How Consistent are Interpretations of the Principle of Proportionality in Situations of Asymmetric Conflicts?

Table of Contents

1. Introduction
1.1 Structure and Argument

2. Literature Review
2.1 What to Balance?
2.2. How to Balance?
2.3. Precision Strikes

3. Methodology
3.1. Approaching the Study
3.2 Limitations
3.3. What I will add

Chapter I: Identifying Inconsistency – An Empirical Analysis
4.1. What to Balance?: A Matter of Distinction
4.2. What to Balance? Military Advantage
4.3. What to Balance?: A Matter of Precautions
4.4. How to Balance?: A Matter of Weighing

Chapter II: The Legal Framework – A Purposive Analysis
5.1 Two Sides of the Same Coin
5.2. Regulators of Norm Conflict

Chapter III: Proportionality Balancing – The Normative Way
6.1. Systemic Interpretation: The Purpose of Protection
6.2. The Crux With US COIN Policy

7. Conclusion

8. Appendix

9. Bibliography

Abstract

In this paper I present a holistic approach to identify how consistent interpretations of jus in bello proportionality regarding precision strikes in the context of counterinsurgency operations are by examining both consistency in substance and process. I hypothesize that there is high inconsistency in proportionality balancing, and will illustrate that it is both a theoretical and practical problem by using a mixed-methods design of empirical and theoretical analysis. Doing so, I will move beyond a positivist reading of the law to present my argument though a normative framework. I will argue that to fully comprehend this inconsistency one needs to understand the normative relationship between IHL and IHRL. Structurally, I begin the thesis with an empirical analysis of case law and of the counterinsurgency policy of the US to demonstrate (in)consistency by rethinking it in four specific categories: scope, content, time, certainty. The purpose of this section is to evidence variance in interpretations of the proportionality variables and of the balancing act in ex-post (jurisprudence) and ex-ante (policy) assessments. Secondly, I will engage in a purposive analysis of the legal framework presenting theoretical approaches on the relationship of IHL with IHRL. This is to dismiss inconsistency in proportionality balancing as a ROL problem by showing that there is a higher (normative) purpose behind it. In the final chapter, I will apply the theoretical findings to the empirical discoveries to demonstrate that inconsistency is due to a “proportionality continuum”. This section will illustrate that inconsistency is not inherently bad as long as it serves the protective purpose intrinsic to both IHL and IHRL, which suggests changing policy rather than the law.

Key Words: Arbitrary deprivation of life, asymmetric conflicts, balancing, civilians, collateral damage, combatants, counterinsurgency, distinction, grey area operations, humanity, international humanitarian law, international human rights law, jurisprudence, policy, precautions, precision strikes, proportionality, protection, targeted killings, security, United States;

List of Abbreviations

illustration not visible in this excerpt

“The greater concern with drones is that because they make it easier to kill without risk to a State’s forces, policy makers and commanders will be tempted to interpret the legal limitations on who can be killed, and under what circumstances, too expansively.”^[1]

Former Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions Phillip Alston on Precision Strikes

“That balancing is difficult when it regards human life. It raises moral and ethical problems. Despite the difficulty of that balancing, there's no choice but to perform it.”^[2]

President (Emeritus) A. Barak on Proportionality Balancing

List of Tables and Figures

Table 1: Proportionality And the Right to Life

Table 2: Comparative Analysis (Case Law/Policy)

Table 3: US COIN Policy At A Glance

Graph 1: IHL Proportionality - An Inverse Relationship

Graph 2: IHRL Proportionality – Law Enforcement (ECHR)

1. Introduction

Much has been written on JBP in conventional AC. It is an IHL principle laid down in AP I, Art. 51/5(b) and Art. 57/2(iii), which necessitates the convergence of military interests with humanitarian values. It stipulates as disproportionate “[...] an(y) attack which may be expected to cause incidental loss of civilian life, injury to civilians, damage to civilian objects, or a combination thereof, which would be excessive in relation to the concrete and direct military advantage anticipated.” The principle aims to humanize warfare, while it also acknowledges that in cases of military necessity the killing of even civilians is warranted. However, the changing nature of warfare and the rapidly evolving means and methods of combat since WW II have started to challenge its interpretation. Particularly, the emergence of NSA and the development of precision technology have produced a certain asymmetry^[3] that has become a feature of modern-day combat. It is in light of these realities that we need to rethink JBP and re-evaluate its fitness for keeping up with the spirits of our time.

Acknowledging this challenge I have begun this paper with the aim of addressing the following question: How consistent are interpretations of PP in situations of asymmetric conflicts? The focus of analysis is on the context of drone strikes in COIN operations. I want to highlight the following: First, I define consistency in terms of substance and process. This means consistency regarding the balancing variables (what?) and the balancing act (how?). Second, COIN operations embody “grey area operations” conveying legal uncertainty. This is due to inconclusiveness about individuals’ status on the ground (e.g. “farmer by day, fighter by night”) (Alston, 2010, para. 61) and to whether a situation amounts to an AC (e.g. Pakistan). (Heynes, 2012, para. 80-81) This poses additional challenges for interpretations on proportionality necessitating an analysis of IL’s normative framework. To fully answer this question I move beyond a purely positivist reading of the law and present a normative outlook on proportionality considering consistency from the angle of systemic coherence. In this sense, my research will be useful to understand what inconsistency in interpretation means in a world that has experienced a trend towards precision technology increasing asymmetry in combat.

1.1 Structure and Argument

In chapter I, I have identified inconsistency in interpretations of JBP by comparing selected case law and US COIN policy according to four specific categories identified in the literature review: scope, content, time, certainty. The purpose of this analysis is to demonstrate if variance can be found with view to the proportionality variables (substance) and the balancing act (process) in a real-world context, with the courts’ jurisprudence embodying ex-post assessments, and the latter comprising ex-ante directives for its evaluation at the time of combat. In chapter II, I have presented a purposive analysis of the legal framework contemplating the ethical foundations of IHL and IHRL, and their relationship in times of absent or conflicting norms. It shows that (in)consistency in interpretation of JBP is subject to a higher (normative) purpose of IL. In chapter III, I have combined the findings from both analyses to demonstrate the limitations and consequences of a purely positivist reading of IL. The purpose of this section is to draw a normative analysis of consistency to show that variance in interpretation of JBP is not inherently bad as long as it upholds IL’s systemic coherence.

Overall, I argue to fully comprehend inconsistency in interpretation of JBP one needs to understand the normative relationship between IHL and IHRL. In this respect, I have demonstrated the following: First, inconsistency is due to a “proportionality continuum” with hard cases in the middle (Schmitt in Barber, 2010, 476) rather than to legal relativism. Second, inconsistency is not bad as long as it serves the protective purpose of both IHL and IHRL. Third, we are confronted with a policy dilemma, not a ROL problem.

2. Literature Review

JBP has been subject of wide debates in the academic literature. Most discussions relate to its applicability in conventional AC. It has only been with the rise of asymmetric conflicts that the literature has started to address the deeper implications proportionality poses. Concerning precision strikes, debates on JBP have just started to emerge featuring a number of concerns for academics and practitioners alike, which prompt further investigation. In this chapter, I have provided an overview of current debates on JBP relating to its constituting elements, its requirement for balancing, and its implications as to targeted killings via drone strikes in COIN operations. I consider these categories specifically as they are the most essential for the subsequent analysis.

2.1 What to Balance?

JBP requires an estimate of CD and an assessment of MA. (Rule 14, ICRC) Most literature addresses the former requirement from two angles: from within the context of discrimination and precautionary measures (Barnidge 2011; Bartels 2013; Craig 2015; Dinstein 2004; Hampson 2010; Henderson 2009). Frequently discussed issues centre on what to include and exclude in the CDE (Fenrick 2000) relating to the difficulty of differentiating between civilians and combatants in asymmetric conflicts including the question of human shields (Bartels 2013, Dinstein 2004, Holland 2004; Wright 2012), and between civilian from military objects given the dual-use nature of some facilities (Benvenuti 2001; Clarke 2012; Henderson 2009; Shue and Wippmann 2002; Wright 2012). Precautionary measures including the choice of weapons (Bartels 2013) as well as identifying an appropriate threshold of harm (Markham and Smith 2013) are discussed as is the closely linked notion of MA. Debates on the latter relate mostly to its content and scope addressing issues such as how “military” or “concrete” and “direct” it must be (Bartels 2013; Henderson 2009; Neuman 2004; Schmitt 2005), the relationship between force protection and MA (Neuman 2004; Geiss 2012), and the level of certainty MA and CD must exhibit. (Schmitt 2005) Despite significant research there is no consensus on most of these issues, which I suggest to group into the dimensions of scope, content, time, and certainty for the empirical part of the analysis.

2.2. How to Balance?

Moreover, JBP necessitates weighing the expected CD with the anticipated MA to establish if their ratio is “excessive”. (Rule 14, ICRC) The majority of scholarly work deals with what is “excessive”, how such evaluation can be made and from what perspective. (Gardam 2004; Green 2008; Henderson 2009; Melzer 2009) There is no consensus on “excessive” leading scholars follow either an objective approach - what a reasonable commander would do (PHPCR 2009), a subjective approach - relative to the commander’s mind (Dinstein 2004; Fleck 2013; Green 2008; Melzer 2009) - or an objective-subjective (hybrid) approach (Henderson 2009; Wright 2012). The academic literature also addresses cognitive factors (Deeks), deals with questions of human rationality in decision-making (Broude 2010), and with challenges the ex-ante requirement to weighing poses. (Bergen and Tiedemann 2010) Moreover, debates relate to the subjective nature of the balancing variables and their incommensurability. Scholars raise the question what relative values the variables would need to be assigned. (Craig 2015; Boothby 2012; Fenrick 2000, Holland 2004) While some scholars suggest there is no common measure applicable to both variables (Boothby 2012; Holland 2004) others point to a “proportionality continuum” to facilitate the balancing act. (Schmitt in Barber, 2010, 476) Altogether, the academic works fail to provide answers on a concrete methodology for balancing in hard cases.

2.3. Precision Strikes

Finally, the subject of JBP regarding precision strikes in COIN operations features a limited body of literature. Most scholarly works focus on the technological advances of precision strikes with Strawser (2010) deriving from it an ethical obligation for their use. While some maintain the use of drones could elevate CD to a zero-tolerance level considering the weapons’ precise nature (Blank 2012) others argue this raises unrealistic expectations due to imperfect information. (Clarke 2012). Findings on drone strikes causing less CD than other weapons (Etzioni 2013), are met with criticism suggesting flawed comparisons across times and weapon categories instead of independent assessments of each strike. (Braun and Brunstetter 2013a) Debates also address the asymmetric nature of precision strikes with scholars suggesting drones add to disproportionality in the means as they guarantee force protection for drone pilots. (Braun and Brunstetter 2013a) While most debates on drones centre on proportionality of the means, much research still needs to be done on proportionality of the methods (Craig 2015); on their usage. Here, the literature is largely confined to case studies on the US conduct. It examines implications of the country’s CDE methodology (McNeal 2014), signature strikes (USCIL 2014; Columbia Law School 2012; Reis Peron 2014), its Target Value Analysis (Melzer 2013; Reis Peron 2014; Sharkey 2009) and the changing nature of COIN operations. (Andresen 2014; Beran 2010; McLeod 2015; Reis Peron 2014, Sitaraman 2009; Stephens 2010)

It has become apparent that much of the literature on JBP is single disciplinary in focus with only some works mixing legal analysis with ethics or theories on cognitive behavior. They also focus on specific matters rather than examine proportionality balancing across the issue areas. Moreover, proportionality as a concept has been neglected in the literature altogether, leaving open the question whether it should have one or different meanings in different legal systems and contexts. Finkelstein et al. (2012), and Fabre (2012) only marginally touch on it in their account of just war theories on targeted killings, while Arai-Takahashi (2002) uncovers its meaning solely under the ECHR. Hence, despite significant research there is no consensus on most of the legal issues. This has been voiced among others by the UN Special Rapporteur on Extrajudicial, Summary or Arbitrary executions and his predecessor (Heynes, 2013; Alston, 2010), and suggests a gap in terms of more holistic analyses. I aim to narrow it with my research by combining the three issue areas subjecting them to a normative evaluation of proportionality.

3. Methodology

In this paper, I have followed a mixed-methods approach (Johnson & Onwuegbuzie, 2004; Denscombe, 2008; Tashakkori & Teddlie, 2010) both theoretical and empirical in nature. Empirically, I have examined six selected court cases covering domestic, regional and international jurisprudence. (Table 2) They represent a range of conflict contexts relating predominantly to COIN and/or drone strikes. (GFCJ, 2010; IHC, 2005; PHC, 2013, ICTY, 2011; ECtHR x2, 2005) I have chosen them in particular as they contain a continuous proportionality analysis (except PHC, 2013) representing best the issue areas under discussion. Also, I have analyzed COIN policy documents of the US. (Table 2, Table 3) I have chosen the US as a policy example for two reasons: 1. it is one of the leading drone nations (Farley, 2015; Tomiuc, 2012), 2. policy documents are publicly available. I have thereby compared and contrasted consistency in interpretation of the proportionality variables along four criteria identified in the literature review: scope, content, time, certainty, and of the balancing act.

Theoretically, I have engaged in a purposive analysis (Aharon, 2005; Bowman, 1995; McLachlan, 2005) of IHL and IHRL presenting theoretical approaches on their relationship and interaction in events of absent or conflicting norms. This shall help clarify questions of applicability of IHL and IHRL in COIN; a detail that is indispensable for interpretations of PP. Consequently, I have applied the theoretical findings to the empirical discoveries moving beyond a purely legal (positivist) approach (Hart, 1961; Austin, Bentham and Kelsen in SEOP, 2003) presenting a normative outlook on proportionality. (Dworkin, 1975) Hence, I have attempted to combine theory with its practical application merging the three issue areas as outlined in the literature review to approach the subject from a more holistic angle including proportionality concerning precision strikes from a methods of combat point of view. As there stated, the majority of studies on JBP have only adopted a singly disciplinary perspective including a narrow focus on an identified issue often by-passing the thematic challenge of asymmetric conflicts, and “grey area operations” like COIN missions, as well as the underlying normative (purposive) struggle between IHL and IHRL.

3.1. Approaching the Study

Understanding the requirement for reconciling two incommensurable values – those of humanity and military necessity - (Andresen, 2014) is crucial for proportionality assessments of precision strikes in COIN. Exploring how the use of force in drone strikes can be reconciled with both ends, it is in the balancing act one should expect answers. Thus, with view to my research question it is the requirement for balancing that will shed light on wether there is consistency in interpretation of JBP. I propose the question is both a theoretical and practical problem with a theoretical and practical solution. By uncovering the normative foundations the principle rests upon while bearing in mind the military and human element of combat operations, namely how the principle is applied in practice under increased pressure and time constraint, I have also raised ethical implications affecting the balancing process. Considering the many challenges the literature review has pointed to, I hypothesize we can expect to find a high degree of inconsistency in interpretation of JBP both in substance and process. It is only through the purposive analysis of the legal framework that we can see how the systemic integrity of IL can be upheld and with it consistency in terms of coherence, while at the same time CD can be minimized on the ground by opening up a pool of protective measures IHL and IHRL contribute to.

3.2 Limitations

The main limitation of this paper lies in its scope. Proportionality is highly contextual with different standards applying in different legal contexts. I have chosen to focus on JBP and have only touched on other spheres of law to the extent strictly required for the analysis. Other than that I have consciously excluded them from this paper. I have, further, restricted my research topic to COIN operations carried out by military forces for two reasons: First, drone strikes by civilian operators such as the CIA point to an even grayer area in IL since they represent private agents requiring an analysis of also ICL - this is beyond the scope of this paper; Second, policy information on precision strikes is hardly published; thus, availability and robustness of information led me to adopt the US approach restricting my choice to COIN. In this respect, counter-terrorism has not been discussed per se, finding relevance only where part of COIN.

3.3. What I will add

I will expand on the academic literature on JBP in the context of drone strikes in COIN operations by discussing proportionality balancing in a more holistic way combining the three issue areas outlined in the literature review (what?, how?, precision strikes). I have thereby focused on proportionality of the methods of combat, an issue that has so far remained insufficiently addressed with regard to precision strikes. Further, I suggest to rethink consistency in substance in terms of the four identified dimensions in the literature review: scope, content, time, certainty along which I have structured the findings of the first part of the proportionality test. Moving from an empirical to a purposive analysis of IL, I have combined practice - proportionality as applied by courts in retrospect and by states ex-ante – with theory on the normative framework of IHL and IHRL. Finally I have turned from a positivist examination of IL to a normative outlook on proportionality, a somewhat neglected approach in mainstream literature. It is in these senses that my paper will add to the understanding of how theory on the legal framework and application of PP can both collide and reinforce each other, exhibiting the power of IL states with drone capabilities choose to either bend or bow to. Politically this will raise policy implications, because in case of high inconsistency in proportionality balancing standards, states will eventually face the challenge to synthesize these different approaches into a coherent policy.

Chapter I: Identifying Inconsistency – An Empirical Analysis

JBP figures part of the legal provisions on distinction and precautions. Even though its wording in Art. 51/5(b) and Art. 57/2(iii) of AP I is identical (Table 1), its meaning differs. Balancing, thus, requires compliance with the provision from both perspectives. JBP being framed as a subsidiary principle, furthermore, stands in an inverse relationship with distinction and precautions: a violation of JBP inevitably constitutes a violation of distinction and/or precautions, while the reverse does not hold true. (Graph 1) This is because military necessity justifies CD under particular circumstances. (Bartels, 2013, 275) Therefore, finding what is proportionate requires, in a first step, to obtain clarity of the balancing variables. Only from there one can turn to the weighing itself. These two requirements are subject to scrutiny in this chapter aiming to uncover (in)consistency in interpretation within and across selected case-law and US COIN policy. (Table 2)

4.1. What to Balance?: A Matter of Distinction

Expected CD comprising civilian lives and objects, and anticipated MA constitute the balancing variables. (Table 1) Giving an estimate thereof requires knowing their content, scope, time frame and certainty, embodying the four criteria along which I have proposed to rethink the multiplicity of issues surrounding the variables. In this respect, it is striking that case-law and US policy don’t systematically deal with all four. (N.A. in Table 2) Whether this is due to their inapplicability or irrelevance in the underlying cases is subject to speculation. It is, however, a first indication of variance in the approach to balancing; e.g. with view to the scope of CD only the PHC considered this category in its jurisprudence declaring it “considerable” respectively “shockingly considerable”, two abstract terms resembling more a vague feeling than a concrete analysis. (2013, para. 5, 22) US policy, quite contrary, specifies a NCV in the ROE, which, if exceeded, requires special authorization (CDM, 2012, Enclosure D para. i(2)). This is problematic as it lacks both an explanation and justification for the threshold.

The example already touches on some of the implications proportionality poses and indicates the inherence of distinction in the CD estimate requiring the separation of combatants from civilians and of military from civilian objects. Distinction is crucial especially for its content and time dimension as IHL stipulates civilians lose their protective status upon DPH (Melzer, 2009), the very notion targetability hinges on. The PHC rules targeted killing lawful when a “combatant” or “fighter” constitutes the target. The same goes for civilians upon DPH. (2013, para. 11) The court, thereby, follows the ICRC’s position advocating a “revolving door” policy, where civilians engaged in DPH retain their right to protection during the resting stage. (ILC, 2014, 158-159) The IHC (2005), however, interprets DPH more widely, listing explicit examples of targetable persons (Table 2). It, thereby, disregards short periods of rest considering them to merely serve the preparation for further hostilities. Nevertheless, it finds in a more restrictive fashion “unlawful combatants” - a term the US has imposed on individuals engaged in terrorist activities - are civilians, which are only legitimate targets upon DPH (2005, para. 26, 35, 36, 39).

The US employs a more expansive interpretation of DPH, whose definition it has not made public. (Alston, 2010, para. 68) Sitaraman proposes “active” participation as explanation for the US approach to distinction. (2009, 31-32) This matches COIN policy as she includes a range of individuals, besides those engaging in DPH in the ICRC’s sense, such as “enemy combatants” (COIN, 2014, para. 13-18) - MOACs exhibiting a “continuous combat function” held targetable any time (ILC, 2014, 16) - associated forces (McNeal, 2014, 691), persons exhibiting certain signatures (e.g. men carrying weapons) (Columbia Law School, 2012, 33), voluntary human shields (CDM, 2012, Enclosure B para. 5) - a targeting category the IHC also agrees with (2005, para. 36) - , and in the case of Afghanistan, also poppy farmers due to the crop’s illicit nature deemed to financially support enemy activities. (COIN, 2014, para. 9-47) She further places all HVT (e.g. militant leaders) on a kill list held to justify higher CD levels. (McNeal, 2014, 702-706) Upon CE, COIN policy, moreover, excludes all military-age males in combat areas from the count classifying them as combatants (ILC, 2014, 164).

Such broad interpretation of DPH strengthens discretion under proportionality (Sitaraman, 2009, 32) and raises serious ethical implications: not only do individuals lose their IHL protections, while being subject to more permissive targeting regulations, but also farmers engaged in domestic criminal offenses find themselves deprived of due process rights and their absolute right to life under US COIN policy. Problematic is, furthermore, that the US does not disclose her criteria for civilian/combatant status including the type of signatures warranting targeting, and fails to define features of “associated forces”. (HRW, 2013) Such opacity as to who is ultimately included in the CD estimate results in variant casualty levels (see i.e. NAF vs. BIJ statistics in Reis Peron, 2014, 86), and with it, differing results on proportionality.

As to civilian and military objects jurisprudence and policy are meagre in interpretation. While the GFCJ considers the fuel tanker in its case a direct military mean because of its logistical support function (2010, 49/50), the ICTY deduces the military nature of the targeted objects (incl. dual-use facilities) from the MA (2011, para. 1899) lacking a proper methodology for distinguishing. US policy acknowledges the existence of dual-use objects but neglects to engage in an extensive analysis on targetable objects It is therefore surprising that the US maintains “positive identification” of targets ensures lower civilian casualty rates^[4] (COIN, 2013, IV-7, VII-14) This further raises the question how certain one has to be of CD. While the ICTY finds the answer in the reasonable expectation of civilians being present (2011, para. 1910), US policy follows a more concrete approach assigning probability thresholds per CDE level. However, it is unclear how she derives at them or what this means for authorities at the STAR Analysis level lacking any provisions thereof. (Table 3) Altogether, jurisprudence and policy focus on civilian lives disregarding the impact of destroyed objects on civilians.

4.2. What to Balance? Military Advantage

In a next step, JBP requires an analysis of MA. Looking at its scope, the courts’ jurisprudence is silent on it. US policy interprets it widely including the broader network effects of targets including value, depth, capacity and recuperation time of the insurgent group. (McNeila, year, 711-716) Content-wise the PHC rules in an IHRL fashion that killing cannot be accepted as the sole aim of an operation (2013, para. 11), which is in line with the ECtHR judgements prioritizing the protection of lives from unlawful violence. (2005, Isayeva II, para. 191) Notice here, that the courts have started taking recourse to IHRL. The GFCJ further deduces the MA from the military objective (2010, 64 (2)) and the ICTY from the HVT (2011, para. 1899) both lacking a proper analysis. The US, on the other hand, with her population-centric approach, has a clear but broad aim: to secure the population and win its support (COIN, 2013, 2014) This leaves open the question how military (scope) the advantage must be. As to certainty solely the GFCJ makes a statement. It considers the achievement of the objective irrelevant with the variable depending merely on the expectations at the time of balancing. (2010, 64(2)) Thus, inconsistency becomes apparent only when looking at the time dimension. While the GFCJ (2010, 64(2)) and the IHC (2005, para. 46) rule in favor of the tactical/direct nature of MA, the US also allows for strategic considerations expanding the horizon of MA, which increases leverage for lethal targeting under JBP. (COIN, 2013, 2014)

[...]

^[1] Israel Supreme Court (sitting as High Court of Justice), The Public Committee Against Torture et al. v. The Government of Israel et al. (HCJ 769/02), Petition for an Order Nisi and an Interlocutory Order submitted January 2002, Judgment of 11 December 2005, para. 46

^[2] Report of the Special Rapporteur on extrajudicial, summary or arbitrary executions, Philip Alston, Study on Targeted Killings, Report to the Human Rights Council, A/HRC/14/24/Add.6, May 2010, p. 24

^[3] I follow Finkelstein et al.’s (2012) understanding of asymmetry, expressed not only in military but also moral terms (e.g. using human shields, deception about civilian/combatant status, deliberate targeting of non-combatants) with proportionality striving to protect, not equalize conflicting parties’ capabilities. (Guiora, 2013, 18)

^[4] Contrary to US claims, McNeal identifies flawed “positive identification” as “the leading cause of harm” in Afghanistan and Pakistan. (2014, 738)