by the United States of a policy of capturing suspected members of terrorist groups and subjecting them to enhanced interrogation (Mazzetti 2013). While this type of detention arose out of the rules relating to armed conflict rather than domestic criminal law, drone strikes in this depiction nevertheless come across as a replacement of law enforcement and the criminal justice process. These depictions are rejected by US officials, who present extraterritorial drone strikes as an element of the open-ended transnational armed conflict involving al-Qaeda and associated forces (Koh 2010). The invocation of criminal justice in the US government narrative occurs in the form of references to trials of individuals captured during this armed conflict (Horwitz 2013). Under the Obama administration, capture, followed by a trial in front of a domestic court or military commission, was presented

6686_Enemark.indd 50

02/03/21 12:22 PM

The Complicated Reality of Drone Strikes for Law Enforcement / 51

as the preference, with lethal force being permitted only where that was impossible (White House 2013). Further, the suggestion that drone strikes are favoured over capture operations has been overtly refuted by US officials (Brennan 2011). Therefore, drone strikes are presented per se as a feature of a military rather than law enforcement response to terrorism. Nevertheless, language suggesting law enforcement certainly does pop up within the narrative of the US government when discussing its drone programmes. President Trump fired off a celebratory tweet in response to a successful lethal strike against a high-level member of an al-Qaeda franchise, describing it as having delivered ‘justice’ (Trump 2019), and US Central Command has described its drone strikes as ‘ongoing counter-terrorism operations’ (US Central Command 2019b), with counterterrorism being very much the preserve of law enforcement. The discourse thus presents somewhat of an occluded picture as to the extent to which drone strikes are actually carried out as law enforcement. Needless to say, that does not mean they are not used for such purposes; indeed it will be contended below that this is one of the principal ways in which they have been used in certain theatres. This chapter will examine, from the perspective of international law and human rights, the complicated reality of law enforcement drone strikes, highlighting how aspects of the law have been used to legitimate and perpetuate the use of military force (in the form of drone strikes) for law enforcement, enabling colonial-era techniques of policing with airpower. Identifying the law enforcement nature of these strikes, it is then possible to argue that these should be regulated not by international humanitarian law (IHL) but by the more restrictive rules of international human rights law (IHRL). The work will proceed in several steps. First, the traditional international law binary between armed conflict and peacetime will be outlined, in which law enforcement is generally viewed as the preserve of the latter. Second, having made this distinction, the chapter will then set out the rules governing peacetime law enforcement drone strikes, and the extremely narrow set of circumstances in which drone strikes may be carried out outside of armed conflict. Third, drone strikes during armed conflict will be considered, and the chapter will challenge the presumption that such drone strikes are necessarily part of the ongoing armed conflict, and so governed by permissive rules of IHL. This key section will illustrate how the categorisation of a situation as an armed conflict can be harnessed to provide ostensible legitimation for the use of military force for law enforcement operations, such as counterterrorism. It will discuss the nature of US drone strikes carried out in Yemen and Somalia, arguing that these should be recognised as acts of law enforcement and practices inspired by colonial-era ‘air control’ that have piggy-backed on pre-existing internal armed conflicts within

6686_Enemark.indd 51

02/03/21 12:22 PM

52 / Max Brookman-Byrne

those states. Finally, the chapter will set out how recognising that drone strikes have been used in this manner transforms the dominant framework of their legal regulation. Such strikes, it is argued, should ultimately be considered through the significantly more restrictive lens of IHRL, rather than IHL, despite occurring within the context of armed conflict. In highlighting the extent to which drone strikes during armed conflicts should be assessed through the lens of IHRL, this work presents an important addition to the literature on drones, providing a foundation upon which further analysis of drone strikes can be undertaken. The work seeks to guide future assessments of this form of combat by rebutting the presumption that military force used to carry out lethal strikes during armed conflict should necessarily be considered to be governed by IHL.

Recourse to Doctrine One way in which to cut through the complexities produced by public discourse on law enforcement drone strikes is to appeal to doctrinal international law, though this can in turn produce its own difficulties. When it comes to conceptualising drone strikes as ‘law enforcement’, the apparent simplicity offered by the recourse to doctrine arises from the fact that international law provides a binary: situations of armed conflict and situations of ‘not armed conflict’. This binary is typically used to assert when drone strikes can be viewed as part of a given armed conflict – in which case they are not law enforcement – and when they fall outside of any armed conflict – in which case they may typically be described as law enforcement operations. From the perspective of international law, the distinction between these two paradigms is vital as it dictates the body of rules applicable to a given activity. During an armed conflict, while human rights protections remain operative, IHL comes into effect (Geneva Conventions I–IV 1949, Articles 2 and 3), changing the nature of rights such as the right to life (Hathaway et al. 2012, 1910–11; Hill-Cawthorne 2015, 313–16), and bringing with it permissions that are otherwise unavailable, such as an increased ability to carry out lethal operations and an acceptance of a ‘proportionate’ number of unintended civilian casualties. As will be shown in the next section, during law enforcement operations drone strikes are significantly more constrained (potentially to the point of being practically impossible to carry out lawfully (Sadat 2012, 225; Nasser 2015, 296–7)). Armed conflicts between states (‘international armed conflicts’ in the language of the Geneva Conventions) are relatively straightforward to identify, it being generally held that any use of force between states will create one (International Committee of the Red Cross 2016, paras. 236–8; Dinstein

6686_Enemark.indd 52

02/03/21 12:22 PM

The Complicated Reality of Drone Strikes for Law Enforcement / 53

2010, 1). Conversely, armed conflicts ‘not of an international character’ (Article 3 Common to the Geneva Conventions) – that is, those involving a state and a non-state actor – arise where there is ‘protracted armed violence between governmental authorities and organised armed groups, or between such groups within a State’ – the criteria of intensity and organisation set down by the International Criminal Tribunal for the Former Yugoslavia (ICTY) in Prosecutor v. Tadić (1995, para. 7). This threshold distinguishes non-international armed conflicts (NIACs) from internal disturbances, ‘banditry, unorganized and short-lived insurrections, or terrorist activities’ (Prosecutor v. Milošević 2004, para .26), which, without satisfying the Tadić criteria, remain ostensibly within a ‘peacetime’ paradigm, and will not result in the application of IHL. To be viewed as part of a NIAC, drone strikes must occur within a situation that satisfies the Tadić criteria. A drone strike will not, in and of itself, produce an armed conflict, despite the severity of the violence involved (Lubell and Derejko 2013, 78). Armed drone operations may therefore be carried out as part of an armed conflict, being used to target members of organised armed groups engaged in sufficiently intense fighting, or outside of armed conflict, in response to internal disturbances, banditry and terrorist activities below the Tadić criteria.1 In the latter context it would be accurate to describe drone use as an aspect of law enforcement, in an international law sense. It is necessary to guard against the presumption that drone strikes are inherently and always part of armed conflict and never purely law enforcement. As I have argued previously, the drone programme of the US in Pakistan, Yemen and Somalia has involved strikes occurring both within and outside of armed conflict (Brookman-Byrne 2017). This chapter will consider two lenses through which ‘law enforcement drone strikes’ can be examined. First, and most obviously, drone strikes occurring outside of armed conflict will be examined, depicting the narrow extent to which they can be used in the paradigmatic sense in which ‘law enforcement’ is discussed within international law doctrine (see, for instance, Gaggioli 2013). Second, the chapter will consider from a more critical angle the extent to which drone strikes can be said to have been carried out for purposes of law enforcement behind the veil of extant armed conflicts in which they appear to be military operations but arguably more accurately resemble extraterritorial law enforcement.

Drone Strikes for Law Enforcement, Outside of Armed Conflict The doctrinal question of the extent to which drone strikes can be used for law enforcement outside of armed conflict has both abstract and concrete relevance: abstract in the sense that drones have the potential to be used

6686_Enemark.indd 53

02/03/21 12:22 PM

54 / Max Brookman-Byrne

outside of armed conflict, and concrete in that this has actually occurred (Brookman-Byrne 2017, 38). A concrete example of this kind of drone strike is that which targeted Ali Qaed Senyan al-Harthi and five others in Yemen in 2002 (Associated Press 2002). Despite claims to the contrary (UN Economic and Social Council 2003, paras. 2–3; Printer 2003, 370–9), it is submitted that the strike certainly occurred outside of ongoing armed conflict (BrookmanByrne 2017, 29–31), a conclusion that is commonly reached by those who have evaluated the strike (Jahangir 2003, para. 39; Lubell 2010, 174–8; Ruys 2005, 22). While US drone strikes have been involved in a NIAC in Yemen, having been invited by the government of Yemen to join fighting between it and the group al-Qaeda in the Arabian Peninsula (AQAP), this conflict cannot realistically be said to have begun (applying the Tadić criteria) earlier than 2011 (Brookman-Byrne 2017, 22–5), long after the strike that killed al-Harthi. Therefore, being situated outside of an armed conflict, the strike can be characterised as a law enforcement operation in a doctrinal sense, with IHL being inapplicable. Instead, this strike, along with other strikes falling within this paradigm of law enforcement, is governed by the much stricter framework of IHRL, provided by treaties and customary international law.2 An initial difficulty faced by any attempt to interrogate extraterritorial, law enforcement drone strikes through the lens of IHRL and the right to life is that of jurisdiction. The US has consistently held that the right to life under Article 6 of the International Covenant on Civil and Political Rights (ICCPR) is restricted in its application to a member state’s physical territory by virtue of the jurisdiction clause in Article 2(1) of the same covenant (‘Each State Party to the present Covenant undertakes to respect and to ensure to all individuals within its territory and subject to its jurisdiction’) (US Department of Defense 2015, 24; UN Human Rights Committee 2005, annex 1; US Department of State 2007, para. 4). The United Kingdom does not take this same strict approach, but nonetheless has held that, without some degree of control over a person killed or the area in which they are situated, the right to life under Article 2 of the European Convention on Human Rights (ECHR) is not implicated in an airstrike (Joint Committee on Human Rights 2016, para. 22).3 While some support these positions (for instance, Schindler 1982, 939; Dennis 2006, 462–3; Paust 2010, 264–5), there is a raft of academic and judicial opinion in favour of applying negative rights (such as the right not to be arbitrarily deprived of one’s life) extraterritorially without restriction (Meron 1995, 79–83; Ben-Naftali and Shany 2003, 60–3; Milanović 2011, 118–228; Da Costa 2013, 91; Van Schaack 2014, 32–3; Lubell 2010, 205; Melzer 2008, 135; US Department of State 2010, 4; UN Human Rights

6686_Enemark.indd 54

02/03/21 12:22 PM

The Complicated Reality of Drone Strikes for Law Enforcement / 55

Committee 2018, para. 63; Legal Consequences of the Construction of a Wall 2004, para. 136; López Burgos v. Uruguay 1981, para. 12.2; Celiberti de Casariego v. Uruguay 1981, para. 10.2). For instance, the European Court of Human Rights has held that human rights jurisdiction will extend to individuals outside a state’s territory, inter alia, where that state exercises effective control of an area or where the acts of its authorities produce effects outside of its territory (Al-Skeini v. United Kingdom 2011, paras. 130–42). The latter can be produced where an intervening state carries out ‘public powers’ with the consent of the territorial state (Al-Skeini v. United Kingdom 2011, para. 135), or where it exercises ‘physical power or control’ over an individual, though it is controversial whether this occurs during acts of lethal force such as airstrikes (Pad v. Turkey 2007; Andreou v. Turkey 2008). Conversely, the Inter-American Commission has more explicitly established a jurisdictional link through the act of a missile strike conducted at a distance, which was held to have ‘placed [those targeted] under [the state agents’] authority’ (Alejandre v. Cuba 1999, para. 25). Ultimately, the reality of the situation in terms of human rights jurisdiction within treaty law remains unclear but, regardless, the right to life is generally accepted to exist as a rule of customary international law (Dinstein 1981, 115; Ramcharan 1985, 3; Rodley 1989, 333; Melzer 2008, 189; Joint Committee on Human Rights 2016, para. 67), and as a result is binding on states generally. In my view, such a customary right to life does not contain the same jurisdictional issues as that under international conventions. As such there is a strong case to be made that where drone strikes are carried out for law enforcement, outside of an armed conflict, they are subject to strict regulation according to laws protecting the right to life. This right is breached where deprivation of life is ‘arbitrary’ or ‘intentional’, as provided by the ICCPR (1966, Article 6(1)) and the ECHR (1950, Article 2(1)) respectively. These two terms have been interpreted as effectively synonymous (Nowak 1993, 111; Melzer 2008, 118–20), with each being referred to in arbitral decisions focusing on the other (Suarez de Guerrero v. Colombia 1982, para. 13.2), and as such it is possible, when discussing the right to life generally, to refer to the basic position that the right is breached where a deprivation of law is arbitrary. Within an armed conflict, lethal force (as part of hostilities within that conflict) is not arbitrary where it is carried out in accordance with the relevant corpus of IHL (Legality of the Threat or Use of Nuclear Weapons 1996, para. 25). Outside of an armed conflict, lethal force is significantly more constrained and will be deemed to be arbitrary on the basis of elements of inappropriateness, injustice, lack of predictability, lack of due process and, importantly, necessity and proportionality (UN Human Rights Committee 2018, para. 12).

6686_Enemark.indd 55

02/03/21 12:22 PM

56 / Max Brookman-Byrne

Specifically in terms of law enforcement, the UN Code of Conduct for Law Enforcement Officials and the UN Basic Principles on the Use of Force and Firearms by Law Enforcement Officials both feature the constitutive right to life elements of necessity, proportionality, the requirement of a basis within law, and precaution (Code of Conduct for Law Enforcement Officials 1979, Article 3; Basic Principles on the Use of Force and Firearms by Law Enforcement Officials 1990, Articles 1 and 5). These requirements balance the use of lethal force against the threat posed by the person targeted; lethal force must be actually required to counteract that threat, and it must be the final option, with all alternatives being insufficient (UN Human Rights Committee 2018, para. 12). Importantly, lethal force can only be used in self-defence or in order to protect the lives of others, in the face of an imminent threat (UN Human Rights Committee 2018, para. 25). Surrounding these requirements that relate to the specific moment in which lethal force is used, there are broader imperatives that govern the context of a use of lethal force for law enforcement. There is a requirement that efforts must be made to limit the resort to the use of firearms (Code of Conduct for Law Enforcement Officials 1979, Article 3(c)), and in cases of operations culminating in lethal force, states have been criticised for failing to give warnings or the opportunity for surrender (Suarez de Guerrero v. Colombia 1982, para. 13.2; Alejandre v. Cuba 1999, para. 42). Further, in considering breaches of the right to life, the entire context of an operation is assessed, including planning and control, to establish whether anything could have been done further ‘upstream’ to limit the chances of a subsequent use of lethal force (Heyns 2014, para. 63; McCann v. United Kingdom 1995, para. 194). Considered alone, the instant use of lethal force may not have breached the right to life of the person killed, but when contextualised within the broader operation it may be that a failure to limit the likelihood of the resort to lethal force results in that right being breached (McCann v. United Kingdom 1995, paras. 213–14; Haász and Szabó v. Hungary 2015, paras. 62–6). In addition to these requirements governing the lead up to and the act of using lethal force, the right to life also presents states with additional subsequent procedural obligations. It requires investigations into potential violations (UN Human Rights Committee 2004, para. 15) which must be effective, meaning that they must be prompt, independent, instigated by relevant authorities, ‘capable of leading to a determination of whether the force used in such cases was or was not justified in the circumstances’, and subject to an element of public scrutiny (Isayeva v. Russia 2005, paras. 209–14). All of these requirements bear very heavily upon the potential use of drone strikes for law enforcement. It is clear that killing is not completely

6686_Enemark.indd 56

02/03/21 12:22 PM

The Complicated Reality of Drone Strikes for Law Enforcement / 57

prohibited by the rules protecting the right to life, so it would be wrong to suggest that drone strikes are inherently incapable of being used for law enforcement without violating that right. However, many commentators have argued that they are pretty well impossible to reconcile with IHRL protections (Sadat 2012, 225; O’Connell 2011, 586; Heyns 2014, para. 136), and this is accurate. Considering the right to life, in order to be lawful a drone strike must be carried out as a last resort with all alternatives having been exhausted, and it must be taken only in response to an imminent threat to life. This is a very specific and narrow set of circumstances, understandably more akin to a police officer using a pistol against a threatening individual than a drone firing a missile in a remote region. The fact that drone strikes target individuals in ‘hard-to-reach’ areas may also weigh against their ability to adhere to the right to life: they bear the hallmarks of operations planned not to limit the possibility of the use of lethal force, but in which applying lethal force is the primary goal, or at least a near certainty (Proulx 2004, 887). Further to all of this, unlike the situation during an armed conflict, any instances of collateral deaths as a result of a drone strike carried out for law enforcement will render the strike contrary to the right to life of those killed other than the intended target (O’Connell 2016, 206). Thus, the only kind of drone strike that would satisfy IHRL protections under the right to life would be one in which it could be demonstrated that the person targeted was in the process of carrying out a threat to life. This dramatically limits (though does not extinguish) the extent to which lethal drone strikes can be used for ‘pure’ law enforcement. The example mentioned above, in which al-Harthi and his associates were killed, presents an instance of a drone strike in which the right to life was almost certainly violated. Though al-Harthi was implicated in the 2000 attack against the USS Cole (BBC 2002a), there is no evidence that either he or the others targeted were involved in an imminent threat, which commentators have asserted renders the strike unlawful (Lubell 2010, 177; Kretzmer 2005, 205). Statements of US officials in relation to the strike support the idea that al-Harthi was targeted due to his identity, not because he posed an immediate threat (Cheney 2002; BBC 2002b), making this a clear case of unlawful targeted killing. As with many US drone strikes, assessment of this operation is stymied by the secrecy that surrounds it; however, based on available information, a picture emerges that portrays the strike as very likely to have violated the right to life of both al-Harthi and his companions. Arbitrary deprivation of life is a key way in which drone strikes carried out outside of armed conflicts for law enforcement may be unlawful. In allowing the targeting of specific individuals who have committed previous violent acts, the capabilities of drones to find and target individuals in

6686_Enemark.indd 57

02/03/21 12:22 PM

58 / Max Brookman-Byrne

otherwise unreachable areas may well engender (and arguably already has) a tendency to target those people with lethal force when the opportunity arises, rather than to pursue less ‘simple’, traditional law enforcement methods. Drones provide opportunities to remove the risk presented by certain individuals, and to satisfy a desire for vengeance for past acts, and these might seem too good for some states to pass up. However, the fact that an individual is a prolific and dangerous criminal, and is in a hard-to-reach area, does not mean that it is permissible to kill them. Ultimately drones are a tool unsuited to what is typically understood as law enforcement, in a doctrinal sense, outside of situations of armed conflict.

Law Enforcement Drone Strikes Under the Veil of Armed Conflict The section above has sketched the very restricted basis upon which law enforcement drone strikes can be ‘properly’ undertaken (in the sense that they occur within the paradigm of law enforcement, outside the context of an armed conflict) and are therefore governed solely by the protections of relevant human rights. This fits with the binary international law paradigms of armed conflict and peace that apply to forcible actions carried out by states. But, when conceptualising how drones have been used, it is perhaps better to attempt to consider them outside this binary in which peacetime drone strikes are law enforcement and those within an armed conflict are not. Although the binary presents doctrinal simplicity, it also potentially serves to hide the reality of how drone strikes are used, particularly where, from a moral perspective at the very least, they occur in contexts that ‘aren’t exactly armed conflict, and . . . aren’t exactly law enforcement’ (Walzer 2016, 14). As will be demonstrated in this section, the categorisation of a situation as armed conflict can serve to provide legal legitimation for the use of military force in what are essentially law enforcement operations, giving states cover for far greater recourse to lethal force. In the case of the US this has arguably resulted in law enforcement aims being pursued with lethal drone strikes, as well as the adoption of tactics of population control through airpower that draw on colonial-era practices. It is correct that many drone strikes have occurred during armed conflict, but it is important to question the presumption that, because a strike occurs during an armed conflict, it is necessarily of that armed conflict. Asserting that a drone programme is part of an armed conflict and is thereby principally governed by IHL provides huge latitude for the use of lethal force. It is one thing to say that drone strikes during an intense period of fighting, such as the campaign against ISIS in Iraq and Syria, are part of an armed conflict; it is

6686_Enemark.indd 58

02/03/21 12:22 PM

The Complicated Reality of Drone Strikes for Law Enforcement / 59

quite another to uncritically apply the same designation to drone strikes carried out as part of a long-term, low-level and sporadic programme of counterterrorism, such as the US is pursuing in Yemen and Somalia. To more fully and critically engage with the reality of extraterritorial drone programmes, in particular those with a counterterrorism mandate, the presumptive legal veil provided by claims of ‘armed conflict’ status must be lifted. It is necessary to confront the fact that law is used to legitimate and privilege violence (Kennedy 2006, 8). In the case of drones, this may mean operationalising the permissive character of IHL by pursuing law enforcement objectives within, and under the guise of, pre-existing NIACs between governments and non-state actors based within their territory. As David Kennedy has argued, law should be recognised as a ‘an instrument which can and is used strategically’ (Kennedy 2012, 165). This manifests in myriad ways, involving less obvious uses of laws relating to contract, finance, information and so on, but of most immediate relevance is the way in which law can be used as a weapon through ‘categorisation’ (Kennedy 2012, 166). This is starkly so in the case of drone strikes: clearly the categorisation of a situation as an armed conflict rather than law enforcement serves to legitimise the level of force involved which would otherwise be unlawful: ‘When this happens successfully, international law confirms the violent expression of sovereign power as right’ (Kennedy 2012, 160). As such, military force for law enforcement becomes an expression of a state’s legitimate – through the imprimatur of lawfulness – actions during a conflict, rather than a blatant violation of human rights, which it would probably be were it not for the legal categorisation as armed conflict. Certainly this is an activity that the US has been involved in, having repeatedly declared that its operations against al-Qaeda constitute a generalised ‘war’ (White House 2002, 5) or armed conflict, seeking to harness the exceptional powers available within activities that can be categorised as such. Harold Koh, speaking as Legal Adviser to the Department of State, explicitly couched this claim in relation to the legal framework, stating that ‘as a matter of international law, the United States is in an armed conflict with al-Qaeda, as well as the Taliban and associated forces, in response to the horrific 9/11 attacks’ (Koh 2010). This assertion, which was stated to include the use of lethal drone strikes, clearly demonstrates the use of legal categorisation to justify recourse to a particular type of (military) force for a particular goal: counterterrorism. That these operations are focused on counterterrorism is important. Counterterrorism is not an activity that can, taken in isolation, produce situations of armed conflict under international law. Indeed, the ICTY has emphasised that ‘[t]he main purpose of the Tadić test is to distinguish an

6686_Enemark.indd 59

02/03/21 12:22 PM

60 / Max Brookman-Byrne

armed conflict from banditry, unorganized and short-lived insurrections, or terrorist activities, all of which are not subject to international law’ (Prosecutor v. Milošević 2004, para. 26, my emphasis). This is categorical: terrorist activities do not produce armed conflicts, therefore the response to terrorism is a law enforcement not a military activity. That is not to say that the violent acts of a non-state actor cannot reach the intensity required for an armed conflict to be said to exist, but arguably at this point the military response cannot reasonably be called counterterrorism. The result of this is that the response of states to terrorism should remain the preserve of law enforcement regulated by strict human rights protections, with the recourse to military force being inappropriate and very probably unlawful. This definitive separation of counterterrorism from military operations is significantly blurred in the historical practice of the US. Though counterterrorism was once considered a purely law enforcement activity, a conceptual tussle in the 1980s between pursuing terrorist groups ‘actively’ (through military means) or ‘passively’ (by relying on law enforcement) resulted in the political and legal understanding that military force could be used to address terrorist threats to US citizens and interests (Gunneflo 2016, 115–16). In 1984 the Reagan administration reconceptualised terrorism as ‘a kind of war’, which accordingly ‘firmly placed the armed forces in the role of combatting [it]’ (Jenkins 1984, 4). This approach began to be operationalised in the late 1990s, with the use of military weapons to pursue al-Qaeda, but it truly crystallised into practice after 9/11 (Gunneflo 2016, 166–92), at which point the US began to detain members of nonstate actors deemed to be part of the conflict (Schoettle 2009, 70) as well as to target them in airstrikes, such as that against Senyan al-Harthi in 2002. Indeed, US Central Command has been explicit in describing its ongoing targeting of AQAP in Yemen as being a campaign of counterterrorism (US Central Command 2019b). Considering this brief and simplified history, we see a picture of counterterrorism transformed, at least in the approach of the US, into a military activity. However, this is not (indeed it cannot be) a replacement of law enforcement with military force; it is rather the continuation of law enforcement by military means. Counterterrorism remains essentially a practice of law enforcement but becomes one that is carried out with military modes of force, in particular through the use of lethal drone strikes. Thus, to avoid appearing prima facie inherently unlawful, this militarised counterterrorism must be presented as falling within a larger armed conflict and thereby governed by the rules of IHL. The majority of US drone strikes have been carried out within NIACs between domestic governments and non-state actors during which US intervention has been invited (Brookman-Byrne 2017).

6686_Enemark.indd 60

02/03/21 12:22 PM

The Complicated Reality of Drone Strikes for Law Enforcement / 61

This makes it relatively straightforward for these strikes to be necessarily characterised as being part of the armed conflict and so governed solely by IHL; it is no logical leap to assume that military force invited during an armed conflict is part of that armed conflict. However, the nature of US-run drone programmes in states such as Pakistan, Yemen and Somalia belies this automatic characterisation and demands a more nuanced approach. First, these drone programmes are predominantly characterised by lowlevel, long-term engagement against emerging targets (signature strikes) and known members of armed groups such as al-Qaeda (personality strikes). Based on data from the Bureau of Investigative Journalism,4 since 2002 there have been 336 reported drone strikes in Yemen, often occurring at a rate of one per month (notable exceptions being May and June 2017, in which 82 strikes were reported) (Bureau of Investigative Journalism n.d.a), while in Somalia they have been carried out at a similarly low level, with 202 since 2007 (Bureau of Investigative Journalism n.d.b). Second, drone strikes are carried out across a wide geographical area, rather than being confined to regions in which fighting between the armed group and government forces is occurring. For instance, in May 2019 a drone strike in Yemen reportedly killed four suspected members of AQAP heading to a farm in Marib province (Agence France-Presse 2019), though this is a separate region from Hadramaut, which was the site of the most temporally proximate AQAP attack (Reuters 2019). Taken together, these factors provide a picture of drone strikes being carried out not as part of intense and ongoing hostilities as might be expected during an armed conflict, but as a more subtle engagement detached from the main conflict. While these characteristics are not incompatible with actions carried out as part of an armed conflict, they militate against the notion that counterterrorism drone strikes should uncritically be considered to be part of the contemporaneous armed conflict to which they are ascribed. Additionally, there are other factors indicating that, while these strikes fall within a particular armed conflict, it may be wrong to consider them to be inherently part of it. Focusing once more on Yemen, a US drone strike on 1 January 2019 targeted and killed Jamal al-Badawi (US Central Command 2019a). This man was an AQAP member who had been indicted by a federal grand jury in 2003 for terrorist offences, including involvement in the 2000 attack on the USS Cole in Aden harbour (US Central Command 2019a). Further, al-Badawi had been convicted by a court in Yemen of helping to organise the Cole attack and had been jailed in Sana’a from 2004 until 2006, when he escaped (Hauser 2006). Additionally, al-Badawi had been the subject of an Interpol Red Notice at the request of the US (Hauser 2006); Red Notices are designed to allow global law enforcement

6686_Enemark.indd 61

02/03/21 12:22 PM

62 / Max Brookman-Byrne

co-operation with a view to the arrest and extradition of a wanted suspect (Interpol n.d.). These facts all point towards al-Badawi being a law enforcement concern who was being pursued not because of engagement in an armed conflict but because of the (admittedly severe) crimes that he had allegedly perpetrated. Ultimately this strongly suggests that the drone strike that killed him was an act of law enforcement. The technical armed conflict during which the strike occurred appears to have provided the legal cover for military force to be used, but it would seem fallacious to argue that the strike was therefore necessarily part of that armed conflict. This strike demonstrates the potential for personality strikes against members of groups like al-Qaeda, wanted by the US, to be examples of military force used to carry out law enforcement actions; they are operationalised for policing in the narrow, liberal sense of fighting crime (Neocleous 2011, 146). However, law enforcement drone strikes within armed conflict go beyond this crime-focused activity and form part of a broader process of ‘policing’, concerned not simply with bringing justice to criminals, but with ‘administration, security, and order’ and the ‘maintenance of the body politic’ (Neocleous 2013, 580). Thus, as well as pursuing law enforcement, the low-level, indefinite nature of campaigns of drone strikes enables the projection of military power by the US to further its own security interests rather than specifically to conduct armed conflicts and support state governments. As demonstrated by critics such as Priya Satia, the use in this manner of precision bombing technology such as drones continues practices of colonial domination through ‘air control’, developed by Britain in the inter-war period (Satia 2014, 14–15). This was an aspect of the broader phenomenon of ‘imperial policing’, concerned with the protection of colonial authority in the face of growing nationalism (Moreman 1996). Air control involved the use of military force for maintaining security outside situations of war; for instance, in Peshawar in 1930, the Royal Air Force conducted five months of aerial bombing against Afridi tribespeople in response to the Redshirt Rebellion, despite there being no extant war (Gwynn 1939, 254–5). Similarly, these methods were used in Iraq in the 1920s against groups resisting British domination non-violently, by refusing to submit to taxation, or harbouring the leaders of rebel groups (Satia 2006, 24). This utilisation of military force was not about winning wars, but about subjugating populations. Air control allowed the ‘possibility of “radiating” British power throughout the country from a handful of fittingly Spartan bases’ (Satia 2006, 28), in much the same way that US drones are able to exert control, including through law enforcement strikes, across wide distances from a handful of bases in Djibouti, Ethiopia, Saudi Arabia and elsewhere. Indeed, it is argued

6686_Enemark.indd 62

02/03/21 12:22 PM

The Complicated Reality of Drone Strikes for Law Enforcement / 63

that the presence of drones and their use in this manner has provided a ‘permanent police presence’ (Neocleous 2013, 590). A further parallel that should not be lost beneath an emphasis on tactics is that highlighted by Alex Edney-Brown, whose research has demonstrated that today’s longterm policing drone programmes replicate the psychological impacts of air control on citizens in colonised states (Edney-Brown 2019, 12–13). These comparisons of colonial air control and US drone use are not mere inference on the basis of apparently similar tactics, but in fact reflect an overt recognition of air control’s potential, having been specifically identified by the US as providing lessons that remain relevant today (Neocleous 2013, 580). Policing operations undertaken with drones are clearly something other than violence carried out within an armed conflict, being ‘far from the traditional understanding of warfare as a contest between enemies’ (Kendall 2016, 94). Instead they bear the hallmarks of colonial-era programmes of population control through the deployment of military airpower. But this utilisation of military force is only legally possible where it occurs within armed conflict. US drone programmes, like those in Yemen and Somalia, thereby appear to piggy-back on extant NIACs as a way of using military force to further US global counterterrorism and security aims. The modern international law binary cleaving peace from armed conflict, and strict human rights protection from IHL, necessitates the existence of an armed conflict for law enforcement drone strikes to be presented as legitimate (or perhaps, to put it more accurately, for them not to be prima facie unlawful). This is precisely what has happened with drone strikes in Yemen, Somalia and, to a lesser extent, Pakistan. In these areas the majority of drone strikes have been carried out within the remit of NIACs that the US has been invited into by host governments, but the nature of the campaigns suggests that not all of those drone strikes are part of those armed conflicts. Instead it seems that the categorisation of these situations as armed conflict is being used to provide a legal veil to enable militarised law enforcement and policing. The above does not mean that we can suddenly declare that all US drone strikes are law enforcement operations and therefore inherently governed by unadulterated human rights protections with no input from IHL. However, by identifying the law enforcement and policing character of drone strikes carried out within these conflicts, we gain critical purchase that would otherwise be hidden behind the veil of legal categorisation. On the one hand, it is possible to present a public and non-legal critique of sustained military law enforcement on the fringes of armed conflict as being inherently problematic, involving as it does the subjection of populations to years of danger, stress and death. On the other hand, from a legal perspective it can

6686_Enemark.indd 63

02/03/21 12:22 PM

64 / Max Brookman-Byrne

be argued that the law enforcement nature of these drone strikes reduces the certainty with which it can be claimed that the permissive lens of IHL is the appropriate means to assess their lawfulness. IHL applies to operations undertaken during armed conflict, but it does not do so in isolation: the protections provided by IHRL continue to apply alongside IHL (UN Human Rights Committee 2018, para. 63). While it is often suggested that the doctrine of lex specialis dictates that the specific rules of IHL will prevail over more general human rights protections (for instance, Watkin 2004, 2; Sassòli 2011, 71) the reality is more complex. Rather than IHL necessarily replacing IHRL, a contextual approach is adopted, which takes account of a particular situation and the content of relevant rules in order to determine the nature and scope of the applicable rules, and the interaction between human rights and IHL (Hathaway et al. 2012, 1910–11; Murray 2016, 89). The result of this is that where an operation does not possess the characteristics of an action carried out pursuant to an armed conflict, human rights protections may continue to provide the dominant mode of regulation, despite the fact that the operation falls geographically and temporally within an overarching ongoing armed conflict. One particularly useful way of conceptualising this is provided by Daragh Murray’s Practitioners’ Guide to Human Rights Law in Armed Conflict, which makes the heuristic distinction (a ‘rule of thumb’ rather than a rule of law) between situations of ‘active hostilities’ and those of ‘security operations’ (Murray 2016, 80). In the former, IHL provides the primary body of applicable rules and IHRL the secondary, while in the latter the hierarchy of rules is reversed (Murray 2016, 80). This is crucially important when considering the use of lethal force: while within ‘active hostilities’ it is governed by the permissive rules of IHL, during ‘security operations’ the powerful protections under IHRL remain in place, limiting the use of lethal force to situations when it is absolutely necessary as a last resort to protect life. This means that certain practices are absolutely prohibited, such as status-based targeted killings (personality strikes) (Murray 2016, 91–2). Activities such as targeting during international armed conflict are argued to constitute ‘active hostilities’ due to the nature of the detailed treaty rules; however, during non-international armed conflict the rules are far less specific and so the relevant framework is less clear, depending instead on the facts of a given situation (Murray 2016, 90–1). According to the Practitioners’ Guide, during NIAC a situation can be classified as ‘active hostilities’ where there is ‘sustained and concerted’ fighting, or where a state lacks effective control of territory (Murray 2016, 96). As discussed above, many law enforcement-type drone strikes carried out by the US have been geographically and temporally

6686_Enemark.indd 64

02/03/21 12:22 PM

The Complicated Reality of Drone Strikes for Law Enforcement / 65

remote from actual fighting despite being assigned to the category of armed conflict. That therefore removes one of the possible bases upon which to ground a claim that they should be governed by IHL. More problematic is the notion that an operation will constitute ‘active hostilities’ if it occurs in a situation in which a state lacks effective territorial control. There is a danger that the confluence of this criterion and the capabilities offered by remote technologies such as drones may produce the conclusion that the use of drones for law enforcement airstrikes in territories where there is weak government control (such as Yemen and Somalia) would per se be governed by IHL due to the lack of government control. This would provide drone-using states with significant latitude to bring all remote strikes carried out in such situations within an ‘active hostilities’ framework, enabling the lawful use of military force for law enforcement under the veil of extant armed conflicts. This would severely erode the human rights protections of all those living within the putatively ungoverned space, exposing civilians to the risk of becoming lawful collateral damage under IHL. The Practitioners’ Guide offers some limitation upon this outcome, holding that areas outside the effective territorial control of a government are those ‘areas of territory under the exclusive control of an armed group’ and those which are ‘actively contested’, that is, subject to ongoing fighting (Murray 2016, 97, emphasis added). This seems to set the threshold relatively high for an area to be deemed ‘active hostilities’, requiring not simply the absence of control by the government but also the active – and exclusive – control of a non-state actor. Nevertheless, this raises the question of whether a situation constitutes ‘active hostilities’ where a non-state actor is present in an area over which a government lacks effective control. It is submitted that this should be read restrictively such that drone strikes carried out for law enforcement in this type of context ought to be assessed by reference to the more restrictive protections under human rights law. Failure to do so would once again grant an unacceptable latitude for the use of military force for law enforcement purposes and should be resisted. Applied to the use of drone strikes during armed conflict, it is clear that a convincing argument can be made that, where they are carried out in a manner that indicates a law enforcement or policing aim, the restrictive rules of IHRL should be the primary lens through which lawfulness is analysed. As a result, drone strikes of that nature will be unlawful in the event that they fall outside of the very narrow set of circumstances in which a lethal drone strike will not violate the right to life. This underscores the importance of rejecting the assumption that lethal force that happens during an armed conflict is necessarily part of that armed conflict, and instead considering the

6686_Enemark.indd 65

02/03/21 12:22 PM

66 / Max Brookman-Byrne

context that surrounds a particular drone strike, or indeed an entire drone campaign. Doing so allows a promising avenue for legal resistance against the unfettered use of drone strikes under the cover of armed conflict.

Conclusion Drone strikes are indisputably an exercise of military force, and yet it is equally indisputable that they have been used for law enforcement operations. This is either in the form of strikes carried out outside of any extant armed conflict, or, less obviously, where they have been used for law enforcement or policing purposes during armed conflicts. A minority of drone strikes fall into the former category and are therefore subject to regulation by IHRL. The nature of the right to life is such that it is very likely that strikes like these will be unlawful, violating that right where, for instance, they are not carried out as a last resort in response to a threat. That this type of law enforcement drone strike will be unlawful in most cases is not a controversial conclusion, and it is indeed a logical one given that this practice represents the naked use of military force for ostensibly nonmilitary purposes. More commonly, drone strikes are used within ongoing armed conflicts, often at the invitation of domestic governments engaged in internal conflicts with non-state actors. While it is possible that these strikes may form part of the armed conflict into which they are invited, US practice suggests rather that many are law enforcement operations carried out under the veil of the armed conflict. These drone strikes represent the radiation of US power extraterritorially, using airpower for counterterrorism, as a method of bringing ‘justice’ to wanted members of groups such as AQAP, while also engaging in practices of ‘policing’ from the air, drawing directly from colonial-era practices developed by Britain. Through the strategic use of legal categorisation, these drone strikes have utilised pre-existing armed conflicts to present the use of military force for law enforcement as legitimate, and governed by the permissive rules of IHL. In this way, law enforcement drone strikes which would otherwise be unlawful, for instance because they involve statusbased targeting or collateral damage (both of which are inimical to human rights), end up being instead evaluated with reference to IHL and appearing legitimate. If unchallenged, this gives drone-using states extensive latitude to conduct lethal drone strikes and has severe consequences for all those affected, dramatically eroding their human rights, and so must be resisted. Having identified the law enforcement nature of certain kinds of drone strikes, this chapter has worked to reground them within the framework of IHRL. By emphasising that, during situations of armed conflict, IHL and

6686_Enemark.indd 66

02/03/21 12:22 PM

The Complicated Reality of Drone Strikes for Law Enforcement / 67

IHRL apply coextensively, it is possible to assert that human rights law provides the primary framework of legal regulation when operations are not carried out during ‘active hostilities’. Ultimately, this chapter has demonstrated the importance of looking beyond the veil of an armed conflict, and to consider the specific context in which drone strikes (or indeed other uses of military force) are used. Where it is apparent that law enforcement aims (such as counterterrorism) are being pursued, it is vital that this is highlighted and the drive to apply IHL resisted. The armed conflict category provides states with a potential cover with which to greatly legitimise the use of military force, and it is crucial not to accept this uncritically and thereby allow states free rein to use military force for law enforcement.

Notes This chapter came out of the DRONETHICS ‘Governing Drone Violence’ workshop held in Southampton, UK, in July 2019, and my thanks go to the attendees present for excellent discussion and very useful feedback. Thanks also to Faye Bird for providing incisive and productive comments on an earlier draft. 1. That is not to say that their use would be necessarily lawful in either context, a determination that must be made on the relevant applicable law, either IHL or IHRL. 2. Customary international law is a source of law, identified under Article 38(1)(b) of the Statute of the International Court of Justice, which arises through the general practice of states that is accepted as law (see International Law Commission 2018). 3. The UK government has elsewhere argued that ‘[i]f armed RPAS [remotely piloted air systems] were to be used outside the scope of an armed conflict, their use must be in accordance with international human rights law’ (UK Mission 2014). 4. Data as available at the time of writing (16 March 2020).

References Agence France-Presse. 2019. ‘Drone Strike “Kills 4 Qaeda Suspects” in Yemen’, Gulf Times, 7 May, https://www.gulf-times.com/story/630902/Drone-strike-kills4-Qaeda-suspects-in-Yemen Alejandre v. Cuba. 1999. Case 11.589, Inter-Am. Comm’n. H.R., Report No. 86/99 (1999), https://www.cidh.oas.org/annualrep/99eng/Merits/Cuba11.589.htm Al-Skeini and others v. United Kingdom. 2011. 2011-VI Eur. Ct. H.R. 99. Amnesty International. 2013. ‘“Will I Be Next?”: US Drone Strikes in Pakistan’, Amnesty International, https://www.amnestyusa.org/files/asa330132013en.pdf Amnesty International. 2017. ‘Key Principles on the Use and Transfer of Armed Drones’, Amnesty International, https://www.amnesty.org/download/Documents/ ACT3063882017ENGLISH.PDF

6686_Enemark.indd 67

02/03/21 12:22 PM

68 / Max Brookman-Byrne

Andreou v. Turkey. 2008. (dec) App no 45653/99, Eur. Ct. H.R., 3 June. Associated Press. 2002. ‘US Kills Al-Qaeda Suspects in Yemen’, USA Today, 11 May, http://usatoday30.usatoday.com/news/world/2002-11-04-yemen-explosion_x.htm Basic Principles on the Use of Force and Firearms by Law Enforcement Officials. 1990. Eighth United Nations Congress on the Prevention of Crime and the Treatment of Offenders, Office of the High Commissioner for Human Rights, United Nations, https://www.ohchr.org/en/professionalinterest/pages/useofforceandfirearms.aspx BBC. 2002a. ‘Profile: Ali Qaed Senyan al-Harthi’, BBC News, 5 November, http:// news.bbc.co.uk/1/hi/world/middle_east/2404443.stm BBC. 2002b. ‘US “Still Opposes” Targeted Killings’, BBC News, 6 November, http:// news.bbc.co.uk/1/hi/world/middle_east/2408031.stm Ben-Naftali, Orna, and Yuval Shany. 2003. ‘Living in Denial: The Application of Human Rights in the Occupied Territories’, Israel Law Review 37 (1): 17–118. Brennan, John O. 2011. ‘Strengthening Our Security by Adhering to Our Values and Laws’, The White House, 16 September, https://obamawhitehouse.archives. gov/the-press-office/2011/09/16/remarks-john-o-brennan-strengthening-oursecurity-adhering-our-values-an Brookman-Byrne, Max. 2017. ‘Drone Use “‘Outside Areas of Active Hostilities”: An Examination of the Legal Paradigms Governing US Covert Remote Strikes’, Netherlands International Law Review 64 (1): 3–41. Bureau of Investigative Journalism. n.d.a ‘Strikes in Yemen’, https://www.thebureauinvestigates.com/projects/drone-war/charts?show_casualties=1&show_ injuries=1&show_strikes=1&location=yemen&from=2002-1-1&to=now (accessed 16 March 2020). Bureau of Investigative Journalism. n.d.b ‘Strikes in Somalia’, https://www.thebureauinvestigates.com/projects/drone-war/charts?show_casualties=1&show_ injuries=1&show_strikes=1&location=somalia&from=2007-1-1&to=now (accessed 16 March 2020). Celiberti de Casariego v. Uruguay. 1981. R.13/56, U.N. Doc. A/36/40(SUPP): 185. UN Human Rights Committee. Cheney, Dick. 2002. ‘Remarks by the Vice President at the Air National Guard Senior Leadership Conference’, The White House, 2 December, https://georgewbushwhitehouse.archives.gov/news/releases/2002/12/20021202-4.html Code of Conduct for Law Enforcement Officials. 1979. G.A. Res. 34/169, 17 December. Da Costa, Karen. 2013. The Extraterritorial Application of Selected Human Rights Treaties. Leiden: Martinus Nijhoff. Dennis, Michael. 2006. ‘Application of Human Rights Treaties Extraterritorially to Detention of Combatants and Security Internees: Fuzzy Thinking All Around?’, ILSA Journal of International and Comparative Law 12 (2): 459–80. Dinstein, Yoram. 1981. ‘Right to Life, Physical Integrity and Liberty’, in International Bill of Rights Covenant on Civil and Political Rights, edited by Louis Henkin. New York: Columbia University Press, 114–37. Dinstein, Yoram. 2010. The Conduct of Hostilities Under the Law of International Armed Conflict, 2nd edn. Cambridge: Cambridge University Press. ECHR. 1950. European Convention on Human Rights. ETS 005, 4 November.

6686_Enemark.indd 68

02/03/21 12:22 PM

The Complicated Reality of Drone Strikes for Law Enforcement / 69

Edney-Brown, Alex. 2019. ‘The Psychological Effects of Drone Violence: Social Isolation, Self-Objectification, and Depoliticization’, Political Psychology 40 (6): 1341–56. Gaggioli, Gloria. 2013. ‘Report: The Use of Force in Armed Conflicts: Interplay Between the Conduct of Hostilities and Law Enforcement Paradigms’, International Committee of the Red Cross, https://www.icrc.org/en/doc/assets/files/publications/ icrc-002–4171.pdf Geneva Convention (I) for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field. 1949. 75 UNTS 31, 12 August. Geneva Convention (II) for the Amelioration of the Condition of Wounded, Sick and Shipwrecked Members of Armed Forces at Sea. 1949. 75 UNTS 85, 12 August. Geneva Convention (III) Relative to the Treatment of Prisoners of War. 1949. 75 UNTS 135, 12 August. Geneva Convention (IV) Relative to the Protection of Civilian Persons in Time of War. 1949. 75 UNTS 287, 12 August. Gordon, Rebecca. 2016. ‘How the US Military Came to Embrace Extrajudicial Killings’, The Nation, 18 July, https://www.thenation.com/article/how-the-usmilitary-came-to-embrace-extrajudicial-killings Gunneflo, Markus. 2016. Targeted Killing: A Legal and Political History. Cambridge: Cambridge University Press. Gwynn, Sir Charles W. 1939. Imperial Policing. London: Macmillan. Haász and Szabó v. Hungary. 2015. App nos 11327/14 and 11613/14, Eur. Ct. H.R., 13 October. Hathaway, Oona A., Rebecca Crootof, Philip Levitz and Haley Nix. 2012. ‘Which Law Governs During Armed Conflict? The Relationship Between International Humanitarian Law and Human Rights Law’, Minnesota Law Review 96 (6): 1883–1944. Hauser, Christine. 2006. ‘Planner of Cole Attack Escapes from Yemeni Jail’, New York Times, 6 February, https://www.nytimes.com/2006/02/06/world/middleeast/ planner-of-cole-attack-escapes-from-yemeni-jail.html Heyns, Christof. 2014. UN Human Rights Council, Report of the Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions. U.N. Doc. A/HRC/26/36 (2014). Hill-Cawthorne, Lawrence. 2015. ‘Humanitarian Law, Human Rights Law and the Bifurcation of Armed Conflict’, International and Comparative Law Quarterly 64 (2): 293–325. Horwitz, Sari. 2013. ‘Somali National Held Aboard Navy Ship Pleaded Guilty, US Officials Say’, Washington Post, 25 March, https://www.washingtonpost.com/ world/national-security/somali-national-held-aboard-navy-ship-pleaded-guiltyus-officials-say/2013/03/25/6c89aff2-956b-11e2-b6f0-a5150a247b6a_story.html International Committee of the Red Cross. 2016. ‘Commentary on the First Geneva Convention: Convention (I) for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field’, International Committee of the Red Cross, https://ihl-databases.icrc.org/ihl/full/GCi-commentary International Covenant on Civil and Political Rights. 1966. 999 UNTS 171, 16 December. International Law Commission. 2018. Draft Conclusions on Identification of Customary International Law, with Commentaries. U.N. Doc. A/73/10.

6686_Enemark.indd 69

02/03/21 12:22 PM

70 / Max Brookman-Byrne

Interpol. n.d. ‘Red Notices’, Interpol, https://www.interpol.int/en/How-we-work/ Notices/Red-Notices Isayeva v. Russia. 2005. App no 57950/00, Eur. Ct. H.R., 24 February. Jahangir, Asma. 2003. UN Economic and Social Council, Report of the Special Rapporteur, Pursuant to Commission on Human Rights Resolution 2002/36. U.N. Doc. E/CN.4/2003/3 (2003). Jenkins, Brian M. 1984. ‘Combating Terrorism Becomes a War’, RAND Corporation, https://www.rand.org/content/dam/rand/pubs/papers/2005/P6988.pdf Joint Committee on Human Rights. 2016. ‘ The Government’s Policy on the Use of Drones for Targeted Killing: Government Response to the Committee’s Second Report of Session 2015–16’, HL 49, HC 747, UK Parliament, 10 May. Kendall, Sara. 2016. ‘Cartographies of the Present: “Contingent Sovereignty” and Territorial Integrity’, in Netherlands Yearbook of International Law 2016: The Changing Nature of Territoriality in International Law, edited by Martin Kuijerand Wouter Werner. Berlin: Springer, 83–105. Kennedy, David. 2006. Of War and Law. Princeton: Princeton University Press. Kennedy, David. 2012. ‘Lawfare and Warfare’, in The Cambridge Companion to International Law, edited by James Crawford and Martti Koskenniemi. Cambridge: Cambridge University Press, 158–84. Koh, Harold H. 2010. ‘The Obama Administration and International Law’, US Department of State, 25 March, https://2009-2017.state.gov/s/l/releases/remarks/139119 .htm Kretzmer, David. 2005. ‘ Targeted Killing of Suspected Terrorists: Extra-Judicial Executions or Legitimate Means of Defence?’, European Journal of International Law 16 (2): 171–212. Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territories, Advisory Opinion. 2004. I.C.J. Rep. 136 (9 July). Legality of the Threat or Use of Nuclear Weapons, Advisory Opinion. 1996. I.C.J. Rep. 226 (8 July) López Burgos v. Uruguay. 1981. R.12/52, U.N. Doc. A/36/40(SUPP): 176. UN Human Rights Committee. Lubell, Noam. 2010. Extra Territorial Use of Force Against Non-State Actors. Oxford: Oxford University Press. Lubell, Noam, and Nathan Derejko. 2013. ‘A Global Battlefield? Drones and the Geographical Scope of Armed Conflict’, Journal of International Criminal Justice 11 (1): 65–88. McCann v. United Kingdom. 1995. Series A Eur. Ct. H.R. 324. Mazzetti, Mark. 2013. ‘A Secret Deal on Drones, Sealed in Blood’, New York Times, 6 April, https://www.nytimes.com/2013/04/07/world/asia/origins-of-cias-not-sosecret-drone-war-in-pakistan.html Melzer, Nils. 2008. Targeted Killing in International Law. Oxford: Oxford University Press. Meron, Theodor. 1995. ‘Extraterritoriality of Human Rights Treaties’, American Journal of International Law 89 (1): 78–82. Milanović, Marko. 2011. Extraterritorial Application of Human Rights Treaties: Law, Principles, and Policies. Oxford: Oxford University Press.

6686_Enemark.indd 70

02/03/21 12:22 PM

The Complicated Reality of Drone Strikes for Law Enforcement / 71

Moreman, Tim R. 1996. ‘“Small Wars” and “Imperial Policing”: The British Army and the Theory and Practice of Colonial Warfare in the British Empire, 1919–1939’, Journal of Strategic Studies 19 (4): 105–31. Murray, Daragh. 2016. Practitioners’ Guide to Human Rights Law in Armed Conflict. Oxford: Oxford University Press. Nasser, Tony. 2015. ‘Modern War Crimes by the United States: Do Drone Strikes Violate International Law? Questioning the Legality of US Drone Strikes and Analyzing the United States’ Response to International Reproach Based on the Realism Theory of International Relations’, Southern California Interdisciplinary Law Journal 24 (1): 289–328. Neocleous, Mark. 2011. ‘ The Police of Civilization: The War on Terror as Civilizing Offensive’, International Political Sociology 5 (1): 144–59. Neocleous, Mark. 2013. ‘Air Power as Police Power’, Environment and Planning D: Society and Space 31 (4): 578–93. Nowak, Manfred. 1993. UN Covenant on Civil and Political Rights: CCPR Commentary. Khel: N. P. Engel. O’Connell, Mary E. 2011. ‘Remarks: The Resort to Drones Under International Law’, Denver Journal of International Law and Policy 39 (4): 585–600. O’Connell, Mary E. 2016. ‘ The Law on Lethal Force Begins with the Right to Life’, Journal on the Use of Force and International Law 3 (2): 205–9. Pad and others v. Turkey. 2007. (dec) App no 60167/00, Eur. Ct. H.R., 28 June. Paust, Jordan J. 2010. ‘Self-Defense Targetings of Non-State Actors and Permissibility of US Use of Drones in Pakistan’, Journal of Transnational Law and Policy 19 (2): 237–80. Printer, Norman G., Jr. 2003. ‘ The Use of Force Against Non-State Actors Under International Law: An Analysis of the US Predator Strike in Yemen’, UCLA Journal of International Law and Foreign Affairs 8: 331–84. Prosecutor v. Milošević. 2004. Case No. IT-02-54, Decision on Motion for Judgment of Acquittal, Int’l Crim. Trib. For the Former Yugoslavia, 16 June. Prosecutor v. Tadić. 1995. Case No. IT-94-1-AR72, Decision on the Defence Motion for Interlocutory Appeal on Jurisdiction, Int’l Crim. Trib. for the Former Yugoslavia, 2 October. Proulx, Vincent-Joël. 2004. ‘If the Hat Fits, Wear it, if the Turban Fits, Run for your Life: Reflections on the Indefinite Detention and Targeted Killing of Suspected Terrorists’, Hastings Law Journal 56 (5): 801–900. Ramcharan, Bertrand. 1985. ‘ The Concept and Dimensions of the Right to Life’, in The Right to Life in International Law, edited by Bertrand Ramcharan. Leiden: Martinus Nijhoff, 1–32. Reuters. 2019. ‘Six Civilians Killed in Yemen in Suspected Al-Qaeda Bombing’, Reuters, 3 May, https://www.reuters.com/article/us-yemen-security/six-civilianskilled-in-yemen-in-suspected-al-qaeda-bombing-idUSKCN1S91UD Rodley, Nigel. 1989. ‘Human Rights and Humanitarian Intervention: The Case Law of the World Court’, International and Comparative Law Quarterly 38 (2): 321–33. Ruys, Tom. 2005. ‘License to Kill? State-Sponsored Assassination Under International Law’, Military Law and Law of War Review 44: 13–49.

6686_Enemark.indd 71

02/03/21 12:22 PM

72 / Max Brookman-Byrne

Sadat, Leila N. 2012. ‘America’s Drone Wars’, Case Western Reserve Journal of International Law 45 (1 and 2): 215–34. Sassòli, Marco. 2011. ‘ The Role of Human Rights and International Humanitarian Law in New Types of Armed Conflicts’, in International Humanitarian Law and International Human Rights Law, edited by Orna Ben-Naftali. Oxford: Oxford University Press, 34–94. Satia, Priya. 2006. ‘ The Defense of Inhumanity: Air Control and the British Idea of Arabia’, American Historical Review 111 (1): 16–51. Satia, Priya. 2014. ‘Drones: A History from the British Middle East’, Humanity: An International Journal of Human Rights, Humanitarianism, and Development 5 (1): 1–31. Scahill, Jeremy. 2015. ‘ The Assassination Complex’, The Intercept, 15 October, https://theintercept.com/drone-papers/the-assassination-complex Schindler, Dietrich. 1982. ‘Human Rights and Humanitarian Law: Interrelationship of the Laws’, American University Law Review 31: 935–43. Schoettle, James A., Jr. 2009. ‘Detention of Combatants and the Global War on Terror’, in The War on Terror and the Laws of War: A Military Perspective, 1edited by Geoffrey S. Corn and Michael W. Lewis. Oxford: Oxford University Press, 67–124. Suarez de Guerrero v. Colombia Communication. 1982. R.11/45, U.N. Doc. A/37/40(SUPP): 137. UN Human Rights Committee. Trump, Donald. 2019. ‘Our Great Military’, @realDonaldTrump, Twitter, 6 January, https://twitter.com/realDonaldTrump/status/1081935373914259458 UK Mission. 2014. ‘UK Statement at the Panel on Remotely Piloted Aircraft Systems (RPAS), Human Rights Council, Geneva’, UK Mission to the WTO, UN and Other International Organisations (Geneva), 22 September, https://www.gov.uk/government/news/uk-statement-at-the-panel-on-remotely-piloted-aircraft-systemsrpas-human-rights-council-geneva UN Economic and Social Council. 2003. Letter dated 14 April 2003 from the Chief of Section, Political and Specialized Agencies, of the Permanent Mission of the United States of America to the United Nations Office at Geneva addressed to the secretariat of the Commission on Human Rights. 22 April. U.N. Doc. E/ CN.4/2003/G/80. UN Human Rights Committee. 2004. General Comment 31. 26 May. U.N. Doc. CCPR/C/21/Rev.1/Add.13. UN Human Rights Committee. 2005. Third Periodic Report: United States of America. 28 November. U.N. Doc. CCPR/C/USA/3 Annex 1. UN Human Rights Committee. 2018. General Comment 36. 30 October. U.N. Doc. CCPR/C/36. US Central Command. 2019a. ‘USCENTCOM Confirms the Death of Jamal alBadawi’, US Central Command, 7 January, https://www.centcom.mil/MEDIA/ STATEMENTS/Statements-View/Article/1725215/uscentcom-confirms-thedeath-of-jamal-al-badawi US Central Command. 2019b. ‘CENTCOM Yemen Strike Summary Jan 1 – Apr 1, 2019’, US Central Command, 1 April, https://www.centcom.mil/MEDIA/PRESSRELEASES/Press-Release-View/Article/1801674/centcom-yemen-strike-summaryjan-1-apr-1-2019

6686_Enemark.indd 72

02/03/21 12:22 PM

The Complicated Reality of Drone Strikes for Law Enforcement / 73

US Department of Defense. 2015. Law of War Manual. Washington, DC: US Department of Defense. US Department of State. 2007. ‘US Observations on Human Rights Committee General Comment 31’, US Department of State, 27 December, https://20012009.state.gov/s/l/2007/112674.htm US Department of State. 2010. ‘Memorandum Opinion on the Geographic Scope of the International Covenant on Civil and Political Rights’, US Department of State, 19 October. Van Schaack, Beth. 2014. ‘ The United States’ Position on the Extraterritorial Application of Human Rights Obligations: Now is the Time for Change’, International Law Studies 90: 20–65. Walzer, Michael. 2016. ‘Just & Unjust Targeted Killing & Drone Warfare’, Dædalus 145 (4): 12–24. Watkin, Kenneth. 2004. ‘Controlling the Use of Force: A Role for Human Rights Norms in Contemporary Armed Conflict’, American Journal of International Law 98 (1): 1–34. White House. 2002. ‘ The National Security Strategy of the United States of America’, The White House, September, https://2009-2017.state.gov/documents/ organization/63562.pdf White House. 2013. ‘Fact Sheet: US Policy Standards and Procedures for the Use of Force in Counterterrorism Operations Outside the United States and Areas of Active Hostilities’, The White House, 23 May, https://obamawhitehouse.archives. gov/the-press-office/2013/05/23/fact-sheet-us-policy-standards-and-proceduresuse-force-counterterrorism

6686_Enemark.indd 73

02/03/21 12:22 PM

FOUR

Drone Violence as Wild Justice: Administrative Executions on the Terror Frontier Christian Enemark

Introduction Outside of Sana’a [Yemen], it’s the Wild West. Vivian Salama, journalist (Woods 2015, 209)

On 12 October 2000, seventeen US Navy sailors were killed when the USS Cole, a guided missile destroyer, was attacked in the Yemeni port of Aden by suicide bombers in a small boat laden with explosives. More than eighteen years later, US President Donald Trump confirmed that a man accused of organising this attack had recently been killed in a US airstrike (probably a drone strike) in Yemen. Trump tweeted in early January 2019 (Browne and Starr 2019): Our GREAT MILITARY has delivered justice for the heroes lost and wounded in the cowardly attack on the USS Cole. We have just killed the leader of that attack, Jamal al-Badawi. Our work against al Qaeda continues.

Soon afterwards, US Central Command, which oversees military operations in the region, tweeted confirmation that al-Badawi was killed in the strike while driving a vehicle alone in the Marib Governorate. A military spokesman later explained that in 2003 al-Badawi had been indicted by a federal grand jury, charged with fifty counts of various terrorism offences, including murder of US nationals. Before the 2019 strike, al-Badawi had been on a list of ‘most wanted terrorists’ maintained by the Federal Bureau of Investigation, and the US State Department’s Rewards for Justice Program had

6686_Enemark.indd 74

02/03/21 12:22 PM

Drone Violence as Wild Justice / 75

offered up to $5 million for information leading to his arrest (Browne and Starr 2019). Although much remains to be discovered about the circumstances surrounding the killing of al-Badawi, it is reasonable to suppose that he was deliberately targeted in a so-called ‘personality strike’. Based on what little is known publicly about the US government’s use of armed drones, the purpose of this kind of drone strike is to eliminate named individuals. Personality strikes are thus distinct from ‘signature strikes’ against ‘individuals who are unidentified but behaving in a manner considered by military and government officials to resemble the actions of combatants’ (Cronin 2018, 108–9). For the purposes of this chapter, there are four features of al-Badawi’s killing worth highlighting: (1) Yemen is not a place where the United States is party to an armed conflict; (2) it is ambiguous, from the known circumstances and the language used by Trump, whether al-Badawi was regarded by the US government as a criminal or a combatant, or both; (3) it is likewise unclear whether the strike against al-Badawi was a preventive response to anticipated wrongdoing (terrorism) or a punitive response to past wrongdoing (the attack on the Cole), or both; and (4) the killing of al-Badawi was apparently done without judicial sanction. These features of al-Badawi’s killing are similar to those of the killing of Ali Qaed Senyan al-Harthi, another man allegedly involved in the 2000 attack on the Cole. In early November 2002, Yemen was the site of the US government’s first personality strike outside a war zone and the first reported instance of an armed drone being used lethally by the civilian Central Intelligence Agency (CIA). On that occasion, anonymous US government sources claimed that six suspected al-Qaeda members (including al-Harthi) died when the vehicle they were travelling in (on a road to the east of Yemen’s capital, Sana’a) was hit by a missile fired from a Predator drone (BBC 2002; Hoyle and Koch 2002). In a television interview on 5 November 2002, US Deputy Defense Secretary Paul Wolfowitz effectively acknowledged US responsibility for the drone strike: ‘it’s a very successful tactical operation . . . And one hopes each time you get a success like that, [you] not only have gotten rid of somebody dangerous, but [you] have imposed changes in their tactics and operations and procedures’ (Crawley and Svitak 2002). On the same day, however, Sweden’s foreign minister Anna Lindh condemned the strike as ‘a summary execution that violates human rights’ (Crawley and Svitak 2002). Ever since, disagreement has persisted about the status, morality and legality of drone strikes conducted against named individuals located outside war zones. Ethicists, lawyers and policy analysts have fiercely debated whether lethal acts of this kind should be judged and governed according to either

6686_Enemark.indd 75

02/03/21 12:22 PM

76 / Christian Enemark

of two moralities of state violence: the morality of the war paradigm, or the morality of the peacetime paradigm of law enforcement. Depending on one’s viewpoint (which depends, in turn, upon one’s view of what counts as ‘war’ and ‘law enforcement’), drone-based personality strikes such as those carried out by the US government are seen either as permissible acts of killing in war, or as impermissible homicides (unlawful acts of lethal law enforcement) outside war. This chapter’s contribution to that debate, focusing on the law enforcement paradigm, is to explore the potential for drone violence of this kind to become permissible. It begins by discussing the difficulty of conceptualising personality strikes outside war zones as acts of war, and it then proposes regarding them instead as a rough practice of lethal, punitive law enforcement: ‘wild justice’ conducted on a US-designated ‘terror frontier’. In this vision, personality strikes arguably feed an appetite for the idea of redemptive violence in US popular and political culture. The US government has claimed that it takes great care in placing named individuals on targeting lists and accordingly conducting what amount to drone-based ‘administrative executions’. However, drone violence of this kind will remain wild (and thus impermissible as a law enforcement practice) for as long as it lacks the fairness required in a proper criminal justice process. This presents a governance challenge: to tame personality strikes by making the authorisation process fairer or, if this cannot be done, to prohibit them.

Drone Violence as Non-War Over the last two decades, the state use of armed drones has received scholarly examination from a variety of disciplinary perspectives: historical, strategic, sociological, legal and ethical. Almost all these studies have proceeded from the assumption that violence of this kind is, in essence, war. Among scholars who address the topic from an ethical perspective, the dominant approach is then to frame armed drones as ushering in a new, militarily revolutionary kind of warfare that occasions the novel application and development of Just War theory (Killmister 2008; Strawser 2010; Boyle 2013; Schulzke 2017). The discourse of ‘drone warfare’ similarly dominates current political thinking, giving primacy to the notion that drone use in general can and should be governed using laws of war that derive from Just War principles. However, there are a variety of circumstances in which states have used and might use an armed drone, and not all uses can plausibly be assigned to the paradigm and morality of war. Sometimes, the conceptualisation of drone violence as war is easier and the applicability of military ethics seems clearer: for example, where an armed drone is used to provide

6686_Enemark.indd 76

02/03/21 12:22 PM

Drone Violence as Wild Justice / 77

close air support in defence of friendly troops who are in contact with (and immediately threatened by) enemy troops. Here, the morality of drone violence is wrapped up in the morality of a broader war effort being pursued by various military means. By contrast, the exclusive use of armed drones as a method of individualised and uncontested violence – in a place where the state using them is not a party to an armed conflict – is harder to conceptualise and judge as the stuff of war. In these circumstances, drone violence might instead more closely resemble (or be better regarded as) a practice of lethal law enforcement. Circumstances and conceptualisations matter because these determine the applicability of certain kinds of rules and, in turn, the degree of operational flexibility that is legally available to a drone-using state. When state violence is exercised in world affairs, there are two distinct and established sets of rules that might apply: humanitarian law, which applies in wartime and is more permissive regarding the taking of human life; or human rights law, which generally applies in peacetime and is more restrictive. Applying the first or second set of rules is critically dependent, however, upon differentiating drone violence as warfare from drone violence as law enforcement respectively. Various international organisations have passed ‘motherhood’ resolutions calling upon states to use armed drones ‘in compliance with international law’ (UNGA 2013; European Parliament 2014; UNHRC 2014; Parliamentary Assemble 2015), but these resolutions have not resolved the uncertainty about which international laws are applicable to drone use in certain kinds of circumstances. In the case of personality strikes conducted against named individuals located in places where the United States is not at war, the US government under President Barack Obama asserted that the law of war is nonetheless applicable. In early 2013 a Justice Department memo was leaked to NBC News that had been sent the previous year to members of the US Senate’s Intelligence Committee and Judiciary Committee. The document addressed ‘the circumstances in which the U.S. Government could use lethal force in a foreign country outside the area of active hostilities against a US citizen who is . . . an al-Qaida leader actively engaged in planning operations to kill Americans’. Among the conditions under which, the memo argued, such a lethal operation would be legal was that: ‘the operation would be conducted in a manner consistent with applicable law of war principles’ (Jaffer 2013, emphasis added). Against this assertion, however, some scholars have suggested that the US government has sometimes wrongly characterised its drone violence as war in order to avail itself of the greater permissiveness of the war paradigm (Alston 2010, 3; Wuschka 2011, 905; Schaller 2015, 195; Meloni 2016, 47).

6686_Enemark.indd 77

02/03/21 12:22 PM

78 / Christian Enemark

If the question of whether a condition of war exists is not merely a subjective matter, at least three kinds of objections could be raised against the categorising of US personality strikes as war. One objection (quantitative) is that this form of violence is of insufficient intensity and/or is too small-scale to count as war. Rather, it is a kind of force short of war (vim) (Brunstetter and Braun 2011). Another objection (legal-territorial) is that, as the condition of war exists only in places where a state is party to an ongoing armed conflict, personality strikes against individuals located outside those places do not count as acts occurring in war (Warren and Bode 2015, 186; Brookman-Byrne 2017, 4).1 A third objection (essentialist) focuses on the radical asymmetry of immediate physical risk (as between killer and victim) inherent in personality strikes conducted remotely across vast distances. If war by its nature involves mutual exposure to risk, and if neither the remote operators of armed drones nor any friendly troops in the vicinity of a strike location are at risk, this kind of drone violence does not count as war (Whetham 2013; Enemark 2014; Chamayou 2015, 17; Gusterson 2016. See also Kahn 2002). Where it is more difficult to conceptualise a state’s violence as war, the permissive morality of war is less obviously available. That violence is then more open to condemnation as impermissible homicide unless there is instead a non-war basis for potentially legitimising it. In the case of personality strikes, perhaps that basis is law enforcement. As an alternative to the paradigm of war, the paradigm of lethal law enforcement comprises two sub-paradigms: preventive (the use of violence by police to neutralise imminent or current threats to public safety) and punitive (the use of violence by executioners to punish criminality). This chapter focuses on the latter sub-paradigm in which there are strong, rights-based restrictions on killing. The most important of these, based on the right to life, is the prohibition on killing arbitrarily. This prohibition exists in international law, for example in the 1966 International Covenant on Civil and Political Rights, which provides (Article 6) that ‘sentence of death . . . can only be carried out pursuant to a final judgement rendered by a competent court’ (ICCPR 1966). And it exists in the domestic law of many states, for example in the US Constitution, which provides (Fifth Amendment) that ‘No person shall be . . . deprived of life . . . without due process of law’ (Kim 2017). In practice, when the punitive law enforcement model applies to non-war violence, respect for the right to life and due process involves the presumption of an accused person’s innocence and the conduct of a fair trial (Ramsden 2011, 395). To follow legal processes carefully is to adhere to the principle of legalism which, in matters of criminal justice, helps to avoid arbitrariness in the

6686_Enemark.indd 78

02/03/21 12:22 PM

Drone Violence as Wild Justice / 79

killing of human beings who bear a presumptive right to life. The essential characteristic of legalism is orderliness. Through the conscientious and thorough application of modern law’s myriad rules and customs, the practitioners and observers of criminal justice processes can achieve a high level of certainty about the correctness of decision-making. By contrast, in the absence of a fair trial and conviction preceding an execution, it is more difficult to be confident that an executed individual really was criminally culpable and deserving of death. A common criticism of the US government’s drone-based personality strikes outside war zones is that they are done without recourse to legalism (Lee 2015, 434; Shane 2011), in which case they are difficult to justify as acts of punitive law enforcement. One way of thinking about this problem is to conclude that the US government, outside the paradigm of war, wilfully perpetrates impermissible homicides (murders). However, this chapter takes a different approach, allowing that there may be scope to conceptualise and perhaps legitimise personality strikes within the law enforcement paradigm. Arguably, these strikes currently count neither as war nor as outright homicide. Rather, it is better to regard them as continuing a US tradition of sometimes inflicting flawed methods of violent law enforcement (‘wild justice’) upon suspected criminals.

Wild Justice and Frontier Violence In their 2010 book Revenge versus Legality, Katherine Maynard and her colleagues explore the contrast and relationship between legal and wild justice. On the one hand, they argue, there is ‘extralegal, vigilante, or “wild” justice based upon revenge and driven by passion and grief’ and, on the other hand, there is ‘tamer, cooler, more rational and institutional legal justice’ (Maynard et al. 2010, 3). As a form of law enforcement violence, wild justice is pursued self-righteously and vengefully. It tends to be untamed by adherence to formal legal constraints, and its wielders typically judge this wildness to be more effective in achieving just outcomes than is the practice of non-wild (legalistic) justice. Moreover, for these reasons wild justice is more likely to occur in situations and territories where the rule of law is weak. In the United States, we can discern an appetite for wild justice driven by a ‘myth of redemptive violence’ which runs deep in its popular and political culture (Maynard et al. 2010, xii). This is revealed by historical or fictional circumstances in which there are certain ‘crimes or injuries that appear so mystifying or disturbing that ordinary, conventional legality – the usual, normative police procedures, judges’ rulings, lawyers’ motions, and jury verdicts – seems inadequate’. Instead, ‘a Dirty Harry is needed to solve or avenge such a crime; cruel or unusual procedures or punishments are

6686_Enemark.indd 79

02/03/21 12:22 PM

80 / Christian Enemark

called for; the law must be bent, broken or at least supplemented to achieve a just revenge’ (Maynard et al. 2010, 5). To the extent that the idea of redemptive violence is an appealing one in the United States, its appeal can be traced back to the country’s early history. In the time of the American Old West (or Wild West), pioneering settlers were extending the US frontier westward. Across this vast and under-governed territory, law enforcement was generally a more violent and disorderly enterprise than that which obtained in the more established, urban settings in the east. Frontier justice was delivered by US government ‘lawmen’ like Wyatt Earp (1848–1929), who, in the Arizona Territory town of Tombstone, took part in the famous ‘Gunfight at the O.K. Corral’ during which lawmen shot and killed three outlaw cowboys (Marks 1989). Such real-life episodes from early American history have since been emulated abundantly in US popular culture. Loren D. Estleman’s 2018 novel Wild Justice, for example, tells the story of Deputy US Marshal Page Murdock, whose job is ‘bringing justice to a wilderness’ on the ‘untamed’ American frontier (Estleman 2018). Between the 1930s and 1960s, Hollywood script writers frequently set narratives of violence in spaces ‘where legality is so weak or non-existent that a good gunfight (or fistfight) is the only credible way to resolve differences or achieve justice’ (Maynard et al. 2010, 167). In ‘westerns’, set in Wild West desert environments, justice was portrayed as most likely to be achieved by characters (played by actors like John Wayne) armed with revolvers or rifles (Maynard et al. 2010, 167). Later, characters played by Clint Eastwood, Charles Bronson and Sylvester Stallone were variously depicted in popular US-made films as ‘heroic avengers ridding the world of punks, thugs, rapists, homicidal maniacs, and other evildoers’. These films contained the clear message that, in confronting evil-inspired dangers, the heroes’ tactics sometimes needed to be ‘ruthless and lawless’ (Maynard et al. 2010, 170). Moreover, this preference for lawlessness sometimes extended to a contempt for legalism as something that gets in the way of true justice. For example, in the 1971 Eastwood film Dirty Harry, the eponymous hero has to contend with criminal and non-criminal antagonists. The latter include ‘a variety of liberal twits and wusses, city and police officials, who . . . castigate him for not following proper legal and police procedures’ (Maynard et al. 2010, 172–3). The likely influence of popular culture valorising wild justice should not, however, be overstated when it comes to US politics and foreign policy. It is important to acknowledge that the US government has historically acted as a champion of legalistic approaches to law enforcement, in contrast to other states (including enemies of the United States) that

6686_Enemark.indd 80

02/03/21 12:22 PM

Drone Violence as Wild Justice / 81

have sometimes preferred to reject legalism. When Hermann Göring was appointed (by Adolf Hitler) as Prussian Minister of the Interior in 1933, he boasted: ‘my measures will not be crippled by any legalistic hesitation’ (Conot 1993, 121). And in Nazi-ruled Germany more generally, due process and the rule of law were swiftly replaced by the straightforward Führerprinzip, the principle that a leader’s commands equate to law. After the end of the Second World War, when justice needed to be served in respect of alleged war crimes, the US government insisted upon restraint and legalism despite urgings to the contrary by its allies. Josef Stalin, for example, had wanted simply to kill up to 100,000 Nazis in revenge for what they had done to the Soviet Union (Maynard et al. 2010, 186). And the British foreign secretary, Anthony Eden, had favoured summary executions of top Nazi leaders, claiming: ‘ The guilt of such individuals is so black’ that ‘they fall outside and go beyond the scope of any judicial process’ (Bass 2000, 185). Regardless, under American leadership, the Nuremburg tribunals proceeded to apply legalism and the judicial method to Nazi defendants (three of whom were acquitted). A major shift in the US government’s attitude to legalism appeared to occur after 11 September 2001, when political violence of external origin came to the US mainland for the first time since the American–British ‘ War of 1812’. America’s people and politicians reacted to al-Qaeda’s ‘9/11’ terrorist attacks with shock and fear, and the political mood that descended upon the United States drew upon the cultural legacy of its frontier era. The popularly appealing myth of redemptive violence surged to the fore as the country set about confronting dreaded enemies inhabiting disorderly settings of civil war, state failure and insurgency in North Africa, the Middle East and Central Asia (Motlagh 2005; Reuters 2008; BBC 2019). One such ‘terror frontier’ is the Federally Administered Tribal Areas of Pakistan, in the northwest of the country bordering Afghanistan, where the national government has little control or meaningful authority over the population or territory. Eventually this became the setting for extensive US use of drone strikes (including personality strikes) against suspected terrorists. For example, in August 2009 the leader of the Pakistani Taliban, Baitullah Mehsud, was killed (along with eleven other people) in a US drone strike in the remote and mountainous region of South Waziristan. Mehsud had allegedly been the organiser of suicide bombings across the border in Afghanistan and of the 2007 assassination of former Pakistani Prime Minister Benazir Bhutto (Reuters 2009; Mazzetti and Walsh 2013). From the beginning of the ‘war on terror’, a dominant political attitude was that American counterterrorism would be more effective if it were conducted aggressively and ruthlessly. In a television interview on

6686_Enemark.indd 81

02/03/21 12:22 PM

82 / Christian Enemark

16 September 2001, US Vice President Dick Cheney declared (White House 2001, emphasis added): I think the world increasingly will understand what we have here are a group of barbarians, that they threaten all of us . . . We also have to work, though, sort of the dark side, if you will. We’ve got to spend time in the shadows . . . That’s the world these [terrorist] folks operate in, and so it’s going to be vital for us to use any means at our disposal, basically, to achieve our objective.

A few days later, President George W. Bush promised the US Congress: ‘whether we bring our enemies to justice, or bring justice to our enemies, justice will be done’ (McCaleb 2001). The bringing of enemies to justice implied the use of legal processes such as arrest and trial, but the alternative of bringing justice to enemies seemed to threaten the use of non-legalistic (wild) methods. The president subsequently made clear his conviction that the formal justice system lacked the capacity to adequately address the problem of terrorism. In his 2004 State of the Union speech, Bush acknowledged: ‘some people question if America is really in a war at all. They view terrorism more as a crime, a problem to be solved mainly with law enforcement and indictments.’ But he insisted: ‘After the chaos and carnage of September 11th [2001], it is not enough to serve our enemies with legal papers’ (Bush 2004). Legalism was largely rejected in favour of wild justice, and so the Bush administration and its successors proceeded ruthlessly to achieve retribution for 9/11. In zealous pursuit of al-Qaeda globally over nearly two decades, suspected terrorists have been indefinitely detained by the US government, sometimes tortured, or sometimes apparently executed without trial. For the last of these, armed drones have been the preferred method, not least because of the capacity they afford to strike an individual in a remote and faraway location and hence to make that strike more secret.

Personality Strikes as Administrative Executions In early 2011 the US Joint Special Operations Command authorised a squadron in control of drones flying over Yemen to kill Anwar al-Awlaki. At the time, ‘Most Wanted’ posters of this US citizen were displayed inside every ground control station at Cannon Air Force Base in New Mexico (home to the 3rd and 33rd Special Operations Squadrons) (Woods 2015, 119). On 30 September 2011 the drones were flown from a base in Saudi Arabia into northern Yemen, where they fired missiles at a car carrying al-Awlaki (Mazzetti et al. 2011). Not until May 2013 did the US Attorney General confirm publicly that al-Awlaki was among four named US citizens who had been killed

6686_Enemark.indd 82

02/03/21 12:22 PM

Drone Violence as Wild Justice / 83

‘outside areas of active hostilities’ in US drone strikes (Woods 2015, 137). More than a year prior to this killing, CIA director Leon Panetta had used a television interview to identity al-Awlaki as a terrorist who posed a threat to the United States. In explaining the man’s placement on a ‘terrorist list’, Panetta said: ‘ You can track Awlaki to the Detroit bomber. We can track him to other attacks in this country that have been urged by Awlaki’ (ABC 2010).2 This highlighting of past wrongdoing suggests that, when al-Awlaki was subsequently killed in a US drone strike, the killing was at least partly punitive in purpose. If so, this personality strike in a non-war setting is an example of drone violence manifesting as wild justice. In 2012 President Obama’s chief counterterrorism adviser, John Brennan, insisted that US use of lethal force was ‘not about punishing terrorists for past crimes; we are not seeking vengeance’ (Brennan 2012). And in 2013 Obama himself claimed that drone strikes in particular are not undertaken ‘to punish individuals – we act against terrorists who pose a continuing and imminent threat to the American people’ (Obama 2013). However, the very phrase ‘continuing and imminent threat’ implies that, when such an individual is targeted for a drone strike, their past actions are at issue. It is a phrase founded on the notion of ‘elongated imminence’, reportedly developed by US State Department legal adviser Harold Koh, which he compared to ‘battered spouse syndrome’. In the same way that a spouse does not have to wait until a hand is being raised to strike before acting in self-defence if there is a pattern of past abuse, Koh argued, the US government does not have to wait to strike at terrorists (Klaidman 2012, 219–20). If such was the thinking behind personality strikes conducted during the Obama administration, there is scope to regard these as instances of punitive drone violence within the law enforcement paradigm. Moreover, the way in which the US government reportedly made personality strike decisions at that time further suggests its application of a law enforcement mentality (albeit a wild one). Wild justice involves a rejection of legalism, but sometimes this rejection manifests as a parodying or mimicking of legalistic approaches to criminal justice. Most obviously, there is the phenomenon of the show trial. In the Soviet Union, for example, the government would often stage public ‘trials’ in which defendants were coerced to make confessions and guilty verdicts (to be used for propaganda purposes) were a foregone conclusion (Maynard et al. 2010, 164). In the case of the Obama administration and its personality strikes, the mimicking of legalism instead involved un-shown quasi-trials. The drone strikes resulting from these secret proceedings could thus be described as administrative executions. The US government under Obama apparently regarded personality strikes as requiring less permissiveness than would be

6686_Enemark.indd 83

02/03/21 12:22 PM

84 / Christian Enemark

allowed in the war paradigm, but not the high degree of restriction on state violence that accompanies full commitment to legalism within the law enforcement paradigm. The White House chief of staff, William Daley, explained in 2011: ‘The president accepts as a fact that a certain amount of screw-ups are going to happen, and to him, that calls for a more judicious process’ (Becker and Shane 2012). And in 2012 John Brennan suggested that the right to life loomed large in Obama’s thinking about personality strikes: ‘the president, and I think all of us here, don’t like the fact that people have to die. And so he wants to make sure that we go through a rigorous [pre-strike] checklist’ (Becker and Shane 2012). Administrative decisions to authorise a drone strike against named individuals (suspected terrorists) located outside war zones were reportedly made only after lengthy deliberation and extensive consultation with officials across the US government. As such, these personality strikes could not be called ‘summary’ executions in the sense of executions performed without delay and hesitation. Even so, they were undoubtedly extrajudicial decisions to kill, despite their bearing a resemblance to a judicial criminal justice process: the involvement of lawyers, the weighing of ‘evidence’, the rendering of a ‘verdict’ and, if a person was ‘convicted’, the imposing of a punishment to fit the crime. In 2012 the New York Times first revealed what it called ‘the strangest of bureaucratic rituals’ in which more than 100 national security officials gathered via videoconference on a weekly basis ‘to pore over terrorist suspects’ biographies and recommend to the president who should be the next to die’ (Becker and Shane 2012). In contrast to the predictability of a Soviet-style show trial, the New York Times reported that ‘participants do not hesitate to call out a challenge, pressing for the evidence behind accusations of ties to Al Qaeda’ (Becker and Shane 2012). Prior to each of these meetings to consider lists of potential targets, the ‘résumés’ of individual terror suspects were put through a process of vetting, validation and evaluation. Then, at the White House, a committee chaired by the president would vote on who, where and when to strike (Cronin 2018, 116). In a 2014 article, Gregory McNeal (2014, 708) described this administrative process as ‘complex and time intensive, usually involving dozens of analysts from different agencies’. To illustrate the high degree of care taken, he noted that ‘target folders are continuously updated to reflect the most recent information regarding a target’s status, [and] the compiled data is independently reviewed by personnel not responsible for its collection’ (McNeal 2014, 721). The occurrence of these practices was largely confirmed towards the end of Obama’s presidency by the 2016 publication of the US government’s drone ‘playbook’: ‘Procedures for Approving Direct Action Against Terrorist Targets Located Outside the United States and Areas of Active

6686_Enemark.indd 84

02/03/21 12:22 PM

Drone Violence as Wild Justice / 85

Hostilities’ (Ackerman 2016; Currier 2016).3 As a basis for conducting personality strikes beyond the paradigm of war, these detailed procedures were possibly intended to be followed also by Obama’s successor, in the hope that she or he might continue his careful approach. And it could be argued that ‘playing’ by the playbook prevented greater injustices that might otherwise have resulted from the US government’s use of armed drones on its terror frontier. Even so, a rigorous administrative approach can still only mimic a legalistic process of guilt-determination and sentencing, and thus Obama’s approach still impoverished drone violence as an exercise in punitive law enforcement. Critically, this was violence controlled entirely by the executive branch of government, allowing for none of the impartiality and fairness that is expected of decision-making by an independent judiciary. This is fundamentally why personality strikes outside war zones amount to wild justice, and it raises the question of whether drone-based law enforcement can or should be ‘tamed’ by involving the courts.

Taming Drone Violence After Donald Trump became president, he quickly sought to distance himself from his predecessor’s approach to global counterterrorism. Whereas Obama had always insisted that his government’s use of force against terrorists was not essentially punitive, in 2017 Trump appeared enthusiastically to favour the exercise of redemptive violence (Trump 2017): Terrorists . . . are nothing but thugs and criminals and predators, and that’s right – losers. . . . These killers need to know they have nowhere to hide, that no place is beyond the reach of American might and American arms. Retribution will be fast and powerful as we lift restrictions and expand authorities.

The president’s clear implication here was that legal and/or administrative restraints would only make US retribution for terrorist acts less effective. Accordingly, senior officials in the Trump administration have reportedly pursued changes to the way drone strikes are conducted, although the secrecy surrounding US drone use makes it hard to know if these changes have been implemented. In contrast to the Obama-era approach, Trump’s reported preferences include expanding the geographic scope for drone strikes to occur ‘outside of areas of active hostilities’, and lowering the evidential thresholds required to conduct strikes against terrorism suspects (Stohl 2018, 5; Jaffe and de Young 2017). If these changes have been enacted in order better to achieve ‘retribution’, the wild justice of US drone

6686_Enemark.indd 85

02/03/21 12:22 PM

86 / Christian Enemark

violence might have become even wilder under Trump than it was under Obama. Moreover, the leaders of some other drone-using states have demonstrated a keenness to emulate the president’s conspicuous adoption of vengeful rationale (Johnson 2018; Gibson 2019). One possible approach to taming that wildness is to prohibit the use of armed drones in non-war settings for a punitive law enforcement purpose (execution). However, such a prohibition could be hard to enforce internationally, as state compliance would be practically difficult to verify. Drone use is generally more conducive to secrecy than is the use of manned aircraft, because no pilot is placed at risk of capture in a foreign territory, and the invisibility of drone violence is deepened when (as is currently typical) drone strikes are conducted in remote parts of the world. In such circumstances, a global prohibition might simply be ignored with impunity by some drone-using states. An alternative approach, then, is to accept that drone violence will sometimes be wielded for a punitive purpose and to allow states to restrain it themselves on law enforcement terms. To the extent that the extrajudicial character of personality strikes is the problem, perhaps the solution is to arrange for judicial sanctioning of such killings. In countries like the United States that already allow the death penalty, judges who are impartial (that is, independent of the executive arm of government) could conduct trials in absentia if an accused terrorist were unable to be extradited. These trials would aim fairly to establish whether the defendant (adequately represented in court) is guilty and, if so, a judge could then sentence that individual to be targeted in a drone strike if no other method of execution were available. In this way, genuine legalism would afford greater public confidence in the probity of drone-based, punitive law enforcement: that the verdict was correct and the punishment was deserved, and that the risk of arbitrariness in the taking of human life was greatly reduced. In a 2013 speech on counterterrorism, President Obama raised the idea of ‘a special court to evaluate and authorize lethal action’ (Obama 2013). And the idea of a ‘drone court’ has attracted some support in the United States as an alternative to what Senator Angus King has described as ‘the executive being the prosecutor, the judge, the jury and the executioner all in one’ (Shane 2013; see also Guoira and Brand 2015). One objection (feasibility) is that courts are not institutionally equipped to be involved in fast-paced military decisions (Katyal 2013; Epps 2013), but this argument is strongest when contemplating drone violence in the war paradigm. Where personality strikes against individuals located outside war zones are instead treated as a law enforcement matter, decisions on whether to authorise placing someone on a target list would not necessarily need to be made quickly. Moreover,

6686_Enemark.indd 86

02/03/21 12:22 PM

Drone Violence as Wild Justice / 87

it could be advantageous from a justice perspective to take the time necessary for the conduct of a fair trial. By going to trial publicly, the government of a drone-using state would effectively put an accused terrorist on notice that they might be targeted. And, while the trial process was under way, that defendant would have a lengthy opportunity to surrender themselves and attend court. Subsequently (if convicted of a capital offence), she or he would be available for execution by conventional means and the possibility of a personality strike would be avoided. Another way of avoiding this possibility would be for the prosecuting government to capture an alleged terrorist, so a judge conducting a ‘drone court’ trial would need to remain satisfied about the continued infeasibility of capture. On this point, however, the government might have to account for the extent to which it is responsible for the infeasibility of capture by arranging for its agents (military, intelligence or police) to be kept at a distance from the area where the accused terrorist is located. If, after all, an individual were unable to be tried in person, the principle of fairness would require that they had access to adequate representation, so a judge would need also to remain satisfied that a secure line of communication was in place between the absent accused and their in-court lawyer. Then, if the court’s verdict was to convict and if capturing the convicted terrorist remained infeasible, a judicially authorised drone strike could be carried out on the condition that this method of punishment would not harm any innocent bystanders (see McMahan 2012, 135–6). Even under such strict conditions for the judicial authorisation of personality strikes outside war zones, there might still be lingering concerns that trials in absentia are not sufficiently fair and that the bystander risk is too high. Arguably, however, these concerns are possibly outweighed by the human rights benefit of reducing the risk of arbitrary killing. As a state’s court system consumed whatever time were needed to maximise confidence in the fairness of a trial, a government acting in accordance with this approach to drone violence would probably itself be unable to conduct personality strikes as frequently as would be the case if these executions were the results of non-judicial (executive) decisions.

Conclusion Drone violence in the form of personality strikes against individuals located outside war zones is difficult to categorise as war, so it cannot easily be justified and governed by reference to war’s moral standards. A better approach is to take seriously the non-war status of that violence. From a law enforcement perspective, US experience on its terror frontier suggests that the redemptive

6686_Enemark.indd 87

02/03/21 12:22 PM

88 / Christian Enemark

violence of personality strikes is itself in need of redemption. Without recourse to genuine legalism, punitive drone use against suspected terrorists will only ever amount to wild justice, and the associated moral problem of arbitrary killings will remain unresolved. If a simple prohibition on this form of drone violence is unlikely to succeed internationally, an alternative worth trying is to consign that violence to a state’s own courts. In operating according to strict rules of legal procedure, a drone-using state is likely to find that judicial killings are less expedient and far more difficult to bring about than extrajudicial ones. However, the countervailing advantage of this non-wild approach to justice is that it would occasion governments to legitimise drone use for a nonwar purpose. Instead of behaving secretly and trying to pretend that personality strikes count as war, the governments of drone-using states could (with the help of independent courts) conduct such strikes openly and carefully as permissible acts of punitive law enforcement.

Notes Research for this chapter was conducted as part of a project funded by the European Research Council under the European Union’s Horizon 2020 research and innovation programme (grant no. 771082): ‘Emergent Ethics of Drone Violence: Toward a Comprehensive Governance Framework’ (DRONETHICS). 1. This viewpoint counters the claim that, in the US-led ‘war on terror’, the battlefield can be anywhere and, therefore, potentially everywhere. 2. The ‘Detroit bomber’ is Umar Farouk Abdulmutallab, a Nigerian man convicted of attempting to detonate explosives hidden in his underwear while on board a flight from Amsterdam to Detroit, Michigan, in late 2009. 3. The word ‘playbook’, used in the context of American football, refers to a document containing descriptions of the different offensive and defensive plays used by a football team.

References ABC. 2010. ‘“This Week” Transcript: Panetta’, ABC News, 27 June, https://abcnews. go.com/ThisWeek/week-transcript-panetta/story?id=11025299 Ackerman, Spencer. 2016. ‘“Drones Playbook” Shows Key Role Played by White House Staff in Deadly Strikes’, The Guardian, 6 August, https://www.theguardian.com/ world/2016/aug/06/drones-playbook-white-house-nsc-obama-clinton-trump Alston, Philip. 2010. Report of the Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions (A/HRC/14/24/Add.6). New York: Human Rights Council, United Nations General Assembly. Bass, Gary J. 2000. Stay the Hand of Vengeance: The Politics of War Crimes Tribunals. Princeton: Princeton University Press. BBC. 2002. ‘CIA “Killed al-Qaeda Suspects” in Yemen’, BBC News, 5 November, http://news.bbc.co.uk/2/hi/2402479.stm

6686_Enemark.indd 88

02/03/21 12:22 PM

Drone Violence as Wild Justice / 89

BBC. 2019. ‘ Why is Libya So Lawless?’, BBC News, 10 April, https://www.bbc.co.uk/ news/world-africa-24472322 Becker, Jo, and Scott Shane. 2012. ‘Secret “Kill List” Proves a Test of Obama’s Principles and Will’, New York Times, 29 May, A1. Boyle, Michael. 2013. ‘ The Costs and Consequences of Drone Warfare’, International Affairs 89 (1): 1–29. Brennan, John O. 2012. ‘ The Efficacy and Ethics of U.S. Counterterrorism Strategy’, Woodrow Wilson International Center for Scholars, 30 April, https://www.wilsoncenter.org/event/the-efficacy-and-ethics-us-counterterrorism-strategy Brookman-Byrne, Max. 2017. ‘Drone Use “Outside Areas of Active Hostilities”: An Examination of the Legal Paradigms Governing US Covert Remote Strikes’, Netherlands International Law Review 64: 3–41. Browne, Ryan, and Barbara Starr. 2019. ‘ Trump: US Military Killed Terrorist Behind USS Cole Bombing’, CNN, 6 January, https://edition.cnn.com/2019/01/04/ politics/uss-cole-al-badawi-killed/index.html Brunstetter, Daniel, and Megan Braun. 2011. ‘ The Implications of Drones on the Just War Tradition’, Ethics & International Affairs 25 (3): 337–58. Bush, George W. 2004. ‘State of the Union Address’, The White House, President George W. Bush, 20 January, https://georgewbush-whitehouse.archives.gov/news/ releases/2004/01/20040120-7.html Chamayou, Grégoire. 2015. Drone Theory, trans. Janet Lloyd. London: Penguin. Conot, Robert E. 1993. Justice at Nuremburg. New York: Carroll & Graf. Crawley, Vince, and Amy Svitak. 2002. ‘Is Predator the Future of Warfare?’, Defense News, 11–17 November, 8. Cronin, Bruce. 2018. Bugsplat: The Politics of Collateral Damage in Western Armed Conflicts. New York: Oxford University Press. Currier, Cora. 2016. ‘White House Finally Releases its “Playbook” for Killing and Capturing Terror Suspects’, The Intercept, 6 August, https://theintercept.com/2016/08/06/ white-house-finally-releases-its-playbook-for-killing-and-capturing-terror-suspects Enemark, Christian. 2014. ‘Drones, Risk, and Perpetual Force’, Ethics & International Affairs 28 (3): 365–81. Epps, Garrett. 2013. ‘ Why a Secret Court Won’t Solve the Drone-Strike Problem’, The Atlantic, 16 February, https://www.theatlantic.com/politics/archive/2013/02/ why-a-secret-court-wont-solve-the-drone-strike-problem/273246 Estleman, Loren D. 2018. Wild Justice. New York: Forge Books. European Parliament. 2014. ‘European Parliament Resolution of 27 February 2014 on the Use of Armed Drones’ (2014/2567(RSP)), EUR-Lex, https://eur-lex. europa.eu/legal-content/EN/TXT/?uri=CELEX%3A52014IP0172 Gibson, Jennifer. 2019. ‘ We’re Quickly Moving Toward a World Where Drone Executions Are the Norm’, Los Angeles Times, 13 November, https://www.latimes. com/opinion/story/2019-11-13/drone-killings-war-syria-turkey Guoira, Amos, and Jeffrey Brand. 2015. ‘ The Establishment of a Drone Court: A Necessary Restraint on Executive Power’, in Legitimacy and Drones: Investigating the Legality, Morality and Efficacy of UCAVs, edited by Steven J. Barela. Farnham: Ashgate, 323–58.

6686_Enemark.indd 89

02/03/21 12:22 PM

90 / Christian Enemark

Gusterson, Hugh. 2016. Drone: Remote Control Warfare. Cambridge, MA: MIT Press. Hoyle, Craig, and Andrew Koch. 2002. ‘ Yemen Drone Strike: Just the Start?’, Janes Defence Weekly, 13 November, 3. ICCPR. 1966. International Covenant on Civil and Political Rights. Adopted and opened for Signature, Ratification and Accession by General Assembly Resolution 2200A (XXI) of 16 December 1966, Office of the High Commissioner for Human Rights, United Nations, https://www.ohchr.org/en/professionalinterest/ pages/ccpr.aspx Jaffe, Greg, and Karen DeYoung. 2017. ‘ Trump Administration Reviewing Ways to Make it Easier to Launch Drone Strikes’, Washington Post, 13 March, https://www. washingtonpost.com/world/national-security/trump-administration-reviewingways-to-make-it-easier-to-launch-drone-strikes/2017/03/13/ac39ced0-07f811e7-b77c-0047d15a24e0_story.html Jaffer, Jameel. 2013. ‘ The Justice Department’s White Paper on Targeted Killing’, American Civil Liberties Union, 4 February, https://www.aclu.org/blog/nationalsecurity/targeted-killing/justice-departments-white-paper-targeted-killing Johnson, Boris. 2018. ‘ Why We Should Chuck Chequers’, The Spectator, 28 July, https://www.spectator.co.uk/2018/07/boris-johnson-why-we-should-chuckchequers Kahn, Paul. 2002. ‘ The Paradox of Riskless Warfare’, Philosophy and Public Policy Quarterly 22 (3): 2–8. Katyal, Neal K. 2013. ‘Who Will Mind the Drones?’, New York Times, 21 February, A27. Killmister, Suzy. 2008. ‘Remote Weaponry: The Ethical Implications’, Journal of Applied Philosophy 25 (2): 121–33. Kim, Jonathan. 2017. ‘Fifth Amendment’, last edited June, Cornell Law School, https://www.law.cornell.edu/wex/fifth_amendment (accessed 13 March 2020). Klaidman, Daniel. 2012. Kill or Capture: The War on Terror and the Soul of the Obama Presidency. Boston: Houghton Mifflin Harcourt. Lee, Steven P. 2015. ‘Human Rights and Drone “Warfare”’, Peace Review 27: 432–9. McCaleb, Ian C. 2001. ‘Bush Vows Justice Will Be Done’, CNN, 21 September, http://edition.cnn.com/2001/US/09/21/gen.president.speech McMahan, Jeff. 2012. ‘Targeted Killing: Murder, Combat or Law Enforcement?’, in Targeted Killings: Law and Morality in an Asymmetrical World, edited by Claire Finkelstein, Jens D. Ohlin and Andrew Altman. Oxford: Oxford University Press, 135–55. McNeal, Gregory. 2014. ‘Targeted Killing and Accountability’, Georgetown Law Journal 102: 681–794. Marks, Paula M. 1989. And Die in the West: The Story of the O.K. Corral Gunfight. New York: Morrow. Maynard, Katherine, Jarod Kearney and James Guimond. 2010. Revenge versus Legality: Wild Justice from Balzac to Clint Eastwood and Abu Ghraib. Abingdon and New York: Birkbeck Law Press. Mazzetti, Mark, and Declan Walsh. 2013. ‘Pakistan Says US Drone Killed Taliban Leader’, New York Times, 30 May, A1. Mazzetti, Mark, Eric Schmitt and Robert F. Worth. 2011. ‘C.I.A. Strike Kills U.S.-Born Militant in a Car in Yemen’, New York Times, 1 October, A1.

6686_Enemark.indd 90

02/03/21 12:22 PM

Drone Violence as Wild Justice / 91

Meloni, Chantal. 2016. ‘State and Individual Responsibility for Targeted Killings by Drones’, in Drones and Responsibility: Legal, Philosophical, and Sociotechnical Perspectives on Remotely Controlled Weapons, edited by Ezio Di Nucci and Filippo Santoni de Sio. London and New York: Routledge, 47–64. Motlagh, Jason. 2005. ‘U.S. Takes Terror Fight to Africa’s “Wild West”’, SFGate, 27 December, https://www.sfgate.com/politics/article/U-S-takes-terror-fight-to-Africa-sWild-West-2555454.php Obama, Barack. 2013. ‘Remarks by the President at the National Defense University’, The White House, President Barack Obama, 23 May, https://obamawhitehouse. archives.gov/the-press-office/2013/05/23/remarks-president-national-defenseuniversity Parliamentary Assembly. 2015. ‘Drones and targeted Killings: The Need to Uphold Human Rights and International Law’ (Resolution 2051(2015)), 23 April, Parliamentary Assembly of the Council of Europe, http://assembly.coe.int/nw/xml/XRef/ Xref-DocDetails-EN.asp?FileID=21746&lang=EN Ramsden, Michael. 2011. ‘ Targeted Killings and International Human Rights Law: The Case of Anwar Al-Awlaki’, Journal of Conflict & Security Law 16 (2): 385–406. Reuters. 2008. ‘A Wild Frontier’, The Economist, 18 September, https://www.economist. com/asia/2008/09/18/a-wild-frontier Reuters. 2009. ‘Obama: We Took Out Pakistani Taliban Chief’, ABC News (Australia), 22 August, https://www.abc.net.au/news/2009-08-21/obama-we-took-outpakistani-taliban-chief/1399370 Schaller, Christian. 2015. ‘Using Force Against Terrorists “Outside Areas of Active Hostilities”: The Obama Approach and the Bin Laden Raid Revisited’, Journal of Conflict & Security Law 20 (2): 195–227. Schulzke, Marcus. 2017. The Morality of Drone Warfare and the Politics of Regulation. London: Palgrave Macmillan. Shane, Scott. 2011. ‘Judging a Long, Deadly Reach’, New York Times, 1 October, A1. Shane, Scott. 2013. ‘A Court to Vet Kill Lists’, New York Times, 9 February, A1. Stohl, Rachel. 2018. An Action Plan on U.S. Drone Policy: Recommendations for the Trump Administration. Washington, DC: Stimson Center. Strawser, Bradley J. 2010. ‘Moral Predators: The Duty to Employ Uninhabited Aerial Vehicles’, Journal of Military Ethics 9 (4): 342–68. Trump, Donald. 2017. ‘Remarks by President Trump on the Strategy in Afghanistan and South Asia’, The White House, 21 August, https://www.whitehouse.gov/ briefings-statements/remarks-president-trump-strategy-afghanistan-south-asia UNGA. 2013. ‘Protection of Human Rights and Fundamental Freedoms While Countering Terrorism’ (A/RES/68/178), New York: United Nations General Assembly, 18 December. UNHRC. 2014. ‘Ensuring Use of Remotely Piloted Aircraft or Armed Drones in Counterterrorism and Military Operations in Accordance with International Law, Including International Human Rights and Humanitarian Law’ (A/HRC/ RES/25/22), Human Rights Council, New York: UN General Assembly, 15 April. Warren, Aiden, and Ingvild Bode. 2015. ‘Altering the Playing Field: The U.S. Redefinition of the Use-of-Force’, Contemporary Security Policy 36 (2): 174–99.

6686_Enemark.indd 91

02/03/21 12:22 PM

92 / Christian Enemark

Whetham, David. 2013. ‘Killer Drones’, RUSI Journal 158 (3): 22–32. White House. 2001. ‘ The Vice President Appears on Meet the Press with Tim Russert’, The White House, President George W. Bush, 16 September, https:// georgewbush-whitehouse.archives.gov/vicepresident/news-speeches/ speeches/vp20010916.html Woods, Chris. 2015. Sudden Justice: America’s Secret Drone Wars. London: Hurst. Wuschka, Sebastian. 2011. ‘ The Use of Combat Drones in Current Conflicts: A Legal Issue or a Political Problem?’, Goettingen Journal of International Law 3: 891–905.

6686_Enemark.indd 92

02/03/21 12:22 PM

FIVE

‘A New Departure’: Britain’s Lethal Drone Policy and the Range of Justice Christopher J. Fuller

Introduction On 21 August 2015, Prime Minister David Cameron authorised the targeted killing of Reyaad Khan, a British citizen and member of the Islamic State of Iraq and Syria (ISIS), by drone strike (BBC 2015). The attack – described by Cameron as ‘a new departure’ for the UK – marked a significant shift in Britain’s lethal drone policy and the state’s late adoption of the controversial interpretation of international law which the United States has employed since launching its ‘war on terror’ in 2001. While never openly opposing the US decision to regard acts of terrorism and the counterterrorism response through a war paradigm, the UK government had, until this strike, treated terrorism as a criminal activity, with responsibility for countering the threat being the domain of the civilian security services. But in deploying the Royal Air Force (RAF) to eliminate a terror suspect, the government set a new precedent in its use of drone violence. Despite the significance of this departure, the government has been purposefully vague in clarifying the detail of its policy, the legal basis which underwrites it, the decision-making process and the accountability mechanisms which exist for strikes. In the absence of a detailed memorandum on the government’s formal policy position, this chapter explores the evolution of the UK’s lethal drone policy through ministers’ statements and government policy documents in order to establish a clearer picture of the ethos which has driven the government’s adoption of lethal drones for counterterrorism. In doing so it reveals that Britain’s drone policy is the product of a very particular set of attitudes towards self-defence, justice and imminence – attitudes which align very closely with the current dominant philosophy

6686_Enemark.indd 93

02/03/21 12:22 PM

94 / Christopher J. Fuller

of American policymakers on how best to respond to emergent terrorist threats. This overlap makes Britain’s policy less of a new departure and more of an adoption of an already well-established form of drone violence.

Self-Defence and the ‘Unable or Unwilling’ Test As important as the technology of remotely piloted aircraft has been in enabling governments to launch strikes such as that executed against Khan, it is the evolving policy which underwrites their use which really empowers drone violence. The United States’ deployment of drones for targeted killing is backed by a complex legal architecture which has evolved over the past four decades, with the initial authorisations dating back to the Reagan administration, when CIA chief counsel Stanley Sporkin argued preemptive strikes against terrorists would be lawful acts of self-defence under Article 51 of the United Nations (UN) Charter. Following the 9/11 attacks the US employed this legal logic, supported by UN Security Council Resolution 1368, Congress’s broad Authorization to Use Military Force Against Terrorists (the ‘AUMF’ of 2001), and a Finding from President George W. Bush to legally sanction America’s drone campaign by classing its counterterrorism actions as part of an ongoing armed conflict against non-state terrorist groups (Fuller 2017, 44). Two decades after the United States adopted its expansive interpretation of self-defence under international law, the UK government adopted a similar legal position to validate its actions. Following consultation with the Attorney General, Cameron cited Article 51 in his justification of the strike against Khan, which he argued was a lawful act of national self-defence against a member of a known terrorist group, who intelligence indicated was involved in the ongoing organisation of terror threats against the British public (Cameron 2015a). Yet, while the UK government’s legal case for the strike may have been similar to that employed by the US, the prime minister held considerably less domestic legal authority to execute such actions than did the American president. First, while the US drone campaign continues to enjoy broad bipartisan support, at the time of the strike the UK Parliament opposed military action in Syria (Chesney 2013). Second, while the American government has faced criticism over the lack of transparency, each US drone strike is briefed to the relevant Congressional oversight committees, providing – in theory at least – the important judgement of an authoritative external entity to validate the executive’s actions (Obama 2013). The strike against Khan was undertaken with no such oversight. Finally, while the US has declared itself engaged in an ongoing state of global armed conflict with extremist groups, at the time of this strike

6686_Enemark.indd 94

02/03/21 12:22 PM

‘A New Departure’: Britain’s Lethal Drone Policy / 95

the UK had made no such assertion. Resultantly, in its analysis of the UK’s use of lethal drones for counterterrorism, Parliament’s Joint Committee on Human Rights (JCHR) criticised the government for failing to clarify precisely what the legal basis of its drone policy is (JCHR 2016, 15). In striking Khan, was the government defending the state, or enforcing the law? Did the prime minister authorise the killing of an enemy, or the punishment of a criminal? Establishing the correct paradigm is vital in ensuring the state has access to the necessary tools to protect itself, while ensuring it does not overextend its authority, and thereby imperil its citizens’ human rights. This balancing act has been further complicated by the extent to which the increasingly well-resourced, organised and aggressive jihadi groups have succeeded in blurring the lines between war (in a traditional sense) and criminal acts of terrorism. The public’s perception also plays an important role in determining how a government responds to terrorism. Yet, while politicians and the media still aim to shape these views, the establishment no longer holds the monopoly over the flow of information it once did. Online tools have empowered terrorists to spread fear and anxiety among the citizens, influencing the populace and limiting the government’s policy options (Stern and Berger 2015, 130–8). US President Barack Obama, for example, sought to downplay the rapid rise of ISIS with the uncharacteristically flippant analogy describing the group as the ‘jayvee team’ to al-Qaeda (Remnick 2014).1 Such language was at odds with the American public’s perception, which, skilfully manipulated by the group’s propaganda machine, saw ISIS as a significant domestic threat. Facing a backlash, the Obama administration was forced to engage in a publicity campaign to promote its counter-ISIS strategy and to reassure the public that the appropriate level of force was being applied to secure the homeland (Obama 2014). Concurrently, should the public regard the terror threat as an essentially criminal enterprise to which the government responds with oppressive measures, that government risks being seen as unduly authoritarian. The backlash against the UK government’s Investigatory Powers Act (nicknamed the ‘Snooper’s Charter’) and protests in the United States against Enhanced Interrogation Techniques serve as warnings to governments about the dangers of overstepping the mark in popular perceptions. Distinguishing enemy from criminal is not just a matter of the state’s national security, but also one of its political and social wellbeing. Disagreements over the nature and identity of the enemy can pull at the fabric of a nation. Should a government label too many as enemies, the state will find itself in a perpetual state of war, draining resources and domestic support. But a failure to identify the enemies a government does

6686_Enemark.indd 95

02/03/21 12:22 PM

96 / Christopher J. Fuller

have leaves the nation overly dependent upon its internal law enforcement mechanisms. Fiercely secular, with simmering tensions between disadvantaged portions of its Muslim community and an active far right (Scott 2007, 40), France was singled out by ISIS in 2014 as a prime target (Weiss 2014). With ISIS operating from safe havens across Iraq and Syria, with access to hundreds of highly motivated European fighters coordinated by a unit dedicated to exporting terror attacks, the UN Security Council classed the threat posed by the group as ‘unprecedented’ (Callimachi 2016; UNSC 2015b). Yet in 2015 the French government was continuing to dedicate just £176 million per year to its security services, relying heavily upon crime fighting structures at a European level – institutions which, given the nature of the threat, were repeatedly found wanting. By contrast, during the same period the United Kingdom spent approximately £1.9 billion per year on its security services – the domestically focused MI5, the foreign service of MI6 and the surveillance operations of GCHQ. This much greater investment provided additional intelligence gathering, monitoring and intervention capabilities to the UK’s domestic law enforcement agencies, adding a significant layer of additional security (Jones and Chassany 2015). Finally, should a nation’s enemies manifest within the country’s borders, a heavy-handed response risks shattering social cohesion, leading to alienation, resentment and internal conflict – something ISIS specifically seek to provoke. In France, the tensions stirred up by the group’s assaults combined virulently with more deep-rooted divisions among Muslim and non-Muslim communities, leading Patrick Calvar (head of France’s domestic intelligence service) to warn in July 2016 that the state had been driven to ‘the brink of civil war’ between right-wing extremists and Islamists (Burke 2016). The strong performance of the right-wing and anti-Islamic presidential candidate Marine Le Pen in France’s 2017 presidential election served to highlight the risks to society posed by such tensions. The risks and challenges associated with identifying the correct response to an ever-evolving terror threat helps explain why the UK government has sought to maintain flexibility over how it applies its drone policy. For the JCHR, however, this deliberate ambiguity risks the erosion of international legal norms. The committee argued in their April 2016 report on the UK’s use of targeted killing that the country’s transparent compliance with the rule of law is vital to its ability to influence fellow nations to comply with other aspects of the global order. Furthermore, they cautioned the government that its failure to formally identify who can be targeted, where, when and why, risked the UK being drawn into the sort of expansive global conflict the US has fought with limited success since 2001. More directly,

6686_Enemark.indd 96

02/03/21 12:22 PM

‘A New Departure’: Britain’s Lethal Drone Policy / 97

committee members warned of the potential legal exposure such ambiguity risked: ‘ We owe it to all those involved in the chain of command for such uses of lethal force’, the members espoused, ‘to provide them with absolute clarity about the circumstances in which they will have a defence against any future criminal prosecutions, including those which might originate from outside the UK’ (JCHR 2016, 25). In spite of such pressure, the government rejected demands for greater transparency on drone strikes (Airwars 2016, 19; Brookman-Byrne 2018, 6). Concerns over the potential exploitation of the nation’s restraint, if the government were to expose its precise thresholds for the use of lethal force, were probably factors in the decision. Perhaps most significant, however, is the fact that policymakers are doubtless reluctant to formally release something as controversial as a targeted killing policy, which would place them and the service personnel involved dangerously close to endorsing what appear to be state-sponsored assassinations (as distinct from acts of war). Despite the government’s refusal to publish a detailed memorandum, careful examination of senior ministers’ public statements and related policy documents enables a clearer picture of the legal basis of the UK’s drone policy to be formed. The first revelation came during the September 2014 Commons debate over authorisation to launch airstrikes against the ISIS forces which were then overrunning Iraqi territory. The debate was triggered by a direct appeal for support from the government of Iraq, with MPs voting overwhelmingly in favour of adding UK airpower to the US-led coalition opposing the self-declared caliphate (BBC 2014). While the UK’s complicity in creating the conditions that enabled ISIS’s emergence in the first place provided motivation for many of the MPs who voted in favour of the airstrikes, Britain’s interventionist past had a wider impact upon the tone and content of the debate. With the bloody spectre of Britain’s 2003 intervention in the same country, as well as the instability wrought by the UK’s role in the violent overthrow of Libya’s Muammar Gaddafi in 2011, MPs from across the political spectrum raised concerns over the nation being dragged into the brutal Syrian civil war. Accordingly, the majority of House of Commons members made clear their vote was to authorise action strictly limited to the defence of the Iraqi government within its own sovereign borders, not a broader campaign to destroy ISIS’s strongholds in Syria (Hansard 2015). During the debate, however, Cameron revealed the UK government’s willingness, under certain conditions such as the need to move swiftly to avert a humanitarian catastrophe, to expand actions into neighbouring Syria without Parliamentary authorisation (Hansard 2015). While it is an exaggeration to suggest the risk posed by Khan’s plotting constituted a potential humanitarian catastrophe, the prime minister’s authorisation of the strike

6686_Enemark.indd 97

02/03/21 12:22 PM

98 / Christopher J. Fuller

against him was consistent with the government’s stated position that it had a legal basis to act over the Iraq–Syria border without a Commons vote if speed of action was believed to be of the essence to save lives. The prime minister further articulated the UK’s new approach to lethal drone use when later informing Parliament of the operation which killed Khan. Cameron (2015b) explained that the policy was a response to the increased threat of ISIS attacks, and was intended to ‘prevent and disrupt plots against [the UK] at every stage’. ‘As part of this counter-terrorism strategy’, he continued, ‘if there is a direct threat to the British people and we are able to stop it by taking immediate action, then, as Prime Minister, I will always be prepared to take that action’ (Cameron 2015b). In making this statement Cameron (2015b) clarified that the operation ‘was not part of coalition military action against ISIS in Syria’ – something which Parliament had not authorised – but was instead ‘a targeted strike to deal with a clear, credible and specific terrorist threat to our country at home’. Defining the scope of this new policy, he stated unequivocally that as prime minister he could, and would, authorise further strikes as and when similar dangers were detected ‘whether the threat is emanating from Libya, from Syria or from anywhere else’ (Cameron 2015b). This bold assertion was promptly undermined, however, by the government’s own permanent representative to the United Nations, Matthew Rycroft, in a letter sent to the UN Security Council that same day. In the brief communication Rycroft specified that the strike was conducted not only in strict self-defence but also as part of the already-authorised collective defence of Iraq (UNSC 2015a). While theoretically there was nothing inherently contradictory in the government drawing upon both individual and collective self-defence as justification for its drone strike in Syria, Rycroft’s position was at odds with Cameron’s claim that the strike was conducted as part of a new counterterrorism approach, independent of the defence of Iraq. Although probably attributable to little more than a miscommunication between Whitehall departments, the discord fuelled uncertainty over exactly what the legal underpinnings of the action against Khan were. The confusion over the government’s approach was to some extent cleared up the following day by the Secretary for Defence, Michael Fallon. Speaking in an interview on BBC Radio 4’s Today programme, Fallon reaffirmed the prime minister’s position that the government had the right to employ lethal drone strikes without Parliamentary authorisation. Reporting that security services had identified at least three other terrorists involved in plots against the UK, Fallon stated that, should these attacks look to be coming to fruition, and if the government had no other way of preventing them, it ‘wouldn’t hesitate to take similar action again’ (Perraudin et al.

6686_Enemark.indd 98

02/03/21 12:22 PM

‘A New Departure’: Britain’s Lethal Drone Policy / 99

2016). Four months later, however, the government’s exact policy position was once more clouded when Fallon was asked, during testimony to the JCHR, to clarify the UK’s policy on targeted killing outside areas of conflict. The minister initially insisted that there was no such policy, before promptly offering a contradictory answer when questioned about how the government would react to future threats emanating from ‘other areas of ungoverned space’. Fallon provided a hypothetical example in which, should a plot emerge from Libya – a state he described as lacking ‘political authority’ but not an area of armed conflict in which the UK was authorised to use military force – the government considered itself justified in launching drone strikes there (Fallon 2015). This double-speak, in which the government maintained that it did not have a targeted killing policy, but at the same time extolled its legal right to undertake such strikes, raised further criticism from the JCHR and led to the committee’s demand for further explanation from the government (JCHR 2016, 59–60). In January 2017 the Attorney General, Jeremy Wright, finally provided a more consistent account of the government’s drone policy when he delivered a public address exploring the evolving nature of self-defence in international law. As the government’s chief lawyer explained, the invitation of the sovereign Iraqi government, coupled with the UN Security Council’s identification of ISIS as an unprecedented threat to international security, provided jus ad bellum (the right to resort to war), thus granting the government the legal right to engage in a non-international armed conflict (NIAC) with the self-declared caliphate (Wright 2017). Thus, while in principle Parliament’s 2014 vote specifically authorised airstrikes against ISIS in Iraq, once the government entered a NIAC the geographical scope of its use of force was not, under international law, confined to the territory of a single state. Under the Laws of Armed Conflict (LOAC) which govern NIACs, the decision over who can be targeted is determined not purely by a potential target’s location, but by their membership of an opposing party in the conflict. Thus, within this context Khan was legitimately targeted under international law as a member of ISIS’s armed forces who were directly engaged in an armed conflict with the UK (Wright 2017). Furthermore, as the legal scholar Aurel Sari explained upon reviewing Wright’s comments, the LOAC do not actually require that lethal targeting of a member of an opposing party in a NIAC be undertaken as a last resort. On the contrary, status-based targeting is permissible as a measure of first resort (Sari cited in Cole 2017). Despite the relative freedom to employ lethal force granted by the UK’s engagement in a NIAC with ISIS, the Attorney General sought to clarify that the government’s drone policy did not reflect the adoption of a ‘Global War

6686_Enemark.indd 99

02/03/21 12:22 PM

100 / Christopher J. Fuller

on Terror paradigm’ (Wright 2017, 19). Instead, Wright stressed that the use of force remained a ‘last resort’, with criminal law enforcement continuing to be the preferred approach (Wright 2017, 1). What, then, made the strike against Khan and the hypothetical strikes in Libya and elsewhere acts of last resort? Here the Attorney General leaned upon the UK’s acceptance of the ‘unable or unwilling’ test – an aspect of international law which became particularly relevant to counterterrorism policy following 9/11, when its aggressive application was championed by the United States (Deeks 2012). Under the conditions of this test, lethal force can be applied in lieu of criminal law enforcement measures when attacks are planned from outside the targeted country’s territory, and where the host state is unable or unwilling to act. In the case of the Khan strike, the government assessed that the Syrian regime lacked ‘effective control of the relevant part of the territory’, and had no distinct motivation to prevent Khan’s threatening actions (Wright 2017, 10–11). Thus, with Syrian law enforcement agents unable or unwilling to act, and UK security services unable to reach Khan to effect an arrest on the streets of the ISIS capital, Raqqa, David Cameron justified the drone strike as the ‘only feasible means of effectively disrupting the attacks planned and directed by this individual’ (Cameron 2015a). The government’s judgement of the Syrian regime as unable or unwilling to act was vindicated four months later. With the security situation in the region worsening, and the consequences being felt across Europe in the form of the burgeoning refugee crisis, and following the bloody terrorist assault on the streets of Paris in November 2015, Parliament voted to expand British airstrikes against ISIS across the porous Iraq–Syria border (Lang 2015).

Hobbes’s State of Nature and the Drone Domain The UK government’s National Security Strategy and Strategic Defence and Security Review (NSS 2015), published in November 2015, provided further clarification of the government’s drone policy. First, the review confirmed the adoption of armed drones as the government’s counterterrorism tool of choice by announcing its intention to double the size of the RAF’s fleet through the procurement of twenty Protector aircraft from the US-based aeronautics manufacturer General Atomics. The successor to the Reaper, this aircraft’s moniker reveals both a subtle concession to some of the negativity associated with the expanded use of Predators and Reapers, and an acknowledgement of the self-defence-oriented argument utilised by the US and UK to justify their drone programmes. The review document also provided important clarification – despite employing vague language similar to that used by ministers – as to the geographical scope of the

6686_Enemark.indd 100

02/03/21 12:22 PM

‘A New Departure’: Britain’s Lethal Drone Policy / 101

policy, and where the ‘unable or unwilling’ test would be applied. Noting that ‘ Terrorism thrives in unstable states and amid civil wars’ – a reference to the rise of ISIS in Syria, as well as other safe havens exploited by associated jihadist groups such as Libya and Somalia – the strategy declared that the UK would maintain a capability to ‘find and track terrorists in these hostile environments’. Where it deemed that these terrorists posed ‘an imminent threat to the UK, British interests abroad or to our allies’, but with ‘no way to bring them to justice’, the UK government clarified its intention to ‘act decisively’ through its ‘global strike capability’ (NSS 2015, 38). While ‘act decisively’ is a broad euphemism, the document leaves no doubt that, within the context of the UK’s strategic approach, the government signalled its willingness to employ lethal force via drones against enemies deemed to be beyond the reach of justice. In attaching a spatial qualifier to justice, and then distinguishing terrorists as being either within its reach or beyond it, the government revealed its belief in two different kinds of domain: one of law and order; and another, brutal, violent and lawless domain, akin to Thomas Hobbes’s state of nature (Hobbes 1996, 89). What does the UK government’s differentiation between a domain of justice and unstable states beyond that space mean? First, it reveals the importance of borders, and political authority within those borders, in British policymakers’ thinking. For the justice paradigm to exist, there must be a bordered and sovereign territory, within which a social contract exists between the government and its citizens. That contract, as the sociologist Max Weber defined it, ensures the government has a monopoly on the legitimate use of violence, and gives it the authority to enforce (sometimes violently) domestic laws over its citizens (Weber 1991). Those within the borders who fail to comply with this contract are labelled criminals and dealt with through the domestic criminal justice system. Beyond the UK’s borders lie its supranational boundaries. These are constructed from membership-based multilateral and global institutions, such as the UN, NATO and the International Criminal Court (ICC). These extended borders emerged from a vision which first manifested itself in the late nineteenth-century idea of peace through law, where states were bound by a legal regime of communication and trade, such as via the 1874 Universal Postal Declaration and the 1899 Hague Convention for the Settlement of International Disputes. In the aftermath of the Second World War, the founding charter of the UN sought to end wars and protect peace through the enshrinement of state sovereignty and the enforcement of individual accountability under a global legal regime, with economic sanctions, peacekeeping actions, and criminal prosecution of leaders who illegally deployed force.

6686_Enemark.indd 101

02/03/21 12:22 PM

102 / Christopher J. Fuller

The ultimate evolution of supranational expansion is the project of human rights law, a universal social contract which transcends state borders. As Paul Kahn (2013, 206–7) has observed, the movement from the League [of Nations] to the [UN] Charter to the International Criminal Court is a single story of the progressive realisation of a global legal order in which the idea of an enemy who is not a criminal ultimately has no place.

Thus, the proponents of this supranational domain believe that wars should be followed by courtroom trials, and warfare should be permitted only as an extension of law enforcement, placing all acts of terrorism firmly within the crime paradigm (Kahn 2013, 206). In his essay ‘ The New Liberal Imperialism’, British scholar-diplomat Robert Cooper labelled the move towards a universal legal regime the ‘postmodern state’ (Cooper 2002). He contended that it was in fact the European Union (EU) leading the world into this new postmodern age (Cooper 1996). The UN Charter’s twentieth-century focus upon state sovereignty no longer aligned with the realities of the post-Cold War world order, in which, Cooper argued, traditional national interests and power politics needed to give way to international law, supranational institutions and pooled sovereignty. Only in this way could the nationalistic, cultural and ethnic divisions that had plagued humankind, in particular Europe over the past century, be dissolved in favour of shared values and combined economic interests. Cooper envisioned the Union would continue to expand as a kind of voluntary, co-operative empire, dedicated to liberty and democracy (Cooper 1996, 37). Although the key architect of the UN and a historical supporter of the EU, the United States has always harboured an extreme suspicion of placing itself under the authority of supranational institutions, preferring to see itself as the hegemonic leader of the system it exhorts others to accept. Since the Second World War, when, as the orthodox American interpretation of history argues, the US reluctantly stepped into the breach to save the democratic world from self-destruction, a guiding principle of American foreign policy has been that no other power, be it a rival or ally, can be trusted to keep the world safe for democratic principles (Kagan 2008, 51–2). The US may invoke Article 51 of the UN Charter (which acknowledges states’ inherent right to self-defence) to justify its pursuit of armed conflict against extremists. However, one only need consider the US rejection of membership of the ICC, its unilateral application of force against Iraq in 2003, or the creation of Guantánamo Bay’s ‘illegal enemy combatants’ category to bypass the universality of human rights laws to see that the United States deems its

6686_Enemark.indd 102

02/03/21 12:22 PM

‘A New Departure’: Britain’s Lethal Drone Policy / 103

own government a higher authority than any international institution. The UK too has long harboured its own mistrust of the postmodern approach to borders and laws, rejecting aspects of the European project such as the 1985 Schengen Agreement and a shared currency, before ultimately withdrawing from EU membership in January 2020. This shared emphasis upon the individual state sovereignty of their own nations further helps explain why the UK has adopted a drone policy so similar to that of the United States. Despite both countries’ globalised interests, each regards its sovereignty and borders as vital aspects of its national identity. Neither the US nor the UK has fully accepted a shift to postmodern politics. Instead, both have chosen to remain as modern, ‘twentiethcentury’ powers. As such, both states continue to bestow the title of ‘enemy’ upon those who threaten to penetrate their borders and bring violence to the homeland from safe havens abroad, beyond the reach of justice. Sovereignty, and defence of the state from external enemies, remains a vibrant concept among American and British policymakers and citizens alike. Should a threat emerge from a terrorist safe haven – a territory the UK government classes as a ‘hostile environment’ in which the host state is unwilling or unable to prevent the projection of violence from within its own borders – then that state is, in accordance with the UK government’s 2015 security strategy, beyond the reach of justice (NSS 2015, 38). If a state fails to maintain a successful monopoly on violence, it has crossed from the civilised world to what Hobbes characterised as the ‘state of nature’ (Hobbes 1996, 89). Within the state of nature there are no laws, thus no criminals, only potential enemies. Consequently, a threat emerging from the state of nature cannot be countered by a nation’s domestic legal system, or by the international laws of the supranational bodies. Instead, the states which exist beyond justice and the enemies which plot within them become part of the drone domain, policed by unmanned aircraft and subject to the LOAC. It is only the application of such force which can control the brutish and violent state of nature. The stark separation between the domestic realm of law and the Hobbesian drone domain was demonstrated in November 2002 with the killing of Kamal Derwish. Derwish, an American citizen, died in a drone strike authorised by the civilian Central Intelligence Agency (CIA) – the first ever conducted outside an official war zone – while travelling with a group of al-Qaeda fighters across the desert of Yemen. ‘ The only way to treat them’, President George W. Bush reasoned bullishly following the strike, ‘is [sic] what they are – international killers’ (Bush 2002). The president’s reference to the international nature of the threat was of particular significance. Derwish had been the ringleader of an al-Qaeda

6686_Enemark.indd 103

02/03/21 12:22 PM

104 / Christopher J. Fuller

sleeper cell in Lackawanna, New York, but had fled to Yemen following the Federal Bureau of Investigation’s (FBI) arrest of his six co-conspirators. While they were tried and sentenced as criminals under US domestic law, Derwish’s getaway to Yemen’s ungoverned region transformed him. He transferred himself from the domestic legal realm to the drone domain, where the LOAC could putatively be applied to protect America’s national security. The Bush administration faced very little backlash to the apparent execution of an American citizen without trial. Despite having been a criminal a month prior to the strike, Derwish’s allegiance to al-Qaeda, his plot against the US homeland and his foreign location qualified him as an enemy (Williams 2013, 42–3; Powell and Priest 2002; Aziz and Lam 2003). The same transformation can be observed with the killing of Anwar al-Awlaki, another American citizen, in 2011. Awlaki had left the US in 2002, and after a brief time in the UK he settled in Yemen in 2004, from where he was linked by the US government to several terror plots. Evidence of his connection to the 2009 Fort Hood shootings, and a failed attempt to bring down an American airliner one month later, saw Awlaki added to the Obama administration’s kill list. Whereas the perpetrators of the plots were arrested on US soil and tried as criminals, Awlaki was killed by drone strike. Once more, the efforts of an American citizen to penetrate the United States’ borders from a domain judged to be beyond the reach of justice was enough to transform him into an enemy combatant, slain by a tool of war (DOJ 2010). The view of a world split between a domestic domain of justice and a separate Hobbesian drone domain also clarifies the thinking behind the UK’s strike against Khan. While domestic ISIS sympathisers have been arrested under the UK’s terrorism legislation, Khan had consciously left the domain of justice, seeking the security offered within the state of nature caused by Syria’s chaotic civil war. In so doing he was transformed from a domestic criminal into an international enemy. Criminals, when caught, are tried and punished through the legal system for their violation of the law, but throughout this process, they have domestic legal rights. They are protected by procedures which fundamentally recognise their status as an individual. In order to maintain its monopoly on the use of violence that Weber’s social contract dictates, the state must justify its actions against the criminal through the basic right to habeas corpus, for example. The enemy, on the other hand, has no such right. They are not recognised as a person, but instead defined by their membership of a specific state, or in the case of the UK’s NIAC, a non-state organisation. When an individual opts to travel to Syria or any other territory regarded as outside the reach of justice, Britain’s drone policy makes clear that in

6686_Enemark.indd 104

02/03/21 12:22 PM

‘A New Departure’: Britain’s Lethal Drone Policy / 105

the government’s view that individual has consciously rejected their right to trial and treatment as a citizen and (alleged) criminal. Importantly for a sustainable counterterrorism policy, the categorisation of foreign-based terrorists posing a threat to the UK as enemies has been supported by the British public. When polled in April 2013 on whether they would support a drone strike to kill a known terrorist overseas to stop an imminent attack upon the UK – a scenario identical to that of the 2015 Khan strike – 75 per cent of poll respondents replied that they would, as long as no innocent civilians would be killed. Even with the risk of two to three civilian casualties, support remained at 64 per cent, and when the casualties rose as high as ten to fifteen civilians, it was still 60 per cent. Key in the public’s calculation, however, is the term ‘imminent’. When this qualification was removed from the polling, support for drone strikes fell significantly: to 67 per cent when no civilian casualties would be incurred, and a mere 32 per cent if ten to fifteen innocents would perish alongside the target (Rogers 2013). Clearly, the timing of strikes under the UK’s drone policy is the final significant factor in need of clarification.

The Shifting Doctrine of Imminence Post-9/11, the Bush administration’s 2002 National Security Strategy established a new American policy of pre-emption against emergent threats. Controversially, the new American doctrine pushed the boundaries of self-defence, authorising the application of lethal force ‘even if uncertainty remain[s] as to the time and place of the enemy attack’ (NSS 2002, 15). This broad interpretation of imminence, combined with the capabilities of armed drones, has locked American policymakers into a ‘whack-a-mole’ cycle, where even the suspicion of a future threat has become enough to see a target added to what has become America’s ever-growing kill list (Maxwell 2012; Cronin 2013). Concerned that the UK’s drone policy could see the country dragged into a similar cycle of violence, the JCHR and a number of non-governmental observers sought confirmation from Downing Street of the exact measure the government had applied to discern that the threat posed by Khan was imminent (Watson 2016). Once again, specific answers were not forthcoming, although statements delivered prior to and following the strike provide some clarification as to what the policy actually is. Critics argued the government had failed to present evidence which proved the danger posed by Khan’s plotting was imminent, citing the legal precedent of the long-established ‘Caroline test’ which defines such a threat as ‘instant, overwhelming, leaving no choice of means, and no moment of deliberation’ (Full Fact 2015). With evidence that the National Security Council autho-

6686_Enemark.indd 105

02/03/21 12:22 PM

106 / Christopher J. Fuller

rised the use of lethal force against Khan three months prior to the strike, it was clear that this traditional measure of imminence was not satisfied (JCHR 2016, 47). However, the conditions established by the nineteenth-century diplomatic correspondence regarding the Caroline incident had already been challenged by the government in 2004, when the Attorney General, Lord Goldsmith, argued in the House of Lords that ‘the concept of what constitutes an “imminent” armed attack’ needed to ‘develop to meet new circumstances and new threats’ (Goldsmith 2004, 370–1). When questioned by the JCHR, Defence Secretary Fallon had sustained this line of argument, stating that in light of modern conditions with threats emerging from difficult-to-monitor safe havens and utilising tools such as encrypted online communication, it was no longer possible to have a ‘hard and fast rule about what you would define “imminent”’. Instead Fallon endorsed a ‘case-by-case’ assessment (JCHR 2016, 46). In January 2016, the Minister of State for the Armed Forces, Penny Mordaunt, responded to the Labour MP David Anderson’s written question on the matter by sharing the Ministry of Defence (MoD) definition of imminence in the application of Article 51 of the UN Charter (state self-defence). Under this interpretation – which Mordaunt stressed had not differed from that of successive UK governments – force could be used ‘in anticipation of an armed attack where such an attack is imminent, provided that such force is both necessary and proportionate to averting the threat’ (Defence 2016). Confirming Fallon’s advocacy for flexibility, the definition added that any assessment ‘would depend on the facts of each case, with consideration likely to include issues such as the nature and immediacy of the threat, the probability of an attack, its scale and effects and whether it can be prevented without force’ (Defence 2016). Despite the government’s efforts, Parliament’s Intelligence and Security Committee, headed by the former Attorney General Dominic Grieve, warned Downing Street in December 2016 that it remained ‘dissatisfied’ with the information made available on its decision to kill Khan (Kerbaj and Shipman 2017). In response, the Attorney General, Jeremy Wright, provided a more detailed explanation of the government’s interpretation of imminence in his January 2017 address. Defending the rejection of the Caroline test as the necessary adaption of international law in the context of the evolving terrorist threat, Wright instead cited a set of principles proposed by Sir Daniel Bethlehem, a former legal adviser to the UK Foreign and Commonwealth Office, in an article published in the American Journal of International Law in 2012. The principles included consideration of the nature and immediacy of the threat; the probability and scale of an attack; whether it formed part of a concerted pattern of continuing armed activity; and the likelihood of other opportunities to undertake effective action

6686_Enemark.indd 106

02/03/21 12:22 PM

‘A New Departure’: Britain’s Lethal Drone Policy / 107

in self-defence. Wright also quoted Bethlehem’s argument that (Bethlehem 2012, 775): [t]he absence of specific evidence of where an attack will take place or of the precise nature of an attack does not preclude a conclusion that an armed attack is imminent for purposes of the exercise of a right of self-defense, provided that there is a reasonable and objective basis for concluding that an armed attack is imminent.

The same article had been endorsed by Brian Egan, the Legal Adviser to the US State Department, in April 2016, and as Wright flagged in his speech, the principles had been discussed at a meeting of the Quintet of Attorneys General (US, UK, Canada, Australia and New Zealand) in Washington, DC, where the principle was broadly agreed upon, effectively creating the first consensus around the reformed interpretation of imminence applied by the UK government (Wright 2017, 18). In listing the criteria considered when assessing the imminence of a threat, Wright set out the legal substance behind the government’s calculations in more explicit detail than ever before. At its core, however, remained the same argument that the UK government had proposed in 2004, and that the US government had set out in 2002. Gone were longestablished ideas about time or place of combat, or even specific timescales to attacks. In their place was the notion that the advantage the terrorist holds is their capacity to threaten anyone, at any place, at any time. Terrorists seek to exploit the tactical benefits of invisibility, fluidity and surprise as against the highly accessible institutions of the modern state. For the US and UK governments, which seek to adopt a proactive counterterrorism strategy, the task is to locate these terrorists and eliminate them before their plots can materialise. It is no coincidence, therefore, that a consensus over the new measure of imminence was reached first between the US, UK, Australia, New Zealand and Canada. They are the members of Five Eyes, the world’s most expansive and penetrative intelligence alliance, which provides unprecedented levels of information sharing and cooperation. Visibility, as Michel Foucault argued in his analysis of Jeremy Bentham’s panopticon, is power, and global surveillance has evolved into a vital aspect of both state governments’ desire and ability to identify and locate targets which operate beyond the reach of justice. As the former Bush administration lawyer John Yoo argued when reflecting upon the American government’s post-9/11 approach, ‘the best defense will be available only during a small window of opportunity when terrorist[s] become visible’ (Yoo 2012, 73). Cameron revealed his government’s agreement

6686_Enemark.indd 107

02/03/21 12:22 PM

108 / Christopher J. Fuller

with this position when he confirmed the UK would utilise drone strikes in self-defence ‘should the right opportunity arise’, thus confirming that imminence is no longer based solely upon the immediacy of the threat, but on the visibility of the target and opportunity to eliminate it with minimal collateral damage (Hansard 2015, 26). The open interpretation of imminence is problematic, given its importance in the British public’s perception of the legitimacy of drone strikes. But it is also reflective of the challenging conflict the United Kingdom is engaged in. Haunted by its failure to take decisive action against al-Qaeda prior to 9/11, the US government determined that the rules of pre-emptive self-defence needed to evolve. Under America’s post-9/11 definition it does not matter whether the terrorist is actually being aggressive at the time of a drone strike; ‘their very status as terrorists’, law professor Russell Christopher explains, ‘qualifies as the conduct of posing an imminent threat’ (Christopher 2012, 256, original emphasis). The philosopher Jeff McMahan has explained this position further, arguing that ‘even while terrorists are sleeping or eating dinner or doing some other innocuous activity, they do not lose their status as terrorists and thus are continuously and invariably constituting imminent threats’ (McMahan 2012, 135). For the UK, the transformation of the foreign-based terrorist from a criminal into an enemy has altered the scope of imminence to match this US mindset. A terrorist engaged in a plot against the UK can, like an enemy soldier in a war, be killed at any time, because they are by their very nature an ongoing threat to the state. This expansive temporal scope grants the British government great latitude when it comes to the decision to unleash drone violence. Such an approach is problematic as it undermines the core basis of the right to self-defence established under Article 51 of the UN Charter, and makes the recourse to lethal force a permissible first response as opposed to a last resort.

Conclusion In advocating the postmodern state, Cooper (1996) warned of the risks the global community’s unequal progress towards such an existence posed. While his pre-9/11 warning concerned nations which still maintained a ‘nineteenth-century’ outlook of ‘every state for itself’ as opposed to the yet-to-materialise threat of non-state terror, he foreshadowed the UK’s spatial distinction of justice by cautioning that, despite progress, ‘the law of the jungle’ would still apply in some places (Cooper 1996, 37–8). ‘ When operating in the jungle’, Cooper remarked, ‘we must also use the law of the jungle.’ Failure to recognise the importance of maintaining the nation’s

6686_Enemark.indd 108

02/03/21 12:22 PM

‘A New Departure’: Britain’s Lethal Drone Policy / 109

defences, he counselled, represented one of the greatest dangers to the postmodern state. As the US and eventually the UK came to realise, states would need to be prepared to revert to the more vigorous methods of an earlier era – characterised by Cooper as ‘force, pre-emptive attack, deception, whatever is necessary’ – in order to protect themselves from those opposed to the spread of the system and values encapsulated in the postmodern state (Cooper 1996, 37–8). Ironically, it would be the very forces of progress Cooper predicted were necessary to end the nationalistic and ideological clashes which had blighted the twentieth century – free markets, universal rights, homogenisation, open borders and secular societies – which would provoke one of the most significant threats to the postmodern state in the early twenty-first century. Extremists such as ISIS seek to drag the world back 1,400 years to their Salafist vision of Islam, and so wage war against the postmodern world on all fronts in an effort to force its retreat. Of course, their goal is impossible to achieve. The vast majority of people in the regions controlled by these extremists do not recognise their warped version of Islam or wish to revert to medieval societies. Furthermore, the fallout of these occupations creates consequences the rest of the world simply cannot ignore. Postmodern states are thus faced with the prospect of a protracted struggle to eliminate an extremist enemy that can never be reasoned with (Kagan 2008, 84). Yet, while the values of postmodern states have contributed to the violence they currently face, those same states and the institutions they belong to have proven incapable of removing the threat. The United Kingdom’s drone policy is paradoxical: carefully crafted to operate within the human rights and international law project of the past century, but in another sense a rejection of it. The UK, like the US before it, has consciously stepped back into a world of Hobbesian division, where the use of lethal force against enemies is the only way to provide national security. Such a policy reveals a belief that the supranational institutions and international laws that bind them are incapable of removing the threat themselves. The UN Security Council has become paralysed by the split between its autocratic and democratic members, while the EU’s counterterrorism structures have proven incapable of preventing devastating attacks upon its members. The UK’s drone policy is as much a condemnation of the supranational organisations’ capabilities as it is an admission of the genuine threat posed by extremism. Such an approach may be at odds with the postmodern world and the march of a universal legal system, but as the realist Reinhold Niebuhr observed during the confrontations of the late twentieth century, to be victorious is not to be innocent (Niebuhr 2008, 5). Yet, while drones are set to remain a key tool of the UK’s counterterrorism

6686_Enemark.indd 109

02/03/21 12:22 PM

110 / Christopher J. Fuller

policy for the foreseeable future, the government must seek a better balance between victory and innocence if it is to ensure the former is not Pyrrhic in nature. It is understandable, given the constantly evolving terrorist threat, that UK policymakers seek to remain flexible in how they employ lethal drones. But for the long-term wellbeing of the nation, a degree of that flexibility should be sacrificed for greater transparency and oversight, managed by a judicial body independent of the government or security services, charged with ensuring that lethal drone use remains firmly within domestic and international law. The current terror threat does indeed pose a danger to the nation. But as the United States’ expansive approach to drone strikes has demonstrated, there are greater long-term perils to a nation’s wellbeing than that posed by jihadists. The seductive nature of remote killing can drag the country into a cycle of infinite warfare. Domestically liberal democracies do not fare well when engaged in open-ended targeted killing campaigns, with the normalisation of assassination as a state policy hardening societal attitudes towards the use of violence, reducing the resilience of the populace in the face of such threats, and creating unrealistic national security expectations which force the threshold for violence to be dramatically lowered. Internationally, the erosion of established legal norms regarding the use of lethal force over borders risks the proliferation of targeted killing as a state-sponsored tool, with the accompanying international tensions and instability posing a much more existential threat to global order than any terrorist group could ever hope to achieve. Ultimately, reduced expediency when it comes to authorising drone strikes is a small price to pay in order to prevent the state from being drawn into a cycle of violence which erodes its legal and moral foundations whilst masquerading as a solution.

Note 1. In American high school and college sporting competitions, a ‘jayvee’ (or ‘JV’) team is a team comprised of junior varsity players. These players are distinct from the main players (the varsity team) in a competition.

References Airwars. 2016. ‘Limited Accountability: A Transparency Audit of the Coalition Air War Against So-Called Islamic State’, Airwars, December, https://airwars.org/wpcontent/uploads/2016/12/Airwars-report_Web-FINAL1.compressed.pdf Aziz, Roya, and Monica Lam. 2003. ‘Profiles: The Lackawanna Cell’, PBS Frontline, 16 October, http://www.pbs.org/wgbh/pages/frontline/shows/sleeper/inside/ profiles.html

6686_Enemark.indd 110

02/03/21 12:22 PM

‘A New Departure’: Britain’s Lethal Drone Policy / 111

BBC. 2014. ‘MPs Support UK Air Strikes against IS in Iraq’, BBC News, 26 September, http://www.bbc.co.uk/news/uk-politics-29385123 BBC. 2015. ‘Cardiff Jihadist Reyaad Khan, 21, Killed by RAF Drone’, BBC News, 7 September, https://www.bbc.co.uk/news/uk-wales-34176790 Bethlehem, Daniel. 2012. ‘Principles Relevant to the Scope of a State’s Right of Self-Defense Against an Imminent or Actual Armed Attack by Non-State Actors’, American Journal of International Law 106 (4): 770–7. Brookman-Byrne, Max. 2018. ‘Falling Short: An Analysis of the Reporting of UK Drone Strikes by the MOD’, Drone Wars UK, July, https://dronewars.net/wpcontent/uploads/2018/07/dw-fallingshort-web.pdf Burke, Jason. 2016. ‘ Why Does France Keep Getting Attacked?’, The Guardian, 15 July, https://www.theguardian.com/world/2016/jul/15/why-does-france-keepgetting-attacked Bush, George W. 2002. ‘Remarks by President Bush at Arkansas Welcome, Northwest Arkansas Regional Airport, Bentonville, Arkansas, 4 November’, The White House, http://georgewbush-whitehouse.archives.gov/news/releases/2002/11/20021104-7 .html Callimachi, Rukmini. 2016. ‘How a Secretive Branch of ISIS Built a Global Network of Killers’, New York Times, 3 August, http://www.nytimes.com/2016/08/04/ world/middleeast/isis-german-recruit-interview.html Cameron, David. 2015a. Prime Minister’s Oral Statement to Parliament, ‘Syria: Refugees and Counter-Terrorism’, UK Government, 7 September, https://www. gov.uk/government/speeches/syria-refugees-and-counter-terrorism-primeministers-statement Cameron, David. 2015b. Statement to House of Commons, 7 September, UK Parliament, cols. 25–7, http://www.publications.parliament.uk/pa/cm201516/ cmhansrd/cm150907/debtext/150907-0001.htm Chesney, Robert M. 2013. ‘Beyond the Battlefield, Beyond Al Qaeda: The Destabilizing Legal Architecture of Counterterrorism’, Michigan Law Review 112 (2): 163–224. Christopher, Russell. 2012. ‘Imminence in Justified Targeted Killings’, in Targeted Killings: Law and Morality in an Asymmetrical World, edited by Claire Finkelstein, Jens D. Ohlin and Andrew Altman. Oxford: Oxford University Press, 253–84. Cole, Chris. 2017. ‘Attorney General Speaks on Legality of UK Preemptive Drone Strikes’, Drone Wars, 12 January, https://dronewars.net/2017/01/12/attorneygeneral-speaks-on-legality-of-uk-pre-emptive-drone-strikes Cooper, Robert. 1996. The Post-Modern State and the World Order. London: Demos, http://www.demos.co.uk/files/postmodernstate.pdf Cooper, Robert. 2002. ‘ The New Liberal Imperialism’, The Guardian, 7 April, https:// www.theguardian.com/world/2002/apr/07/1 Cronin, Audrey. 2013. ‘ The “War on Terrorism”: What Does it Mean to Win?’, Journal of Strategic Studies 37 (2): 174–97. Deeks, Ashley S. 2012. ‘“Unwilling or Unable”: Toward a Normative Framework for Extraterritorial Self-Defense’, Virginia Journal of International Law 52 (3): 483–550.

6686_Enemark.indd 111

02/03/21 12:22 PM

112 / Christopher J. Fuller

Defence. 2016. Written Question 23242, UK Parliament, 28 January, http://www. parliament.uk/business/publications/written-questions-answers-statements/ written-question/Commons/2016-01-20/23242 DOJ. 2010. Department of Justice White Paper: ‘Lawfulness of a Lethal Operation Directed Against a US Citizen Who Is a Senior Operational Leader of Al-Qa’ida’, Penn Law, https://www.law.upenn.edu/live/files/1903-doj-white-paper,%201 Fallon, Michael. 2015. ‘Oral Evidence: The UK Government’s Policy on the Use of Drones for Targeted Killing’ (Q34), Joint Committee on Human Rights. UK Parliament, 16 December, http://data.parliament.uk/writtenevidence/committeeevidence.svc/evidencedocument/human-rights-committee/the-uk-governments-policy-on-the-use-of-drones-for-targeted-killing/oral/27633.html Full Fact. 2015. ‘International Law and the Death of Reyaad Khan’, 8 September, https://fullfact.org/law/international-law-and-death-reyaad-khan Fuller, Christopher. 2017. See It/Shoot It: The Secret History of the CIA’s Lethal Drone Program. New Haven: Yale University Press. Goldsmith, Peter. 2004. Attorney General Lord Goldsmith, House of Lords, 21 April, UK Parliament, cols. 370–1, http://www.publications.parliament.uk/pa/ ld200304/ldhansrd/vo040421/text/40421-07.htm#40421-07_spmin0 Hansard. 2015. House of Commons, 7 September, UK Parliament, col. 26, http:// www.publications.parliament.uk/pa/cm201516/cmhansrd/cm150907/ debtext/150907-0001.htm Hobbes, Thomas. 1996. Leviathan, edited by Richard Tuck. Cambridge: Cambridge University Press. JCHR. 2016. ‘ The Government’s Policy on the Use of Drones for Targeted Killing’, Joint Committee on Human Rights, 27 April, UK Parliament, http://www.publications.parliament.uk/pa/jt201516/jtselect/jtrights/574/574.pdf Jones, Sam, and Anne-Sylvaine Chassany. 2015. ‘Paris Killings Show the Challenges for Security Services’, Financial Times, 15 January, http://www.ft.com/ cms/s/0/77c73f3c-9cc2-11e4-a730-00144feabdc0.html#axzz4G669Dlke Kagan, Robert. 2008. The Return of History and the End of Dreams. New York: Atlantic Books. Kahn, Paul W. 2013. ‘Imagining Warfare’, European Journal of International Law 24 (1): 199–226. Kerbaj, Richard, and Tim Shipman. 2017. ‘Spies Told to Come Clean on Cameron’s Order to Kill’, Sunday Times, 19 February, https://www.thetimes.co.uk/article/ spies-told-to-come-clean-on-camerons-order-to-kill-ppfnlwsdd Lang, Arabella. 2015. ‘Legal Basis for UK Military Action in Syria’, House of Commons Briefing Paper Number 7404, 1 December, UK Government, https://www.gov. uk/government/uploads/system/uploads/attachment_data/file/478933/52309_ Cm_9161_NSS_SD_Review_web_only.pdf McMahan, Jeff. 2012. ‘ Targeted Killing: Murder, Combat or Law Enforcement?’, in Targeted Killings: Law and Morality in an Asymmetrical World, edited by Claire Finkelstein, Jens D. Ohlin and Andrew Altman. Oxford: Oxford University Press, 135–55.

6686_Enemark.indd 112

02/03/21 12:22 PM

‘A New Departure’: Britain’s Lethal Drone Policy / 113

Maxwell, Mark. 2012. ‘Rebutting the Civilian Presumption: Playing Whack-a-Mole Without a Mallet’, in Targeted Killings: Law and Morality in an Asymmetrical World, edited by Claire Finkelstein, Jens D. Ohlin and Andrew Altman. Oxford: Oxford University Press, 31–59. Niebuhr, Reinhold. 2008. The Irony of American History. Chicago: University of Chicago Press. NSS. 2002. National Security Strategy (2002), The White House, September, https:// georgewbush-whitehouse.archives.gov/nsc/nss/2002 NSS. 2015 ‘National Security Strategy and Strategic Defence and Security Review’, 23 November, UK Government, https://www.gov.uk/government/publications/ national-security-strategy-and-strategic-defence-and-security-review-2015 Obama, Barack. 2013. Remarks of President Barack Obama, National Defense University, 23 May, The White House, https://www.whitehouse.gov/the-pressoffice/2013/05/23/remarks-president-barack-obama Obama, Barack. 2014. ‘Statement by the President on ISIL’, 10 September, The White House, https://obamawhitehouse.archives.gov/the-press-office/2014/09/ 10/statement-president-isil-1 Perraudin, Frances, Patrick Wintour and Nicholas Watt. 2016. ‘UK Prepared to Carry Out More Drone Strikes against British Jihadists, Says Fallon’, The Guardian, 8 September, https://www.theguardian.com/world/2015/sep/08/uk-would-nothesitate-to-carry-out-more-strikes-against-british-jihadis Powell, Michael, and Dana Priest. 2002. ‘US Citizen Killed by CIA Linked to NY Terror Case’, Washington Post, 9 November, https://www.washingtonpost. com/archive/politics/2002/11/09/us-citizen-killed-by-cia-linked-to-ny-terrorcase/697a3a3e-72cf-4485-bdbf-1a00b9c3c95c Remnick, David. 2014. ‘Going the Distance: On and Off the Road with Barack Obama’, New Yorker, 27 January, http://www.newyorker.com/magazine/2014/01/27/ going-the-distance-david-remnick Rogers, Joel F. 2013. ‘British Attitudes on Drones’, YouGov, 3 April, https://yougov. co.uk/news/2013/04/03/british-attitudes-drones-and-targeted-killing Scott, Joan Wallach. 2007. The Politics of the Veil. Princeton: Princeton University Press. Stern, Jessica, and J. M. Berger. 2015. ISIS: The State of Terror. New York: Ecco. UNSC. 2015a. Letter from the Permanent Representative of the UK to the UNSC, 7 September, Security Council Report, http://www.securitycouncilreport.org/atf/ cf/%7B65BFCF9B-6D27-4E9C-8CD3-CF6E4FF96FF9%7D/s_2015_688.pdf UNSC. 2015b. UN Security Council Resolution 2249, 20 November, Security Council Report, http://www.securitycouncilreport.org/atf/cf/%7B65BFCF9B-6D274E9C-8CD3-CF6E4FF96FF9%7D/s_res_2249.pdf Watson, Abigail. 2016. ‘Great Britain Needs to Figure Out Its Killer Drone Rules’, War is Boring, 25 September, https://warisboring.com/op-ed-great-britain-needsto-figure-out-its-killer-drone-rules-acb5a2a93308#.adcotthdc Weber, Max. 1991. ‘Politics as a Vocation’, in From Max Weber: Essays in Sociology, edited by H. H. Gerth and C. Wright Mills. London: Routledge, 76–128.

6686_Enemark.indd 113

02/03/21 12:22 PM

114 / Christopher J. Fuller

Weiss, Caleb. 2014. ‘Islamic State Spokesman Again Threatens West in New Speech’, Long War Journal, 21 September, https://www.longwarjournal.org/archives/2014/09/ islamic_state_spokesman_again.php Williams, Brian G. 2013. Predators: The CIA’s Drone War on al Qaeda. Washington, DC: Potomac Books. Wright, Jeremy. 2017. ‘ The Modern Law of Self-Defence’, speech at the International Institute of Strategic Studies, 11 January, Just Security, https://www.justsecurity. org/wp-content/uploads/2017/01/United-Kingdom-Attorney-General-Speechmodern-law-of-self-defense-IISS.pdf Yoo, John. 2012. ‘Assassination or Targeted Killing After 9/11’, New York Law School Review 56: 57–79.

6686_Enemark.indd 114

02/03/21 12:22 PM

SIX

Ethics for Drone Operators: Rules versus Virtues Peter Olsthoorn

Introduction Militaries are by tradition rule-guided organisations, and mostly for good reasons: setting rules makes clear to military personnel what they can and cannot do, and it provides outsiders to the organisation, say the local population in a mission area, with some security regarding the way they are treated. An example is the prohibition of torture, a ban that is to be maintained regardless of how expedient it might be not to do so. Some decisions, and the matter of torture is again a good illustration of that (but so is the use of certain types of weapons, such as chemical and biological weapons, or expanding bullets), we do not leave to the discretion of the individual soldier. Rule-based ethics point to the importance of having universal, categorically binding moral norms. On the other hand, such rule-based approaches have as an important downside that rules lack flexibility and are often mostly ineffective when there are no witnesses around. Also, rule-following can impede the ability to see the moral aspect of what one is doing, while that ability is evidently essential to morally sound decision-making. Hence the notion that rules should leave soldiers with some leeway in that decision-making, if only to keep them from committing so-called ‘crimes of obedience’ (Kelman and Hamilton 1989). It is probably for that reason that one textbook on military ethics, meant for educating military personnel, unambiguously states that ‘in any situation where law and ethics set different standards, a member of the military profession will follow the higher standard, inevitably the one required by ethics’ (Coleman 2013, 268). Making good use of this leeway presupposes a good disposition, though, and it is at least partly for that reason that many militaries see a virtue-based

6686_Enemark.indd 115

02/03/21 12:22 PM

116 / Peter Olsthoorn

approach to teaching military ethics as an important complement to rules imposed from above in an effort to make their personnel behave ethically. Where rule-based approaches, interpreted narrowly, mainly aim at securing compliance, virtue ethics asks for a lot more, including the kind of supererogatory acts the military depends on. What is more, virtue ethics assumes that character can be developed: virtues are to be understood as dispositions that can be acquired through training and practice.1 This appeals to many military trainers and educators because such an approach sits rather well with the way most militaries see themselves: as being in the business of character-building. Finally, virtue ethics is in keeping with the tendency of many Western militaries to move away from a largely functional approach in their ethics education and towards a more aspirational approach that aims at making soldiers better persons, mainly based on the view that bad persons are not likely to become morally good soldiers – although they could of course still be effective ones (Robinson 2007; Wolfendale 2008, 164). In theory, utilitarianism is a possible third candidate for underpinning the ethics education for military personnel, but in practice it is seen by most as being particularly unfit for that purpose, mostly because ‘an outcomecentred approach may lead all too easily to military expedience as the sole guide to actions in war’ (Bonadonna 1994, 18). Utilitarianism not only holds that we should base our judgement of whether an act is morally right or wrong (and hence also whether it should be done or not) upon the foreseen consequences, but also, much more revolutionary, that everyone’s life and happiness should weigh equally. Its critics seem to hold that utilitarianism is not bad per se, but that the utilitarian calculus is likely to be misapplied in a self-serving way.2 Military ethicist and political philosopher Michael Walzer has pointed out, as have many others, that the valuing of ‘each and every person’ in the same way will not work when ‘solidarity collapses’ (Walzer 2004, 39). Precisely that is what happens in war, where we cannot but expect to see little willingness to take the consequences to all parties into account equally. The ethics (and specifically the Just War theory) that is to guide politicians and military decision-makers during armed conflict consists of a mix of rule-based elements (such as the prohibition of certain weapons, and discrimination between civilians and combatants) and, despite Walzer’s misgivings, more utilitarian ones (proportionality, chance of success). However, as we have already noted, most militaries today consider an aspirational virtue ethics approach as the best way to underpin the ethics education of military personnel that are actually carrying out military operations (Robinson 2007). The question is whether this approach is also the best one for the moral education of drone operators, and if it is, what virtues

6686_Enemark.indd 116

02/03/21 12:22 PM

Ethics for Drone Operators: Rules versus Virtues / 117

should hold central place in that virtue-based education. The more general, underlying question is whether what counts as a military virtue is place and time dependent; at first sight, a convincing argument can be made that this is not the case. Some military virtues are valued at all times and in all places – mainly because they perform an important function in or for the military. Martial courage is, of course, the obvious example here, being the quintessential military virtue. But if we take a closer look at courage, it also becomes clear that armed drones bring us to a whole new ball game: its operators do not seem to need any physical courage at all. Regarding other important military virtues appearing on military lists (see Robinson 2008), such as loyalty, discipline or obedience, it is at the minimum unclear what beneficial role these could have for operating armed drones. If we for the moment assume that traditional martial virtues such as physical courage are not the most relevant for drone operators, there are at least three possible answers to the question of what we do need. One could argue: (1) that the virtue approach is the right one, but that we need virtues that are better suited for military personnel flying drones than are the traditional, rather bellicose ones; or (2) that not only is the virtue approach the right one, but the traditional virtues by and large suffice – with the caveat that the use of drones does ask for new interpretations of these virtues; or, finally, (3) that virtues are of little help here and that we need something different altogether, presumably rule-based or utilitarian ethics, or a combination of both.

‘New’ Virtues for New Tasks If we assume that the existing conceptions of current militaries’ virtues are of little use in regulating the conduct of drone operators, devising a new list of virtues would be a first possible way forward. In this line of thought, drone operators do need virtues, but not necessarily the traditional military ones. The virtues we teach military personnel are to fit their particular job, and the virtues that drone operators need are most likely to be more about exercising restraint than about demonstrating virtues such as courage, loyalty and discipline. Such virtues of restraint are less military-specific and could, for instance, be found among the more ‘general’ cardinal virtues. Interestingly, of the four cardinal virtues of courage, wisdom, temperance and justice, only courage has hitherto made it to the traditional lists of military virtues and values, albeit not only in its Aristotelian form of physical courage on the battlefield, but also as moral courage. Wisdom, temperance and justice have not made it to most lists of military values but are today probably as necessary as courage is (see Skerker et al. 2019). These virtues

6686_Enemark.indd 117

02/03/21 12:22 PM

118 / Peter Olsthoorn

have a wider scope than the traditional military virtues, and incorporating them could have the collateral benefit that it would bring those parts of the military that would subscribe to these cardinal virtues into closer alignment with society at large. Respect, incidentally, is a virtue that is listed by many militaries, but is evidently also deemed important in larger society. Opting for a set of virtues that is closer to the four cardinal virtues would also give us a set of virtues that does justice to the ancient idea that one cannot have one virtue without the others, and that all the virtues are interrelated. Being just is of little value if one lacks the courage to defend justice, for instance, while courage is not of much use without practical wisdom to guide it. The haphazard lists of virtues that militaries now subscribe to (see Robinson 2008) miss these interconnections. Yet one could also argue that devising a new list of virtues from scratch is perhaps a bridge too far for what is, for the most part, a relatively traditional organisation. That the existing military virtues are ill-aligned to the operation of armed drones does not necessarily mean we have to opt for different ones. Instead, one could also identify the weaknesses of the existing virtues and see if the way militaries interpret these traditional virtues can be improved. Although most militaries today cling to fairly traditional interpretations of the virtues, other readings are of course possible. The question is then not which new virtues the military should promote, but in what form the existing ones should best be understood.

Interpreting the Old Virtues in New Ways A second way forward would be to interpret the existing virtues somewhat differently; more precisely, less narrowly than is commonly the case (see Schulzke 2016, 195–6). The gist of the codes, oaths and values (which, even if their actual influence is limited, at the minimum do communicate what an organisation thinks is important) as currently formulated in most militaries mainly pays attention to the organisation and colleagues. There is little in them that regulates the behaviour of soldiers towards civilian populations. The virtue of courage is especially interesting in the context of unmanned warfare. As was noted above, at present most conceptions of military courage include moral courage, instead of being limited to more martial (physical) forms of courage. So it seems that courage means different things in different contexts. Aristotle, for instance, famously defined courage in his Nicomachean Ethics as the mean between rashness and cowardice, and he thought that this virtue is especially needed in battle – the idea that a brave man does not fear a noble death in war (Aristotle 1962). This conception of courage as a mean was well suited to the ancient Greek

6686_Enemark.indd 118

02/03/21 12:22 PM

Ethics for Drone Operators: Rules versus Virtues / 119

phalanx formation, as either an excess or a deficiency of courage would destroy the organised whole that it was. But this martial notion of courage is clearly worlds apart from what Mahatma Gandhi envisioned when he pleaded for courageous but non-violent resistance to British colonial rule in India – that plea was about moral courage. Some years ago, Jesse Kirkpatrick (2015a; 2015b) and Robert Sparrow (2015a) had an interesting although somewhat semantic discussion on the question of whether drone operators possessed the virtue of courage. If a definite conclusion had to be drawn from that discussion, it would be that these operators do need courage, but more in the form of moral courage than of martial courage. Seeing that drone operators run no significant physical risk (at least not in the current asymmetric conflicts), the term ‘martial courage’ is out of place here. Moral courage is an important subspecies of the virtue of courage as it asks us to uphold our principles even if others disagree and perhaps hold us in contempt for sticking to them. As Ian Miller defines it, moral courage is ‘the capacity to overcome the fear of shame and humiliation in order to admit one’s mistakes, to confess a wrong, to reject evil conformity, to denounce injustice, and to defy immoral or imprudent orders’ (Miller 2000, 254). It is virtues such as these that have to provide guidance to military personnel in morally ambiguous situations, seeing that providing general rules and guidelines for such complex situations will not work – militaries seem now and then even disinclined to provide them (see, for instance, Whetham 2017). Peter Lee (2019) gives an excellent example when he describes how an acting sergeant on her first day in a supervisory role overseeing a Reaper drone crew stuck to her judgement, against the opinion of all present, that an alleged parcel placed on the back seat of a motorbike piloted by a Taliban target was in fact a child – which in the end it turned out to be. (She would have been equally courageous, of course, if the supposed parcel turned out to be precisely that – a parcel. But in that case it would perhaps have been more difficult to muster that same amount of moral courage at another time.) What is interesting here is that physical courage is primarily (though certainly not only) something one’s superiors and colleagues benefit from. Moral courage has a wider reach and is, in line with the aforementioned distinction between an aspirational and a functional approach, more about being a better person than about being an effective soldier (compare: Robinson 2007, 22; 2008, 1). This form of courage is important to the military not only because it needs personnel who will dare to blow the whistle if necessary, but also because it benefits from having soldiers who dare to correct colleagues who they think are acting wrongly, or even to report

6686_Enemark.indd 119

02/03/21 12:22 PM

120 / Peter Olsthoorn

them if necessary. Its beneficiaries, today, are not only military colleagues, as is predominantly the case with physical courage, but, as Lee’s example of the acting sergeant clearly shows, also the outsiders (for example, civilians in the vicinity of a planned drone strike) the military is there to protect. Moral courage can only perform that function if militaries allow room for it, however. The good thing is that militaries today rarely fail to at least pay lip service to moral courage, and they generally claim to deem it a plus to have among their personnel principled people who dare to blow the whistle if necessary or to stop a colleague who is about to commit a wrong. But although their definitions of courage include moral courage, in reality military organisations all too often offer a fairly unfriendly environment for acting on moral principles, especially when adherence to these principles appears to conflict with organisational interests or mission success. That drone pilots are probably less subject to the forces of peer pressure and group loyalty than other members of the military could make it easier for them to gather the moral courage that is needed to make right decisions or to display loyalty to principle instead of group loyalty (Lee 2012, 15). Loyalty is as often mentioned on lists of military virtues as courage is. But that same loyalty that militaries value so much is at the same time a cause of both unethical conduct and attempts to cover that conduct up. To improve matters, militaries could interpret loyalty in such a way that it includes loyalty to a profession or principle, not just loyalty to one’s group and organisation, as we now often see (see Olsthoorn 2011). Loyalty to one’s professional ethic, instead of to one’s organisation and colleagues, is nothing more or less than what is commonly understood to be one of the key characteristics of a professional, something military personnel claim to be.3 Of course, the position of a professional in a civilian occupation is essentially different from military personnel in a conflict zone. Civilian professionals are able to put the interest of their clients above everything else without putting themselves in harm’s way, whereas soldiers cannot at all times act in the interest of the local population without incurring more risks to themselves. For that reason, especially when we consider the fact that in the eyes of many the predominant task of most military organisations is still the defence of national territory, the emphasis on loyalty to the organisation is not that surprising. The interesting thing here is, of course, that drone operators can take the interests of outsiders into account with no extra risk to themselves, and in that way the fact that the risks for this category of military personnel is effectively nil might open the door to a more professional – in the meaning of impartial – attitude. One could even wonder whether, at a time when many armed forces consider the promotion of universal principles as their main ground for existence, the development of

6686_Enemark.indd 120

02/03/21 12:22 PM

Ethics for Drone Operators: Rules versus Virtues / 121

a truer professionalism, with the main focus of loyalty being the soldier’s professional ethic instead of their organisation, is still too far-fetched. Respect, finally, was mentioned previously as a virtue valued by both the military and society at large. A closer look, however, reveals that respect in the military is now and then limited to respect towards colleagues. The US Army describes respect as, among other things, ‘trusting that all people have done their jobs and fulfilled their duty’ (US Army n.d.). This definition seems to implicitly limit respect to colleagues. Here, too, a less narrow interpretation seems in place.4

Rules and Utility Instead of Virtues A final way ahead would be to reconsider whether virtues in fact form the best underpinning for the ethics education of drone operators in the first place. We have seen that the traditional military virtues are of themselves already more inward looking than the cardinal virtues, but also on a more theoretical level that virtue ethics is fairly self-regarding. Virtue ethics focuses on the agent and his or her character and flourishing, even in situations (and war is probably such a situation) where an outcome-centred approach would seem to be more appropriate. Aristotle’s idea of virtue is on the whole a lot less attentive to the needs of others than is the utilitarian notion of the greatest happiness for the greatest number or the rule-based maxim to treat others in the same way you want them to treat you.5 Now, as we have seen, military ethicists often criticise utilitarianism because it would make military expedience outweigh all other concerns (see, for instance, Snow 2009, 560), but in fact the consequentialist precept that the consequences to all persons should weigh equally could, if taken seriously, lead to a fairer distribution of the right to life. Utilitarianism does not condone the maximising of our own utility, as some seem to hold, but that of all. This means that soldiers should take as much care, and run as much risk, to avoid casualties among enemy civilians as they would do for their own civilians (Shaw 2005). Although the holding of such an impartial view might be expecting too much from regular soldiers in a regular war in defence of one’s own country, in many of today’s operations in which armed drones are used one probably should be able to do so a bit more easily. Even when enemy forces do little to avoid civilian casualties, or even target them deliberately, utilitarianism is not about fairness, but about minimising the damage of warfare (Shaw 2005, 139–40). As to rule-based ethics, in recent years, both Schulzke (2016) and Renic (2018, 194) have argued that more emphasis on rule-based approaches is in place in the case of drone operators – the first-mentioned mainly on the

6686_Enemark.indd 121

02/03/21 12:22 PM

122 / Peter Olsthoorn

ground that alternative interpretations of the military virtues that would fit drone warfare are so different from the current ones. In defence of the somewhat more rudimentary form of rule-based ethics (rudimentary because it disregards the good intention most forms of rule-based ethics ask for)6 that we encounter in most militaries, one could argue that pointing out what is permitted and what is not, and what the consequences of transgressing these rules are, should also have a role in the ethics education of military personnel operating drones. Failing to do so can be costly for perpetrator and victim alike – something that can be overlooked in an ethics education that focuses too much on character development. That universal rules lack flexibility is not always a problem; torture, as we already noted, is at present forbidden under every circumstance, and flexibility here could quickly bring us onto a slippery slope. As said in the introduction to this chapter, we do not leave the decision on these matters to individual soldiers, however virtuous they might be. The Just War tradition is primarily founded on an ethic that stresses the importance of such universal, categorically binding moral rules (though, as we have noted before, there are unmistakably also some consequentialist elements within the Just War tradition). Clear rules have the additional benefit that they can turn potential moral dilemmas into tests of integrity: it is clear what is the correct way to proceed, yet there is pressure (from peers, or the prospect of furthering one’s own interest) to follow a different course of action (Coleman 2009, 105–6).7 Recent research has shown that military personnel who lack guidelines to deal with such morally critical situations experience more moral dilemmas, increasing the likelihood of moral injury (Schut 2015). This risk of moral injury is something that drone operators face too. Although in the past some argued that killing might get a bit easier with the increased physical and psychological distance between soldiers and the battlefield (see, for instance, Olsthoorn 2011, 126), we now know that ‘[p]hysical separation from the combat zone does not [. . .] automatically lead to emotional disconnection. The crew of a Tornado flying at low level above an enemy contact may be more emotionally disengaged than the Reaper crew’ (Lee 2012).

Risk and the Military Profession In earlier days, bows, catapults and firearms have been vilified for being the weapon of choice of cowards, yet it seems that armed drones push things even further by doing away with risk altogether – which raises interesting questions about the extent to which risk is fundamental to the (image of the) military profession and whether the elimination of risk will change it. Although the use of drones is at first sight not very different (as long

6686_Enemark.indd 122

02/03/21 12:22 PM

Ethics for Drone Operators: Rules versus Virtues / 123

as such systems are not fully autonomous, that is) from using a manned aircraft to drop a bomb from a high altitude, their rise makes it possible to engage the enemy from such a safe distance that it reduces the risks for their military operators to about zero. This reduction in risk to personnel could make one wonder whether the military profession becomes a less honourable one as a consequence, as honour often involves acting against one’s own self-interest (including the preserving of life and limb) to further a higher interest. The difference between running a limited risk and running no risks is perhaps not merely gradual: ‘For men to join in battle is generally thought to be honourable, but not if they are so situated as to be able to kill others without exposing themselves to danger whatever’ (Welsh 2008, 4). The Time magazine journalists Mark Thompson and Bobby Ghosh (2009) have observed how people in Waziristan (the region in Pakistan where US drones have killed many Taliban leaders) see the use of drones as dishonourable and cowardly.8 And, according to military ethicist George R. Lucas (2016, 175), ‘the removal of any risk of harm to the military [. . .] seems grotesquely unfair, persecutory, oppressive, abusive, and therefore morally repugnant’, reminding us of ‘the Death Star from Star Wars’. Some militaries are aware of that problem and attempt to draw the use of armed drones into the realm of honour. A 2016 article in the New York Times reported, for example, how (Schmidt 2016): [f]or years, the military’s drone pilots have toiled in obscurity from windowless rooms at bases in suburban America, viewed by some in the armed forces more as video game players than as warriors. But in a reflection of their increasingly important role under President Obama, the drone operators will now be eligible for military honors akin to those given to pilots who flew over the battlefields of Iraq and Afghanistan.

This new award can also be conferred upon US military personnel who launch a cyber-attack, the article continued. Although all of this – honouring the courage of what are sometimes somewhat derogatively called ‘cubicle warriors’ – might sound somewhat odd to many people, it fits well with the rise of ways of warfighting in which soldiers are exposed to less risk. In the same article, we read that ‘[a]ccording to the Pentagon, the first seven Medal of Honor awards for service in Iraq and Afghanistan were given to those who had died. But since 2010, all 10 people who have received the Medal of Honor have been living at the time it was awarded’ (Schmidt 2016). Most observers, however, will still feel that the use of armed drones is difficult to reconcile with what is commonly understood by the term ‘honourable’: incurring risk to oneself seems to be a vital part of that concept. Fighting one’s

6686_Enemark.indd 123

02/03/21 12:22 PM

124 / Peter Olsthoorn

adversary from a low-flying manned aircraft would indicate an acceptance of risk to oneself, but if that would also increase the risk to the local population one might ask what the point is. Dismissing the use of drones because their use is free of risk for the attacking side and thus dishonourable might, as a result, boil down to accepting higher risks to oneself and the local population just to prove your honourableness (see Strawser 2010). As that would be a distinctly unsatisfying option, we have to ask ourselves whether the language of honour is suited to describe drone warfare to begin with (see Goldstein 2015, 75). That the drawbacks of military honour are as numerous as the advantages might form another reason to leave honour out of the equation (see Sparrow 2015b, 390). What defines warriors is not so much their acceptance of risk as the restraint with which they exercise violence (see Renic 2018) – which brings us back to our earlier conclusion that for drone operators abiding by the rules is perhaps as important as exercising virtue.9 Contrasting supposedly risk-averse drone operators with the assumed death wish of, for instance, a suicide bomber is not going to be particularly helpful, though. Such a comparison echoes the occidentalist rhetoric of a feminine West that al-Qaida and ISIS are always eager to embrace. Buying into that ‘you love Pepsi, we love death’ rhetoric implies not only that drone operators are not honourable because they do not put their life at risk, but possibly also that we should deem their ‘opposites’ (the deathseeking suicide bombers) honourable – not a conclusion many of us would want to accept. Risk aversion is not bad or dishonourable in itself; it is only a problem in so far as it comes at the cost of increased risk to the outsiders (civilians in foreign territories) that the military should defend if it wants to live up to its professed ambition to be a force for good. Perhaps the real issue deserving of our attention is that we are generally more concerned about casualties among our fellow countrymen and countrywomen than among unknown persons in faraway countries – this is perhaps to some extent understandable and natural, but certainly not moral. Or honourable, for that matter.

Conclusion Apart from some academics who have a clear preference for virtue ethics, rule-based ethics or consequentialist ethics, in real life most people tend to see a role for both virtues and rules, and they consider the consequences of an action as well (see Nagel 1986, 166). They are probably quite right in doing so, and one could even argue that those involved in professional ethics education are more or less duty-bound to take a fairly comprehensive approach towards teaching ethics. That means paying attention to rules,

6686_Enemark.indd 124

02/03/21 12:22 PM

Ethics for Drone Operators: Rules versus Virtues / 125

virtues and consequences, but also to situational factors that make unethical conduct more likely to occur.10 What is clear, though, is that some of the traditional martial virtues such as courage and loyalty are in their current interpretations less relevant for today’s drone operators. In particular, the obsoleteness of physical courage – the willingness to incur risk to life and limb – in unmanned warfare has led to some not too helpful comments on its honourableness. Nonetheless, we do need to look for alternatives to the traditional military virtues, and in this chapter a few possible ways ahead have been outlined.

Notes 1. Virtues are typically described as stable character traits that are worth having, often working as correctives to our self-regarding inclinations (Foot 2002, 8–12). Most virtue ethicists hark back to the time-proven work of Aristotle, who held that we become virtuous by actually performing virtuous acts. Performing courageous deeds grows courage, for instance. It is this Aristotelian view on virtues that underlies most literature on military virtues too. As Robinson has explained: ‘ The approach adopted in most armed forces is that of “virtue ethics”, with their philosophical origins found in Aristotle. Essentially, virtue ethics seeks to ensure moral behavior by instilling certain virtues (loyalty, honesty, and courage) to create good character. Consequently, many military academies have adopted an approach based on Aristotelian virtue ethics’ (Robinson 2007, 29). 2. As one author writes, ‘utilitarianism would lend itself to abuse in precisely those kinds of situations in which ethical safeguards are most needed, and should, for this reason, be stricken from the list of viable alternatives for the military’ (Snow 2009, 560). According to Stephen Deakin (2008, 84): ‘the utilitarianism ethic often does not work in a military community. It is not the ethic of a virtuous person desiring to do good in every circumstance [. . .]. Rather, it is the ethic of a highly educated rational calculator who is constantly considering whether an action is harmful to the Army or not, and, inevitably, what he can get away with.’ 3. Loyalty to the organisation is the main aspect of military professionalism that is somewhat at odds with what a ‘regular’ professional ethic entails and, in the past, some have for that reason maintained that the military profession was ill suited to develop into a ‘true’ profession (see, for instance, Doorn 1975). Armed forces socialise their employees thoroughly into the organisation, which contributes to the strong loyalty military personnel feel towards each other and their employer. The fact that military personnel are predominantly trained inhouse, whereas other professionals (such as doctors) as a rule receive most of their formal professional training before entering their job, makes this socialisation into the organisation (instead of into a profession) easier. As a consequence, different militaries have different organisational values (often still service-specific), but there are as yet no values of ‘the military profession’ as

6686_Enemark.indd 125

02/03/21 12:22 PM

126 / Peter Olsthoorn

4.

5.

6.

7.

8.

9.

10.

6686_Enemark.indd 126

such. By contrast, the values and standards of ‘regular’ professionals stem from universities and professional associations, not from their particular hospital or law firm (Mintzberg 1983, 192). Military ethicist Timothy Challans describes how ‘early drafts of the Army’s 1999 leadership manual included the notion of respect; in fact, the key feature of respect was that of respecting the enemy on the battlefield. That idea did not survive the staffing process, and even a cursory check of the manual today will reveal that only Americans are mentioned as being recipients of this important value of respect’ (Challans 2007, 163). For Aristotle, the good life took precedence over the moral life (Nagel 1986, 195, 197). The virtue of justice is a possible exception; Aristotle deemed it the most complete virtue because the best person is not one ‘who practices virtue toward himself, but who practices it toward others, for that is a hard thing to achieve’ (Aristotle 1962, 1030a). Although not asking anyone to go beyond the call of duty, rule-based ethics, especially as conceived by its main protagonist Immanuel Kant, can demand quite a lot from military men and women. In this understanding, moral duties are to be followed because one accepts them by choice, not because they are imposed from the outside and backed by sanctions (see also Martinelli-Fernandez 2006, 56–7). Although the distinction between ethical dilemmas and tests of integrity is an important and meaningful one, the situations in which this apparently straightforward distinction is blurred are the most interesting. For instance, it is generally thought that the loyalty one feels towards colleagues is nothing more than a pressure that can create a test of integrity. Loyalty is then viewed as the suspension of independent judgement, or the ‘willingness not to follow good judgment’ (Ewin 1992, 412). But if loyalty amounts to a value, and for most members of the military it does, then there might be a dilemma again (see also Coleman 2009, 112). Two weeks after the 9/11 attacks, Susan Sontag (2001) made a somewhat similar (and much criticised) remark: ‘If the word “cowardly” is to be used, it might be more aptly applied to those who kill from beyond the range of retaliation, high in the sky, than to those willing to die themselves in order to kill others. In the matter of courage (a morally neutral virtue): whatever may be said of the perpetrators of Tuesday’s slaughter, they were not cowards.’ In addition, one could argue, as Michael Robillard has, that ‘the value of military effectiveness should be held in higher esteem than that of mere physical risk in battle [. . .]. [T]he ultimate mission of the military should be effective national defense and not heroism for heroism’s sake. Accordingly, the strategically effective logistician, drone pilot, or informational officer should be held in equal if not higher regard in the military of the twenty-first century than that of the tactical level infantryman’ (Robillard 2017, 217, original emphasis). Like other military personnel, drone operators are subject to situational forces that are much stronger than those most of us will ever encounter. Their ethics education should aim not only at furthering virtues, or respect for rules, but

02/03/21 12:22 PM

Ethics for Drone Operators: Rules versus Virtues / 127

also at giving insight into the factors that make unethical conduct more likely to take place. Factors such as negative peer pressure, dehumanisation, stress, sleep deprivation, the national and organisational culture, and also the amount and kind of training and education received perhaps influence our conduct more than our character does (Doris and Murphy 2007). The social psychologist’s advice to avoid morally challenging situations is clearly not very helpful for drone operators, but with more knowledge about the influence of these factors, militaries can do more to make the erosion of moral standards less likely to occur. The insights social psychology offers should hence have a place in the moral education of UAV (unmanned aerial vehicle) personnel. However, some of these factors might have less influence on drone operators than on regular military personnel.

References Aristotle. 1962. Nicomachean Ethics. Indianapolis: Bobbs Merrill. Bonadonna, Reed R. 1994. ‘Above and Beyond: Marines and Virtue Ethics’, Marine Corps Gazette 78 (1): 18–20. Challans, Timothy L. 2007. Awakening Warrior: Revolution in the Ethics of Warfare. Albany: State University of New York Press. Coleman, Stephen. 2009. ‘ The Problems of Duty and Loyalty’, Journal of Military Ethics 8 (2): 105–6. Coleman, Stephen. 2013. Military Ethics. Oxford: Oxford University Press. Deakin, Stephen. 2008. ‘Education in an Ethos at the Royal Military Academy Sandhurst’, in Ethics Education in the Military, edited by Paul Robinson, Nigel de Lee and Don Carrick. Aldershot: Ashgate, 15–30. Doorn, Jacques A. A. van. 1975. The Soldier and Social Change. London: Sage. Doris, John M., and Dominic Murphy. 2007. ‘From My Lai to Abu Ghraib: The Moral Psychology of Atrocity’, Midwest Studies in Philosophy 31: 25–55. Ewin, Robert E. 1992. ‘Loyalty and Virtues’, Philosophical Quarterly 42 (169): 403–19. Foot, Philippa. 2002. Virtues and Vices and Other Essays in Moral Philosophy. Oxford: Clarendon Press. Goldstein, Cora. 2015. ‘Drones, Honor, and War’, Military Review, November– December, 70–6. Kelman, Herbert C., and V. Lee Hamilton. 1989. Crimes of Obedience: Toward a Social Psychology of Authority and Responsibility. New Haven: Yale University Press. Kirkpatrick, Jesse. 2015a. ‘Drones and the Martial Virtue Courage’, Journal of Military Ethics 14 (3–4): 212–19. Kirkpatrick, Jesse. 2015b. ‘Reply to Sparrow: Martial Courage – or Merely Courage?’, Journal of Military Ethics 14 (3–4): 228–31. Lee, Peter. 2012. ‘Remoteness, Risk and Aircrew Ethos’, Air Power Review 15 (1): 1–19. Lee, Peter. 2019. ‘Case Study 2: Moral Courage’, in Military Virtues, edited by Michael Skerker, David Whetham and Don Carrick. Havant: Howgate, 111–17.

6686_Enemark.indd 127

02/03/21 12:22 PM

128 / Peter Olsthoorn

Lucas, George R. 2016. Military Ethics: What Everyone Needs to Know. Oxford: Oxford University Press. Martinelli-Fernandez, Susan. 2006. ‘Educating Honorable Warriors’, Journal of Military Ethics 5 (1): 55–66. Miller, Ian. 2000. The Mystery of Courage. Cambridge, MA: Harvard University Press. Mintzberg, Henry. 1983. Structure in Fives. Englewood Cliffs: Prentice Hall. Nagel, Thomas. 1986. The View from Nowhere. New York: Oxford University Press. Olsthoorn, Peter. 2011. Military Ethics and Virtues: An Interdisciplinary Approach for the 21st Century. New York: Routledge. Renic, Neil C. 2018. ‘UAVs and the End of Heroism? Historicising the Ethical Challenge of Asymmetric Violence’, Journal of Military Ethics 17 (4): 188–97. Robillard, Michael. 2017. ‘Risk, War, and the Dangers of Soldier Identity’, Journal of Military Ethics 16 (3–4): 205–19. Robinson, Paul. 2007. ‘Ethics Training and Development in the Military’, Parameters, Spring: 22–36. Robinson, Paul. 2008. ‘Introduction: Ethics Education in the Military’, in Ethics Education in the Military, edited by Paul Robinson, Nigel de Lee and Don Carrick. Aldershot: Ashgate, 1–12. Schmidt, Michael S. 2016. ‘Pentagon Will Extend Military Honors to Drone Operators Far from Battles’, New York Times, 6 January, https://www.nytimes. com/2016/01/07/us/pentagon-will-extend-military-honors-to-drone-operatorsfar-from-battles.html Schulzke, Marcus. 2016. ‘Rethinking Military Virtue Ethics in an Age of Unmanned Weapons’, Journal of Military Ethics 15 (3): 187–204. Schut, Michelle. 2015. Soldiers as Strangers: Morally and Culturally Critical Situations During Military Missions. Doctoral dissertation, University of Nijmegen. Shaw, Martin. 2005. The New Western Way of War: Risk-Transfer War and its Crisis in Iraq. Cambridge: Polity. Skerker, Michael, David Whetham and Don Carrick (eds.). 2019. Military Virtues. Havant: Howgate. Snow, Nancy E. 2009. ‘How Ethical Theory Can Improve Practice: Lessons from Abu Ghraib’, Ethical Theory & Moral Practice 12 (5): 555–68. Sontag, Susan. 2001. ‘ Talk of the Town’, The New Yorker, 24 September. Sparrow, Robert. 2015a. ‘Martial and Moral Courage in Teleoperated Warfare: A Commentary on Kirkpatrick’, Journal of Military Ethics 14 (3–4): 220–7. Sparrow, Robert. 2015b. ‘Drones, Courage, and Military Culture’, in Routledge Handbook of Military Ethics, edited by George R. Lucas. New York: Routledge, 380–94. Strawser, Bradley J. 2010. ‘Moral Predators: The Duty to Employ Uninhabited Aerial Vehicles’, Journal of Military Ethics 9 (4): 342–68. Thompson, Mark, and Bobby G. Ghosh. 2009. ‘ The CIA’s Silent War in Pakistan’, Time, 1 June. US Army. n.d. ‘ The Army Values’, https://www.army.mil/values (accessed 22 April 2020). Walzer, Michael. 2004. Arguing about War. New Haven: Yale University Press.

6686_Enemark.indd 128

02/03/21 12:22 PM

Ethics for Drone Operators: Rules versus Virtues / 129

Welsh, Alexander. 2008. What is Honor? A Question of Moral Imperatives. New Haven: Yale University Press. Whetham, David. 2017. ‘ABCA Coalition Operations in Afghanistan, Iraq and Beyond: Two Decades of Military Ethics Challenges and Leadership Responses’, in Military Ethics and Leadership, edited by Peter Olsthoorn. Leiden: Brill, 86–103. Wolfendale, Jessica. 2008. ‘ What Is the Point of Teaching Ethics in the Military?’, in Ethics Education in the Military, edited by Paul Robinson, Nigel de Lee and Don Carrick. Aldershot: Ashgate, 161–74.

6686_Enemark.indd 129

02/03/21 12:22 PM

SEVEN

Drone Warriors, Revealed Humanity and a Feminist Ethics of Care Lindsay C . Clark and Christian Enemark

Introduction I see mothers with children, I see fathers with children, I see fathers with mothers, I see kids playing soccer. Anonymous operator of a US Air Force Reaper drone flown over Afghanistan (Bumiller 2012)

How should the operators of armed drones conduct drone strikes? Specifically, how should those individuals make decisions (about killing another human being) in a way that avoids unjust harm to people within a strike zone? Our chapter approaches this question with an expanded notion of what constitutes ‘just’ drone warfare, using the feminist idea of an ethics of care to move beyond the traditional confines of Just War morality. Such expansion is worthwhile because it has the potential to illuminate injustices resulting from violent drone use which might otherwise remain obscure and unaddressed. It is an approach premised upon recognition of two morally significant claims. The first is a general claim that, when violence in war is targeted against one person, the potential unjust effects upon innocent others are not necessarily limited to immediate injury and death. Survivors of that violence can still suffer the lasting and non-physical harm of, for example, being deprived of a caregiver (the physical victim of lethal violence) upon whom they depend. The second claim is drone-specific: that, prior to a long-range targeted killing, an armed drone equipped with a satellite-linked video-camera has the capacity to reveal to its operator the prosaic humanity of a targeted individual. The features of that humanity include, critically, the human relationships in which a particular individual is embedded. In combination, these two claims

6686_Enemark.indd 130

02/03/21 12:22 PM

Drone Warriors, Revealed Humanity, a Feminist Ethics of Care / 131

generate a distinct ethical concern: that a drone operator might decide to kill an individual in circumstances where he or she would thereby knowingly deprive dependent civilians (for example, family members) of a vital source of care.1 In the conduct of war, jus in bello requires discrimination and proportionality, and these principles for decision-making rest upon the abstract concepts of ‘combatant’ and ‘non-combatant’ (or ‘civilian’) (see Frowe 2011, 95–117). However, to rely exclusively upon abstract moral reasoning is potentially insufficient in the context of a drone-based targeted killing. This is because such reasoning cannot account adequately for the emotions felt by a conscientious drone operator who is put on notice (by means of a video-camera) that the destructiveness of a drone strike would extend to the destruction of unique human relationships. Our chapter therefore advances an idea for improvement: an operator’s jus in bello decisions should be augmented by an ethics of care, an approach to moral reasoning originating in feminist thought. In contrast to the way Just War decision-making is traditionally approached by reference to abstract principles, care ethics focuses on people not as abstract others but as real individuals with particular needs, and its ontological emphasis is on persons in relation to each other. So, for the operator of an armed, camera-equipped drone who richly perceives the humanity he or she has in common with a targeted individual, and whose actions are guided also by care ethics, the issue for decision becomes: for whom should I care, and what is the caring response, when I am deciding whether to kill? The answer, we argue, should draw upon a thickened concept of ‘harm’ to civilians; one which recognises the potential for relationships of care to be destroyed by wartime violence. The chapter begins by acknowledging the problem of non-physical harm to civilian survivors of violence, which is so often overlooked by Just War thinkers. It then explains why a drone operator’s prolonged observation of a targeted individual’s prosaic humanity (including the features of her or his personal relationships) presents a peculiar moral challenge for that operator. The second section describes the origins, character and distinctiveness of an ethics of care as a mode of moral reasoning, and the chapter concludes with a discussion of care ethics being applied in violent contexts including the use of armed drones.

Just War, Drone Violence and Non-Physical Harm Just War morality is the centuries-old framework according to which war is traditionally judged and governed (Walzer 2006). It assesses violence by reference to two sets of principles: jus ad bellum (the justice of resorting to

6686_Enemark.indd 131

02/03/21 12:22 PM

132 / Lindsay C . Clark and Christian Enemark

war) and jus in bello (the just conduct of war). For present purposes, the moral agency of drone operators is the central concern. And, as they make no decisions on whether war should be resorted to in the first place, the focus is on jus in bello. This set of Just War principles includes the ethical requirements that instances of violence during war should: (1) discriminate between combatants (who may legitimately be targeted) and civilians (who may not); and (2) be anticipated to generate a level of harm (including unintended harm to civilians) that is proportional (not excessive) in relation to the expected military benefit. In ethics-oriented literature on armed drones, when attention turns to in bello matters, scholars have tended to inquire into whether, how and by how much drone technology increases or reduces the amount of unjust human harm caused by political violence. One possibility is that drone technology, incorporating powerful targetidentification capabilities, has the moral advantage of enabling a degree of discrimination and proportionality in the use of force that is greater than what is achievable using other weapon platforms (Vogel 2010). Or, drone technology might instead induce morally inferior uses of force because drone operators, if they are morally disengaged from the killing process, are less likely to act with restraint, resulting in the exposure of civilians to more risk of harm (Alston and Shamsi 2010; Calhoun 2015). Such application of Just War morality traditionally rests upon a thin conceptualisation of harm. That is, when ethical judgements are made – about whether harmful effects are being directed discriminately, and whether expected harms are proportional to expected benefits – ‘harm’ is generally understood to mean physical injury or death. It follows from this that reducing the risk of civilian harm occurring in a drone strike could simply be approached as a technical challenge. By arming a drone with a smaller munition, for example, a narrower blast radius would be likely to endanger fewer innocent bystanders in the vicinity of a person being targeted (Marks 2012). Or, such harm could be reduced by the taking of ‘exceptionally rigorous steps’ in the targeting process to avoid exposing civilians to risk (see Rodin 2006). For example, former US president Barack Obama insisted that, prior to any drone strike, ‘there must be near-certainty that no civilians will be killed or injured’, and he described this as ‘the highest standard we can set’ (Obama 2013). Setting aside the obvious objection that certainty is the highest standard, the same ‘standard’ problem can be seen to persist here: that the traditional concept of in bello harm is a thin one. If a thicker concept of harm were adopted, there would be more scope to discern and address the range of possible injustices arising in war. One area worthy of greater consideration in moral reasoning is the causing of non-physical harm to civilians. Regarding the use of armed drones,

6686_Enemark.indd 132

02/03/21 12:22 PM

Drone Warriors, Revealed Humanity, a Feminist Ethics of Care / 133

evidence has emerged that the mere threat (and fearful anticipation) of physical harm can itself harm civilians psychologically. In parts of Pakistan, for example, civilian victimhood has manifested in a constant and debilitating fear of being struck by ever-present US drones (Friedersdorf 2012; Amnesty International 2013). Beyond this, we argue that it is important also to extend the concept of harm to the experience of a civilian who survives the actuality of violence against a targeted individual. That civilian should still be considered a victim if they have thereby been violently deprived of a caregiver’s support. The value of such consideration is that it pays greater attention to humans as relational beings, in contrast to the way Just War thinking has traditionally envisaged harm and humans in abstract terms (see Kellison 2019). Human relationships, often involving the giving and receiving of care, have always been vulnerable to destruction in war. However, the operator of an armed drone is arguably better positioned than any other kind of warrior in history to observe the features of particular relationships among the people who inhabit prospective strike zones. In making moral decisions from this unprecedented vantage point, he or she is potentially put on notice that the targeted killing of one individual would adversely affect innocents who are apparently and critically dependent upon a targeted individual’s care. A remotely controlled drone equipped with missiles is an instrument of violence, but it is morally significant also as an instrument of observation. Although the aircraft is geographically far removed from its ground-based operator, the targeted victim of a drone-launched missile is visually brought close to that operator by a mounted camera transmitting video imagery back to base via satellite. And visual proximity, rather than physical remoteness, is what matters more for the purposes of our discussion. Often, the images a drone operator sees include both friendly troops and the enemies with whom they are engaged, because armed drones are frequently used to provide air support during the conduct of ground-based combat missions. We will return briefly to this scenario towards the end of the chapter, but for now our focus is directed towards drone use in places outside what would traditionally be considered a conflict zone. Here, the US government in particular uses drones for the ‘targeted killing’ of individuals identified as terrorists. In this circumstance, before the planned killing occurs, a drone’s video-camera will have enabled its operator to observe closely, and for a prolonged period, a prospective target. Sometimes, the operator will have witnessed behaviour that is somehow linked to a military threat against the drone-using state, and more often he or she will have abundantly perceived (over consecutive hours, days or weeks) many of the everyday features of the targeted individual’s life. In this way, the humanity – prosaic and immediately unthreatening – of that individual is revealed and illuminated,

6686_Enemark.indd 133

02/03/21 12:22 PM

134 / Lindsay C . Clark and Christian Enemark

such that any effort by a drone operator to dehumanise a ‘target’ is made more difficult. Constantly watching the ordinariness of a particular person’s life being lived can sometimes eventually produce a feeling of ‘knowing’ that person. For example, former US drone operator Matt Martin has recalled of one experience: ‘Sometimes I thought I knew as much about this man [the target], my nemesis, as I knew about my wife’ (Martin 2010, 296). And, as Mark Coeckelbergh has argued (2013, 97), that ‘knowledge of the opponent’ has the effect of confirming ‘his humanity, personality, embodiment, and vulnerability’. The significance of such ‘tele-intimate’ violence in terms of its rebounding impact on drone operators is only beginning to be understood (see Williams 2015; Clark 2019). One possibility is that the taking of a closely observed human life might sometimes be so upsetting as to cause ‘moral injury’. That is, after executing a drone strike, drone operators might judge themselves so harshly as to be undone by their own sense of virtue, becoming victims of a debilitating conviction that they have done wrong and have ceased to be a good person (Enemark 2019). For present purposes, a drone operator’s emotional response to what he or she sees and does is relevant, but only in the context of a broader moral concern for human relationships. What matters most here is that the existence and uniqueness of those relationships are sometimes made obvious to a drone operator. This occurs when, for example, drone operators ‘watch targets spend time with family and friends and even playing with their children’ (Lee 2020). At other times, the prospective target of a drone strike might be observed playing with a dog, drinking tea, making love to his wife or attending a wedding or funeral (Freeman 2015; Abé 2012; Hurwitz 2013; Daggett 2015, 371). Then, if the strike goes ahead, its effects in the context of human relationships are made obvious too. Former US drone operator Heather Linebaugh (2013) has recalled how she ‘watched dozens of military-aged males die in Afghanistan, . . . some right outside the compound where their family was waiting for them to return home from the mosque’. Sometimes, in the conduct of post-strike damage assessments, operators see that the wives and children of a targeted individual are also among the dead. Or, they behold from on high the lamentations of grieving family members below. As one UK drone operator explained in an interview with Peter Lee (2019, 179): we may watch ‘ Target A’ for weeks, building up a pattern of life for the individual: know exactly what time he eats his meals; drives to the Mosque; or uses the ablutions . . . what we also see is the individual interacting with his family – playing with his kids and helping his wife around the compound.

6686_Enemark.indd 134

02/03/21 12:22 PM

Drone Warriors, Revealed Humanity, a Feminist Ethics of Care / 135

When a [missile] strike goes in, we stay on station and see the reactions of the wife and kids when the body is brought to them. You see someone fall to the floor and sob so hard their body is convulsing.

The civilian survivors of drone violence can be seen to have suffered a non-physical harm: the destruction of a relationship characterised by care and support. And yet the thin concept of harm that traditionally informs Just War thinking would lead to the conclusion that no injustice occurred: civilians were spared injury and death. This, we argue, is not enough in the peculiar context of drone warfare. Rather, moral reasoning needs to account for the way a drone’s powerful video-camera reveals a prospective target’s humanity as manifested in unique human relationships. Accordingly, we propose augmenting Just War thinking in this context with an approach to morality that originates in feminist theory: the ethics of care. In establishing its suitability, three aspects of care ethics are especially worth emphasising: its feminist rejection of a gendered hierarchy of modes of moral thought; the challenge it presents to abstract moral reasoning; and its relational ontology.

Ethics of Care The origin of care ethics is usually traced to the late-twentieth-century work of American psychologist Carol Gilligan. This work was inspired by deficiencies she perceived in her colleague Lawrence Kohlberg’s investigations into moral development in children. Kohlberg had claimed that his test results showed that girls progress more slowly than boys towards acquiring moral maturity. However, the prevailing assumption behind this finding was that moral maturity was signified by a moral agent’s capacity to detach herself or himself from the context of the specific ethical dilemma being confronted, and to make a moral judgement by reference to what general rules or values ought to apply to any similar situation. Gilligan denied this assumption about the superiority of abstract reasoning, arguing that reasoning focused instead on the dynamics of particular relationships was a fully developed moral perspective rather than an intermediate one to be surpassed. In her 1982 book In a Different Voice, Gilligan claimed that her own empirical studies into the moral development of children revealed the existence of two different moral ‘languages’: a language of ‘justice’ characterised by abstraction and impartiality, and a language of ‘care’ characterised by a concern for responsibility within personal relationships. The ‘different voice’ was that of care, and Gilligan found it to be mainly associated with females (Gilligan 1982). She argued that women’s moral development is largely established through their relationships with men and through caring for their children, and her aim was to

6686_Enemark.indd 135

02/03/21 12:22 PM

136 / Lindsay C . Clark and Christian Enemark

see ‘female’ ethics taken seriously (Gilligan 1982, 17). This aim was shared by the philosopher Nel Noddings, who, in her 1984 book Caring: A Feminine Approach to Ethics and Moral Education, distinguished the traditional ‘approach of the father’ from the ‘approach of the mother’ (Noddings 1984, 36–7): The first moves immediately to abstraction where . . . thinking can take place clearly and logically in isolation from the complicating factors of particular persons, places, and circumstances; the second moves to concretization where . . . feelings can be modified by the introduction of facts, the feelings of others, and personal histories.

As pioneers of an ethics of care, Gilligan and Noddings were both clearly advancing a feminist claim of gender equality: that ‘male’ and ‘female’ modes of moral reasoning are equally valid. However, the problem with their claim was that it risked reinforcing social prejudices that care itself is a women’s activity and that only females are essentially disposed to be caring. For this reason, later theorists of care ethics have deliberately moved away from an essentialist logic that ‘feminises’ care. Instead, they have argued that a care-oriented approach to moral reasoning, which has historically been subordinated by gendered hierarchies in society, should now be equally open to women and men (Held 2006, 20; Tännsjö 2008, 115; Levine 2010, 143; Sander-Staudt and Hamington 2011, x; Thomas 2011, 137). It is important also to emphasise how care ethics presents a challenge not only to the presumed superiority of abstract moral reasoning but also to its presumed adequacy. The moral language of ‘justice’ that Gilligan referred to is the language which features in the two mainstream perspectives of Western moral theory: consequentialist reasoning as reflected, for example, in the utilitarianism of Jeremy Bentham and J. S. Mill, and deontological reasoning as exemplified by the ideas of Immanuel Kant. The first mode of reasoning sees moral decision-making driven by cost–benefit analyses and an aim to maximise good consequences, and the second requires decisionmaking that discharges set duties and prioritises respect for human rights. And, although the Kantian rejects the morality of instrumental reasoning while the utilitarian embraces it, each kind of thinker is rationalistic in their approach. That is, both perspectives tend to be supported by a traditional assumption that ‘the more abstract the reasoning about a moral problem the better because the more likely to avoid bias and arbitrariness, the more nearly to achieve impartiality’ (Held 2006, 11). And yet, for care ethicists, the problem with applying abstract and universal principles like Bentham’s Principle of Utility or Kant’s Categorical Imperative is that this involves assuming that people are interchangeable units and that moral judgements

6686_Enemark.indd 136

02/03/21 12:22 PM

Drone Warriors, Revealed Humanity, a Feminist Ethics of Care / 137

can be derived in a manner akin to deducing a mathematical problem. It reflects, perhaps, what Jenny Edkins (2016, 2) has condemned as ‘a politics that misses the person, a politics that objectifies and instrumentalises’, because it treats people as objects to be classified and managed rather than as unique human beings. Moreover, the pursuit of perfect impartiality when applying abstract moral reasoning is also potentially a problem if it requires us to ‘reject emotion in determining what we ought to do’ (Held 2006, 24). By contrast, the application of an ethics of care ‘relies upon contextualized knowledge and wisdom, informed by principles, but honed by practical experiences and empathetic appreciation of and response to others’ (Sander-Staudt and Hamington 2011, ix). In this way, care ethics rejects as simplistic any notion that moral life is a matter of following specific abstract rules, and instead it emphasises consideration of ‘the relationships and roles within which people find themselves, and . . . the somewhat fluid and context-dependent nature of moral responsibilities’ (Groenhout 2014). Critically, then, care ethics values rather than rejects human emotion, although this is not to say that every emotion is valued or is equally valuable. Rather, as Virginia Held (2006, 10) has argued, ‘such emotions as sympathy, empathy, sensitivity, and responsiveness’ are ‘the kind of moral emotions that need to be cultivated not only to help in the implementation of the dictates of reason but to better ascertain what morality recommends’. People tend to feel these emotions towards other people, which brings our discussion of care ethics to the third relevant aspect: its relational ontology. An ethics of care (as given by one human being to another) is necessarily founded on a concept of the self as relational. This involves a recognition that a person’s identity and subjectivity are developed not in isolation but rather through interactions with others. Since birth, each of us has depended for our very survival upon care received from others, and throughout our lives we arguably need relationships with others ‘as a basis on which we can grow, learn and experience well-being’ (Barnes et al. 2015, 3). Acknowledging this need leads to an understanding that relations of interdependence remain a fundamental feature of human existence. In a sense, people are their relationships with one another; they are essentially socially enmeshed. It follows that moral decisions regarding any and every individual must be made in the context of someone’s embeddedness in familial and other social relations. Morality, then, can be seen to exist not in a series of universal principles that can guide action but ‘in the practices of care through which we fulfil our responsibilities to particular others’ (Robinson 2011, 4). In contrast to a mode of moral reasoning that envisages ‘abstract others’, an ethics of care ‘allows focus on real individuals with needs that differ from other real

6686_Enemark.indd 137

02/03/21 12:22 PM

138 / Lindsay C . Clark and Christian Enemark

individuals’ (Widdows 2011, 266). In addition, by making relationships rather than individuals-in-isolation the key consideration, a care ethics perspective illuminates sources of human suffering that might otherwise remain obscure. From a security perspective, according to Fiona Robinson (2011, 10), it is vital to understand the human subjects of security as ‘beings in-relation’ for whom relations of care occupy a position of centrality in moral life and in day-to-day activities. When those relations are damaged or destroyed, she argues, ‘security is threatened’ (Robinson 2011, 10). Thus, to be deprived of a caregiver is to be rendered insecure and, if the deprivation is deliberate and violent, to suffer a moral harm.

Care and Political Violence An ethics of care – characterised by a feminist origin, a challenge to abstraction and a relational ontology – has been applied in a variety of spheres of human endeavour: for example, medicine and healthcare, environmental conservation, the legal profession and business (Held 2006, 9; SanderStaudt and Hamington 2011; Adhariani et al. 2017). However, it might be difficult at first glance to see how a care-oriented morality could gain traction within the context of political violence. This context rather seems fated to be a care-free zone in human affairs. Historically, much of the violence wielded in wartime has been inspired and sustained by destructive gender stereotypes that, for example, promote the idea that combat is a valid way for men to accomplish masculinity aggressively. Conceptualising human beings in the abstract can facilitate processes of dehumanisation that often make warriors less reluctant to kill. And war in general is probably the gravest example of human relationships breaking down and being replaced by mutual hostility. Against such a backdrop, it would be implausible indeed to claim that violence can be dealt with ‘simply by caring’ (Held 2010, 116). Instead, the application of care ethics to problems of violence would involve a kind of thinking about Others that differs from the kind that traditionally has helped legitimise war as a political practice. Even if this different kind of thinking is unable wholly to delegitimise and thereby eliminate war itself, it carries the potential at least to improve the justness of wartime methods of violence. Sara Ruddick’s notion of ‘maternal thinking’, for example, relies not only on abstract moral principles but mainly on a moral agent’s recognition of the needs of others in particular contexts. For Ruddick, the problem with ‘analytic fictions’ in Just War thinking is that they work to make ‘“enemies” abstract enough to be killable’ (Ruddick 1989, 150). And yet this criticism of the traditional approach to wartime morality does not necessarily denote

6686_Enemark.indd 138

02/03/21 12:22 PM

Drone Warriors, Revealed Humanity, a Feminist Ethics of Care / 139

a preference for pacifism; care ethics is not inevitably subsumed into a pacifist morality. Rather, as Ruddick has argued, a ‘peacemaker’ who ‘will never celebrate violence’ may nevertheless ‘act violently in careful, conscientious knowledge of the hurt she inflicts and the costs to her as well as her victim’ (1989, 138). Accordingly, a violent actor in war who is guided by an ethics of care will retain their ‘capacity to empathize with the victims of violence’ (Held 2010, 122). Doing this is compatible with an acceptance of jus in bello norms such as the requirement for violence in war to be proportional. But the protective effect of adhering to those norms would be strengthened, arguably, by a care ethics emphasis on ‘the overriding context of caring relations between human beings that are so obviously shattered by war’ (Held 2010, 125). One type of wartime activity in which the dual application of Just War morality and care ethics potentially makes some sense is a counterinsurgency (COIN) campaign. According to Jillian Terry, the problem with theorising COIN exclusively by reference to Just War principles (for example, jus in bello) is that ethical analysis of violence becomes disconnected from how COIN is actually experienced by insurgents and local civilian populations. She therefore advocates an ethical reorientation towards ‘a feminist ethics premised on care, empathy, and relationality’ (Terry 2015). Thinking only in abstract terms about ‘discrimination’ between ‘civilians’ and ‘combatants’ is less meaningful (and less useful) than thinking also in relational terms, because (Terry 2015): Counterinsurgents have complex relationships with both civilians (who are not simply bystanders but rather active members of the community to be coerced and enlisted by both the counterinsurgents and insurgents themselves) and combatants (who often are not simply enemies but can be sources of information or even allies, and are sometimes members of the same community the counterinsurgents have been tasked with helping).

With ethical decision-making focused on these ‘complex relationships’, the potential result is that counterinsurgent violence to neutralise threats is wielded less readily and more carefully than would be the case in a more straightforward wartime scenario. COIN conducted in this way could involve, as Daniel Levine (2010, 148) has argued, counterinsurgents ‘cultivating a mindset that approaches the [local] population as people with meaningful needs and aspirations’ in a way that tends ‘to “humanize” them and make the use of violence against them more distasteful’. Moreover, to fulfil an associated commitment to those people to avoid violence whenever possible, the caring counterinsurgent might willingly endure additional physical risk to themselves (Levine 2010, 152). Thus, traditional adherence

6686_Enemark.indd 139

02/03/21 12:22 PM

140 / Lindsay C . Clark and Christian Enemark

to jus in bello restraint, combined with a caring disposition (which intuitively seems anathema to successful warfighting), could generate two moral advantages: a reduced risk of harm to civilians, and an increased likelihood of COIN success (assuming the latter is morally important). Even so, a downside risk must also be acknowledged: that applying an ethics of care in wartime might, like the historical practice of caregiving itself, become subordinated through instrumentalisation. In this regard, there is a cautionary tale in the occasional issuing of ‘condolence payments’ to local civilians affected by violence during the post-2001 US-led wars in Iraq and Afghanistan. According to the US government, these discretionary payments are intended as an ‘expression of sympathy for death, injury, or property damage caused by coalition or US forces generally during combat’ rather than an ‘admission of legal liability or fault’ (GAO 2007, 13). A US legal officer in Iraq, for example, ‘could offer up to $2,500 to the families of dead civilians, up to $1,500 for serious injuries, and up to $200 for minor injuries’ (Gregory 2020, 160). Such amounts seem scarcely commensurate with the suffered loss, and yet they appear at least to constitute some acknowledgement that human suffering in war extends beyond physical injury and death. For this reason, some scholars claim there is a moral case for genuinely compensating civilian survivors of wartime violence (Holewinski 2013; Bazargan-Forward 2017, 173), provided that human lives are not thereby cheapened and perpetrators of violence are not enabled to ‘buy their way out of moral guilt’ (Schulzke 2017, 159). The greater disadvantage that potentially remains, however, is that condolence payments will reinforce the notion that objectifying (lost) human life can be a means to an end. According to Thomas Gregory (2020, 175), the use of these payments has become ‘a weapon of war, a tool that could be used to manage or mitigate the strategic costs of civilian casualties’. If so, this practice does not signal the advent of an escalating effort to see caring for civilians as an inherent good, and nor does it herald a more expansive approach to taking moral responsibility for civilian harm. Rather, Gregory claims, there has been ‘no great awakening of humanitarian sentiment’ among coalition forces in Iraq and Afghanistan, who have expressed no commitment to ‘learn from their mistakes’ and make meaningful changes to the way in which war is conducted (Gregory 2020, 175). If, then, greater care-oriented effort towards moral reform of warfare is needed, the question arises: how might the practical conduct of war be improved if jus in bello morality were genuinely augmented with an ethics of care? In the next section, we address this question by focusing on how the operator of an armed, camera-equipped drone might exercise a greater degree of care for civilians in advance of making life-and-death decisions.

6686_Enemark.indd 140

02/03/21 12:22 PM

Drone Warriors, Revealed Humanity, a Feminist Ethics of Care / 141

Revealed Humanity and a Drone Warrior’s Ethics of Care A caring drone operator needs first to be a feminist in the sense that he or she recognises care ethics as different from but not inferior to Just War morality. Such recognition allows a caring disposition and jus in bello principles to be applied in combination, and the intended effect of this is to achieve a greater degree of restraint in the wielding of drone violence than would otherwise be exercised. To the contrary, it could be argued that achieving more restraint requires Just War morality to be displaced by care ethics. In Ruth Groenhout’s view, for example (Groenhout 2014): whatever it means to be a truly caring person . . . it most certainly does not mean that one always rigidly follows certain abstract rules of action. . . . a care ethics account of moral reasoning rejects abstract principles because they often get in the way of responding in . . . caring ways.

For present purposes, however, we maintain that it is more useful to regard care ethics as an added layer of morality; a complement to a traditional, principled ‘justice’ stance rather than an alternative in competition with it. This viewpoint accords with that of other care ethicists who claim that the ‘care’ and ‘justice’ approaches to morality, as originally presented by Gilligan, are neither mutually exclusive nor essentially antagonistic. Although an ethics of care rejects dogmatic and exclusive reliance upon abstract principles as guides to action, ‘it need not promote aversion to principles altogether’ (Sander-Staudt and Hamington 2011, xv). And it may be, as Held has argued, that ‘an adequate, comprehensive moral theory will have to include the insights of both the ethics of care and the ethics of justice’ (Held 2006, 16). When it comes to the exercise of drone violence by individual drone operators, the ambition of greater restraint would be pursued by augmenting jus in bello adherence with an ethics of care, and it would be founded on a thick conception of non-combatant harm. A care-oriented jus in bello would require the avoidance of physical and non-physical harms to civilians, and the latter would include the harm of being deprived of a caregiver. Such a conception of harm is made available, and is arguably required, by the manner in which the humanity of potential strike victims is revealed to a drone operator via video-camera. In effect, the drone’s camera helps the agent of violence to concretise and contextualise the object of violence, dispelling notions of Otherness in war. In the past, some drone operators have reportedly attempted to counteract the humanising effect of the drone’s camera. For example, in 2015 the former US Air Force (USAF) airman

6686_Enemark.indd 141

02/03/21 12:22 PM

142 / Lindsay C . Clark and Christian Enemark

Michael Haas explained why operators of US drones flying over foreign territories sometimes referred to young children appearing in video footage as ‘ TITS’ (terrorists in training): ‘ You did anything you could to remove their humanity’ (Walker 2015). By contrast, a caring disposition is undoubtedly morally superior to a dehumanising one, so the operator of an armed drone should instead seek to respond to rather than resist what the video-camera reveals about humanity and its relational essence. The justifiability of using armed drones will probably continue to depend upon being able to sustain the claim that these weapons are highly discriminate, so there is a military-ethical imperative for the associated camera technology to improve. And as the revelation of others’ humanity to drone operators thereby becomes richer and more detailed, the difficulty of dehumanisation seems likely to increase. Considered in the abstract, the ‘target’ of a drone strike is conceptually isolated from potential civilian ‘collateral damage’. But in reality, the two can be (and be seen to be) connected by interpersonal relationships. Moving beyond the abstraction that characterises a strict jus in bello approach to wartime morality, incorporating an ethics of care would enable greater sensitivity to ‘contextual nuance and particular narratives’ (Held 2006, 10). The drone’s video-camera is highly conducive to such sensitivity because, in the circumstances of a named individual’s ‘targeted killing’, it confers upon the drone operator an unprecedented ability to observe the everyday life of a prospective target playing out in unique fashion. Often, an individual being targeted for a drone strike will continue to live among the broader population, remaining materially enmeshed within a family and/or neighbourhood. After engaging in activities which the drone-using observer might judge as posing a threat, that individual will return to participate in ‘regular’ family life, before perhaps leaving to engage again in those activities (see Levine 2010, 146). This means that, prior to a planned strike, a drone operator will have witnessed at length the various prosaic acts that confirm the target’s humanity. And, more importantly, the spectacle of that humanity will sometimes include the ordinary transactions of family and/or neighbourly relationships. In turn, this might occasion the drone operator to be reminded of her or his own relationships with parents, children or friends, and the moral value thereof. Where the relationships between a prospective target and other humans nearby incorporate a dynamic of caregiving (for example, parenting), the prolonged observation of this dynamic puts the conscientious drone operator on notice: even if civilian injuries and deaths could be avoided in a strike against the targeted individual, the strike would still inflict upon related civilians the non-physical harm of being deprived of a caregiver. A child who survives will have lost a

6686_Enemark.indd 142

02/03/21 12:22 PM

Drone Warriors, Revealed Humanity, a Feminist Ethics of Care / 143

parent, and a surviving wife will have lost a husband (or vice versa). Thereafter, it is reasonable to expect those innocent individuals to be cast perhaps into the care of any surviving relations or, worse, into a life of being uncared for altogether. The material consequence of this form of harm is barely recognised in the aforementioned practice of making condolence payments to civilian survivors of wartime violence. A US military lawyer interviewed by Gregory (2020, 163) explained: If a man dies, and the woman is left with three kids, well I’m going to give her $2,500 because she needs everything she can get; she can’t work because she’s got four mouths to feed, and the money earner is dead.

Clearly, though, this ‘caring’ response to actualised victimhood is inadequate in the long term. From a care ethics perspective, a better response to civilians’ care needs, if they are observed in advance by a prospective perpetrator of violence, is to refrain from knowingly severing caregiving relationships in the first place. In the acquisition of knowledge (through seeing) that a targeted individual is caring for family members on an ongoing basis, a drone operator’s moral responsibility to avoid civilian harm (a loss of care) acquires greater weight as against the moral responsibility of neutralising a military threat. In which case, even when it would be permissible (purely as a jus in bello matter) to kill someone, a drone operator should refrain from doing so out of concern that innocent dependents should continue to be cared for. Critically, this is not a caring response to the targeted individual qua military threat. Rather, it is a caring response to non-threatening others whose vulnerability to the loss of a caregiver has been made clear (via video-camera) to the drone operator. If, for example, the target is evidently the father of a child who relies on him for care, the drone operator in turn should care for that child by sparing the father’s life. In different circumstances, the moral requirement to exercise care in this way would not arise. Where a targeted individual’s interpersonal relationships are not observed in detail by a drone operator, the impulse and duty to care about preserving the caregiving function of those relationships will be absent. Here, it is worth acknowledging that a commonplace use of armed drones is to provide close air support (CAS) to friendly troops who are in contact with the enemy. In this circumstance, any pre-strike observation of a threatening enemy combatant by a drone operator is likely to be fleeting rather than prolonged. Consequently, there is little or no opportunity for the operator to be prompted morally to account for any evident relationships of care that might exist as between a particular enemy combatant and particular

6686_Enemark.indd 143

02/03/21 12:22 PM

144 / Lindsay C . Clark and Christian Enemark

local civilians. Even so, in this context an ethics of care could still be applied in recognition of a drone operator’s relationship with fellow members of her or his military organisation. Reportedly, CAS provision tends to be what drone operators find most satisfying about their work as military professionals. For example, as USAF psychiatrist Colonel Kent McDonald told the New York Times in 2011: ‘ These guys are up above firing at the enemy. They love that, they feel like they’re protecting our people. They build this virtual relationship with the guys on the ground’ (Bumiller 2011). This account accords with Levine’s observation that ‘the warrior ethos’, embodying a strong notion of interdependence, ‘resonates with care ethics’ because warriors are encouraged to ‘look on their comrades-in-arms as something like family’ (Levine 2010, 153). In a COIN context, moreover, Levine (2010, 154) has insisted that ‘the caring relationship’ between counterinsurgents and local civilians does not replace ‘the relationships that counterinsurgents have with their comrades-in-arms’. However, if this is so, it serves as a caution that ‘care’ might not be regarded as ‘innocent’ in wartime in the sense of being necessarily and only for the benefit of potential civilian victims. In combining care ethics with jus in bello morality, it is therefore important to be mindful of a countervailing risk: one warrior’s excessively caring disposition towards her or his comrades might lead to a moral overvaluing of forceprotection. If, in calculating the in bello proportionality of using an armed drone, the value of preserving friendly troops’ lives were afforded too much weight, the zealous wielding of excessively ‘protective’ violence could become harmfully detrimental to any nearby civilians.

Conclusion Most discussions of missile-armed drones focus on their capacity to enable killing across vast geographical distances. In this chapter, we have instead focused on the drone as a camera-equipped instrument of war, because the moral significance of drone violence lies in the potential for someone not only to be killed from afar but also to be closely observed. The video-camera mounted on a drone has an unprecedented power to reveal a distant target’s humanity and its relational essence. This, we argue, compels drone operators to think differently and more deeply about the morality of their violence. The video-powered revelation of humanity puts them on notice that, sometimes, a drone strike would cause physical harm to a targeted individual and non-physical harm (the loss of a caregiver) to innocents in a relationship with that individual. A feminist ethics of care reminds us that, ethically, relationships matter. In contrast to Just War morality, care ethics focuses on people not as abstract

6686_Enemark.indd 144

02/03/21 12:22 PM

Drone Warriors, Revealed Humanity, a Feminist Ethics of Care / 145

others but as real individuals with unique needs, and its emphasis is on persons in relation to each other. A care-oriented mode of moral reasoning, which just happens to be historically associated with feminine stereotypes, is not a lesser mode, just a different one, and it is as valid as other modes. Indeed, to care for others is normal. In the exceptional circumstance of war, however, this ingrained social impulse is largely suppressed through the brutalising process of military training and by ongoing practices of dehumanisation that purportedly make it easier to harm others. When it comes to the wartime use of camera-equipped drones, the restoration of humans’ caring responses is arguably both possible and necessary. In the augmenting of principled jus in bello reasoning with an ethics of care, there may be scope to achieve a greater degree of humane restraint in the exercise of drone violence. For a caring drone operator, who must somehow respond to the on-screen spectacle of others’ caregiving relationships, it would not be enough to act violently in the expectation that innocents would be spared from injury or death. Rather, that operator would be disposed to refrain from taking action if it would clearly deprive those innocents of a critical source of ongoing care. In this way, he or she would honour a concept of civilian harm which is thicker than that which traditionally informs Just War thinking. Provided that a caring disposition is only exercised for the benefit of civilians and not at their expense, the application of a drone warrior’s ethics of care potentially constitutes a moral improvement to the conduct of war.

Notes Research for this chapter was conducted as part of a project funded by the European Research Council under the European Union’s Horizon 2020 research and innovation programme (grant no. 771082): ‘Emergent Ethics of Drone Violence: Toward a Comprehensive Governance Framework’ (DRONETHICS). 1. For present purposes, a ‘drone operator’ is one of the two people directly involved in remotely controlling the aircraft and its payload from a ground station: the sensor operator, who controls the mounted video-camera, and the pilot (seated adjacent) who controls flight. Typically, the sensor operator focuses the camera and aims the targeting laser before the pilot pulls a trigger releasing a missile from the drone, and he or she then maintains aim as the missile descends.

References Abé, Nicola. 2012. ‘Dreams in Infrared: The Woes of an American Drone Operator’, Der Spiegel, 14 December, http://www.spiegel.de/international/world/pain-continuesafter-war-foramerican-drone-pilot-a-872726.html

6686_Enemark.indd 145

02/03/21 12:22 PM

146 / Lindsay C . Clark and Christian Enemark

Adhariani, Desi, Nick Sciulli and Robert Clift. 2017. Financial Management and Corporate Governance from the Feminist Ethics of Care Perspective. Cham: Palgrave Macmillan, https://link.springer.com/book/10.1007%2F978-3-319-33518-6 Alston, Phillip, and Hina Shamsi. 2010. ‘A Killer above the Law?’, The Guardian, 8 February, http://www.theguardian.com/commentisfree/2010/feb/08/afghanistan-drones-defence-killing Amnesty International. 2013. ‘ Will I Be Next?’: US Drone Strikes in Pakistan. London: Amnesty International. Barnes, Marian, Tula Brannelly, Lizzie Ward and Nicki Ward. 2015. ‘Introduction: The Critical Significance of Care’, in Ethics of Care: Critical Advances in International Perspective, edited by Marian Barnes, Tula Brannelly, Lizzie Ward and Nicki Ward. Bristol: Policy Press, 3–19. Bazargan-Forward, Saba. 2017. ‘Compensation and Proportionality in War’, in Weighing Lives in War, edited by Jens Ohlin, Larry May and Claire Finkelstein. Oxford: Oxford University Press, 173–87. Bumiller, Elisabeth. 2011. ‘Air Force Drone Operators Report High Levels of Stress’, New York Times, 19 December, A8. Bumiller, Elisabeth. 2012. ‘A Day Job Waiting for a Kill Shot a World Away’, New York Times, 30 July, A1. Calhoun, Laurie. 2015. We Kill Because We Can: From Soldiering to Assassination in the Drone Age. London: Zed Books. Clark, Lindsay C. 2019. Gender and Drone Warfare: A Hauntological Perspective. New York: Routledge. Coeckelbergh, Mark. 2013. ‘Drones, Information Technology, and Distance: Mapping the Moral Epistemology of Remote Fighting’, Ethics and Information Technology 15: 87–98. Daggett, Cara. 2015. ‘Drone Disorientations: How “Unmanned” Weapons Queer the Experience of Killing in War’, International Feminist Journal of Politics 17 (3): 361–79. Edkins, Jenny. 2016. Missing: Persons and Politics. Ithaca: Cornell University Press. Enemark, Christian. 2019. ‘Drones, Risk, and Moral Injury’, Critical Military Studies 5 (2): 150–67. Freeman, C. 2015. ‘Armchair Killers: Life as a Drone Pilot’, The Telegraph, 11 April, http://www.telegraph.co.uk/culture/film/film-news/11525499/good-killdrone-pilot-true-story.html Friedersdorf, Conor. 2012. ‘“Every Person Is Afraid of the Drones”: The Strikes’ Effect on Life in Pakistan’, The Atlantic, 25 September, https://www.theatlantic. com/international/archive/2012/09/every-person-is-afraid-of-the-drones-thestrikes-effect-on-life-in-pakistan/262814 Frowe, Helen. 2011. The Ethics of War and Peace: An Introduction. London: Routledge. GAO. 2007. ‘ The Department of Defense’s Use of Solatia and Condolence Payments in Iraq and Afghanistan’ (GAO-07–699), United States Government Accountability Office, https://www.gao.gov/new.items/d07699.pdf

6686_Enemark.indd 146

02/03/21 12:22 PM

Drone Warriors, Revealed Humanity, a Feminist Ethics of Care / 147

Gilligan, Carol. 1982. In a Different Voice: Psychological Theory and Women’s Development. Cambridge, MA: Harvard University Press. Gregory, Thomas. 2020. ‘ The Costs of War: Condolence Payments and the Politics of Killing Civilians’, Review of International Studies 46 (1): 156–76. Groenhout, Ruth. 2014. ‘Virtue and a Feminist Ethics of Care’, in Virtues and Their Vices, edited by Kevin Timpe and Craig A. Boyd. Oxford Scholarship Online, https://www. oxfordscholarship.com/view/10.1093/acprof:oso/9780199645541.001.0001/ acprof-9780199645541-chapter-23 Held, Virginia. 2006. The Ethics of Care: Personal, Political, and Global. New York: Oxford University Press. Held, Virginia. 2010. ‘Can the Ethics of Care Handle Violence?’, Ethics and Social Welfare 4 (2): 115–29. Holewinski, Sarah. 2013. ‘Do Less Harm: Protecting and Compensating Civilians in War’, Foreign Affairs, January/February, https://www.foreignaffairs.com/articles/2012-01-01/do-less-harm Hurwitz, Elijah S. 2013. ‘Drone Pilots: “Overpaid, Underworked, and Bored”’, Mother Jones, 18 June, https://www.motherjones.com/politics/2013/06/dronepilots-reaper-photo-essay Kellison, Rosemary. 2019. Expanding Responsibility for the Just War: A Feminist Critique. Cambridge: Cambridge University Press. Lee, Peter. 2019. Reaper Force: The Inside Story of Britain’s Drone Wars. London: John Blake. Lee, Peter. 2020. ‘Iran Attack: How Reaper Drones Really Carry Out Airstrikes’, The Conversation, 7 January, http://theconversation.com/iran-attack-how-reaper-dronesreally-carry-out-airstrikes-129411 Levine, Daniel H. 2010. ‘Care and Counterinsurgency’, Journal of Military Ethics 9 (2), 139–59. Linebaugh, Heather. 2013. ‘I Worked on the US Drone Program. The Public Should Know What Really Goes On’, The Guardian, 29 December, http://www. theguardian.com/commentisfree/2013/dec/29/drones-us-military Marks, Paul. 2012. ‘Smaller, Better, Deadlier?’, New Scientist, 11 August, 26–7. Martin, Matt. 2010. Predator: The Remote-Control Air War over Iraq and Afghanistan: A Pilot’s Story. Minneapolis: Zenith Press. Noddings, Nel. 1984. Caring: A Feminine Approach to Ethics and Moral Education. Berkeley: University of California Press. Obama, Barack. 2013. ‘Remarks by the President at the National Defense University’, The White House, 23 May, https://obamawhitehouse.archives.gov/the-pressoffice/2013/05/23/remarks-president-national-defense-university Robinson, Fiona. 2011. The Ethics of Care: A Feminist Approach to Human Security. Philadelphia: Temple University Press. Rodin, David. 2006. ‘ The Ethics of Asymmetric War’, in The Ethics of War: Shared Problems in Different Traditions, edited by Richard Sorabji and David Rodin. Aldershot: Ashgate, 153–68.

6686_Enemark.indd 147

02/03/21 12:22 PM

148 / Lindsay C . Clark and Christian Enemark

Ruddick, Sara. 1989. Maternal Thinking: Towards a Politics of Peace. London: Women’s Press. Sander-Staudt, Maureen, and Maurice Hamington. 2011. ‘Introduction: Care Ethics and Business Ethics’, in Applying Care Ethics to Business, edited by Maurice Hamington and Maureen Sander-Staudt. Dordrecht: Springer, vii–xxii. Schulzke, Marcus. 2017. Just War Theory and Civilian Casualties: Protecting the Victims of War. Cambridge: Cambridge University Press. Tännsjö, Torbjörn. 2008. Understanding Ethics: An Introduction to Moral Theory, 2nd edn. Edinburgh: Edinburgh University Press. Terry, Jillian. 2015. ‘Ethical Encounters – Care in Counterinsurgency: Feminist Ethics and the Morality of “Winning Hearts and Minds”’, The Disorder of Things, https://thedisorderofthings.com/2015/05/01/ethical-encounters-carein-counterinsurgency-feminist-ethics-and-the-morality-of-winning-heartsand-minds Thomas, Alan. 2011. ‘ Virtue Ethics and an Ethics of Care: Complementary or in Conflict?’, Eidos 14: 132–51. Vogel, Ryan R. 2010. ‘Drone Warfare and the Law of Armed Conflict’, Denver Journal of International Law and Policy 39 (1): 101–38. Walker, Lauren. 2015. ‘Death from Above: Confessions of a Killer Drone Operator’, Newsweek, 19 November, https://www.newsweek.com/confessions-lethal-droneoperator-396541 Walzer, Michael. 2006. Just and Unjust Wars, 4th edn. New York: Basic Books. Widdows, Heather. 2011. Global Ethics: An Introduction. Durham: Acumen. Williams, John. 2015. ‘Distant Intimacy: Space, Drones, and Just War’, Ethics & International Affairs 29 (1): 93–110.

6686_Enemark.indd 148

02/03/21 12:22 PM

EIG HT

Armed Drone Systems: The Ethical Challenge of Replacing Human Control with Increasingly Autonomous Elements Peter Lee

Introduction On 3 January 2020, Iranian General Qasem Soleimani was killed outside Baghdad International Airport by a missile fired from an American drone. In his coverage of the event, BBC reporter Jon Sopel provided an excellent description of what a report of an autonomous drone strike might look like: ‘Once the order was given [by President Trump], an American Reaper MQ-9 drone based in the Gulf state of Qatar was launched at the touch of a button by an operator 7,000 miles away in Nevada’ (Sopel 2020). The impression of a drone acting on its own without active human input was reinforced when he continued: ‘In position over Baghdad airport, it [the drone] fired two Hellfire missiles into a convoy carrying Soleimani’ (Sopel 2020). In reality, a vast human and organisational infrastructure was necessary to get an MQ-9 Reaper into position for a missile strike. Military personnel would service, fuel, arm, pre-flight check, and get the Reaper airborne using a local Launch and Recovery crew. All this would occur before control was transferred via satellite link-up to the US. In addition, this strike on a moving target also required a labour-intensive intelligence system, airspace management, an authorising officer, a joint terminal attack controller, the functions of the chain of command, and the actions of a pilot and sensor operator in firing and controlling each missile (Lee 2020). This chapter explores the practical and ethical challenges involved in moving from the current, human-intensive, remotely piloted military drone strike of the kind that killed Soleimani, towards future artificial intelligence (AI)-enabled drone systems with increasingly autonomous elements as can be imagined from Sopel’s reportage. The first part of the chapter sets out the

6686_Enemark.indd 149

02/03/21 12:22 PM

150 / Peter Lee

practical and ethical challenges of using remotely piloted military drones. The remaining sections will address a range of anticipated practical and ethical challenges as some human functions are increasingly replaced by autonomous elements in the successors to the MQ-9 Reaper and similar military drone systems. These challenges relate to: the contested meaning of ‘autonomy’; degrees of autonomy; AI-based potential bias in lethal autonomous weapon systems (LAWS); and the distribution of accountability in such systems. The discussion concludes with some suggested principles to address in future ethical debates about the development and use of lethal autonomous drones and associated constraints. At the outset, however, it is worth briefly setting out the evolution from piloted aircraft to remote flying, highlighting attempts in the 1920s to constrain bombing from aircraft, as a precursor to contemporary concerns about bombing from remotely piloted or autonomous military drones.

A History of Bombing and Imprecision For more than a century, air warfare has been dominated by pilots and piloted aircraft. The myth of the heroic pilot emerged during the First World War (Harris 1958, 12) and legends were built and sustained in the press, in books and in official propaganda. On 1 April 1918 Lord Rothermere, the British Air Minister, wrote of how pilots of the Royal Flying Corps (RFC) and Royal Naval Air Service (RNAS) had redefined bravery and daring (Rothermere 1918, 8). In that war of national survival, where trench warfare and massive artillery bombardments were resulting in obscene numbers of casualties, the pilot’s ethic was to kill the enemy while avoiding his own death. Major James McCudden, RFC – and later Royal Air Force (RAF) – was one of Britain’s most successful fighter ‘aces’. He wrote of his experience in killing the enemy from behind: ‘I hate to shoot the Hun down without him seeing me, for although this method is in accordance with my doctrine, it is against what little sporting instincts I have left’ (McCudden 2009, 253). In this way, aerial combat began as a personal, almost intimate act conducted at close distances. Bombing of targets on the ground slowly developed through the First World War but the aircraft and munitions used had limited capabilities. Biddle summarised those early developments in bombing from the air as ‘a history of the tension between imagined possibilities and technical realities’ (Biddle 2002, 11). Despite those early limitations, the strategic possibilities attributed to bombing were subsequently prioritised by major powers, to such an extent that there was a concerted international attempt to limit – or ban – bombing from aircraft. A Commission of Jurists met in The Hague between December 1922 and February 1923 to agree ‘Rules

6686_Enemark.indd 150

02/03/21 12:22 PM

Armed Drone Systems: The Ethical Challenge / 151

for Aerial Warfare’ which would provide protection for civilians and focus bombing solely on military targets. Two articles demonstrate the intention of the Rules (Commission of Jurists 1922): Article 24.1. An air bombardment is legitimate only when is [sic] directed against a military objective, i.e. an objective whereof the total or partial destruction would constitute an obvious military advantage for the belligerent; Article 24.3. Any bombardment of cities, towns, villages, habitations and building which are not situated in the immediate vicinity of the operations of the land forces, is forbidden.

However, the potential military usefulness and strategic benefit to the states that had the capacity to develop and use large bombers prevailed against widespread political, social and ethical concerns, and so the Rules were never signed into international law. Gray summarises the eventual outcome: ‘ The conventional view, therefore, was (and remains) that they were a political and legal failure’ (Gray 2010, 16). This failed attempt to limit bombing from aircraft bears some similarity to international discussions in Geneva from 2017 to 2019 by the Group of Governmental Experts on Lethal Autonomous Weapons Systems (GGE on LAWS) (United Nations 2019), to be revisited later in the chapter. The development of mass bombing in the Second World War was not constrained by the efforts of the Commission of Jurists and their 1922–3 deliberations. From the firebombing of Hamburg and Cologne through to the destruction of Dresden, the legacy of the ethical choices and actions involved in the use of widespread aerial bombing have been extensively described and analysed (Wells 2000; Biddle 2002; Gray 2010; Lee 2013). Those mass bombings of cities in a global war have given way over time to the capability for more precise use of airpower in the twenty-first century. Despite that potential for missile and bomb precision, the word ‘precise’ in this context should not distract from the nature of high explosive warheads and their capacity to generate massive destruction and indiscriminate death. Airwars monitors civilian deaths from aircraft bombing in four conflicts, with an emphasis on Syria and Iraq since 2014. At the time of writing, it had recorded 54,109 alleged civilian deaths, thus highlighting the risks to civilians that still result from aerial bombing – precise or otherwise (Airwars 2020).

Remotely Piloted Aircraft and Challenges for Autonomy In the aftermath of the 9/11 attacks, the newly armed, experimental MQ-1 Predator remotely piloted aircraft fulfilled a US operational requirement

6686_Enemark.indd 151

02/03/21 12:22 PM

152 / Peter Lee

for extensive Intelligence Gathering, Surveillance and Reconnaissance (ISR) over Afghanistan (Woods 2015). By 2006 the more advanced MQ-9 Reaper was ready for mass roll-out, providing enhanced loiter time and greater offensive capability than the Predator. In that year the UK acquired the Reaper for the RAF, and it has since been involved in military operations from Afghanistan to Syria and Iraq (Lee 2019b). During its time in RAF service the remotely piloted Reaper drone has operated alongside conventionally piloted aircraft like the Tornado and the Typhoon, and under the same legal, operational and intelligence framework – as confirmed by Jim Shannon MP in a UK Parliamentary debate (UK Parliament 2015): All UK drone pilots have to follow the law of armed conflict and the rules of engagement in exactly the same manner as pilots of traditional manned aircraft. The decision-making process leading to the identification and engagement of targets is identical to that for conventionally manned aircraft.

When the RAF attacked Colonel Gaddafi’s armed forces in Libya in 2011, several of these historic discourses around heroism, precision bombing and a desire to protect civilians were brought together. One UK tabloid newspaper headline and comment made several connections (Wheeler 2011): RAF Top Guns Launch Libya Raids British Top Guns last night launched a series of precision bombing raids on Colonel Gaddafi’s armoured vehicles as they were poised to attack civilians.

Meanwhile, on 11 March 2011 a laser-guided bomb attack on a Taliban truck in Afghanistan by an RAF Reaper resulted in four civilian deaths. This time, the corresponding headline had a different emphasis (Hopkins 2011): Afghan Civilians Killed by RAF Drone Four Afghan civilians were mistakenly killed and two others injured in an attack by a remotely controlled RAF ‘drone’ targeting insurgent leaders in Helmand province.

In the first newspaper the focus was on the people – the ‘ Top Gun’ pilots and navigators (or Weapon System Officers) – who fired the weapons. In the second headline the focus was on the drone, even though the weapon was also fired by a pilot and controlled by a sensor operator. It also does not capture the extent to which the civilian deaths occurred despite the Reaper providing some operational advantages – and potential for precise strikes – over its faster-moving, manned counterparts. As the UK Parliament was reminded in 2015 (UK Parliament 2015):

6686_Enemark.indd 152

02/03/21 12:22 PM

Armed Drone Systems: The Ethical Challenge / 153

Our drone pilots have greater access to information, through a combination of the aircraft’s on-board sensors and the ability to access off-board information, they are the best informed and least pressurised of all our aircrew who have to make critical decisions about when to strike.

A first-hand account of the aforementioned four Afghan civilian deaths by one of the Reaper drone crew members involved highlights the challenges of decision-making in air operations and the need for accurate and timely intelligence information (Lee 2019b). Some elements of that incident also reveal the difficulty of trying to replicate human capabilities with AI-enabled elements in a potential future autonomous drone system. According to the account, two Taliban trucks carrying home-made explosives (HME) were to be targeted in a dry ravine at night. After hitting the first truck with a 500-pound laser-guided bomb the crew orbited the Reaper round into position to bomb the second truck, which was silhouetted against the burning shell of the first. As they approached for the bombing run, the mission intelligence coordinator (MIC) spotted a man silhouetted against the flames on his screen. And beside the man he saw a person half the man’s size. It is worth reading a section of the account to gain a perspective on the challenge of replacing a human’s insight with an ‘intelligent’ machine’s (Lee 2019b, 100–1): ‘Is this a CHILD?’ For a split second he doubts what he is seeing. The picture on the screen does not match the picture he expects to see from the intelligence they have been given. What if he is wrong? Don’t panic. But he could see what he could see, even if it didn’t make sense. ‘Abort. Abort.’ He interrupted the pilot who was focused on the final few seconds of the weapon release, while the [sensor operator] was tense, waiting to guide the next bomb. ‘ What’s the problem?’ asked the pilot, his grip on the trigger snapping loose at the same time. The release process stopped instantly, the pre-strike tension suddenly broken. ‘ There’s a kid down there. A child down there.’ [The MIC] only had the perspective of seeing one person versus another, but even with the limitations of the infrared picture, that comparison between the two figures was clear. He might not have noticed if the child was standing on his own.

The crew had operational authorisation (including legal authorisation) to strike the second truck, just as they had to strike the first truck, described above, causing the civilian deaths (Lee 2019b, 98). Their intelligence briefing was that each truck contained one Taliban driver, but subsequent inquiries found that their decisions and actions rested upon incomplete

6686_Enemark.indd 153

02/03/21 12:22 PM

154 / Peter Lee

or inaccurate or outdated information. The need for ‘accurate, timely and relevant information’ (Ministry of Defence 2017, 30) should be borne in mind as the next section begins to explore the challenges of developing and using autonomous drone systems, or even just drone systems with discrete AI-based autonomous elements within them.

Towards Autonomy In 2018 the US Department of Defense (DoD) took a significant and public step towards autonomous armed drones when it sent out a funding call for a research project entitled ‘Automatic Target Recognition of Personnel and Vehicles from an Unmanned Aerial System Using Learning Algorithms’ (Department of Defense 2018). Its objective would be to (Department of Defense 2018, emphasis added): Develop a system that can be integrated and deployed in a class 1 or class 2 Unmanned Aerial System (UAS) to automatically Detect, Recognize, Classify, Identify (DRCI) and target personnel and ground platforms or other targets of interest. The system should implement learning algorithms that provide operational flexibility by allowing the target set and DRCI taxonomy to be quickly adjusted and to operate in different environments.

The five italicised actions in this statement of requirement – detect, recognise, classify, identify and target – all currently rely on extensive human input and insight. The Reaper MIC in the example above would have been involved in all of these activities in the build-up to the strike that killed the four civilians. A continuation of those activities also ensured that the second strike did not take place and result in the deaths of several more civilians. The Reaper crew had been directed to the location using shared coalition intelligence. The decision to target the trucks would have been taken at a higher level and authorisation for the strike given by the UK Red Card holder at the Combined Air Operations Centre (CAOC). This sequence of events demonstrates the scale of technological challenge in delegating several or all of these processes to AI-enabled automated or autonomous elements within an overall system. One crucial element in this chain of activities is the need for automated facial recognition systems that can provide accurate and predictable identifications, from an airborne drone looking down at an angle, of faces potentially obscured by coverings or adornments. The US DoD Project MAVEN was conceived specifically to address this technological challenge and Google, with its AI capabilities, was contracted to undertake the development. However, the social and political difficulties involved came to

6686_Enemark.indd 154

02/03/21 12:22 PM

Armed Drone Systems: The Ethical Challenge / 155

public prominence when 3,000 Google scientists wrote to their employer and demanded that the company withdraw from Project MAVEN due to its potential contribution to a drone-based LAWS (BBC 2018). Within weeks, Google announced it would not participate in future AI-related research for the DoD (Wakabayashi and Shane 2018). Beyond these public ethical and commercial concerns, difficulties must be overcome to achieve the level of autonomous capability for recognition and targeting that the US DoD is pursuing. The development of fully autonomous drones – and autonomous weapons more broadly – faces many challenges but the following are addressed here: the semantics of ‘autonomy’ (including the issue of degrees of autonomy); and bias avoidance in LAWS.

Autonomy: A Contested Concept Definitions of autonomy have important operational, legal, ethical and political dimensions which go far beyond simple semantics. For example, the UK government’s official position on ‘fully’ autonomous weapons was set out by the Parliamentary Under-Secretary of State for Foreign and Commonwealth Affairs in a 2013 Parliamentary debate on Lethal Autonomous Robotics (UK Parliament 2013, col. 733): [T]he Government of the United Kingdom do not possess fully autonomous weapon systems and have no intention of developing them. Such systems are not yet in existence and are not likely to be for many years, if at all. Although a limited number of defensive systems can currently operate in automatic mode, there is always a person involved in setting the parameters of any such mode. As a matter of policy, Her Majesty’s Government are clear that the operation of our weapons will always be under human control as an absolute guarantee of human oversight and authority and of accountability for weapons usage.

The Under-Secretary went on to state: ‘ We cannot develop systems that would breach international humanitarian law (IHL), which is why we are not engaged in the development of such systems’ (UK Parliament 2013, col. 735). It is worth noting, however, that the UK government’s ‘intention’ did not exclude several possibilities: development of ‘fully autonomous weapon systems’ at some point in the future; the development and deployment of systems that fall short of being ‘fully autonomous’ (UK Parliament 2013, col. 735); and the development of fully autonomous non-weapon systems. Despite these caveats – and the lack of a definition of what ‘fully autonomous’ means – there is also reference to weapons ‘always be[ing] under human control’; not in the sense of individual weapon engagements but ‘as

6686_Enemark.indd 155

02/03/21 12:22 PM

156 / Peter Lee

an absolute guarantee of human oversight and authority and of accountability for weapons usage’. Neither conventional Just War ethics nor compliance with IHL specifies a requirement for continuous human control of weapons (ICRC 2010). However, the point is politically and culturally sensitive, and the lack of direct human control of weapon release is a key component of the argument by the Campaign to Stop Killer Robots (CSKR) that LAWS are ‘abhorrent, immoral, and an affront to the concept of human dignity and principles of humanity’ (CSKR 2018). While there is not the scope here to engage with this ‘human dignity’ argument, Amanda Sharkey provides a useful starting point for further debate (Sharkey 2019, 85; see also Birnbacher 2016; Ulgen 2016): If it is accepted that there are many weapons, artefacts, and human behaviours that are held to be against human dignity, then this itself becomes a reason for not relying too heavily on human dignity in arguments against LAWS, as distinct from other means and weapons of warfare.

Human dignity has been violated through enslavement, rape, torture and myriad abuses throughout the history of war, and it is hard for me to see why LAWS are a special case trumping all other indignities. The CSKR also used similar language to that of the UK government when, in 2013, it called for a ‘pre-emptive and comprehensive ban on the development, production, and use of fully autonomous weapons’ (CSKR 2013). Such fully autonomous weapons were those which had the capability to ‘choose and fire on targets on their own’ (CSKR 2013). It is not clear if the Campaign was simply demanding what the UK government appeared to be offering – human control of weapon systems – or was driven in part by a broader pacifist ideology which opposes all military violence, with ‘fully autonomous weapons’ serving as a campaign focus. Every government, however, will adopt its own position and will not necessarily be open to human control of LAWS, so the CSKR provides a timely global challenge. There are further semantic points of which to be aware in the burgeoning literature on autonomy, which extend to the context of LAWS. For now, however, I suggest that a crucial terminological battle is – or should be – over the word ‘fully’ and what it means in relation to autonomy in (or of ) weapon systems. I have written elsewhere about what can be referred to as AI and autonomy in an idealised philosophical sense and also in the sense of more ‘limited machine autonomy’ (Lee 2018, 302). In the former – idealised autonomy – AI achieves or exceeds human equivalence (the ‘singularity’) in functioning, reasoning and decision-making, and is represented in science

6686_Enemark.indd 156

02/03/21 12:22 PM

Armed Drone Systems: The Ethical Challenge / 157

fiction films. But even these science fiction autonomous robots do not claim or allude to the next and ultimate level of artificial intelligence – artificial general intelligence (AGI). AGI occurs when computer-driven AI reaches a state of self-learning that exponentially increases its knowledge on a continuous, positive-feedback learning loop (see Goertzel and Pennachin 2007; Goertzel 2007; Bostrom 2014). Limited machine autonomy based on AI more closely reflects the practical experience of AI theorists and practitioners. In a military context, a UK Joint Doctrine Note captured that limited scope in 2011 (Ministry of Defence 2011, 2–3): An autonomous system is capable of understanding higher level intent and direction. From this understanding and its perception of its environment, such a system is able to take appropriate action to bring about a desired state. It is capable of deciding a course of action, from a number of alternatives, without depending on human oversight and control, although these may still be present. Although the overall activity of an autonomous unmanned aircraft will be predictable, individual actions may not be.

Recent developments and testing of military technology in the UK indicate the future trajectory when it comes to LAWS. In 2016, the Royal Navy hosted Exercise UNMANNED WARRIOR which, in conjunction with forty industry and other partners, enabled the testing of ‘unmanned and autonomous vehicles . . . to remove human operators from the most tedious, unpleasant and dangerous activities, such as mine-laying and recovery and anti-submarine operations’ (Royal Navy 2016). These practical applications of emerging technology are consistent with the UK government’s stated aims set out above. In 2018, the British Army hosted Exercise AUTONOMOUS WARRIOR with the specific aim of ‘test[ing] and evaluat[ing] the effectiveness of robotic and autonomous systems (RAS) on the battlefield’ (Ministry of Defence 2018). Defence Minister Mark Lancaster reinforced the vision for the future place of autonomy in UK military doctrine and practice when he said: ‘Autonomous Warrior sets an ambitious vision for Army operations in the 21st Century as we integrate drones, unmanned vehicles and personnel into a world-class force for decades to come’ (Ministry of Defence 2018). While the language of autonomy is used more prominently here than in reference to Exercise UNMANNED WARRIOR in 2016, there is still no explicit mention of LAWS that could be deployed in an attack role, let alone a ‘fully autonomous weapon system’ of the kind that both UK policy and the CSKR reject.

6686_Enemark.indd 157

02/03/21 12:22 PM

158 / Peter Lee

Yet, despite this flurry of activity – a mere glimpse of worldwide autonomous systems development – agreed definitions are no closer. A small sample of relevant literature highlights the extent of the challenges, definitions and practicality of LAWS. In a US military-focused context, Massie has discussed the extent of autonomy, the role of AI, machine–human cooperation and the importance of trust in systems (Massie 2016). In 2014, the Birmingham Policy Commission examined drone use in the UK and discussed many of the challenges of LAWS at great length. The Commission challenged the UK government to ‘take a leading role in discussions to build an international consensus around a set of norms to regulate, if not ban, LAWS’ (Birmingham Policy Commission 2014, 7). It helpfully defined LAWS as ‘ones that have the following properties: automation, volition, and intention’ (Birmingham Policy Commission 2014, 19, original emphasis). These properties were sourced from William and McNail, who expand further (William and McNail 2013, 1150–1, original emphasis): Autonomy also requires some decision-making agency, which is captured by volition, or ‘choice in action or thought’, and intent, or deliberate ‘pursuit of goals’. Truly autonomous machines may also actually be able to learn, meaning they can draw conclusions based on past experience and incorporate these lessons into future actions. This baseline distinction between automation and autonomy offers a useful starting point.

These concepts of automation, volition and intention get even more interesting – and have attributed to them philosophical capabilities that are typically identified as human – when traced back even further to Clark’s 1999 theorising on ‘cyborged ecosystems’ (Clark 1999, 113–14). The agency required for entity autonomy includes ‘independence of comportment’ and ‘a sufficiently conscious mind’; the necessary automation includes ‘the capacity to operate without outside intervention’; volition extends to ‘defining its own goals and then formulating and executing strategies for attaining them’; before, finally, ‘in order to be significantly autonomous an entity must be intentful, and actually exercise its volition’ (Clark 1999, 113–14). Perhaps increasing the usefulness of these concepts is Clark’s contention that ‘autonomy should be measured on a continuous scale’ (Clark 1999, 113). At one extreme we would find ‘truly autonomous’ systems or, to use more recent terminology, ‘fully autonomous’ systems referred to above by the UK government, the CSKR and the Birmingham Policy Commission, as well as by Human Rights Watch (Human Rights Watch

6686_Enemark.indd 158

02/03/21 12:22 PM

Armed Drone Systems: The Ethical Challenge / 159

2012, 42). Lower levels of autonomy would still be available but would go beyond mechanised automation. Clark’s insights undermine an important statement in the Birmingham Policy Commission report, where it says: ‘Put simply, a weapons system is either autonomous or it is not – there is no spectrum of autonomy’ (Birmingham Policy Commission 2014, 66). The proposition that emerges from this brief discussion is that it is both practically and conceptually possible to have LAWS that fall short of being ‘fully autonomous’, however that contested phrase is understood. Regardless of how autonomy is defined or the degree of autonomy that any particular system is anticipated to achieve, other ethical challenges emerge from the technical challenges involved in trying to replicate human function and performance in a machine. One of these challenges is potential bias in any future autonomous system.

Bias Avoidance in LAWS A significant challenge for operationalising AI in LAWS is the propensity for human bias to be programmed – inadvertently or otherwise – into its component algorithms. There are therefore two linked tasks for software engineers, psychologists and ethicists: understanding the nature and potential for human bias; and coding operational software in such a way as to avoid inputting that bias into a LAWS. Such bias could have serious repercussions when distinguishing, for example, between combatants and noncombatants, or even in making judgements about the legitimacy of LAWS in the first place. Perhaps the greatest bias that a person might have – if they are even aware of the human propensity for bias – is a sense that it does not affect them; that their own rationality can keep bias at bay. When it comes to LAWS, bias operates on different levels: at a policy level there is the question of whether they should be allowed to exist, while at an operational level there are questions about how they can be used in ethical ways. Almost fifty years ago, Tversky and Kahneman described how bias in imagining the unknown can inform the extent to which an activity might be perceived as risky (Tversky and Kahneman 1973, 18): The risk involved in an adventurous expedition, for example, is evaluated by imagining contingencies with which the expedition is not equipped to cope. If many such difficulties are vividly portrayed, the expedition can be made to appear exceedingly dangerous . . . Conversely, the risk involved in an undertaking may be grossly underestimated if some possible dangers are either difficult to conceive, or simply do not come to mind.

6686_Enemark.indd 159

02/03/21 12:22 PM

160 / Peter Lee

These words are worth considering in the context of potentially building and operating a LAWS. Discussion around LAWS necessarily involves imagination because future systems that are being conceived and developed do not exist yet, even though the legal, ethical and operational challenges must be considered during the ongoing developmental process. Take two possible opposed views. On the one hand there is implacable opposition to LAWS, where they are ‘made to appear exceedingly dangerous’ (Tversky and Kahneman 1973, 18), drawing on science fiction tropes and imaginings that are informed by films like Terminator and I, Robot. On the other hand, technical experts and experienced military figures might be less concerned about the potential of LAWS, perhaps because ‘some possible dangers are either difficult to conceive, or simply do not come to mind’ as a result of familiarity with the use of lethal force in a military context and the multiple legal and practical constraints that they operate within (Tversky and Kahneman 1973, 18). It seems highly unlikely that either position, as they have been exaggerated here, is without bias. When it comes to using LAWS, bias has the potential to surface in different guises. One potential widespread risk is that the subjective bias of the software coder is somehow encoded into the system through the particular lines of code that are used as the building blocks of that system’s autonomous elements. As long ago as 1996, Friedman and Nissenbaum highlighted three different categories of bias in computer systems: ‘preexisting bias, technical bias, and emergent bias. Preexisting bias has its roots in social institutions, practices, and attitudes. Technical bias arises from technical constraints or considerations. Emergent bias arises in a context of use’ (Friedman and Nissenbaum 1996, 332). Each of these types of bias is a field of study on its own, so we should consider the implications for LAWS if biased coding was to influence their operations. As Clark observes, ‘complete independence in an entity requires a structure that is free of any implicit design objectives or behavioral biases that might influence the definition or pursuit of goals’ (Clark 1999, 114). One of the potential challenges of LAWS is the adoption of sensor systems that have different degrees of recognition effectiveness across different age, gender and racial characteristics. Introna and Wood point out algorithms that display evidence of bias in facial recognition: males being more accurately recognised than females, and older people being easier to recognise than young people (Introna and Wood 2004, 190). These authors were developing the work of Givens and his colleagues, whose study found that Asians, African Americans and ‘other race members are [all] easier to recognise than whites’ (Givens et al. 2003, 8). Even setting aside additional practical difficulties like face coverings and the wearing of spectacles, there are clear

6686_Enemark.indd 160

02/03/21 12:22 PM

Armed Drone Systems: The Ethical Challenge / 161

ethical consequences for the potential uses of LAWS and the identification of targets if the reliability of facial recognition is greater in some parts of the world than in others. In an assessment of the Metropolitan Police Service’s (MPS) trial of live facial recognition (LFR) technology in London, Fussey and Murray highlighted failures and potential risks in the system which have implications for LAWS (Fussey and Murray 2019). Watch-list accuracy poses such a risk, for example where the system correctly identifies a person on the list but for a minor offence which would not normally be deemed serious enough to warrant a place on the watch-list in the first place (Fussey and Murray 2019, 11). This may have an impact on the liberty of a criminal in London, but the consequences are potentially more lethal for anyone on a watch-list to be targeted by a LAWS (for example, an autonomous armed drone). Even more problematic is the reported low-level accuracy of the system in recognising faces (Fussey and Murray 2019, 75): Across the six test deployments MPS officers engaged with 22 individuals as a direct result of a computer-generated match judged to be credible by a human operator. Fourteen of these (63.64%) were verified incorrect matches, eight were verified correct matches (36.36%).

Raising further ethical questions about using AI to identify human targets in a LAWS, an AI system used by American judges to predict if an offender is likely to reoffend in the future is alleged to be biased towards identifying racial minorities: ‘ The formula was particularly likely to falsely flag black defendants as future criminals, wrongly labelling them this way at almost twice the rate as white defendants’ (Angwin et al. 2016). Transposing this degree of potential bias into an intelligence gathering context has obvious risks in terms of ensuring that individuals are not wrongly criminalised or targeted. Even if a misidentification was recognised and cancelled as part of the process of positively identifying targets, any such errors in an AI system would create and increase time inefficiency. Despite these numerous potential drawbacks, in 2020 the Metropolitan Police in London announced that it intended to go ‘live’ with active facial recognition (Dodd 2020). The challenges of obtaining accurate facial recognition have not yet been fully resolved even in static, street-level police cameras where head-on, direct images and video footage of faces can be obtained. There will be even greater difficulty in acquiring consistently accurate facial recognition from a moving drone flying hundreds or thousands of feet above ground level, where surveillance cameras are looking down at acute angles.

6686_Enemark.indd 161

02/03/21 12:22 PM

162 / Peter Lee

Future Challenges of Accountability in Increasingly Autonomous Systems The technical difficulties of incorporating operational and ethical accountability into LAWS are vast, and the associated ethical challenges are deeply complex. In 2018 the GGE on LAWS suggested ten possible guiding principles. The first three principles are enough to indicate the extent of the challenges that states face if they are to develop and use IHL-compliant LAWS or even systems with increasingly autonomous elements that fall short of full autonomy (GGE 2018, 4): 26. It was affirmed that international law, in particular the United Nations Charter and international humanitarian law (IHL) as well as relevant ethical perspectives, should guide the continued work of the Group. Noting the potential challenges posed by emerging technologies in the area of lethal autonomous weapons systems to IHL, the following were affirmed, without prejudice to the result of future discussions: (a)

(b)

(c)

International humanitarian law continues to apply fully to all weapons systems, including the potential development and use of lethal autonomous weapons systems. Human responsibility for decisions on the use of weapons systems must be retained since accountability cannot be transferred to machines. This should be considered across the entire life cycle of the weapons system. Accountability for developing, deploying and using any emerging weapons system in the framework of the CCW [UN Convention on Certain Conventional Weapons] must be ensured in accordance with applicable international law, including through the operation of such systems within a responsible chain of human command and control.

As general principles these appear at first glance to be unproblematic: IHL, the UN Charter and relevant ethical perspectives apply; human responsibility and accountability continue; and a suitable chain of command and control is needed. However, as the preceding discussion has highlighted, some aspects of AI make accountability more difficult. For example, if a self-learning algorithm is part of an autonomous element of a future drone system – say a facial recognition programme – where will ethical and legal responsibility lie for a misidentification that results in a civilian death? With the programmer? The manufacturer? The military officer who deploys the system or the political leader who approved the deployment? Taking a UK example, one potential hierarchy is suggested by Table 8.1, in which a lethal autonomous drone system (third column) is set alongside current and historical comparators (Lee 2019a).

6686_Enemark.indd 162

02/03/21 12:22 PM

Armed Drone Systems: The Ethical Challenge / 163

Table 8.1 Military-political hierarchies of airpower command. Adapted from Lee (2019a). British Second World War bombing command hierarchy

UK Reaper MQ-9 remotely piloted aircraft system command hierarchy

Theoretical autonomous weapon system command hierarchy

1. Prime Minister 2. Defence Minister 3. Chairman of the Chiefs of Staff Committee 4. Chief of the Air Staff 5. Air Officer Commandingin-Chief, Bomber Command 6. Bomber squadron commanders 7. Bomber captain

1. 2. 3. 4. 5.

1. 2. 3. 4. 5.

Prime Minister Defence Minister Chief of Defence Staff Chief of the Air Staff Air Officer Commanding 1 Group 6. Reaper squadron commanders 7. Reaper captain (with continuous remote access to multiple support elements and resources)

Prime Minister Defence Minister Chief of Defence Staff Chief of the Air Staff Air Officer Commanding 1 Group 6. Lethal Autonomous Aircraft System squadron/ military commanders 7. Lethal Autonomous Aircraft System – offensive and/or countermeasures (plus constellation of ethically implicated actors)

While much of the military-political hierarchy would remain unchanged in a future system, the elements that would enable an autonomous system – or a system comprised increasingly of autonomous functions – to be deployed become more complex and distributed away from a linear hierarchy. The Canberra Working Group has compared the military infrastructure, system and mission checks required for conventional manned military bombers and remotely piloted aircraft (RPA) with missile or bomb capability, such as the MQ-9 Reaper (see Table 8.2). The third column of the table (under ‘System’) sets out the elements of the military infrastructure, system checks and mission checks that would still be required for an autonomous aircraft system to operate (Canberra Working Group 2019, 6). The third column (under ‘Mission’) also begins to identify the areas where ethical responsibility diverges from a conventional military hierarchical approach to implicate individuals who contribute to the operation of autonomous elements within the overall system (for example, ‘the Authorising Officer’). In current RPA systems the crew has final ethical, legal and operational responsibility for killing designated human targets or destroying objects (Lee 2019b). Other humans in that decision-making process and ‘kill chain’ include intelligence officers, lawyers, commanders and attack controllers. The prospect of autonomous drones making their own decisions – within predefined, programmed parameters – and targeting enemy combatants would change that dynamic radically. But that does not mean that ethical responsibility would somehow just disappear if human Reaper (or equivalent) drone pilots and

6686_Enemark.indd 163

02/03/21 12:22 PM

6686_Enemark.indd 164

02/03/21 12:22 PM

Mission • Weapon payload • Meteorology report for operational area • Operations Intelligence update • Aircraft tasking (ISR/weapon strike) • Special instructions (warnings about other aircraft activities in the area) • Aircraft ‘signed out’ to the Captain (pilot) by the Authorising Officer • Crew walks to the aircraft

Conventional military bomber aircraft preflight checks, authorisations and ‘sign-out’ System • Engine serviceability and service record • Fuel status • Airframe serviceability • Avionics • Sensor systems (including cameras) • Communications systems • Control systems • Navigation systems • IT updates confirmed

Mission • Weapon payload • Meteorology report for operational area • Operations Intelligence update • Aircraft tasking (ISR/weapon strike) • Special instructions (warnings about other aircraft activities in the area) • Aircraft ‘signed out’ to the Captain (pilot) by the Authorising Officer • Crew walks to the Ground Control Station

Military RPA pre-flight checks, authorisations and ‘sign-out’ System • Engine serviceability and service record • Fuel status • Airframe serviceability • Avionics • Sensor systems (including cameras) • Communications systems • Control systems • Navigation systems • IT updates confirmed Mission • Weapon payload • Meteorology report for operational area • Operations Intelligence update • Aircraft tasking (ISR/weapon strike) • Special instructions (warnings about other aircraft activities in the area) • Operation parameters programmed into operating system, incorporating engagement criteria • System backups confirmed • Aircraft ‘signed out’ by the Authorising Officer (who is responsible for ensuring that all of the above checks and procedures have been carried out prior to deployment)

Military autonomous aircraft pre-flight checks, authorisations and ‘sign-out’ System • Engine serviceability and service record • Fuel status • Airframe serviceability • Avionics • Sensor systems (including cameras) • Communications systems • Control systems • Navigation systems • IT updates confirmed

Table 8.2 Elements of system checks and mission checks. Canberra Working Group (2019, 6).

Armed Drone Systems: The Ethical Challenge / 165

sensor operators were removed. Ethical, legal and operational responsibility would be increasingly distributed to other people. These include AI scientists, self-learning algorithm writers, software coders, systems engineers, optics experts, image recognition developers, squadron commanders and others in the military hierarchy. Even with autonomous elements in a drone system, the human and organisational infrastructure will still be substantial.

Future Directions Recognising the scale of future challenges, and against the backdrop of the GGE on LAWS discussions at the UN, in 2019 I was one of eleven technical, legal, applied ethics and military specialists who drafted a future-looking discussion document entitled ‘Guiding Principles for the Development and Use of LAWS’ (Canberra Working Group 2019). The six principles outlined by the Canberra Working Group as a contribution to current and future debate about the development and use of armed autonomous drones and other LAWS are (Canberra Working Group 2019): Principle 1. International Humanitarian Law (IHL) Applies to LAWS Principle 2. Humans are Responsible for the Employment of LAWS Principle 3. The Principle of Reasonable Foreseeability Applies to LAWS Principle 4. Use of LAWS Should Enhance Control over Desired Outcomes Principle 5. Command and Control Accountability Applies to LAWS Principle 6. Appropriate Use of LAWS is Context Dependent

The difficulties involved in replacing even some elements of currently remotely piloted Reapers or other large armed drones with AI vary in complexity and should not be underestimated. The most straightforward AI enhancements would involve basic machine functions, such as taking off and landing. In 2019 a team from the Technical University of Munich demonstrated a light aircraft landing autonomously without control either from its pilot or from instrument landing systems that enable airliners to make automated landings (Coldeway 2019). A much more difficult challenge is to create LAWS that can make their own AI-based decisions to fire weapons while conforming to conventional war ethics, IHL and Laws of Armed Conflict (LOAC). In 2019 Turkey was reported to be on the verge of becoming ‘the first nation to use drones able to find, track and kill people without human intervention’ (Hambling 2019). This development highlights the importance of the ongoing GGE on LAWS discussions in Geneva and of wider debate about the principles which will shape future governance of lethal

6686_Enemark.indd 165

02/03/21 12:22 PM

166 / Peter Lee

autonomous weapon systems. The speed and extent of military drone proliferation generate social, military, ethical and legal challenges for states, non-state groups and international organisations. By 2019, ninety-five countries were identified as having ‘an active drone inventory’, representing a 58 per cent increase in ten years (Gettinger 2019, viii). The example from Turkey encapsulates the future challenge of lethal autonomous drones in microcosm: the mass proliferation of advanced weapon technology with actual or potential autonomous capabilities, the political will to use such weapons, a military context for their deployment, and the absence of an internationally agreed ethical or legal framework for their development and use. If lethal autonomous drones – large or small – emerge and proliferate without recourse to constraining principles that are rooted in the ethics and laws of war, it might not mark the end of humanity, but it will surely reflect the worst of humanity.

Note In this chapter, the section headed ‘Autonomy: A Contested Concept’ first appeared in an earlier article (Lee 2019a), published under the UK’s Open Government Licence v3.

References Airwars. 2020. Airwars, updated April, https://airwars.org Angwin, Julia, Jeff Larson, Surya Mattu and Lauren Kirchner. 2016. ‘Machine Bias’, ProPublica, May, https://www.propublica.org/article/machine-bias-riskassessments-in-criminal-sentencing BBC. 2018. ‘Google Staff Challenge Pentagon Project’, BBC News, 5 April, https:// www.bbc.com/news/business-43656378,. Biddle, Tami Davis. 2002. Rhetoric and Reality in Air Warfare: The Evolution of British and American Ideas about Strategic Bombing, 1914–1945. Princeton and Oxford: Princeton University Press. Birmingham Policy Commission. 2014. The Security Impact of Drones: Challenges and Opportunities for the UK. Birmingham: University of Birmingham. Birnbacher, Dieter. 2016. ‘Are Autonomous Weapon Systems a Threat to Human Dignity?’, in Autonomous Weapons Systems: Law, Ethics, Policy, edited by Nehal C. Bhuta, Susanne Beck, Robin Geiß, Hin-Yan Liu and Klaus Kreß. Cambridge: Cambridge University Press, 105–21. Bostrom, Nick. 2014. Superintelligence: Paths, Dangers, Strategies. Oxford: Oxford University Press. Canberra Working Group. 2019. ‘Guiding Principles for the Development and Use of LAWS’, UNOG, https://www.unog.ch/__80256ee600585943.nsf/(httpPages)/5535 b644c2ae8f28c1258433002bbf14?OpenDocument&ExpandSection=6#_Section6

6686_Enemark.indd 166

02/03/21 12:22 PM

Armed Drone Systems: The Ethical Challenge / 167

Clark, O. Grant. 1999. Characterization of Cyborged Ecosystems. Montreal: McGill. Coldeway, Devin. 2019. ‘ Watch a Plane Land Itself Truly Autonomously for the First Time’, TechCrunch, 5 July, https://techcrunch.com/2019/07/05/watch-a-planeland-itself-truly-autonomously-for-the-first-time Commission of Jurists. 1922. ‘Rules Concerning the Control of Wireless Telegraphy in Time of War and Air Warfare’, International Committee of the Red Cross, https:// ihl-databases.icrc.org/applic/ihl/ihl.nsf/INTRO/275 CSKR. 2013. ‘Urgent Action Needed to Ban Fully Autonomous Weapons’, Campaign to Stop Killer Robots, 23 April, https://www.stopkillerrobots.org/wp-content/ uploads/2013/03/KRC_LaunchStatement_23Apr2013_fnl.pdf CSKR. 2018. ‘Majority Call to Negotiate a New Treaty’, Campaign to Stop Killer Robots, 31 August, https://www.stopkillerrobots.org/2018/08/sixthmeeting Department of Defense. 2018. ‘Automatic Target Recognition of Personnel and Vehicles from an Unmanned Aerial System Using Learning Algorithms’, SBIR STTR: America’s Seed Fund, https://www.sbir.gov/sbirsearch/detail/1413823 Dodd, Vikram. 2020. ‘Met Police to Begin Using Live Facial Recognition Cameras in London’, The Guardian, 24 January, https://www.theguardian.com/technology/2020/ jan/24/met-police-begin-using-live-facial-recognition-cameras Friedman, Batya, and Helen Nissenbaum. 1996. ‘Bias in Computer Systems’, ACM Transactions on Information Systems 14 (3): 330–47. Fussey, Pete, and Daragh Murray. 2019. Independent Report on the Metropolitan Police Service’s Trial of Live Facial Recognition Technology. Colchester: Human Rights Centre, University of Essex, https://48ba3m4eh2bf2sksp43rq8kk-wpengine. netdna-ssl.com/wp-content/uploads/2019/07/London-Met-Police-Trial-ofFacial-Recognition-Tech-Report.pdf Gettinger, Dan. 2019. The Drone Databook. New York: Center for the Study of the Drone, Bard College, https://dronecenter.bard.edu/files/2019/10/CSD-DroneDatabook-Web.pdf GGE. 2018. ‘Report of the 2018 Session of the Group of Governmental Experts on Emerging Technologies in the Area of Lethal Autonomous Weapons Systems’ (CCW/GGE.1/2018/3). Geneva: United Nations, https://www.unog. ch/80256EDD006B8954/(httpAssets)/20092911F6495FA7C125830E003F9A5 B/$file/CCW_GGE.1_2018_3_final.pdf Givens, Geof, J. Ross Beveridge, Bruce A. Draper and David Bolme. 2003. ‘A Statistical Assessment of Subject Factors in the PCA Recognition of Human Faces’, 2003 Conference on Computer Vision and Pattern Recognition Workshop, https:// ieeexplore.ieee.org/document/4624359 Goertzel, Ben. 2007. ‘Human-Level Artificial General Intelligence and the Possibility of a Technological Singularity: A Reaction to Ray Kurzweil’s The Singularity Is Near, and McDermott’s Critique of Kurzweil’, Artificial Intelligence 171: 1161–73. Goertzel, Ben, and Cassio Pennachin. 2007. Artificial General Intelligence. Berlin: Springer. Gray, Peter. 2010. ‘ The Gloves Will Have to Come Off: A Reappraisal of the Legitimacy of the RAF Bomber Offensive Against Germany’, Air Power Review 13 (3): 9–40.

6686_Enemark.indd 167

02/03/21 12:22 PM

168 / Peter Lee

Hambling, David. 2019. ‘Autonomous Killer Drones Set to Be Used by Turkey in Syria’, New Scientist, 20 September, https://www.newscientist.com/article/2217171autonomous-killer-drones-set-to-be-used-by-turkey-in-syria Harris, John Norman. 1958. Knights of the Air. London: Macmillan. Hopkins, Nick. 2011. ‘Afghan Civilians Killed by RAF Drone’, The Guardian, 5 July, https://www.theguardian.com/uk/2011/jul/05/afghanistan-raf-drone-civiliandeaths Human Rights Watch. 2012. ‘Losing Humanity: The Case against Killer Robots’, Human Rights Watch, https://www.hrw.org/sites/default/files/reports/arms1112 ForUpload_0_0.pdf ICRC. 2010. ‘ The Geneva Conventions of 1949 and Their Additional Protocols’, International Committee of the Red Cross, 29 October, https://www.icrc.org/en/ doc/war-and-law/treaties-customary-law/geneva-conventions/overview-genevaconventions.htm Introna, Lucas, and David Wood. 2004. ‘Picturing Algorithmic Surveillance: The Politics of Facial Recognition Systems’, Surveillance & Society 2 (2/3): 177–98. Lee, Peter. 2013. ‘Return from the Wilderness: An Assessment of Arthur Harris’s Moral Responsibility for the German City Bombings’, Air Power Review 16 (1): 70–91. Lee, Peter. 2018. ‘Armed Drones: Automation, Autonomy, and Ethical DecisionMaking’, in The Political Economy of Robots: Prospects for Peace and Prosperity in the Automated 21st Century, edited by Ryan Kiggins. Basingstoke: Palgrave Macmillan, 291–316. Lee, Peter. 2019a. ‘An Ethics Framework for Autonomous Weapon Systems’, Air and Space Power Review 22 (3): 60–81. Lee, Peter. 2019b. Reaper Force: The Inside Story of Britain’s Drone Wars. London: John Blake. Lee, Peter. 2020. ‘Iran Attack: How Reaper Drones Really Carry Out Airstrikes’, The Conversation, 7 January, https://theconversation.com/iran-attack-how-reaperdrones-really-carry-out-airstrikes-129411 McCudden, James T. B. 2009. Flying Fury. Newbury: Casemate. Massie, Andy. 2016. ‘Autonomy and the Future’, Strategic Studies Quarterly 10 (2): 134–47. Ministry of Defence. 2011. ‘Joint Doctrine Note 2/11: The UK Approach to Unmanned Aircraft Systems’, Development, Concepts and Doctrine Centre, UK Ministry of Defence, UK Government, https://assets.publishing.service.gov.uk/ government/uploads/system/uploads/attachment_data/file/644084/20110505JDN_2-11_UAS_archived-U.pdf Ministry of Defence. 2017. ‘UK Air and Space Doctrine (Joint Doctrine Publication 0–30), Second Edition’, Development, Concepts and Doctrine Centre, UK Ministry of Defence, UK Government, https://assets.publishing.service.gov.uk/ government/uploads/system/uploads/attachment_data/file/668710/doctrine_ uk_air_space_power_jdp_0_30.pdf

6686_Enemark.indd 168

02/03/21 12:22 PM

Armed Drone Systems: The Ethical Challenge / 169

Ministry of Defence. 2018. ‘British Army Set to Redefine Warfare with Joint Autonomous Warrior’, UK Government, 20 June, https://www.gov.uk/government/ news/british-army-set-to-redefine-warfare-with-joint-autonomous-warrior Rothermere, Lord. 1918. ‘British Airmen’s Daring’, The Times, 1 April. Royal Navy. 2016. ‘Unmanned Warrior’, Royal Navy, https://www.royalnavy.mod. uk/unmannedwarrior Sharkey, Amanda. 2019. ‘Autonomous Weapons Systems, Killer Robots and Human Dignity’, Ethics and Information Technology 21: 75–87. Sopel, Jon. 2020. ‘ Trump’s Decision to Keep UK in the Dark about Soleimani Killing Shows Contempt for the Special Relationship’, Daily Mail, 5 January, https:// www.dailymail.co.uk/news/article-7852845/BBC-NORTH-AMERICA-EDITORJON-SOPEL-Decision-Trumps-White-House.html Tversky, Amos, and Daniel Kahneman. 1973. ‘Judgement Under Uncertainty: Heuristics and Biases’, Oregon Research Institute Research Bulletin 3 (1): 1–33. UK Parliament. 2013. ‘Lethal Autonomous Robotics’, House of Commons Hansard Volume 564. Westminster: UK Parliament, https://hansard.parliament.uk/Commons/2013-06-17/debates/13061744000002/LethalAutonomousRobotics UK Parliament. 2015. ‘Armed Drones Debate’, House of Commons Hansard Volume 603. Westminster: UK Parliament, https://hansard.parliament.uk/ Commons/2015-12-01/debates/15120142000001/ArmedDrones#contribut ion-15120142000090 Ulgen, Ozlem. 2016. ‘Human Dignity in an Age of Autonomous Weapons: Are We in Danger of Losing an Elementary Consideration of Humanity?’, European Society of International Law Conference Paper Series 8 (9): 1–19. United Nations. 2019. ‘2019 Group of Governmental Experts on Lethal Autonomous Weapons Systems (LAWS)’. Geneva: United Nations, https://www.unog. ch/80256EE600585943/(httpPages)/5535B644C2AE8F28C1258433002BBF14 ?OpenDocument Wakabayashi, Daisuke, and Scott Shane. 2018. ‘Google Will Not Renew Pentagon Contract that Upset Employees’, New York Times, 1 June, https://www.nytimes. com/2018/06/01/technology/google-pentagon-project-maven.html Wells, Mark K. 2000. Courage and Air Warfare: The Allied Aircrew Experience in the Second World War. London: Frank Cass. Wheeler, Virginia. 2011. ‘RAF Top Guns Launch Libya Raids’, The Sun, 24 March, https://www.thesun.co.uk/archives/news/448608/raf-top-guns-launch-libya-raids William, Marra, and Sonia K. McNail. 2013. ‘Understanding the Loop: Regulating the Next Generation of War Machines’, Harvard Journal of Law and Public Policy 36 (3): 1139–85. Woods, Chris. 2015. Sudden Justice: America’s Secret Drone Wars. Oxford: Oxford University Press.

6686_Enemark.indd 169

02/03/21 12:22 PM

NINE

Autonomous Armed Drones and the Challenges to Multilateral Consensus on Value-Based Regulation Thompson Chengeta

Introduction In multilateral discussions on arms control, non-governmental organisations (NGOs) – believing in the purity and inviolability of the values they seek to salvage – are often pitted against powerful states that seem certain of the supremacy of their own cause. In each party’s view, no other cause is more noble and, before long, discussions degenerate into a ‘them versus us’ tug of war. Yet those who normally bear the brunt of the outcome of such discussions, the ordinary people of the world, are far removed from the discussion table, sometimes even unaware that such discussions are taking place. Such may be the case in the current United Nations debate on autonomous armed drones (AADs). Over the years, the use of armed drones in armed conflict and law enforcement situations has become a common occurrence (Akande et al. 2016). While far removed from the battlefield or point of impact, drone pilots project lethal and harmful force on their targets. Targeted killings through armed drones have also resulted in deaths of civilians (Columbia Law School 2012). Whereas humans remotely pilot armed drones, on the horizon are AADs – also known as killer robots, lethal autonomous weapon systems (LAWS) or simply autonomous weapon systems (AWS) – which, once activated, are able to decide who to kill or harm without any further human control (Heyns 2013a). AADs raise far-reaching legal, ethical and security concerns. On the legal front are questions about whether AADs are lawful weapons; whether they can be used in compliance with international humanitarian law (IHL) and international human rights law (IHRL), and

6686_Enemark.indd 170

02/03/21 12:22 PM

Autonomous Armed Drones and the Challenges to Consensus / 171

whether the use of AADs will create an accountability gap in violation of international laws on responsibility (Heyns 2013a). Some ethicists have questioned whether it is morally acceptable for AADs to make decisions about who lives or dies (Asaro 2016). Furthermore, it is argued that the advent of AADs will make it too easy for states to use force, thereby undermining international peace and security (Heyns 2013a). In view of the above concerns, in 2013 the issue of AADs was tabled for discussion in the UN Human Rights Council (Heyns 2013a). States resolved that discussions on AADs should occur within the framework of the UN Convention on Certain Conventional Weapons (CCW) which subsequently established the UN Group of Governmental Experts (GGE) on AADs. The CCW is a treaty that requires consensus among its member states for any form of regulation to be adopted. Since 2014, there has not been consensus on major issues relating to AADs, and progress towards addressing concerns has been slow. The Campaign to Stop Killer Robots (CSKR) has criticised the CCW process on AADs for ‘moving forward at a snail’s pace’ (CSKR 2019; Delcker and Gray 2020). The lack of consensus and slow progress have largely been attributed to a lack of shared definitions and common understandings of AADs. Given that some of the critical concerns on, and objections to, the use of AADs are anchored on our values – values that are arguably subjective – what are the chances of a regulatory framework emerging from this UN multilateral process? In addressing that question, this chapter discusses factors that may better explain the current lack of progress and consensus in the CCW. Importantly, the chapter’s discussion is premised on the position that a new, legally binding instrument is required to address the novel challenges that are posed by AADs (Chengeta 2020). The fundamental question is whether such an instrument can be attained through multilateral consensus given the current challenges, such as: the recent threats to multilateralism in arms control; the arms race in artificial intelligence (AI) military technologies; AADs’ implications for military power and national security; Geneva diplomats’ ‘banphobia’ and anti-NGO sentiments; and relativity and different interpretations of values applicable to AADs. Of course, for a very long time, what constitute values or whether such values are relative or universal has been a subject of debate in sociology, political science and other studies. Suffice to say at this point that, on the question of relativism or universalism of values that are relevant to AADs discussions, I adopt an expansive approach. Nevertheless, before the discussion of values that are relevant to AADs, it is important to locate the discussion within the context of recent threats to multilateral decision-making.

6686_Enemark.indd 171

02/03/21 12:22 PM

172 / Thompson Chengeta

Recent Threats to Multilateral Decision-Making On 22 January 2020, UN Secretary-General António Guterres identified the issue of AADs as one of the four current main threats to world security and recommended high levels of multilateralism and co-operation on the matter (Guterres 2020). Yet recent years have seen a threat to multilateralism from major powers. On 30 January 2020, the Trump Administration announced the United States president’s cancellation of his predecessor’s policy not to use anti-personnel landmines (US Press Secretary 2020). While the US is not party to the 1997 Ottawa Treaty that bans anti-personnel landmines, the Obama administration had made it a policy not to use them. Just as one would criticise AADs for their unpredictability, anti-personnel landmines have been condemned for being unpredictable and indiscriminate (ICRC 2018). Some commentators have strongly condemned the cancellation of the policy by the Trump administration and its impact on multilateralism in disarmament matters (see, for example, HRW 2020). Likewise, on 30 November 2017, the Trump administration – through a Department of Defense policy memo – decided to end the US’s long-standing policy, issued by the Obama administration in 2008, not to use unreliable cluster munitions (HRW 2017), which are prohibited by the 2008 Convention on Cluster Munitions. There has also been an erosion of treaties such as the 1987 Intermediate Range Nuclear Forces Treaty, the falling out with Iran in relation to the 2015 Joint Comprehensive Plan of Action, and strains in the nuclear arms control treaty relationship between the US and Russia. Furthermore, the departure of the US from the UN Human Rights Council, one of the most powerful and influential organs of the UN as far as protection of human rights is concerned, remains a considerable dent in multilateralism. Back in 2013 and before the US withdrawal, the discussion on AADs was first introduced to the UN through the UN Human Rights Council, in which the US participated (Heyns 2013a). One can only wonder what could have been the content of those discussions had the issue been introduced after the US’s departure from the Council. While the United Kingdom has explicitly noted its concerns regarding the US’s unilateral decisions on matters of international importance (House of Lords 2018), its own recent departure from the European Union is also not favourable to the idea of multilateralism. The European Parliament has already made resolutions calling for a ban on AADs (European Parliament 2018) and, with the UK’s departure from the EU, such resolutions may be of diminished importance to the UK government. The current breakdown in multilateralism and strained geopolitics in the area of arms control makes it difficult to reach consensus on AADs.

6686_Enemark.indd 172

02/03/21 12:22 PM

Autonomous Armed Drones and the Challenges to Consensus / 173

In the face of these threats to multilateralism, particularly in the area of AADs, Germany and France initiated the Alliance for Multilateralism (AFM), which is ‘an informal network of countries united in their conviction that a rules-based multilateral order is the only reliable guarantee for international stability and peace’ (AFM 2019a). As one of its initiatives, in 2019 the AFM made a declaration emphasising the importance of multilateralism on AADs to produce an effective, comprehensive, normative and operational framework (AFM 2019a). The declaration also repeats the idea that the CCW is the appropriate forum to discuss the issue of AADs because it ‘seeks to strike a balance between military necessity and humanitarian considerations’ (AFM 2019b). Yet that underlying balance is being threatened by seemingly irreconcilable differences among participating states. It is becoming more apparent that powerful states prefer discussions to occur within the CCW because of the mode by which decisions are made in that forum. As already indicated, any decision on AADs within the CCW will have to be made by consensus. It can also be argued that France and Germany’s 2018 proposal for a political declaration on AADs (instead of a legally binding instrument) was offered in the pursuit of consensus with states such as the US, the UK, Russia and Australia which are strongly opposed to a new legally binding instrument on AADs (PAX 2018). What remains to be seen is whether the desire to salvage multilateralism in the CCW on the issue of AADs at all costs will result in ‘negative consensus’, or whether the pursuit of consensus in international organisations ‘may lead to paralysis and anarchy’ (Claude 1984, 120). It is important to avoid such institutional paralysis in order ‘to ensure the effective working of international organisations’ (Kissack 2010, 59). The CCW has already been criticised for not producing any outcomes on issues of disarmament (Brzozowski 2019), so it is important to highlight how decisions are made in the CCW and in other multilateral institutions where AADs are currently being discussed or may be discussed in the future.

Multilateral Decision-Making on Armed Drones and AADs For a long time, multilateralism through multilateral institutions like the UN, the EU and the African Union (AU) has been at the core of solving issues of international concern such as those relating to international security, disarmament and human rights. While in institutions like the UN Security Council, the most powerful states have the privilege of deciding the course of action on critical matters, majoritarian votes carry the day in institutions like the UN General Assembly (Kissack 2010). By contrast, institutions like the CCW that operate by consensus are in a way considered

6686_Enemark.indd 173

02/03/21 12:22 PM

174 / Thompson Chengeta

to be the middle ground between majoritarianism and privilege (Kissack 2010). Consensus, majoritarianism and privilege as modes of multilateral decision-making originate from ‘the equalitarianism of traditional international law, the majoritarianism of democratic philosophy, and the elitism of European great power diplomacy’ (Claude 1984, 118). While the issue of privilege constantly arises in discussions and decision-making in AADs, the most relevant ones are majoritarian- and consensus-based institutions.

Majoritarian-Based Institutions Majoritarian-based institutions refer to those institutions whose decisions are made on the basis of majority votes by member states. Such institutions, including the UN General Assembly and the UN Human Rights Council, operate on a normative pattern of democracy and are guided by the legal assumption that all sovereign states are equal (Claude 1984, 118). Majoritarianism in multilateral decision-making is considered critical because ‘decision-making by majority vote emphasises the interests of actors affected by the decision’ (Rittberger and Zangl 2006, 89). As such, decisions that are made through majoritarianism ‘do not have to reflect the interests of all the powerful actors involved but rather the interests of a majority of these actors’ (Rittberger and Zangl 2006, 89). There are multilateral treaties banning certain weapons that have been concluded outside multilateral institutions that operate on consensus. For example, the 2017 Treaty on the Prohibition of Nuclear Weapons was adopted by the UN General Assembly following a process outside the CCW. This treaty would probably not have succeeded in the CCW given that the five officially recognised nuclear-armed states – the US, Russia, UK, France and China – had openly rejected the idea of a new treaty prohibiting nuclear weapons. In the case of AADs, the US, Russia and the UK have clearly stated in the GGE that they oppose the negotiation of a new treaty on AADs (PAX 2018). One could make a comparison with how the issues of armed drones and AADs have been handled so far in other multilateral institutions like the UN General Assembly, the UN Human Rights Council, the European Parliament, the African Commission on Human and Peoples’ Rights and the African Group on Disarmament (AGD). When reports on the issues of AADs and armed drones were submitted to both the UN Human Rights Council and the UN General Assembly, the great powers using armed drones and those developing AADs insisted that the issues were in the wrong forum and should rather be discussed in the CCW. Yet the majority of states considered both issues relevant to the mandate of the UN Human Rights Council and proceeded – through majority vote – to adopt reports, containing regulation recommendations, issued by the UN Special Rapporteur on the Promotion

6686_Enemark.indd 174

02/03/21 12:22 PM

Autonomous Armed Drones and the Challenges to Consensus / 175

and Protection of Human Rights and Fundamental Freedoms while Countering Terrorism and the UN Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions (Emmerson 2014; Heyns 2013a; 2013b). The adoption of these reports was a huge step in putting the issues of armed drones and AADs on the agenda of the UN. Yet since 2014, when the issue of AADs was placed before the CCW, not much has been achieved as states struggle to reach consensus even on basic issues like whether the notion of human control of AADs should be on the agenda. Other multilateral institutions that make decisions through majority votes, like the European Parliament, have been able to adopt resolutions requesting the banning of AADs. The AGD and the African Commission on Human Rights have also adopted declarations and statements on AADs (AGD 2018). These institutions took up the issue of AADs only some years after the CCW had started its discussions, yet they have made considerable progress. Nevertheless, it has been pointed out that the weakness of making decisions through majority votes is that it alienates powerful states. Some scholars argue that international laws that are made through majority voting may be useless if they alienate powerful states, even if they are in the minority (Buzan and Jones 1981, 326). The powerful states tend, after all, to be the ones most deeply involved in the development of AADs. As such, treaties that are made with the intention of binding certain states that choose not to be part of the process are liable to be seen as harmless thunderbolts. Furthermore, it is argued that while outputs of majoritarianism such as those from the UN General Assembly, the UN Human Rights Council and the European Parliament on AADs may ‘have a significant normative value in the development of standards of behaviour among the society of states, they are ineffective when it comes to implementing concrete policy change in the short-term’, as they are ‘non-binding and unable to influence the behaviour of states unless they voluntarily accept them’ (Kissack 2010, 59). It is in this sense that in February 2020 the UN High Representative for Disarmament Affairs, Izumi Nakamitsu, noted (Delcker and Gray 2020): If a possible treaty [on AADs] is pushed through by voting, that could happen – thanks to a majority of those countries that do not possess such weapons anyway. The question is what effect that would have? It might be a good expression of what people think, but can you consider that as something effective?

However, there is a counter-argument: even if powerful states choose not to be part of such treaties, the adoption of new treaties creates new international norms that can be used as a basis to pressure (or name and shame) those who are not part of it. Furthermore, such norms may, in the long

6686_Enemark.indd 175

02/03/21 12:22 PM

176 / Thompson Chengeta

run – especially where there is no persistent objector – end up acquiring the status of customary international law, thus becoming binding on everyone. An example can be found in the case of the US and the 1997 Ottawa Treaty banning anti-personnel landmines. Although the US is not party to that treaty, the Obama administration felt the public pressure and ended up making it a policy not to use such weapons. It is in this sense that Miles Kahler, in responding to the criticism that majoritarianism in multilateral decision-making is a toothless force, notes that the fact that powerful states bitterly contest or criticise international organisations that make decisions on the basis of majority vote is proof of the importance of majoritarianism (Kahler 1992, 703). Furthermore, since majoritarianism is characterised by states’ attempts to form majorities through coalition building, it encourages long-lasting multilateralism (Rittberger and Zangl 2006, 89).

Consensus-Based Institutions In criticising majoritarianism, it is often argued that decision-making on matters of importance such as international security should reflect power dynamics in the real political world (Claude 1984, 118). As such, critics suggest that consensus in multilateralism ensures support for decisions, particularly in a highly divided international system (Buzan and Jones 1981, 326). To the same end, Robert Kissack has explained that consensus is critical in multilateral decision-making because ‘states remain closely guarding of their sovereignty and unwilling to be bound by international laws and treaties they have not freely agreed to be party to’ (Kissack 2010, 59). As already indicated, the debate on AADs was first tabled in the UN Humans Rights Council and later moved to the CCW, which subsequently formed the GGE on AADs. The GGE has since discussed several important themes and expressed converging and diverging views. Some of these prominent views of states are reflected in the eleven guiding principles of the GGE that are meant to guide discussions (GGE 2019, 13). Among these is the principle that the CCW – an institution that operates on consensus – is the appropriate forum within which the issue of AADs must be discussed (GGE 2019, 13). Arguably, consensus in a multilateral process like that of the CCW is meant to protect the privilege and interests of powerful states. For example, while the majority of states prefer to negotiate a legally binding instrument on AADs, the US and Russia have blocked the GGE from making such a recommendation. It is understandably within the sovereign rights of the US and Russia to choose not to be bound by a particular international treaty. Yet what may be more difficult to understand is why they can prevent other states choosing which treaties to negotiate or enter. This ability shows

6686_Enemark.indd 176

02/03/21 12:22 PM

Autonomous Armed Drones and the Challenges to Consensus / 177

powerful states’ privilege of holding the world to ransom, even beyond the powers they enjoy in the UN Security Council. Moreover, this counters the argument by critics that decisions made by the majority (without the most powerful states) are ineffective, since those states are clearly worried about such decisions. A good example of the privilege of powerful states in the AADs debate relates to the notion of meaningful human control (MHC), introduced by the NGO Article 36 as a possible solution to the challenges posed by AADs. As a participant in the UN debate on AADs, I have observed the notion of MHC gaining traction, with the majority of states and NGOs involved citing it as the crux of the discussions on AADs. Some scholars consider that the notion of MHC is inherent in existing international law, while others note that it encompasses ethical standards and values (Asaro 2016). The US has criticised the notion of MHC and instead uses the term ‘appropriate levels of human judgment’ (US 2018). Others have used terms such as ‘sufficient human control’. Nevertheless, it does not matter what name or term is used to describe human control because what matters is the substance and standards of that control. In the August 2019 meeting of the GGE, the majority of states proposed that the notion of MHC should be elevated to the status of a guiding principle of the GGE. Once again, the US resisted this idea, leading to the exclusion of the term MHC from the 2019 report of the GGE and the use of the term ‘human–machine interaction’ instead (GGE 2019). While it is understandable that consensus only works with some level of compromise, it is important to note here that the majority of states in the GGE had not suggested that the US’s preferred term ‘appropriate levels of human judgment’ should be excluded from the GGE report. Their suggestion was that the term MHC, supported by the majority, should be included in the report. A reasonable compromise would have been, perhaps, to include both terms. Yet it was the wishes of the US that carried the day. Likewise, an earlier version of the 2019 GGE Report included a paragraph that noted that the use of AADs may compound or worsen social injustices such as racial and gender discrimination. During the discussions, no state representative contested that paragraph. Later in the evening of the same day when another version of the report was provided, the paragraph had been removed. The delegations from South African and Canada questioned why this had occurred, but no remedy was provided and the text addressing discrimination risks remained excluded. The above examples illustrate Inis Claude’s observation that the rule of consensus and unanimity in international organisations as far as decision-making is concerned ‘confers upon a minority of one the procedural competence and the moral authority to determine policy in a negative

6686_Enemark.indd 177

02/03/21 12:22 PM

178 / Thompson Chengeta

fashion’ (Claude 1984, 125). Consensus-based multilateral institutions are indeed sometimes permeated with the privilege of powerful states. It is no wonder, then, that Kissack referred to ‘the immiscibility of privilege and multilateralism’ when questioning whether ‘privilege and multilateralism can be talked about in the same sentence, other than to point out that on first appearances the former appears to be the antithesis of the latter’ (Kissack 2010, 99). In the same sense, CSKR coordinator Mary Wareham noted in February 2020 that ‘the challenge with the CCW is that its method of consensus decision-making ultimately results in compromise text and weaker standards, if agreement is even possible’ (Delcker and Gray 2020). Thus, while it would be ideal to have a ‘universal treaty on killer robots with all the major military powers on board’, there is a concern that the outcome may be too weak to address the concerns that are raised by AADs (Delcker and Gray 2020).

Lack of Consensus on Values Applicable to AADs The challenges raised by AADs are not only about whether these weapons systems are compatible with international law but also about their acceptability in terms of our moral values (ICRC 2018, 7). Indeed, the call for a pre-emptive ban on AADs includes the ethical concern that their use will result in loss of human agency, diffusion of moral responsibility and the violation of human dignity (ICRC 2018, 22). Thus, ethical values are at the crux of the current debate on AADs (ICRC 2018, 20). Nevertheless, there is no consensus among debaters on the application and interpretation of ethics and other values such as those contained in the Martens Clause.1 The Clause stipulates that where existing law is inadequate, fighters and civilians remain under the protection of the laws of humanity and the dictates of public conscience (Koutroulis 2017). The GGE has since stipulated the need for further discussions and clarifications when it comes to consideration of values such as ethics (GGE 2019, 4). On the one hand, there are those who argue that human values such as the principles of humanity and public conscience are either inapplicable or too vague to be useful to the AADs debate as they are prone to endless fluctuations (Evans 2014, 731). In its 2019 working paper to the GGE, Argentina questioned whether, as a substantive matter, values such as the principles of humanity and public conscience should be taken into consideration in attempts to regulate AADs (Argentina 2019, 4). And Australia noted that, while it will consider these values, it takes a narrow view when interpreting what is meant by the principles of humanity or the dictates of public conscience (Australia 2019, para. 17). On the other hand, there are

6686_Enemark.indd 178

02/03/21 12:22 PM

Autonomous Armed Drones and the Challenges to Consensus / 179

those who note that the above-mentioned values are universal and timeless, regardless of ‘developments in types of situations or technology’ (ICRC 1987, 39; 2016, 413). Antonio Cassese notes that the Martens Clause is an ‘ingenious blend of natural law and positivism’ that ‘has responded to a deeply felt and widespread demand in the international community: that the requirements of humanity and the pressure of public opinion be duly taken into account when regulating conflict’ (Cassese 2000, 212). Along the same lines, Michel Veuthey considers the Martens Clause as the moral law that serves as a safety net that guarantees ‘the survival and fundamental dignity of humankind’ (Veuthey 2004, 609). After all, as observed by Mariëlle Matthee, the ‘human face is the special character of international humanitarian law itself’ (Matthee 2013, xvi). Some note the relevance of the Martens Clause in its role as an interpretative tool for directing customary international human rights law into IHL (Mihr and Gibney 2014, 98). The International Committee of the Red Cross (ICRC) notes that ‘the Martens Clause embodies a moral framework whereby in the absence of a necessity to kill, lethal force should not be used even against lawful targets’ (ICRC 2014, 16). Human Rights Watch (HRW) argues that AADs are unlawful as they contravene the principles of humanity and the dictates of public conscience (HRW 2012, 26), and it considers the idea of AADs ‘being given the power of life and death over human beings’ to be ‘unethical, shocking and unacceptable’ (HRW 2012, 35). Using such weapons is like ‘setting a mousetrap for human being’ (Johnson 2013, 134) and treating ‘our enemies like vermin’ (Sparrow 2011, 11). Christof Heyns notes that the use of AADs ‘risks taking humanity out of the loop’ (Heyns 2013a, 16), and he thus views the use of AADs as ‘some kind of mechanized pesticide’ – contrary to principles of humanity (Heyns 2013a, 18). Furthermore, Bradley Strawser has argued that those who use AADs ‘fail to satisfy reflexive duties to respect their own rationality, autonomy or dignity, they fail to take responsibility for their own actions’ (Strawser 2013, 237). In explaining the relevance of human values applicable to AADs and as contained in the Martens Clause, Peter Asaro notes that international law is a human construct that codifies social values and shared beliefs (Asaro 2016). In this sense, not only is the Martens Clause the ‘point where social and moral values weigh on IHL – it is in many ways the whole point of IHL’ as it embodies human values, written and unwritten norms, customs, shared moral sensibilities and collective interests (Asaro 2016, 372). Thus, international law is deemed to contain our moral inflections and conscience, which ‘includes the felt forces of duty and obligation, and the moral sentiments attached to the processes of moral deliberation in their totality’ (Asaro 2016, 375). Along the same lines and from a perspective

6686_Enemark.indd 179

02/03/21 12:22 PM

180 / Thompson Chengeta

that states are part of the international community, John Geer notes that such moral conscience is characterised by moral overtones that are shaped by values and interests shared among nations, and by cohesive state behaviour to protect the nation’s image or avoid international isolation (Geer 2004, 505). As such, ‘the weight afforded to the principles of humanity and the dictates of the public conscience, will determine how the Clause can impact – or prevent – the development or use of [AADs] in armed conflicts’ (Evans 2014, 723).

Understanding the Lack of Consensus on Values Relevant to AADs The lack of consensus on AADs and the diverging interpretations given to the values that are applicable have thus far been largely explained in terms of lack of shared definitions and understanding of AADs, so it is worthwhile now to discuss some of the factors that may more fully explain the lack of consensus. Values – and their interpretation – may be informed by the different interests, beliefs and attitudes of different societies and communities that are currently represented by governments and other organisations in the AADs debate. As such, the lack of consensus may be explained in terms of relativity of values, and the diverging political and humanitarian interests that inform the default positions of stakeholders. Of course, values remain contested because there is no agreement as to what they are and what they do. Yet, notwithstanding such lack of consensus, it is indeed ‘values [that] give direction to the way that individuals, organizations, and societies act; what they strive for; and what they deem important’ (Jaspers 2016). Values as influenced by interests, attitudes and beliefs may differ in importance depending on context (Jaspers 2016). Without necessarily being conscious of it, often our responses as to whether a particular use of force is morally right may be influenced by: whether we perceive the victim to be innocent or guilty; whether the lives lived by the victims reflect our own lifestyles; whether the purpose of the violent action in question was or is for our own self-preservation; and so on. Thus, while there may be standards or even universal values, our reaction when these are violently violated may well depend on such factors. The same applies to the positions that are taken by states and other stakeholders in the current debate on AADs.

Default Positions of Debaters and Interpretation of Values Stakeholders that are participating in the AADs debate can be divided into four kinds: the states that are developing AADs; powerful states that are

6686_Enemark.indd 180

02/03/21 12:22 PM

Autonomous Armed Drones and the Challenges to Consensus / 181

not developing AADs and are interested in multilateralism on the matter; smaller or less powerful states that are not developing AADs or have no capacity to do so; and the NGOs that want AADs to be pre-emptively prohibited. Among the last is the ICRC, whose crucial role is to encourage belligerents to respect IHL during armed conflict. The positions of these different stakeholders on AADs are influenced by different values and interpretations of certain values. The US has made it clear that it has taken a deliberate decision to permit weapons that are programmed to make decisions that relate to targeting (US 2018, 2). The US – like Russia and Australia – has taken the position that no new regulation on AADs is required at the moment. The UK, Switzerland, France and Germany have taken the position that they do not intend to develop fully autonomous weapon systems and emphasise the importance of complying with existing international law (Switzerland 2017). Yet without codification of certain ethical values applicable to AADs, such a position rests on quicksand, since existing international law is inadequate to govern AADs. China’s position is that development of AADs should be allowed while their use must be prohibited, so it has suggested an additional legal instrument prohibiting the use of AADs (CSKR 2019). Clearly, there is no consensus among the powerful states on the way forward. These states have largely attributed the lack of consensus on AADs to a lack of shared definitions and understandings of AADs, but some scholars have also explained that the different interpretation of applicable values depends on whether one approaches the issue from a deontological or a consequentialist standpoint. The main argument stemming from deontological ethics is (as highlighted above) that AADs contravene human dignity, and as such humans must retain moral responsibility over the use of force (Scharre 2018, 287; ICRC 2018). It is in this sense that UN SecretaryGeneral Guterres views AADs as morally repugnant (Guterres 2019), and AADs are further considered to weaken moral responsibility and prevent considerations of humanity (ICRC 2018). By contrast, some consequentialists argue that AADs may make better decisions than humans and that using them may thus save more lives over time (ICRC 2018; Arkin 2013, 6; Marchant et al. 2011, 280). In addition to their speed of operation, AADs do not have emotional weaknesses such as anger, frustration, desire for revenge, fear or hysteria, which so often cause humans to make bad decisions in war (Marchant et al. 2011, 280). For these reasons, there may be a consequentialist basis to an argument in favour of an ethical obligation to use AADs (Herbach 2012, 14). Consequentialists also tend to prefer the idea of belligerents winning a war as quickly as possible and thereby minimising casualties. The ‘unnecessary’ exercise of empathy and mercy by

6686_Enemark.indd 181

02/03/21 12:22 PM

182 / Thompson Chengeta

fighters towards lawful targets could thus be seen as delaying the ending of a war and thereby increasing the overall number of casualties – another ‘weakness’ that AADs do not have (Scharre 2018, 273).

The Arms Race in AI Military Technologies Beyond ethics, power politics and diverging interests largely explain the different interpretations of values that are relevant to the regulation of AADs. The current rivalry in the race for dominance in the field of military AI – in particular, that including major powers like the US, China and Russia – is clearly not conducive to regulation of AADs (Imbrie et al. 2020). Major powers are aware that AI research and military technologies such as AADs have the potential to enhance geopolitical influence and impact the current balance of power (Imbrie et al. 2020, 27). China, for example, is particularly aware that leadership in AI carries the potential to position it as a global power with the capacity to control and influence norms (Imbrie et al. 2020, 28). While China is leading on AI commercial applications, the US leads on AI applications relating to national defence and military power (Imbrie et al. 2020). Thus, while China currently seeks to ‘leapfrog’ the US in innovation of AI military applications, the US is eager to maintain its lead in AI military technologies (Khan 2019). In this race for AI power among the powerful states, ethical values such as human dignity may have little if any influence. Multilateralism and consensus often require trust or at the very minimum transparency. Yet, while not a new phenomenon, the current heightened mistrust between US and China in the area of AI technologies – as exacerbated by Trump’s anti-China rhetoric – makes co-operation and transparency on AADs regulation difficult, as there is always a fear of potential ‘exploitation of access to sensitive information and targeting of intelligence efforts against experts, scientists, or officials’ (Imbrie and Kania 2019).

The Implications of AADs for Military Power and National Security To the minority powerful states, AADs present enormous military and national security advantages, such as force multiplication (Scharre 2014), being faster and sparing the use of a state’s own soldiers (Horowitz 2019, 764). For these reasons, such states would probably be reluctant to forego such conceived or perceived military benefits (Davis 2019, 115). Furthermore, in the global fight against terrorism that constantly and mainly threatens the national security of states like the US, armed drones have played and continue to play a very important if not a decisive role. Armed drones have been referred to as

6686_Enemark.indd 182

02/03/21 12:22 PM

Autonomous Armed Drones and the Challenges to Consensus / 183

a weapon of choice against terrorists (Marcus 2015), and since AADs are a more advanced version of armed drones, it is reasonable to expect that some states may conceive of them as an important counterterrorism tool in the near future. Accordingly, just as the US creatively introduced new terminologies and standards in international law, such as ‘elongated imminence’ in ‘anticipatory self-defense’, to justify its attacks through armed drones, the US also creatively interprets values and standards that are applicable to AADs to justify their development and use (Akande et al. 2016; US 2018). Yet the discussions on AADs also include smaller or less powerful states, some of which already feel that their territorial sovereignty has been and continues to be violated through the use of armed drones (Akande et al. 2016). For states in the Muslim world, armed drones that have been effectively – and sometimes recklessly – deployed in Muslim-majority territories stand as an anti-Islam symbol and a ‘global war on tribal Islam’ (Ahmed 2013). And if remotely piloted armed drones are perceived as a symbol of Western antiIslam sentiments and perversion, AI-piloted (autonomous) versions might be seen as being even worse. Many other states that do not have the capacity to develop AADs are aware that the current global imbalance of power and historical invasion and colonisation of smaller states were a result of the military prowess of minority powerful states whose powers largely depend on their weapon arsenals. The venture of minority states to develop AADs might therefore be perceived as a quest for continued dominance and neocolonialism. This may, to some extent, explain the position of African states and members of the Non-Aligned Movement in the UN’s Geneva-based debates about AADs.

Anti-NGO Sentiments and ‘Banphobia’ in Geneva On the basis of humanitarian values such as consideration and preservation of human dignity in armed conflict, NGOs and other international organisations have campaigned for the banning of landmines, cluster munitions and nuclear weapons, and for the negotiation of the 2013 Arms Trade Treaty. Without doubt, NGOs and other humanitarian organisations like the ICRC have played a significant role in the adoption of important treaties in arms control and disarmament. Yet it cannot go unnoticed that in recent years, some state delegations and diplomats have developed a ‘banphobia’ and an anti-NGOs sentiment. Some scholars, like Tyler Evans, view NGOs as hell-bent on taking every weapon from the arsenals of states and usurping their legislative powers (Evans 2014, 725; see also Meron 2000, 98). This has led to, or at least intensified, some states’ opposition to NGO initiatives in disarmament. In this regard, Mark Perry (2020, 1) notes, for example,

6686_Enemark.indd 183

02/03/21 12:22 PM

184 / Thompson Chengeta

that ‘landmines were never important to the military – until NGOs tried to ban them’ (Perry 2020). Arguably, this feeling among states and certain stakeholders further negatively impacts the chances of achieving consensus on the regulation of AADs.

Conclusion Notwithstanding the differences highlighted above, there are some powerful states (for example, Germany) that are keen to see a tangible output on AADs through multilateral processes. Furthermore, because AADs present, to some degree, shared challenges, states will be forced to continue co-operating. For example, AADs present capabilities that may undermine the survivability of nuclear arsenals and potentially render them vulnerable, resulting in overwhelming threats to global security (Keir 2017). This is a concern for all states, and especially those that possess nuclear weapons. Even so, in multilateral processes that are characterised by consensus in decision-making, such as the current CCW process on AADs, the risk remains that the political interests of a few mighty powers may overshadow common values that the majority of states seek to codify. While the current lack of consensus in the CCW process on AADs has been attributed to a lack of shared definitions and common understandings of AADs, this chapter concludes that the lack of consensus is largely explained by diverging political interests in power politics and national security. Such political interests are rooted in the significance of AADs for global power and dominance, national security, and the territorial sovereignty of smaller or less powerful states. In turn, these diverging political interests influence whether a given state views the ethics of AADs from a deontological or consequentialist perspective. Given the CCW’s consensus rule, it seems unlikely that a regulation on AADs will emerge from the current negotiation process. It is hoped, however, that such a regulation will emerge instead from majoritarian institutions like the UN General Assembly.

Notes Research for this chapter was conducted as part of a project funded by the European Research Council under the European Union’s Horizon 2020 research and innovation programme (grant no. 771082): ‘Emergent Ethics of Drone Violence: Toward a Comprehensive Governance Framework’ (DRONETHICS). 1. The ‘Martens Clause’ was introduced for the first time in the preamble of the 1899 Hague Convention (Convention with Respect to the Laws and Customs of War on Land), on the proposal of Fyodor Fyodorovich Martens, a Russian lawyer and member of the Russian delegation to the 1899 International Peace Conference (Koutroulis 2017).

6686_Enemark.indd 184

02/03/21 12:22 PM

Autonomous Armed Drones and the Challenges to Consensus / 185

References AFM. 2019a. ‘ What is the “Alliance for Multilateralism”?’, Alliance for Multilateralism, https://multilateralism.org/the-alliance AFM. 2019b. ‘Declaration by the Alliance for Multilateralism on Lethal Autonomous Weapons Systems’, Alliance for Multilateralism, https://multilateralism.org/ declaration-on-lethal-autonomous-weapons-systems-laws.pdf AGD. 2018. ‘Statement by the African Group’, African Group on Disarmament, Third Meeting of the Convention on Conventional Weapons Group of Governmental Experts on Lethal Autonomous Weapons Systems, Geneva, 27 August, Reaching Critical Will, http://reachingcriticalwill.org/images/documents/Disarmament-fora/ ccw/2018/gge/statements/29August_AfricanGroup2.pdf Ahmed, Akbar. 2013. The Thistle and the Drone: How America’s War on Terror Became a Global War on Tribal Islam. Washington, DC: Brookings Institution Press. Akande, Dapo, Christof Heyns, Lawrence Hill-Cawthorne and Thompson Chengeta. 2016. ‘ The International Law Framework Regulating the Use of Armed Drones’, International and Comparative Law Quarterly 65 (4): 791–827. Argentina. 2019. ‘Argentina’s Questionnaire on the Legal Review of Means and Methods of Warfare, Submitted to the Group of Governmental Experts on Emerging Technologies in the Area of Lethal Autonomous Weapons Systems’, Geneva, 25–9 March and 20–1 August, CCW/GGE.1/2019/WP.6. Arkin, Ronald. 2013. ‘Lethal Autonomous Systems and the Plight of the NonCombatant’, AISB Quarterly137 (July): 1–9. Asaro, Peter. 2016. ‘Jus Nascendi, Robotic Weapons and the Martens Clause’, in Robot Law, edited by Ryan Calo, Michael Froomkin and Ian Kerr. Cheltenham: Edward Elgar, 367–86. Australia. 2019. ‘Australia’s System of Control and Applications for Autonomous Weapon Systems’, CCW/GGE.1/2019/WP.2/Rev.1, Group of Governmental Experts on Emerging Technologies in the Area of Lethal Autonomous Weapons Systems, United Nations Office in Geneva, https://www.unog.ch/80256EDD006B8954/(httpAssets)/ 16C9F75124654510C12583C9003A4EBF/$file/CCWGGE.12019WP.2Rev.1.pdf Brzozowski, Alexandra. 2019. ‘No Progress in UN Talks on Regulating Lethal Autonomous Weapons’, Euractiv, 22 November, https://www.euractiv.com/section/ global-europe/news/no-progress-in-un-talks-on-regulating-lethal-autonomousweapons Buzan, Barry, and Barry Jones. 1981. Change and the Study of International Relations: The Evaded Dimension. London: Frances Pinter. Cassese, Antonio. 2000. ‘ The Martens Clause: Half a Loaf or Simply Pie in the Sky?’, European Journal of International Law 11 (1): 187–216. Chengeta, Thompson. 2020. ‘Is Existing Law Adequate to Govern Autonomous Weapon Systems?’, South Africa Forum for Artificial Intelligence Research, 3–6 December 2019, Cape Town, CEUR Workshop Proceedings, http://ceur-ws.org/ Vol-2540/FAIR2019_paper_9.pdf Claude, Inis. 1984. Swords into Plowshares: The Problems and Progress of International Organization. New York: McGraw-Hill.

6686_Enemark.indd 185

02/03/21 12:22 PM

186 / Thompson Chengeta

Columbia Law School. 2012. The Civilian Impact of Drone Strikes: Unexamined Costs, Unanswered Questions, https://www.law.columbia.edu/human-rights-institute/ counterterrorism/drone-strikes/civilian-impact-drone-strikes-unexaminedcosts-unanswered-questions CSKR. 2019. ‘Alarm Bells Ring on Killer Robots’, Campaign to Stop Killer Robots, 15 November, https://www.stopkillerrobots.org/2019/11/alarmbells Davis, Zachary. 2019. ‘Artificial Intelligence on the Battlefield: Implications for Deterrence and Surprise’, Prism 8 (2): 115–31. Delcker, Janosch, and Andrew Gray. 2020. ‘ Top UN Official: It’s Not Too Late to Agree Ban on Killer Robots’, Politico, 11 February, https://www.politico.eu/article/top-un-official-its-not-too-late-to-curb-ai-powered-weapons Emmerson, Ben. 2014. ‘Report of the Special Rapporteur on the Promotion and Protection of Human Rights and Fundamental Freedoms while Countering Terrorism’, 11 March, United Nations Human Rights Council, A/HRC/25/59. European Parliament. 2018. ‘Resolution of 12 September 2018 on Autonomous Weapon Systems’ (2018/2752(RSP). Evans, Tyler. 2014. ‘At War with the Robots: Autonomous Weapon Systems and the Martens Clause’, Hofstra Law Review 41: 697–733. Geer, John. 2004. Public Opinion and Polling Around the World: A Historical Encyclopaedia. Santa Barbara, CA: ABC-CLIO. GGE. 2019. ‘Report of the 2019 Session of the Group of Governmental Experts on Emerging Technologies in the Area of Lethal Autonomous Weapons Systems’, CCW/GGE.1/2019/3. Guterres, António. 2019. ‘Remarks at the 2019 Web Summit’, United Nations, 5 November, https://www.un.org/press/en/2018/sgsm19332.doc.htm Guterres, António. 2020. ‘Remarks to the General Assembly on Priorities for 2020’, United Nations, 22 January, https://www.un.org/sg/en/content/sg/statement/ 2020-01-22/secretary-generals-remarks-the-general-assembly-his-priorities-for2020-bilingual-delivered-scroll-down-for-all-english-version. Herbach, Jonathan. 2012. ‘Into the Caves of Steel: Precaution, Cognition and Robotic Weapons Systems under the International Law of Armed Conflict’, Amsterdam Law Forum 4 (3): 3–20. Heyns, Christof. 2013a. ‘Report on Lethal Autonomous Weapon Systems by the Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions’, United Nations Human Rights Council, A/HRC/23/47. Heyns, Christof. 2013b. ‘Report on Armed Drones by the UN Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions’, United Nations General Assembly, A/68/382. Horowitz, Michael. 2019. ‘ When Speed Kills: Lethal Autonomous Weapon Systems, Deterrence and Stability’, Journal of Strategic Studies 42 (6): 764–88. House of Lords. 2018. ‘UK Foreign Policy in a Shifting World Order’, Select Committee on International Relations, 5th Report of Session 2017–19, HL Paper 250, 18 December, https://publications.parliament.uk/pa/ld201719/ldselect/ ldintrel/250/250.pdf

6686_Enemark.indd 186

02/03/21 12:22 PM

Autonomous Armed Drones and the Challenges to Consensus / 187

HRW. 2012. ‘Losing Humanity: The Case Against Killer Robots’, Human Rights Watch, 19 November, https://www.hrw.org/report/2012/11/19/losing-humanity/caseagainst-killer-robots HRW. 2017. ‘US Embraces Cluster Munitions’, Human Rights Watch, 1 December, https://www.hrw.org/news/2017/12/01/us-embraces-cluster-munitions HRW. 2020. ‘US: Trump Administration Abandons Landmine Ban’, Human Rights Watch, 31 January, https://www.hrw.org/news/2020/01/31/us-trump-administration-abandons-landmine-ban ICRC. 1987. ‘Commentary on the Additional Protocols of 8 June 1977 to the Geneva Conventions of 12 August 1949’, International Committee of the Red Cross, 17 October, https://www.icrc.org/en/publication/0421-commentary-additionalprotocols-8-june-1977-geneva-conventions-12-august-1949 ICRC. 2014. ‘Autonomous Weapon Systems: Technical, Military, Legal and Humanitarian Aspects’, International Committee of the Red Cross, 1 November, https:// www.icrc.org/en/document/report-icrc-meeting-autonomous-weapon-systems26-28-march-2014 ICRC. 2016. ‘Commentary to the First 1949 Geneva Convention’, International Committee of the Red Cross, 22 March, https://ihl-databases.icrc.org/ihl/full/GCicommentary ICRC. 2018. ‘Ethics and Autonomous Weapon Systems: An Ethical Basis for Human Control?’, International Committee of the Red Cross, 3 April, https://www.icrc.org/en/ document/ethics-and-autonomous-weapon-systems-ethical-basis-human-control Imbrie, Andrew, and Elsa Kania. 2019. ‘AI Safety, Security, and Stability Among Great Powers: Options, Challenges, and Lessons Learned for Pragmatic Engagement’, CSET Policy Brief, Georgetown University, https://cset.georgetown.edu/wpcontent/uploads/AI-Safety-Security-and-Stability-Among-the-Great-Powers.pdf Imbrie, Andrew, Elsa Kania and Lorand Laskai. 2020. ‘ The Question of Comparative Advantage in Artificial Intelligence: Enduring Strengths and Emerging Challenges for the United States’, CSET Policy Brief, Georgetown University, https:// cset.georgetown.edu/wp-content/uploads/CSET-The-Question-of-ComparativeAdvantage-in-Artificial-Intelligence-1.pdf Jaspers, Eva. 2016. ‘ Values’, Oxford Bibliographies (Sociology), 27 October, https:// www.oxfordbibliographies.com/view/document/obo-9780199756384/obo9780199756384-0182.xml Johnson, Aaron. 2013. ‘ The Morality of Autonomous Robots’, Journal of Military Ethics 12 (2): 129–41. Kahler, Miles. 1992. ‘Multilateralism with Small and Large Numbers’, International Organization 46 (3): 681–708. Keir, Lieber. 2017. ‘ The New Era of Counterforce: Technological Change and the Future of Nuclear Deterrence’, International Security 41 (4): 9–49. Khan, Saif. 2019. ‘Maintaining the AI Chip Competitive Advantage of the United States and its Allies’, CSET Policy Brief, Georgetown University, https://cset.georgetown. edu/wp-content/uploads/CSET-Maintaining-the-AI-Chip-Competitive-Advantageof-the-United-States-and-its-Allies-20191206.pdf

6686_Enemark.indd 187

02/03/21 12:22 PM

188 / Thompson Chengeta

Kissack, Robert. 2010. Pursuing Effective Multilateralism. London: Palgrave Macmillan. Koutroulis, Vaios. 2017. ‘Martens Clause’, Oxford Bibliographies (International Law), 16 August, https://www.oxfordbibliographies.com/view/document/obo9780199796953/obo-9780199796953-0101.xml Marchant, Gary, Braden Allenby, Ronald Arkin, Edward Barrett, Jason Borenstein, Lyn Gaudet, Orde Kittrie, Patrick Lin, George Lucas, Richard O’Meara and Jared Silberman. 2011. ‘International Governance of Autonomous Military Robots’, Columbia Science and Technology Law Review 12: 271–315. Marcus, Jonathan. 2015. ‘Drone Wars: Today’s Weapon of Choice?’, BBC News, 14 September, https://www.bbc.co.uk/news/world-34219417 Matthee, Mariëlle. 2013. ‘A Lit Beacon in the Dark’, in Armed Conflict and International Law: In Search of the Human Face, edited by Mariëlle Matthee, Brigit Toebes and Marcel Brus. The Hague: TMC Asser Press, xv–xxi. Meron, Theodor. 2000. ‘ The Martens Clause, Principles of Humanity, and Dictates of Public Conscience’, American Journal of International Law 94 (1): 78–89. Mihr, Anja, and Mark Gibney. 2014. The SAGE Handbook of Human Rights. London: Sage. PAX. 2018. Crunch Time: European Positions on Lethal Autonomous Weapon Systems. Utrecht: PAX Netherlands. Perry, Mark. 2020. ‘Not Even the Military Thinks Landmines Are a “Vital Tool”’, The American Conservative, 6 February, https://www.theamericanconservative.com/ articles/not-even-the-military-thinks-landmines-are-a-vital-tool Rittberger, Volker, and Benhard Zangl. 2006. International Organization: Polity, Politics and Policies. New York: Palgrave Macmillan. Scharre, Paul. 2014. ‘Robotics on the Battlefield Part II: The Coming Swarm’, Center for a New American Security, 15 October, https://www.cnas.org/publications/ reports/robotics-on-the-battlefield-part-ii-the-coming-swarm Scharre, Paul. 2018. Army of None: Autonomous Weapons and the Future of War. New York: W. W. Norton. Sparrow, Robert. 2011. ‘Robotic Weapons and the Future of War’, in New Wars and New Soldiers: Military Ethics in the Contemporary World, edited by Paolo Tripodi and Jessica Wolfendale. Farnham: Ashgate, 117–36. Strawser, Bradley J. 2013. ‘Introduction: The Moral Landscape of Unmanned Weapons’, in Killing by Remote Control: The Ethics of an Unmanned Military, edited by Bradley J. Strawser. Oxford and New York: Oxford University Press, 3–24. Switzerland. 2017. ‘A Compliance-Based Approach to Autonomous Weapon Systems’, Group of Governmental Experts on Emerging Technologies in the Area of Lethal Autonomous Weapons Systems, CCW/GGE.1/2017/WP.9. US. 2018. ‘Human–Machine Interaction in the Development, Deployment and Use of Emerging Technologies in the Area of Lethal Autonomous Weapons Systems’, CCW/GGE.2/2018/WP.4. US Press Secretary. 2020. ‘Statement from the Press Secretary’, The White House, 31 January, https://www.whitehouse.gov/briefings-statements/statement-presssecretary-107

6686_Enemark.indd 188

02/03/21 12:22 PM

Autonomous Armed Drones and the Challenges to Consensus / 189

Veuthey, Michel. 2004. ‘Public Conscience in International Humanitarian Law Today’, in Krisensicherung und Humanitärer Schutz/Crisis Management and Humanitarian Protection, edited by Horst Fischer, Ulrike Froissart, Wolff Heinegg von Heintschel and Christian Raap. Berlin: Berliner Wissenschaftsverlag, 611–42.

6686_Enemark.indd 189

02/03/21 12:22 PM

C ONCLUSION

Christian Enemark

In the context of a worldwide increase in state capacities to wield drone violence, each contribution to this volume is driven by a common concern: that ethical deliberation and human moral progress should not be outpaced by the accelerating technological advancement and global proliferation of lethal drone technology. Collectively, the chapters assembled here do not converge towards a consensus judgement about the morality of using armed drones. Rather, each chapter offers a distinct perspective on how drone violence works in different circumstances, the effects it generates and the moral justification for restraining it. The volume as a whole thus demonstrates how a variety of disciplinary perspectives, theoretical approaches and normative arguments can be brought to bear upon the persistently contentious topic of drone use. The aim of this concluding chapter is neither to settle nor to adjudicate all the disagreements and differences of emphasis that readers may have discerned along the way. A resolution of these and other disharmonies should instead await the fruits of a further broadening and deepening of the moral debate over drone violence. As a precursor to that process, though, it may be helpful to conclude the volume by highlighting three themes of enduring ethical concern that run through it: the significance of risk to the wielder of violence; the categorisation of violence; and the need for better governance of violence. Like most ‘emergent’ phenomena in world affairs, drone violence is not entirely new (in terms of technology and function) compared to pre-existing forms of violence. Even so, it can often seem sufficiently novel as to occasion fresh consideration of when and why violence ought to be justified or condemned. Extreme asymmetry of physical risk has long been a theme of ethical concern in the contemplation of airpower, especially when an aircraft

6686_Enemark.indd 190

02/03/21 12:22 PM

Conclusion / 191

pilot’s release of weapons from a high altitude endangers civilians on the ground. However, the elimination of risk on one side – achieved by removing onboard pilots and replacing them with remote-control technology – is at the heart of what makes armed drones distinctively interesting to many ethical thinkers. Here, the key issue attracting moral interest is not the idea that drone violence avoids all kinds of risk. It does not. For example, a droneusing state might be exposed to the strategic risk of inspiring popular hatred in foreign territories where its drone strikes occur and of provoking terroristic retaliation (‘blowback’) against its own citizens. Rather, ethical attention tends to focus upon the immediate experience of the individual drone operator who, at the moment of a strike, kills without physical risk. Then, the question arises: what, if anything, is morally wrong with this? To some extent, assuming that a given instance of drone violence is a war action, the answer depends on whether one prefers to engage in classical or revisionist Just War thinking. In Chapters 1 and 2 respectively, Sparrow and Braun discussed these contending approaches to military ethics, the broader reconciliation of which is beyond the aims and scope of this volume. For present purposes, from a classical viewpoint, to kill remotely with an armed drone seems wrong because it offends the principle of the moral equality of combatants; a targeted enemy is effectively deprived of any ability to exercise their equal right to inflict reciprocal harm. By contrast, a revisionist is less likely to regard physical risk to warriors as important in the ethics of war, such that one warrior’s complete avoidance of risk might present no moral problem. What rather matters is whether that warrior is fighting for a just cause, in which case their unjust enemy’s inability (by reason of radical remoteness) to harm a ‘just combatant’ is a permissible and desirable condition in war. Regardless of which version of Just War thinking is applied, however, it is important also to consider the moral relationship between risk to the wielder of violence (be they a just or unjust combatant) and the risk to innocents in the vicinity of an attack. As Olsthoorn warned in Chapter 6, risk-avoidance measures designed to protect combatants (including the use of armed drones rather than manned aircraft) present a moral problem if they effectively increase the risk of harm to civilians living in targeted territories. The wielding of drone violence has the potential to generate such an increase in unjust harm, but the absence of physical risk to the drone operator might also be a factor in instead enabling foreign civilians to be better protected. Although, in Chapter 1, Sparrow rejected the idea that risk-avoidance makes drones morally distinctive, he was more interested in their unique capacity to loiter above prospective targets. This capacity, combined with onboard video-camera technology, arguably affords a would-be wielder of drone violence more time and opportunity (than is available to

6686_Enemark.indd 191

02/03/21 12:22 PM

192 / Christian Enemark

the pilot of a fast-moving bomber) to acquire information and act carefully. As Clark and Enemark explained in Chapter 7, there is moral significance in the potential that armed drones hold for someone to be killed from afar after being closely observed. The camera-equipped drone, they argued, rather than necessarily facilitating callous or reckless violence by drone operators, enables and therefore morally demands the exercise of a greater degree of concern for the fate of civilians living near a prospective strike zone. If so, this could form part of an answer to Olsthoorn’s complaint that some militaries dishonour themselves by demonstrating too little concern for casualties ‘among unknown persons in faraway countries’ (Chapter 6). Moreover, according to Olsthoorn, who focused in this volume on virtue ethics rather than Just War, the absence of physical risk to drone operators potentially confers a moral advantage: an ability for them thus to be more impartial and committed to principles (rather than to organisation or mission) when foreign civilian victimhood is in prospect. The traditional military virtue of courage in the face of mortal danger is, by contrast, unnecessary and irrelevant for the safely remote wielder of drone violence. Thus, in Chapters 1 and 6 respectively, Sparrow and Olsthoorn rightly raised the question of what it means to be a warrior – in terms of military culture and professionalism – when one no longer needs physical courage to participate in war. Being a virtuous drone operator is, according to Olsthoorn, most likely to involve ‘exercising restraint’ rather than exhibiting other virtues such as courage, loyalty and discipline. And yet, once the martial image of the ‘warrior’ is transformed in this way, it begins to present a challenge to traditional expectations about what it means to wage war. Just War principles and military virtues are exclusively applicable to the category of political violence called ‘war’. However, if drone violence sometimes takes a form that cannot plausibly be assigned to that category, different modes of ethical judgement may apply. In some circumstances, the use of armed drones is readily regarded as effecting a change in the character of war, but in other circumstances their use is more difficult to identify as war. Accordingly, several contributors to this volume were inspired to test an alternative categorisation proposition: that drone violence is, essentially, violent law enforcement. They tended to focus on drone-based ‘targeted killings’ of individuals, and their arguments on the potential for injustice inevitably extended beyond the traditional bounds of Just War thinking. Here, though, it is worth recalling that the distinction between the categories of war and law enforcement – as reflected in contemporary international law, for example – has not always been a clear one, at least as a matter of moral theory. As Braun explained in Chapter 2, the medieval theologian Thomas Aquinas wrote of ‘war’ in reference to a sovereign

6686_Enemark.indd 192

02/03/21 12:22 PM

Conclusion / 193

authority’s imposition of the death penalty, and he would probably have used that term also to describe domestic police uses of force today. It is currently the case, however, that ethicists and lawyers are usually keen to establish whether a condition of war or law enforcement exists, so that appropriate judgements can be rendered about the permissibility of certain acts of state violence. Yet sometimes, when observing the exercise of drone violence, clarity on this point has been elusive. For example, as Fuller showed in Chapter 5, a categorical uncertainty persists regarding the UK government’s use of an armed drone to kill British citizen Reyaad Khan in Syria in 2015: was the government defending state security by neutralising an enemy, or was it enforcing state law by punishing a criminal? In Chapter 3, Brookman-Byrne warned that a drone-using state might be tempted to pursue a political advantage by allowing such uncertainty to remain unresolved. Insisting that drone strikes are neither inherently nor always part of armed conflict, he observed that these are sometimes (when conducted outside of armed conflict) essentially acts of law enforcement. A moral problem arises, then, when a state chooses to treat those acts as acts of war, because the legal restrictions on killing (based on international human rights law) that ought to apply are thereby displaced by the more permissive regime of international humanitarian law. Human harm resulting from drone-based law enforcement, of a kind which is illegitimate according to a rights-based judgement, ends by being unjustly ‘legitimised’ because it was categorised – erroneously and deliberately – as wartime harm. The solution, Brookman-Byrne argued, is to lift ‘the presumptive legal veil’ provided by a state’s claims of armed conflict status, to take seriously the status of drone violence as violent law enforcement, and to govern it properly according to human rights principles. An assessment of drone use so categorised might, however, reveal other moral problems, as was argued in Chapter 4. Even when a state implements extensive and rigorous administrative procedures to guide drone-based, punitive law enforcement actions, such arrangements for drone violence arguably still amount to a kind of ‘wild justice’ that offends rights-based rules against extrajudicial killing. Other authors expressed similar concerns that ethical standards for violent law enforcement might not be upheld when armed drones are used in remote foreign territories designated as special zones of violence. Chapter 4 termed such zones ‘the terror frontier’, Brookman-Byrne warned of drones enabling a return to ‘colonial-era techniques of policing with airpower’ (Chapter 3), and in Chapter 5 Fuller described a ‘drone domain’ (beyond the domain of justice) where criminals are transformed into enemies to be destroyed. In these and all the other accounts of drone violence in this volume, there is a common underlying concern that unjust killing might occur with

6686_Enemark.indd 193

02/03/21 12:22 PM

194 / Christian Enemark

impunity. For example, as Fuller (Chapter 5) and Chengeta (Chapter 9) observed, the temptation armed drones offer states to resort to force with undue haste is a potential jus ad bellum problem. And, as Lee explained in Chapter 8, the incorporation of an autonomous facial recognition function into a drone strike system might introduce biases (regarding different age, gender and racial characteristics) that degrade the system’s ability to discriminate between combatants and civilians. In response to such ethical challenges, some chapters included specific suggestions for how the state use of armed drones could be improved. Consistent with Brookman-Byrne’s call to resist the muddled ‘continuation of law enforcement by military means’ (Chapter 3), I advocated in Chapter 4 finding some way for drone-using states (assisted by independent courts) instead to conduct drone strikes openly as permissible acts of punitive law enforcement. And Braun too, reasoning from a ‘retributive just war perspective’, favoured the conduct of judicial trials in absentia prior to (or to negate the case for) the use of an armed drone ‘as a tool of justice’ against culpable terrorists in remote locations (Chapter 2). On the issue of AI enhancement, where the need for governance innovation is arguably most urgent, Lee (Chapter 8) recommended a set of ethical principles to guide the development and future use of armed autonomous drones. In Chapter 9, by contrast, Chengeta preferred to approach this governance challenge as one to be addressed at the level of international arms control law. Here, as he explained, the prevailing diplomatic circumstances do not favour the implementation of a proposed ban on ‘fully autonomous weapon systems’, so it seems likely that alternative ways and degrees of governing AI-controlled drones will be explored. More generally, the prospect of agency and moral responsibility devolving from humans to intelligent machines is such a profound change in the dynamics of political violence as to demand intense and sustained normative inquiry. In the years ahead, when more and better drones become available, there will be a need for further ethical assessments in response to shifting patterns of drone use, especially as AI technology advances. Beyond the assessments assembled in this volume, inquiries into the morality of drone violence could be of greatest benefit if they inform the preparation of practical mechanisms for preventing or mitigating injustice.

6686_Enemark.indd 194

02/03/21 12:22 PM

INDEX

aerial warfare Commission of Jurists, 150–1 imperial policing via airpower, 51–2, 58, 62–3 military-political hierarchies of, 162–3 piloted bombing missions, 150–1 systems and mission checks, 163–5 Alliance for Multilateralism (AFM), 173 al-Qaeda, 82, 124 al-Qaeda in the Arabian Peninsula (AQAP), 54, 61 Aquinas, Thomas, St., 35–6, 40, 42, 43 Aristotle, 117, 118–19, 121 armed, uninhabited aerial vehicles (UAVs) see drones artificial general intelligence (AGI), 157 asymmetric warfare asymmetry rationale of policing, 36, 39–40 combatants and, 39, 41 objections to drone strikes, 37, 39, 41

6686_Enemark.indd 195

personality strikes, 78 riskless warfare and, 12, 14, 16–18, 21, 190–1 autonomous armed drones (AADs) accountability safeguards for, 171, 194 arms race in AI military technologies, 182 automated facial recognition systems, 154–5, 194 consequentialist arguments for, 181–2 continuous human control over, 155–6, 177 for counterterrorism activities, 182–3 decision-making on by majoritarian-based institutions, 174–6 human values applicable to, 178–9 international humanitarian law and, 155, 162, 165, 170–1 lack of consensus on values relevant to, 178–82 Martens Clause applied to, 178–80

02/03/21 12:22 PM

196 / Index

autonomous (cont.) military power applications, 182–3 military-political hierarchies of, 162–3 NGOs stance on, 170, 177, 181, 183 privilege of powerful states, 180–1 systems and mission checks, 163–5 UK policy on, 155, 157, 181 United Nations Group of Governmental Experts (GGE), 171, 176, 177 US development of, 154, 181 violation of human dignity, 156, 178, 181 see also Lethal Autonomous Weapons Systems (LAWS) autonomy cyborged ecosystems, 158 definitions of, 155, 158–9 ‘fully’ autonomous concept, 156, 159 idealised autonomy, 156–7 limited machine autonomy, 157 al-Awlaki, Anwar, 82–3, 104 al-Badawi, Jamal, 61–2, 74–5 Beauchamp, Zack, 23 bellum generale, 35, 36, 38, 43 bellum particular, 35, 36, 38, 43 Bethlehem, Sir Daniel, 106–7 bias bias avoidance in lethal autonomous weapons systems, 150, 159–61 in facial recognition systems, 160–1, 194 human bias within programmed systems, 159, 160 risk perception and, 159–60

6686_Enemark.indd 196

bin Laden, Osama, 45 Braun, Megan, 3, 19, 34 Brennan, John, 83, 84 Brunstetter, Daniel, 3, 19, 31, 33, 34 Bush, George W., 82, 94, 103 Cameron, David, 93, 94, 97–8, 107–8 Campaign to Stop Killer Robots (CSKR), 156, 171 Canberra Working Group, 163–4, 165 Cassese, Antonio, 179 Chamayou, Grégoire, 3 Cheney, Dick, 82 Clark, O. Grant., 158–9, 160 Claude, Inis, 177–8 colonialism, 51–2, 58, 62–3, 193 combatants asymmetric warfare and, 39, 41 drone warfare and reduced risk to, 2, 3, 4 moral equality of, 39–40, 41 mutual risk of, 39 in riskless warfare, 12, 13, 19–20, 22 see also drone operators condolence payments, 140, 143 Cooper, Robert, 102, 108–9 counterterrorism autonomous armed drones as tool for, 182–3 for continuing and imminent threats, 83, 94, 105, 108 correct responses by the state, 95–6 excessive redemptive violence, 85 imminence doctrine and, 107–8 law enforcement drone strikes for, 59–60

02/03/21 12:22 PM

Index / 197

militarised counterterrorism strategies, 60–2, 63–4, 66, 194 public perceptions of, 95 self-defence justifications, 94, 98 UK’s lethal drone policy, 93–5, 96–100 unable or unwilling test, 100, 101 US global war on terror, 18, 23–4, 32, 81–2, 94 war paradigm lens for, 93 wild justice over legalism, 82, 193 see also personality strikes cyborged ecosystems, 158 death penalties, 33, 36, 43, 78–9; see also personality strikes Derwish, Kamal, 103–4 drone courts, 86–7 drone operators close air support (CAS) work, 143–4 decision-making principles of, 153 deprivation of dependent civilians of care, 130–1, 133, 134–5, 143 an ethics of care for, 131, 141–3 jus in bello decisions, 131 military awards for, 123 military ethics education for, 116–17 military honour and, 123–4 new virtues for, 117–18, 192 observation of targets, 130, 131, 133–5, 142, 143–4, 191–2 post-strike damage assessments, 134

6686_Enemark.indd 197

reinterpretation of old virtues, 118–21 riskless warfare of, 122–4, 191 rule-based ethics, 121–2, 124 UK press coverage of, 152–3 utilitarian approaches for, 121 virtue of courage, 119 virtue-based ethics approach for, 116–17 drone strikes asymmetry objection to, 37, 39, 41 civilian casualties, 21–2, 25 for continuing and imminent threats, 83, 94, 105, 108 and escalation to war, 46 within humanitarian interventions, 23 infrastructure behind, 149 within Just War theory, 2–3, 76 non-war form of state violence, 76–7 as preferable to capture operations, 50–1 state capacities, 1–2, 165–6 Tadić criteria applied to, 53, 59–60 see also personality strikes drones Hellfire missile, 45 Predator aircraft, 1, 100, 150–2 Reaper aircraft, 1, 2, 100, 152 unarmed drones, 15 use by non-state actors, 38 see also autonomous armed drones (AADs); Lethal Autonomous Weapons Systems (LAWS) Edkins, Jenny, 137 Edney-Brown, Alex, 63

02/03/21 12:22 PM

198 / Index

ethics assessments of drone warfare, 2–3 atomic bombing of Japan, 3–4 NATO campaign in Kosovo, 4, 14 of piloted bombing missions, 149–50 in revisionist Just War theory, 16 rule-based ethics, 115, 121–2, 124 utilitarianism, 116, 121, 136 see also military ethics education ethics of care applied to problems of violence, 138 counterinsurgency campaigns, 139–40, 144 dependents of a targeted individual, 142–3 drone operators’ adoption of, 131, 141–3 emphasis on personal relationships, 131 feminist origin of, 135–6 in relation to abstract moral reasoning, 136–7, 138–9 relational ontology, 133, 137–8, 139 Evans, Tyler, 183 facial recognition systems, 154–5, 160–1, 194 Fallon, Michael, 98–9, 106 Five Eyes, 107 Fort Hood shootings, 104 France, 96, 173, 181 Frowe, Helen, 31, 34, 36 Gaddafi, Muammar, 97, 152 Galliott, Jai, 39–40, 41

6686_Enemark.indd 198

Germany, 173, 181 Gilligan, Carol, 135–6, 141 Great Britain (GB) see United Kingdom (UK) Gregory, Thomas, 139 Groenhout, Ruth, 141 Gusterson, Hugh, 3 Guterres, António, 172, 181 al-Harthi, Ali Qaed Senyan, 54, 57, 75 Hobbes, Thomas, 101, 103 human dignity approach, 156, 178, 181 human rights interventions, 13–14; see also international human rights law (IHRL) humanitarian interventions, 12–13, 14, 23 imminence doctrine applied to the killing of Reyaad Khan, 105–6 ‘Caroline test’, 105, 106 continuing and imminent threat, 83, 94, 108 interpretations of, 105–8 for law enforcement drone strikes, 57 self-defence justifications, 105 in UK drone policy, 101 in US foreign policy, 83, 94, 105 intelligence for drone strikes, 153–4 security service spending, 96 International Committee of the Red Cross (ICRC), 181, 183 International Criminal Tribunal for the Former Yugoslavia (ICTY), 53

02/03/21 12:22 PM

Index / 199

international human rights law (IHRL) alongside IHL, 64–5, 66–7, 193 in relation to autonomous armed drones, 170–1 the right to life, 54–5, 57, 78–9 for security operations, 64–5 state violence and, 77 supranational expansion of state borders, 102 international humanitarian law (IHL) for active hostilities, 64–5 alongside IHRL, 64–5, 66–7, 193 armed conflict categorisation and, 58–9, 63–4, 77 law enforcement drone strikes and, 52–8, 64–5, 77, 192, 193 Martens Clause and, 179–80 militarised counterterrorism strategies, 60–2, 66 in relation to autonomous armed drones, 155, 162, 165, 170–1 the right to life, 54, 55, 66, 78–9 the Tadić criteria and, 53, 54, 59–60 international law borders and sovereign territory, 101–2 legal basis of the UK’s drone policy, 96–100 meaningful human control (MHC) concept, 177 self-defence justifications, 99 unable or unwilling test, 100, 101 Introna, Lucas, 160 Iran, 23, 149, 172 Iraq, 20–1, 32, 62, 97–8, 99–100, 140

6686_Enemark.indd 199

ISIS, 25, 38, 95, 96, 97–8, 99–100, 101, 109, 124 jus ad bellum (the right to resort to war) applied to ISIS, 99 asymmetric warfare and, 39 ethics of armed drones, 2, 19, 131–2, 194 jus ad vim in relation to, 19, 34, 36, 41 reasonable prospects of success, 16–17 jus ad vim (the right to resort to force short of war) as a distinct third moral framework, 32–3, 34 within Just War theory, 31–2, 35–7 in relation to jus ad bellum, 34, 36 revisionist critiques of, 31, 32 in riskless warfare, 19, 20 Walzerian support for, 31 within the war of ethics, 33–5 jus in bello (the just conduct of war) alongside an ethics of care, 140, 141, 142, 144, 145 in asymmetric warfare, 11–12, 17 decision-making principles of, 131 ethics of armed drones, 2–3 non-physical harm to civilians, 131–2, 134–5, 139 principles of distinction and proportionality, 11, 18, 37, 131 retaliation, 17–18, 19 revisionist Just War theory and, 16

02/03/21 12:22 PM

200 / Index

Just War theory applied to counterinsurgency campaigns, 139–40, 144 applied to drone warfare, 2–3, 76 classical, 14, 15–16, 31–2, 35, 38, 40, 42, 131–2, 191 jus ad vim (force short of war) within, 31–2, 35–7 just cause, 42–3 moral rules for, 122 principles of distinction and proportionality, 11, 18, 131 reductivist arguments, 34 revisionist, 14, 15, 16, 31, 33–4, 35, 36, 39, 40, 41, 42, 191 Thomistic idea of, 35–6, 38 Walzerian notion of, 33–4, 35 justice paradigm borders and sovereign territory, 101–2 excessive redemptive violence, 85 legal justice, 79, 80–1, 82, 86 moral language of ‘justice’, 135, 136 retribution/vengeance distinctions, 44 sham trials, 83 state of nature and, 101, 103 of the war on terror, 81–2 wild justice, 79–80, 82–3, 86–7, 195 Kahler, Miles, 176 Kahn, Paul, 10, 11–14, 20–1, 23, 24–5, 39, 102 Kahneman, Daniel, 159–60 Kennedy, David, 59 Khan, Reyaad, 93, 94, 95, 97–8, 99–100, 104, 105–6, 192 Killmister, Suzy, 12

6686_Enemark.indd 200

Kirkpatrick, Jesse, 119 Kissack, Robert, 176, 178 Koh, Harold, 59, 83 Kohlberg, Lawrence, 135 Kosovo, 4, 14 law enforcement drone strikes armed conflict categorisation and, 58–9, 63–4, 77 counterterrorism mandate of, 59–60 drone courts, 86–7 drone strikes as militarised counterterrorism activities, 60–2, 66, 194 as extraterritorial law enforcement, 58–66 imminent threat to life criteria, 57 as imperial policing via airpower, 51–2, 58, 62–3 within international law doctrine, 52–8, 64–5, 66–7, 77, 192, 193 narratives of, 50–1 public perceptions of, 95 the right to life, 54–7, 66, 78–9 as wild justice, 79–85, 193 see also counterterrorism; personality strikes Lee, Peter, 119 legalism principle of, 78–9 in relation to personality strikes, 79, 86 in US foreign policy, 80–1 Lethal Autonomous Weapons Systems (LAWS) accountability guiding principles, 160, 165–6, 194

02/03/21 12:22 PM

Index / 201

accountability safeguards for, 160–5, 171 automated facial recognition systems, 160–1, 194 autonomous system development, 157–8 bias avoidance, 150, 159–61 context of use of lethal force, 56 future directions for, 165–6 international humanitarian law and, 155, 162, 165 systems and mission checks, 163–5 see also autonomous armed drones (AADs) Levine, Daniel, 139, 144 Libya, 21, 23 Luban, David, 44 Maynard, Katherine, 79 McCudden, James, 150 McNail, Sonia K., 158 military ethics education for drone operators, 116–17 utilitarian approaches, 116, 121 virtue-based approaches, 115–16 military virtues adoption of the cardinal virtues, 117–18 courage, 117, 118–19 moral courage, 117, 118, 119–20 professional loyalty, 117, 120–1 respect, 118, 121 rule-based ethics, 115, 116, 121–2, 124 virtue-based military education, 116–17 Miller, Ian, 119 morality abstract moral reasoning, 136–7, 138–9

6686_Enemark.indd 201

of drone violence, 2–3 moral challenge of observation of a target, 130, 131, 133–5, 142, 143–4, 191–2 moral communities, 12–13, 14 moral courage, 117, 118, 119–20 moral equality of combatants, 39–40 moral language of ‘justice’, 135, 136 moral maturity, 135 reductivist arguments, 34 rule-based ethics and, 115, 116, 121–2, 124 Mordaunt, Penny, 106 multilateralism Alliance for Multilateralism (AFM), 173 consensus-based institutions, 174, 176–8 current breakdown in, 172–3 majoritarian-based institutions, 174–6 privilege of powerful states, 173–4, 176, 177–8 the Trump administration and, 172 Murray, Daragh, 64, 65 Noddings, Nel, 136 non-international armed conflicts (NIACs) active hostilities during, 64–5 drone strikes within, 53, 59 IHL and IHRL in tandem for, 64–5, 66–7 against ISIS, 99–100 Tadić criteria applied to, 53, 54, 59–60 US counterterrorism strategies within, 60–2, 63

02/03/21 12:22 PM

202 / Index

Obama, Barack, 77, 83–5, 86, 95, 104, 131, 172, 176 Pakistan, 21, 53, 61, 63, 81 Panetta, Leon, 83 Perry, Mark, 183–4 personality strikes as administrative executions, 83–4 armed conflict categorisation and, 77–8 characteristics of, 61, 75 curbs for wild justice, 86–7 domain of justice/ drone domain distinction, 100–5, 109, 193 drone courts, 86–7 ethical debates on, 75–6 judicial authorisation of, 86, 87 killing of Ali Qaed Senyan alHarthi, 54, 57, 75 killing of Anwar al-Awlaki, 82–3, 104 killing of Jamal al-Badawi, 61–2, 74–5 killing of Kamal Derwish, 103–4 killing of Qasem Soleimani, 149 killing of Reyaad Khan, 93, 94, 95, 97–8, 99–100, 104, 105–6, 192 legalism and, 79, 86 as lethal law enforcement, 78 militarised counterterrorism strategies, 62 under the Obama administration, 83–5 observation of targets, 133–5, 142, 143–4, 191–2 as retributive targeted killings, 44–5 under the Trump administration, 85–6

6686_Enemark.indd 202

US administrative process for, 84–5, 94 policing asymmetry rationale of, 36, 39–40 imperial policing via airpower, 51–2, 58, 62–3 law enforcement drone strikes as, 62 paradigm of, 32, 33 the right to life, 56–7 riskless warfare as, 11, 14 use of lethal force, 56 Project MAVEN, 154–5 Reichberg, Gregory M., 41 retributive punishments, 44–5 right intention criteria, 44–6 right to life, 54–7, 66, 78–9 risk and the ethics of violence, 11 and humanitarian interventions, 12–13 perceptions of and bias, 159–60 reduced risk to personnel, 2, 3, 4 riskless warfare asymmetric warfare and, 12, 14, 16–18, 21, 190–1 civilian casualties, 21–2, 25 combatant honour, 122–4 democratic legitimacy of wars, 12, 13, 22 humanitarian interventions, 24–5 impact on military culture and organisation, 26 jus ad vim considerations, 19, 20 jus in bello retaliation, 11–12, 17–18, 19 mutual imposition of risk, 11 as policing, 11, 14

02/03/21 12:22 PM

Index / 203

and the protection of personnel, 12, 13, 19–20, 22 pursuit of, 11–13 in relation to Just War theory, 14, 191 sovereignty considerations, 13–14, 18–19, 23–4 thresholds for war, 13 US military operations, 20–1 as ‘virtue-less’, 122–3 rixa, 38 Ruddick, Sara, 138–9 Rycroft, Matthew, 98 Satia, Priya, 62 Savulescu, Julian, 23 self-defence justifications counterterrorism strikes, 94, 98 ethics, 11, 13 imminence doctrine, 105, 108 within international law doctrine, 99–100 just cause, 42–3 postmodern states, 108–9 UN Charter, Article 51, 94, 102, 106 Sharkey, Amanda, 156 Soleimani, Qasem, 149 Somalia, 21, 25, 53, 61, 63 sovereign authority aim of the common good, 38, 42 within the justice paradigm, 101 as non-applicable to terrorist actors, 38, 40–1 postmodern states, 102, 108–9 right to wage war/authority criterion, 35–6, 37–41 riskless warfare as threat to, 13–14, 18–19, 23–4

6686_Enemark.indd 203

Sparrow, Robert, 119 state of nature, 101, 103 Strawser, Bradley, 39, 41 Syria, 6, 23, 38, 97–8, 100, 104 Terry, Jillian, 139 Trump, Donald, 74, 85, 172 Tversky, Amos, 159–60 United Kingdom (UK) borders and sovereign territory of, 103 Brexit, 172 domain of justice/ drone domain distinction, 104–5, 109, 193 drone strikes by, 152 enemy identification, 96 geographical scope of the drone policy, 100–1 interpretations of imminence, 105–8 legal basis of the UK’s drone policy, 96–100 new lethal drone policy, 93–4, 100–1, 105–6, 107–8, 109–10 non-international armed conflict with ISIS, 99–100 policy on autonomous armed drones, 155, 157, 181 press coverage of drone strikes, 152–3 security service spending, 96 strike against Reyaad Khan, 93, 94, 95, 97–8, 99–100, 104, 105–6, 192 validation of lethal drone policy, 94–5 United Nations Convention on Certain Conventional Weapons (CCW), 171, 173–4, 175

02/03/21 12:22 PM

204 / Index

United Nations Group of Governmental Experts (GGE) on AADs, 171, 176, 177 United Nations Human Rights Council, 171, 172, 176 United Nations Security Council, 173 United Nations (UN), 170 United Nations (UN) Charter, 42, 94, 101, 102, 106, 162 United States of America (USA) administrative process for personality strikes, 84–5, 94 anti-personnel landmine policy, 172, 176 armed conflict categorisation and, 59, 63–4, 77–8 asymmetric military operations, 20–5 cluster munitions policy, 172 condolence payments, 140, 143 extraterritorial drone strike narratives, 50–1 militarised counterterrorism strategies, 60–2, 63–4, 66, 194 pre-emption policy, 105, 108 Project MAVEN, 154–5 response to 9/11, 81 retaliation against Iran, 23 suspicion of supranational institutions, 102–3

6686_Enemark.indd 204

tradition of violent law enforcement, 79–80 war against terror, 18, 23–4, 32, 81–2 see also Yemen USS Cole, 57, 61–2, 74–5 Veuthey, Michel, 179 virtue ethics, 115–16, 118 Walzer, Michael, 14, 16, 31, 32–3, 39, 40, 116 war against terror, 18, 23–4, 32, 81–2; see also counterterrorism Weber, Max, 101 William, Marra, 158 Wolfowitz, Paul, 75 Wood, David, 160 Wright, Jeremy, 99–100, 106–7 Yemen killing of Ali Qaed Senyan alHarthi, 54, 57, 75 killing of Anwar al-Awlaki, 82–3, 104 killing of Jamal al-Badawi, 61–2, 74–5 killing of Kamal Derwish, 103–4 US drone strikes in, 21, 23, 25, 32–3, 53, 54, 61–2, 63, 74–5 Yoo, John, 107

02/03/21 12:22 PM