Proportionality in Customary International Law: An Argument Against Aspirational Laws of War

I. Introduction

Entertain the following hypothetical: The year is 2020 and Bashar Al-Assad has been removed from power. Syria is governed by a nascent democratic regime. In an effort to prevent destabilizing recriminations by the now ascendant Sunni population against Assad’s Shiite sympathizers, the United Nations Security Council passes a resolution calling for the creation of an international tribunal for the prosecution of persons responsible for serious violations of international humanitarian law committed in Syrian territory since 2011.1 As the tribunal canvasses the claims that require adjudication, one of the most frequently recurring alleged violations of international humanitarian law is the use of disproportionate force.2 In the process of adjudicating these alleged violations, the jurists on this International Criminal Tribunal are tasked with giving legal content to the war crime of disproportionate force. Searching for the applicable definition of proportionality, the tribunal will find itself facing little in the way of settled law. No positive international law with respect to proportionality applies to crimes committed during the conflict.3 Syria is not a signatory to the Rome Statute of the International Criminal Court (Rome Statute).4 Moreover, the prohibition of disproportionate force in the 1978 Additional Protocol I of the Geneva Conventions (AP I) is inapplicable to intrastate conflicts.5 Consequently, the tribunal will have to determine what, if any, customary international law (CIL) of proportionality can be applied to the Syrian conflict. Because case law on the subject is sparse,6 the tribunal will likely face a matter of first impression.7

This hypothetical illustrates the importance that proportionality as CIL may have for the adjudication of international law. It also underscores that proportionality as a matter of CIL is in need of clarification. This Comment provides an answer to the question that the hypothetical tribunal would face. The definition of proportionality under CIL that should be adopted is proportionality as it is defined in the Rome Statute, as opposed to the definition provided in AP I of the Geneva Convention.

The central thesis of this Comment is that the Rome Statute definition of proportionality is preferable on doctrinal, realist, and policy grounds. Because the legal status of proportionality is particularly important in the context of human shields and warfare against non-uniformed insurgencies, emphasis is placed on the strength of the Rome Statute definition of proportionality in maintaining the credibility of the laws of armed conflict while allowing for the effective and humane prosecution of modern asymmetrical war.

The Comment proceeds as follows: Section II describes the current international regime of proportionality, both in terms of CIL and positive international law and concludes with a survey of the various options a hypothetical tribunal would have before it as it considers the question of proportionality as CIL. Section III establishes the meaningful differences between the definitions of proportionality in the AP I and Rome Statute. Section IV argues—on doctrinal, realist, and policy grounds—that the Rome Statute definition of the war crime of disproportionate force should be adopted as CIL.

II. Background

Proportionality is a moral and legal norm that forms one part, along with the principles of distinction and necessity, of the holy triad of the modern law of armed conflict.8 Contrary to popular misunderstandings,9 proportionality is not a principle that limits the number of casualties one party can inflict on the other party by reference to the number of casualties they have suffered. In principle, it is entirely possible for a party that has suffered no casualties to engage in a proportional strike that results in the death of hundreds or thousands of casualties. Proportionality, rightly understood, is a principle that limits the acceptable amount of destructive secondary (non-targeted) effects an attack can produce given the anticipated military advantage from the attack. Secondary effects can include both traditional collateral damage, as well as more attenuated effects of an attack, such as power outages or environmental devastation.10 Although proportionality is sometimes seen as a logical subcategory of the principle of distinction—the obligation of belligerents to distinguish between combatants and noncombatants—it is conceptually distinct.11 The principle of distinction requires that a belligerent not aim at an illegitimate target. Proportionality, on the other hand, limits the conditions under which the trigger can be pulled even when that bullet (or bomb) is heading toward a legitimate target. Indeed, adherence to the principle of proportionality may require military forces to expose themselves to greater risk in order to avoid excessive collateral damage.12

Given the vast variety of combat circumstances, a single detailed explication of proportionality would be unworkable.13 By necessity, the principle is general and, by some commentators’ lights, irretrievably vague.14 Nonetheless, international law purports—primarily in the Rome Statute and the Additional Protocol I to the Geneva Convention—to provide a definition of proportionality that guides military behavior and serves as a legal and moral standard of accountability for states in general, and military commanders in particular, during armed conflicts. A degree of ambiguity is to be expected for such an expansive legal principle. But the fact that the current legal status of proportionality is so befogged is in part attributable to the fact that international tribunals have not yet provided direct case law on the topic.15 Which definition of proportionality constitutes CIL remains an open question. As is discussed Section II(A)(2) below, this means that the binding law for many of the world’s largest and most conflict-prone nations is currently unknown.

Escalating violence around the world16 underscores the importance of developing a sound understanding of the legal principle of proportionality. Ideally, proportionality could be defined in such a way that both reflects and reinforces a genuine international consensus and is also sensitive to the realities of twenty-first-century battlefields. This Comment’s approach is informed by a decided skepticism about a morally aspirational approach to international humanitarian law that does not adequately grapple with geopolitical and military realities. The increasing proportion of civilian casualties in modern conflicts17 in particular underscores the importance of clarifying the legal liability for the war crime of disproportionate force so that the norm of proportionality exerts real influence in current and future military operations.

Proportionality as a principle of international law is understood to function at two levels: state liability and individual criminal liability for military commanders. Both the academic and international legal communities have grappled in recent years with this conceptual distinction18 —some arguing that there should be no distinction at all. The central focus of this Comment, however, concerns the definition of CIL regarding individual liability, though the analysis may contain implications for how CIL should be understood with respect to state liability.

A.  Current State of Proportionality in Customary International Law

1. The Historical Origins of Proportionality.

The origins of proportionality provide the necessary backdrop for understanding its prevalence as a state practice and legal norm. Historical scholarship varies in identifying the specific date that proportionality emerged alongside necessity and distinction as a core feature of the laws of armed conflict.19 The contemporary iteration of proportionality in its focus on preventing civilian losses, however, is of relatively recent vintage. Current historians maintain that the modern concept of proportionality emerged sometime in the 1970s.20

The development of the principle of proportionality in the second half of the twentieth century followed the advent of high altitude bombing as a tactic of war, which brought the population into the battlefield in a way that had not typically been true of historical wars.21 In past conflicts, when noncombatants were killed in war, it was typically because they were either targeted, victims of a fairly unusual accident, or broadly targeted as a class, such as in the case of a siege.22 When the axis and allied powers of World War II began deploying high altitude nighttime aerial bombing raids, they were not (necessarily) targeting noncombatants, but nonetheless the strikes incurred a severe human toll.23 Indeed, the now ubiquitous term “collateral damage” was coined in the post-War period.24

What should be made of this? While it is true that proportionality is conceptually separable from the principle of distinction, the development of proportionality was motivated by the same underlying concern—protecting civilians from the machinations of war. Proportionality as a doctrine, however, was formulated as a response to a particular set of historical circumstances. Current articulations of the principle should thus bear in mind that the proportionality was crafted as a consequence of a recognition that abstract principles—in this case the principle of distinction applied to novel historical settings of high altitude aerial bombing—can fail to fulfill their intended purposes. As will be discussed infra in Section IV(C) advocates of applying the most expansive formulation of the proportionality principle in asymmetric conflicts may be missing the crucial lesson that principles need to be adapted to circumstances.

2. The Current Scope of Proportionality in Customary International Law.

While the definitions of proportionality contained within the AP I and the Rome Statute bind the signatory states to those treaties for the scope of activities specified in the treaties,25 CIL is the only source of legal accountability for proportionality violations for a broad range of military activity undertaken by nations. Significant non-signatories to AP I (for example India, Israel, Pakistan, and the U.S.) have declared that they are bound by the AP I only insofar as it reflects CIL.26 This may be a distinction without a difference, as the International Committee of the Red Cross (ICRC) maintains that the AP I definition of proportionality is binding as CIL,27 though others have expressed skepticism about the ICRC’s claim.28

Which definition of proportionality is binding as CIL—if either—is also important because CIL legally binds all nations.29 Additionally, very little positive treaty law regarding proportionality has legal force with respect to non-international armed conflicts, such as the ongoing conflict in Syria. The AP I applies only where regular armed forces engage the regular armed forces of a foreign state or enter the territory of a foreign state without permission.30 Thus, the aborted attempt by American forces to rescue diplomatic personnel from Iran in April of 1980 would qualify as an international armed conflict, but ongoing NATO operations in Afghanistan conducted with the permission of the Afghan government would not. The Rome Statute is likewise limited with respect to non-international conflicts. The Rome Statute only offers applicable law for the war crime of disproportionate force in international armed conflicts.31 Thus, for any judicial body determining the legal accountability for a military commander charged with a proportionality violation as a part of the NATO coalition in Afghanistan, the only relevant international law will be CIL. Only Common Article 3 of the 1949 Convention and Additional Protocol II of the Geneva Convention—neither of which makes reference to proportionality—have legal force with respect to non-international armed conflicts.32

Additionally, many major countries are not signatories to the AP I or the Rome Treaty—such as India, China, the United States, Russia, etc. All of these countries, however, are bound by CIL.33 Proportionality, as it is established in CIL, then, constitutes the only the binding law of proportionality for states governing at least two billion people, virtually all nuclear weapons, and many of the most active militaries. Because the definition of proportionality in CIL is ambiguous, for many of the countries that are likely to be involved in non-international conflict, (arguably the conflicts in Ukraine, Syria, Libya, Iraq, Afghanistan, and Sri Lanka, to name just a few, all qualify in some respect as non-international conflicts), the matter of which principle of proportionality, if any, is binding is up for grabs.34

3. Defining Proportionality within Customary International Law.

Granting the importance of determining CIL, it follows that all of the plausible definitions of proportionality should be considered. The following represent the realistic candidates for CIL proportionality as a war crime:

a) The AP I definition without qualification. This appears to be the position of the International Committee of the Red Cross in its study of CIL proportionality.35 Under the grave crimes doctrine, this would mean that even the domestic courts of the United States, China, India, and Israel should hold their military commanders criminally liable for conduct within the scope of the AP I definition.

b) The AP I definition with qualification. Perhaps the most technically accurate assessment of international legal practice would endorse this option. A great many signatories to the AP I included signing statements that qualified their assent to the open-textured language of Article 51(5)(b).36 Moreover, the fact that the U.S., India, Israel, and Pakistan have refused to sign onto the AP I and have indicated that their compliance extends only to CIL indicates that they see some gap between the AP I and CIL with respect to proportionality.37 The difficulty is in pinning down some definition that can capture the multitudinous formulations, signing statements, actual state practices, etc.38 If we expect CIL to constitute a reasonably clear and administrable rule, then a “rule” including the various formulations of proportionality found in the interstices of the AP I and the Rome Statute seems an implausible candidate for CIL.

c) The Rome Statute. Using the Rome Statute as the CIL definition of proportionality has the benefit of being a clearer and more administrable rule. But using the Rome Statute as CIL also has this advantage over using the AP I: Article 120 disallows states to make reservations to the Statute.39 Moreover, once the effect of signing statements on the AP I definition is considered, the result starts to look in many cases like the definition provided in the Rome Statute.41 The former may be desirable, as a normative matter. But it is too far out of step with actual state practice to be a realistic account of CIL. A CIL definition of proportionality more permissive than the Rome Statute, meanwhile, would be dramatically at odds with the international community’s understanding of the relationship between CIL and international humanitarian law. 43

This section has provided an overview of the historical background and current scope of proportionality in international law, as well as the potential future options for CIL proportionality. The only viable options involve the total or partial adoption of either the AP I or Rome Statute definitions of proportionality. The task of Section III will be to analyze the difference between the AP I and Rome Statute definitions of proportionality before arguing in Section IV for the adoption of the Rome Statute as proper definition of proportionality in CIL.

III. The Differences Between The Rome Statute And Geneva Convention Regarding Proportionality

Article 51(5)(b) of the AP I prohibits attacks which “may be expected to cause” injuries or damage to civilians “which would be excessive in relation to the concrete and direct military advantage anticipated” as violations of the proportionality principle.44 The Rome Statute by contrast, offers a different definition of proportionality violations, prohibiting attacks “intentionally launch[ed]…in the knowledge that such attack will cause” injuries or damage to civilians “which would be clearly excessive in relation to the concrete and direct overall military advantage anticipated.”45 Though the two definitions are structurally similar, the terms of the Rome Statute create a higher threshold mens rea, specifically through the additional requirements of intention and knowledge. Moreover, the Rome Statute definition requires injury or damage to civilians that is clearly excessive rather than merely excessive in the AP I definition. Lastly, the Rome Statute measures the degree of injury or damage to civilians in relation to “the concrete and direct overall military advantage” (emphasis added).46 The deliberate inclusion of the term “overall” indicates a greater degree of flexibility in the proportionality calculation.

A. The Differences between the Definitions

Recent scholarship has suggested that there is little practical difference between the AP I and Rome Statute definitions.47 The analysis below argues that this conclusion is incorrect. Taken seriously as legal standards, the difference in the AP I and Rome Statute definitions could have broad and deep implications in the laws of war.

First, the different functions of the Rome Statute and the AP I need to be clarified. The Rome Statute was drafted as a means of providing individual criminal liability for violations of international crimes over which the International Criminal Court (ICC) asserts subject matter jurisdiction, including war crimes, crimes against humanity, genocide, and crimes of aggression.48 The AP I, meanwhile, serves largely as a prescriptive, action-guiding document that lays out step-by-step instructions for commanders and soldiers to respect international law while engaged in military conduct.49 But the AP I has a grave breach provision that obliges signatory nations to repress and treat as war crimes certain violations of the AP I.50 Thus while the Rome Statute and the AP I serve different general purposes as legal regimes, both have a retrospective criminal law function.

As a preliminary matter, the text of the AP I may not contain a freestanding grave breach provision regarding proportionality. Article 85, which outlines the criteria for grave crimes, seemingly includes proportionality by making explicit reference to the language of proportionality as it is defined in Article 57. However, the reference to Article 57 is preceded by the following language, “launching an indiscriminate attack affecting the civilian population or civilian objects in the knowledge that such attack will cause excessive loss of life . . . .”51 Thus, a lawyer defending an alleged war criminal accused of a proportionality violation under the AP I could very well argue that only indiscriminate disproportionate attacks qualify as grave breaches. As noted above, the crime of indiscriminate military conduct is a related but conceptually distinct grave breach. Proportionality violations generally presuppose that the military commander is discriminating between combatants and noncombatants, but asks the further question of whether the anticipated harm to noncombatants of a given attack is justified in light of the anticipated military advantage to be gained. By limiting criminal liability for proportionality to the universe of indiscriminate attacks, this reading of the AP I diminishes the legal basis upon which a tribunal could act to convict military commanders who used disproportionate force.

On the whole, the AP I is generally read to create more liability for proportionality violations than the Rome Statute, but at least in this respect, a major difference between the two could be that the AP I is entirely lacking a grave breach provision regarding proportionality qua proportionality. This textual difficulty notwithstanding, the AP I’s grave breach provision is broadly understood to include violations of proportionality regardless of the question of discrimination.52

Another place where the elements of the war crime of a proportionality violation differ between the Rome Statute and the AP I is in their respective mens rea requirements. Under Article 85 of the AP I, for a violation of proportionality to qualify as a “grave breach,” it must be willful and done with knowledge.53 The International Criminal Tribunal for the former Yugoslavia applied this language and interpreted “willfully” in the context of war crimes as “incorporat[ing] the concept of recklessness, whilst excluding mere negligence, and therefore that [t]he perpetrator who recklessly attacks civilians acts ‘willfully’.”54

The Rome Statute, in contrast, requires a showing of knowledge. The material and mental elements for proportionality violations are as follows:

1. The perpetrator launched an attack.

2. The attack was such that it would cause incidental death or injury to civilians . . . and that such death [or] injury . . . would be of such an extent as to be clearly excessive in relation to the concrete and direct overall military advantage anticipated.

3. The perpetrator knew that the attack would cause incidental death or injury to civilians . . . and that such death [or] injury . . . would be of such an extent as to be clearly excessive in relation to the concrete and direct overall military advantage anticipated.55

Taken in combination with the plain language of Article 8(b)(2)(i)-(iv), the Rome Statute raises the bar for criminal liability by requiring evidence of both intention (in launching the attack) and knowledge (of the proportionality violation). The mens rea requirement of knowledge is a source of doctrinal difficulty and raises questions of whether knowledge should be assessed subjectively (the defendant knew) or objectively (the defendant should have known). Some scholars have argued that the Rome Statute’s mens rea requirement of knowledge can be expanded to include an unreasonable erroneous belief about factual circumstance, thus “objectifying” the knowledge requirement and lowering the bar for liability to include conduct that was done with a reckless or negligent state of knowledge.56 However, this reading is rendered implausible by the presence of Article 30 of the Rome Statute which states that “[u]nless otherwise provided, a person shall be criminally responsible and liable for punishment for a crime within the jurisdiction of the Court only if the material elements are committed with intent and knowledge.”57

Another important distinction with respect to mens rea is that the Rome Statute’s formulation indicates that proportionality liability is necessarily an ex ante question. As such, the inquiry into liability focuses on the decision of the military commander rather than the effect of the attack, while the AP I formulation has been interpreted to include a results-oriented, ex post analysis.58 Because of the unpredictable nature of conflict in real time, there are doubtless attacks that appear to be disproportionate ex post that were not disproportionate ex ante. Consequently, an ex post analysis would likely affect the legal outcome in adjudications of the war crime of disproportionate force. This distinction could also have serious ramifications for drafting military policy and for the potential behavior of military commanders,59 and could therefore change the decision-making calculus of those potentially liable for a proportionality violation—for better or worse.

Another salient distinction between the Rome Statute and the AP I definitions of proportionality can be seen in the Rome Statute’s “clearly excessive” language preceding “overall military advantage” where the AP I says only “excessive.” The limitation of proportionality liability to cases of clearly excessive force opens a range of possible justifications for a particular attack; this appears to be have been a deliberate choice by the drafters.60 As a scholar studying the differences between the definitions has noted: “[t]he word ‘clearly’ implies a margin of appreciation, such that only in cases where the disparity between military advantage and collateral damage is somewhat gross and obvious will the offence have been committed.”61 The significance of this terminological difference in practice can be seen in the contrast between the ICC Office of the Prosecutor’s handling of allegations of proportionality violations by coalition forces in Iraq in 2006 and the U.N. Goldstone report applying the AP I definition of proportionality to Israeli attacks in “Operation Cast Lead.”62 In the former case, the Office of the Prosecutor justified not initiating an investigation partially on the following grounds:

Article 8(2)(b)(iv) draws on the principles in Article 51(5)(b) of the 1977 Additional Protocol I to the 1949 Geneva Convention, but restricts the criminal prohibition to cases that are “clearly” excessive . . . with respect to Article 8(2)(b)(iv) allegations, the available material with respect to the alleged incidents was characterized by . . . a lack of information indicating clear excessiveness in relation to military advantage.63

The U.N. Goldstone Report, in contrast, concluded that Israeli attacks targeting Hamas police constituted a disproportionate use of force under the AP I definition of proportionality even though:

the Mission has earlier accepted that there may be individual members of the Gaza police that were at the same time members of the al-Qassam Brigades . . . [e]ven so, the Mission concludes that] the deliberate killing of 99 members of the police at the police headquarters and three police stations . . . failed to strike an acceptable balance between the direct military advantage anticipated (i.e. the killing of those policemen who may have been members of Palestinian armed groups) and the loss of civilian life (i.e. the other policemen killed and members of the public who would inevitably have been present or in the vicinity).64

The Report allows that an acceptable balance can be struck, but gives no indication that a “margin of appreciation” ought be given to the military commanders.65 While these two instances of alleged proportionality violations were undoubtedly distinct in many ways, it is clear that the decision by the Rome Statute’s drafters to render only those attacks that are clearly excessive in relation to the overall expected military advantage limits the realm of liability for disproportionate use of force.

B.  Implications of the Definitional Difference

Perhaps most importantly for the modern military era, the difference in definitions has an effect on the legality of military responses to the use of human shields in asymmetrical conflict. While the Rome Statute does not provide carte blanche permission for attacks on targets with human shields, its definition certainly expands a military commander’s latitude in such engagements relative to the AP I’s definition.66

As noted above, the intent element of the Rome Statute’s definition of a proportionality violation changes the analysis of a commander’s decision, raising the bar of liability. However, while liability is less expansive under the Rome Statute’s definition, an attack on a human-shielded target could meet the definition of “intentionally launching an attack with knowledge that such attack will cause” injury or damage to civilians. While this formulation may protect commanders in situations where it is unclear whether or not human shields are being used, just looking to the mens rea requirements, the Rome Statute and the AP I may yield the same conclusion on a prototypical human shield case.

A more categorical difference, however, may emerge with regard to the inclusion of “overall military advantage” language in the Rome Statute.67 This language uniquely invites a consideration of what tactics are appropriate in light of a broader view of the conflict. First, commanders facing human shields can argue that not being able to fire upon targets shielded by civilians provides their foe an enormous tactical advantage of immunity from retaliatory attack.  Showing that the inability to target militant positions renders a battle unwinnable or a broader military campaign unachievable could offer compelling grounds for allowing, at least in limited circumstances, attacks on human shielded targets. Second, it may be argued that a policy of consistently firing upon military targets with human shields will protect more civilians, because it will reduce or remove the incentive for opposing militants to use human shields as a battle tactic. If the foregoing is true, it raises a possible argument for granting commanders latitude to attack human shielded targets grounded in the secondary effects doctrine of proportionality. The secondary effects doctrine of proportionality requires military commanders to take into consideration not only immediate harms, but broader harms such as depriving a population of access to potable water or electricity. If a secondary effect of strenuously avoiding incidental harm to civilians is an enhanced incentive on the part of insurgent forces to use human shields, then an overly strict definition of proportionality could itself be disproportionate.68

The AP I definition provides less of an opening for flexible application of the proportionality principle in light of asymmetric conflict. By expanding mens rea liability to recklessness and localizing the calculation of the overall military advantage gained by a strike, the AP I formulation makes deliberate strikes against human shields difficult, if not impossible, to justify.69

These distinctions matter because while the principle of proportionality is itself vague, there is evidence that the principle, and thus its formulation, has a concrete affect on military strategy.70 For example, U.S. military planners at the outset of the first Gulf War, in order to comply with international law of armed conflict, followed a multi-step protocol that required: 1) an analysis of whether a target was legitimate; 2) whether a legitimate target would result in the disproportionate use of force; and 3) whether the weapon proposed for the target would result in disproportionate force.71 Lest one think that such a process amounts to little more than a paperwork hurdle, multiple targets were declined because of proportionality concerns.72 The language used in international law is reflected in national and military law, including even the law of nations that are non-signatories.73 The heightened bar of liability under the Rome Statute could very well affect the outcome of legal proceedings. For all the reasons stated above, it is clear that there are significant differences between the Rome Statute and the AP I definition of proportionality. Those differences have the consequence that the AP I’s definition would hold more military conduct liable for the war crime of disproportionate force than would the Rome Statute’s definition.74

IV.  Rome Statute Proportionality As Customary International Law

A consideration of proportionality through doctrinal and realist frameworks shows that the Rome Statute’s definition of proportionality should be regarded as CIL.

A.  Doctrinal framework

1. First Principles of Customary International Law.

CIL is traditionally defined as follows: “[i]nternational jurists speak of a custom when a clear and continuous habit of doing certain actions has grown up under the aegis of the conviction that these actions are, according to international law, obligatory or right.”75 The Statute of the International Court of Justice more tersely defines CIL as a “general practice accepted as law.”76 Though there is a vast and discordant literature on what constitutes international custom as a source of international law, it is broadly agreed that CIL involves 1) some convergence or regularity in practice among states; 2) that convergence of practice is necessary, but not sufficient to constitute custom; 3) that there must be convergence of deliberate practice, not induced by force, fraud, or mistake; and 4) that the deliberate and volitional practice must be accompanied by a certain attitude, belief, intention, or disposition, which is called opinio juris.77 This final element, often called the “psychological” element of demonstrating CIL, proves to be the trickiest.78 But as a legal formula, a practice by nations becomes a rule of CIL when it is accepted by nations as stemming from a sense of legal obligation rather than as an exercise of policy making discretion.79

However, this formulation of CIL presents further problems when one attempts to determine if a particular practice or principle qualifies as CIL. For instance, does it matter if nations have recognized a legal obligation by explicitly codifying the principle, but violate the principle in practice?80 Some courts have answered that a CIL prohibition on torture can exist while acknowledging at the same time that torture is an ongoing practice in many countries.81 Moreover, just how universally accepted must a practice be before it can qualify as CIL?82 Canvassing all 190 or so of the world’s nations at one time is a difficult proposition and has not been required for the establishment of CIL.83 Frequently, statements by government officials or ratification of a treaty that contains a legal norm that is similar to the proposed CIL norm will suffice as a demonstration of opinio juris.84

2. Treaties.

The doctrinal approach to CIL is typically defined as a “customary practice of states followed from a sense of legal obligation.”85 123 nations are states party to the Rome Statute of the International Criminal Court.86 The AP I has 174 signatories, though 40 of the signatories attached signing statements qualifying the assent a signature would otherwise provide.87 As an example, Canada’s signing statement explicitly qualifies the AP I definition of proportionality by stating that “military commanders and others responsible for planning, deciding upon or executing attacks have to reach decisions on the basis of the information reasonably available to them at the relevant time and that such decisions cannot be judged on the basis of information which has subsequently come to light.”88 Therefore, if the quantity of signatories to the respective treaties is taken as evidence of broader opinio juris for the AP I definition, that conclusion is undermined by the reservations expressed by signatories.

3. State practice.

Looking for general state practice on the battlefield is a problematic undertaking.  First, it is not clear that a practice contrary to a potential CIL principle is relevant when there is broad opinio juris.89 Second, military practices and disciplinary procedures for commanders are notoriously opaque.90 Nonetheless, working with the imperfect information that is available, the ICRC’s study of CIL proportionality established that states almost universally accept the wisdom of the proportionality principle as a matter of public policy.91 Yet the study did not reveal a universal consensus regarding how states interpret and operationalize the proportionality principle92 ­—which is essential to determining whether the AP I’s definition of proportionality is generally practiced.

Scholar A.P.V. Rogers’ review of the history of proportionality and its predecessor concepts in the late 19th and early and mid-20th century concludes that even in cases where the AP I definition has been technically employed as the basis of CIL, courts have included mens rea requirements.93 Rogers concludes that the Rome Statute’s definition of proportionality provides a far better account of the historical and current state of CIL than the AP I definition.94

There is also a purely analytical reason that state practice supports the Rome Statute. The Rome Statute’s definition of proportionality is more likely to be a basis for CIL because of its more limited scope. Logically, every military act that violates the Rome Statute’s definition of proportionality also violates the AP I definition. In that respect, the two definitions are co-extensive. But violations of proportionality as defined by the AP I do not necessarily violate the Rome Statute’s definition. To the extent that there is support in state practice and opinio juris for the AP I definition, it works as support for the Rome Statute, but not vice versa. Assuming  state practice generally deviates downward from the AP I definition of proportionality, the Rome Statute version of proportionality has a greater degree of support as CIL.

B. Realist framework.

While the doctrinal approach is ostensibly the method by which courts determine CIL, academic commentators have observed that in reality courts rarely attempt to genuinely satisfy the doctrinal elements of CIL.95 Instead, judicial practice with respect to CIL evinces disinterest in an empirical canvassing of state practice. Courts often conflate practice with opinio juris or disregard an absence of state practice when opinio juris is present. Courts also appear to selectively employ materials to demonstrate CIL, and occasionally even hold that CIL is binding while acknowledging that the traditional criteria are absent.96

Recent prominent scholarship on the topic of CIL advances the thesis that CIL is, on its own terms, a fiction.97 On this view, courts do not really attempt to determine universal practices or opinio juris98 To support this supposition scholars cite the assumption by courts in CIL formation that the silence of nations in light of a declared CIL is evidence of the validity of the norm.99 In reality, many nations are not aware of these putative customary laws unless and until their interests are directly affected by them, making a mockery of the legal principle that consent without knowledge is impossible.100 Scholars advancing this view of CIL as non-consensual point to the practice of judges establishing new CIL on the basis of no more than mere analogy to the principles of non-binding treaties.101 Notwithstanding these observations, CIL jurisprudence is faithful to its first principles in some respects, such as in the case of diplomatic immunity.102 But for CIL skeptics, this is just the exception that proves the rule.

Beyond arguing that CIL jurisprudence does not take its own doctrine seriously, recent scholarship by Professors Jack Goldsmith and Eric Posner has endeavored to provide a realist, functional explanation of CIL. They argue that CIL regarding human rights and laws of armed conflict is in truth a mere collection of the promulgated preferences and interests of powerful nations.103 Moreover, those preferences are subject to enforcement only when the costs for the powerful nations are low relative to the gains in enforcement. This, they argue, explains the difference in the willingness of the U.S. to impose human rights based sanctions on Myanmar and Cuba on the one hand, while China and Saudi Arabia are free from such sanctions. Meanwhile international treaties and covenants, when effective, are best explained through game theory as a solution to the prisoner’s dilemma.104 But CIL does not work within the game theory framework. Because it derives its force from the practices and legal obligations of nations absent any coordination or negotiation between parties, CIL lacks the strategic basis of, for instance, the policy of diplomatic immunity.

An alternative, but not mutually exclusive account of CIL suggests that CIL’s function is emotive and value-laden.105 This approach may make more sense of CIL than a focus on the mutual self-interest of states under the Westphalian model of international law. The function of “codifying” these values as international law would be to “brand” outlier states—say those states that tolerate slavery or torture—in a way that creates both a deterrent and a potential basis for coercive measures such as sanctions.106

Taking both of these views together—that CIL is the promulgation of the preferences of the powerful, and that those preferences may be emotionally or morally motivated efforts to negatively brand nations that deviate from core moral principles—we gain a clearer sense of the realist view of CIL.107 However, the picture gets more complicated. Powerful nations are far from a homogenous group. Some nations with outsized military and economic power do not evince the emotive, value-laden desire to brand human rights violating “outliers” (for instance, China). Meanwhile, other nations with significant economic clout and less military influence are politically invested in the political project of branding moral outliers (Germany).108

To determine which definition is more likely to qualify as CIL under the realist lens, we must ask which practices are favored by global powers, and would not place undue restrictions on the ability of those nations to act in their self-interest. But as noted above, this requires some disaggregation.

In realist terms, powerful nations without an investment in the political branding project, such as China, prefer to adopt as few limitations on their military conduct as possible. A limitation may be justified only if some other benefit—such as avoiding the costs associated with being branded as a human rights violator—outweighs the harm of reduced military discretion. Clearly, this group of countries would prefer the Rome Statute formulation of proportionality to the AP I definition. For many of these countries it is consistent with their internal policies, such that the Rome Statute comes at little to no cost.109 For others, it may require a tightening of military policy, but the lost military discretion is made up for by the avoided losses of being branded an “outlier” in the human rights regime. The military practice of some nations may be significantly curtailed by the Rome Statute, and as such they may wish there were no proportionality in CIL whatsoever. But as between the AP I and the Rome Statute, the Rome Statute’s formulation of proportionality is a far likelier candidate for adoption as CIL among this class of countries.

Those nations that are largely inactive on the military stage, but whose domestic politics emphasize moral and emotional concerns do not face significant costs by the adoption of AP I as CIL. The Rome Statute definition of proportionality is more likely to garner support from this power center as well. Part of the reason for this is that these nations are economically and militarily interdependent with militarily active nations that also share the moral commitment to branding outliers (for example the U.S., U.K., and Australia) but they would nevertheless resist the adoption of an overly stringent definition as CIL.110

Meanwhile, powerful countries will not enforce something that binds them, and the AP I’s definition has the potential to bind the activities of many of the world’s active military powers.111 The Rome Statute, on the other hand, while providing a real constraint, is not so vague as to open military operations up to broad condemnation unless military commanders are clearly operating outside the normal rules of engagement.

This is particularly true as nations look to the future of conflict during the twenty-first century. Asymmetric conflict presents unique strategic, tactical, and moral challenges for modern militaries.112 The conflicts are often waged outside of traditional battlefield settings. Non-state belligerents are typically ununiformed and make efforts to blend in with non-combatants. As noted above, insurgents actively use the civilian population tactically in order to immunize their bases of attack from reprisals.113 The Rome Statute definition of proportionality is well formulated to adjust to these changing realities. In particular, it could more plausibly justify direct reprisals against targets with human shields, which may be necessary to diminish the tragic tactical advantage gained by belligerents through the use of human shields.114

The Rome Statute approach to proportionality is also more likely to garner broad buy-in and actually impact domestic, political decisions—an important realist consideration. There is evidence that international human rights law can have an impact on the development of internal practices, even without an external enforcement mechanism.115 A frequent element of international human rights laws that fail to accomplish anything more than symbolism is a combination of excessive ambition and ambiguity.116 The Rome Statute definition of proportionality, in contrast to the AP I, creates criminal liability for a more restricted and less ambiguous range of military behavior. Consequently, the Rome Statute is less likely to face widespread opposition117 and stands a better chance of changing internal policies and practices with respect to proportionality.

C. Policy considerations

Having argued that, as a descriptive matter, the Rome Statute’s approach to proportionality is a more plausible CIL candidate on both doctrinal and realist grounds, the next question is whether, as a normative matter, this is a positive development. The conclusion reached by this Comment is that it is, largely because the Rome Statute enables asymmetrical war to be more effectively and humanely prosecuted.

a) Efficacy. The more permissive Rome Statute definition of proportionality will render military operations more effective. There is no doubt that, relative to the AP I definition, the Rome Statute definition of proportionality expands the range of military conduct that is permissible under CIL. However, a strong case can be made that giving militaries more tactical latitude can harm the ultimate success of the mission.118 This is particularly true in military engagements where gaining the trust and cooperation of the domestic population is an essential condition for victory. Look no further than General Stanley McCrystal’s decision to unilaterally tighten the “Rules of Engagement” for NATO coalition troops in Afghanistan for evidence that a restriction on military choices can be strategically advantageous.119

Yet several considerations suggest that, on balance, the relative permissiveness of the Rome Statute allows for more effective military engagements. First, while there are circumstances that may call for a military to impose strict proportionality rules of engagement on itself in order to serve particular goals, such as increased goodwill among the domestic population, those circumstances will vary, even within in specific operational contexts. There is a difference between, for instance, trying to ensure goodwill in a district that has historically resisted the Taliban, such as the non-Pashtun northern regions of Afghanistan, and trying to ensure goodwill in a region like Kandahar, the population of which is more supportive of the Taliban. The strategic advantage of more or less restrictive proportionality rules of engagement would likely differ in a commander’s approach to these regions. If effectively implemented, the AP I definition of proportionality, and in particular interpretations of its broad language that expand liability, would undoubtedly affect military behavior.120 As a consequence, military commanders would work from a more limited tactical toolbox.121 Insofar as we can expect military commanders to be the best situated and incentivized actor to pursue the most efficacious tactics, and given the vast variety of circumstances individual commanders face, it stands to reason that the greater the number of tactical tools available, the greater the chances of an effective approach.122

Additionally, there is reason to believe that restrictive proportionality regimes lend critical support to insurgencies. Restrictive proportionality regimes effectively immunize areas imbued with civilians from attack, thus presenting insurgents—who are often strategically incapable of winning in open battle—an attractive point from which to launch attacks.123 In the event that military forces simply abstain from attacking those areas, the insurgent forces have gained a significant tactical advantage. But if the military forces do attack, even in a limited way that tries to reduce civilian casualties, the insurgent forces can count a tactical loss as a strategic win because of the civilian losses. Under the AP I definition, the military commander may well be held out as a war criminal.124 Because insurgencies often seek to delegitimize the ruling government over the long term, these condemnations can present a greater victory than assassinations, acts of terror, or military confrontation.125

b) Humanity. The Rome Statute’s definition of proportionality, if accepted as CIL, will better ensure the protection of civilian lives than the AP I. This position may seem counterintuitive, but is rendered plausible by considering the following factors: 1) clarity; 2) enforceability; and 3) participation. The Rome Statute definition of the war crime of disproportionate use of force provides clarity for the legal prosecution of the war crime and corollarily provides guidance to military commanders. Knowingly attacking targets that would be reasonably viewed as clearly and obviously failing to balance military advantage against civilian losses may not create a bright line, but it does create a legal standard that lines up with an unambiguous and strong moral intuition. Military commanders aware of the legal standard of the Rome Statute would both have reasonable discretion to conduct military operations and be incentivized, when presented with information indicating the possibility of civilian harm, to stop and consider the magnitude of the damage in relation to the military advantage.

Yet it could be that increased clarity in what constitutes a proportionality violation will only serve as a roadmap for evasion. When commanders suspect that they are approaching or planning an attack that might present serious proportionality concerns, one might worry that commanders would deliberately reduce the amount of information accepted about the target so as to avoid the “knowledge” mens rea required under the Rome Statute.

Though this is a possibility, several practical considerations militate against the widespread practice of deliberate evasion. First, military commanders would need just enough information to know that there are proportionality concerns, but not too much that the commander would qualify as having knowledge of the potential civilian losses. If such an epistemic space exists, it is very small indeed. Moreover, it is doubtful that many commanders would feel confident that they are occupying it when making consequential decisions. Secondly, military commanders have a multitude of reasons for wanting as much intelligence as possible regarding a potential target.126 Commanders want to be sure of the enemy’s location in targeting an attack, so as not to waste resources or risk military personnel unnecessarily. And for reasons stated above, commanders are often independently motivated to avoid excessive civilian casualties for strategic reasons.127

In terms of enforceability, the Rome Statute’s definition of proportionality has several advantages over the AP I definition. Though courts ought always to feel some discomfort in evaluating the real time decisions of military commanders from the comfort of a courtroom, proportionality is particularly challenging in that it requires a comparison of two distinct qualities: military advantage against harm to civilians.128 While the quantity of harm to civilians can be calculated by some objective measure, the quality of “military” advantage that justifies the risk of killing 20 as opposed to 70 civilians is almost non-judiciable. No artificial formula can answer this question (as some courts have found out).129 Given that fact, the Rome Statute definition provides jurists with a less ambiguous legal question. The Rome Statute’s inclusion of a clear mens rea requirement and a margin for appreciation make the legal task more manageable. Therefore, when the criminal elements are present, they more likely to be undertaken with confidence and vigor by prosecutors and courts.130

There is a threshold question of which of the two definitions stands a better chance of exerting influence on military conduct. Because of the buy-in advantage of the Rome Statute outlined above,131 it seems likely that if either standard achieves widespread legitimacy in regulating military conduct, it would be the Rome Statute’s definition. Endless discussion of the merits and demerits of a more stringent proportionality regime is ultimately meaningless if the legal standard is met with indifference or hostility by nations and their militaries. The greatest flaw of the AP I definition of proportionality can be seen in the signing statements of so many of the signatories.132 Nations simply are not willing to constrain their military conduct to the extent that the AP I definition of proportionality would prescribe.133 As such, it remains a morally aspirational standard, which, however pleasing to its authors and advocates, undermines its own chance to guide military conduct.

The Rome Statute’s definition, meanwhile, presents a legal standard that can command broader and deeper support.134 For a nation to refuse to hold its military to the Rome Statute definition of proportionality is tantamount to saying openly to the international community that that nation reserves the right to purposefully and knowingly attack targets when the damage to civilians clearly outweighs the overall military advantage to be gained. That is, simply stated, a politically—not to mention morally—untenable position, and one that few countries are likely to take. The Rome Statute definition, then, can function as a broadly supported common denominator for military conduct.

V. Conclusion

This Comment has shown that the distinctions between the AP I and Rome Statute matter for international law because the international law of proportionality affects the way wars are evaluated and ultimately fought. Furthermore, this Comment has argued that the Rome Statute’s less capacious reading of proportionality liability better meets the relevant criteria for what qualifies as customary international law. Lastly, this Comment has suggested that this descriptive state of affairs has advantages that may be welcomed as a normative matter.

Laws of war must not be too aspirational, lest they become little more than morally self-congratulatory gestures that lack purchase during conflicts. Aspirations declared in peacetime international meeting rooms are too easily disregarded under the pressure of war. Those who would bemoan this development as an evisceration of the traditional principle of proportionality must remember that proportionality was created to grapple with a specific set of historical circumstances, and so it is entirely appropriate for the application of the principle to adjust in light of a new form of conflict. Yet the concern of those who advocate for a more restrictive proportionality does not fall upon deaf ears; it is of profound importance that the rule of proportionality does not become so permissive that it describes military operations without delimiting them. By acknowledging that this law of armed conflict must straddle the Scylla of aspiration and the Charybdis of cynicism, we can see that in the laws of war, efforts must constantly be made to correctly calibrate long-standing moral principles to transient legal and military realities.

• 1Modeled after the International Criminal Tribunal for the Former Yugoslavia (ICTY).
• 2Human Rights Watch, World Report 2015: Syria, https://www.hrw.org/world-report/2015/country-chapters/syria (last visited Feb. 16, 2016)(“Between February and July [of 2014], there were over 650 new major impact strikes in Aleppo neighborhoods held by armed opposition groups. Most of the strikes had damage consistent with barrel bomb detonations. One local group estimated that aerial attacks had killed 3,557 civilians in Aleppo governorate in 2014.”).
• 3Assuming there is no ICC referral by the Security Council. In the case of referrals, ICC law applies irrespective of the signatory status of the referred nation. This is why Sudan was legally obligated to cooperate with the ICC’s order to arrest Al Bashir despite its not being a signatory to the Rome Statute. See Dapo Akande, Legal Nature of ICC referrals to the ICC and Its Impact on Al Bashir’s Immunities, 7 J. Int’l. Crim. Just. 333, 335 (2009).
• 4Rome Statute of the International Criminal Court, UN Doc A/CONF.183/9 (July 17, 1988). States Parties to the Rome Statute—Asia-Pacific States, https://www.icc-cpi.int/en_menus/asp/states%20parties/asian%20states/Pages/asian%20states.aspx (last visited, Feb. 16, 2016); Rome Statute of the International Criminal Court, Jul. 17, 1998, 2187 U.N.T.S. 90 [hereinafter “Rome Statute”].
• 5See Protocol Additional to the Geneva Conventions of 12 August 1949, and Relating to the Protection of Victims of International Conflicts (Protocol I), 8 June 1977, 1125 U.N.T.S. 3 [hereinafter “AP I”] and Protocol Additional to the Geneva Conventions of 12 August 1949, and Relating to the Protection of Victims of Non-International Armed Conflicts (Protocol II), 8 June 1977, 1125 U.N.T.S. 609; see also William Fenrick, The Rule of Proportionality and Protocol I in Conventional Warfare, 98 Mil. L. Rev. 91, 98 (1982).
• 6See Rogier Bartels, Dealing with the Principle of Proportionality in Armed Conflict in Retrospect: The Application of the Principle in International Criminal Trials, 46 Isr. L. Rev. 271, 272 (2013) (pointing out that “no case law exists to which the International Criminal Court (ICC) . . . could turn were it to be seized of a case concerning alleged disproportionate attacks”).
• 7This situation is not entirely hypothetical. Sri Lanka is, as of this writing, grappling with the process of establishing such a tribunal. Like Syria, Sri Lanka’s conflict with the Tamil Tigers was intrastate and Sri Lanka is a signatory to neither AP I nor the Rome Statute.
• 8Robert D. Sloane, Puzzles of Proportion and the “Reasonable Military Commander”: Reflections on the Law, Ethics, and Geopolitics of Proportionality, 6 Harv. Nat’l Sec. J. 299, 310 (2015).  
• 9See, for example, Eugene Robinson, It’s Disproportionate. . ., Wash. Post, (July 25, 2006), http://www.washingtonpost.com/wpdyn/content/article/2006/07/24/AR2006072400810_pf.html.
• 10ICTY, Final Report to the Prosecutor by the Committee Established to Review the NATO Bombing Campaign Against the Federal Republic of Yugoslavia, ¶ 19 (June 13, 2000). See also Commentary on the HCPR Manual on International Law Applicable to Air and Missile Warfare, http://ihlresearch.org/amw/Commentary%20on%20the%20HPCR%20Manual.pdf (“[W]hen a military objective is attacked, and expected collateral damage is assessed compared to the anticipated military advantage, the proportionality analysis also needs to take into account the expected collateral damage to the natural environment.”).
• 11Sloane, supra note 8, at 311. (“[T]he truth is that proportionality . . . .imposes a more onerous, and qualitatively distinct, constraint: arguably, it requires military forces to subject their forces to greater risks of death and injury in an effort to reduce collateral damage.”).
• 12See, for example, David Luban, Risk Taking and Force Protection, in Reading Walzer 277 (Itzhak Benbaji & Naomi Sussman eds., 2013).
• 13As William Fenrick puts it, the problem is not whether or not the principle exists, “but what it means and how it is to be applied.” William Fenrick, Attacking the Enemy Civilian as a Punishable Offence, 7 Duke J. Comp. & Int’l L. 539, 545 (1997).
• 14W. Hays Parks, Air War and the Law of War, 32 A.F. L. Rev. 1, 173 (1990) (“By American domestic law standards, the concept of proportionality [in the law of armed conflict] would be constitutionally void for vagueness.”).
• 15See Bartels, supra note 6, at 272 (pointing out that “no case law exists to which the International Criminal Court (ICC) . . . could turn were it to be seized of a case concerning alleged disproportionate attacks”).
• 16Death Toll in 2014’s Bloodiest Wars Sharply Up on Previous Year, Project for the Study of the 21st Century, https://projects21.files.wordpress.com/2015/03/death-toll-in-2014-62415-update-1.pdf (last visited Feb. 16, 2016).
• 17In World War I, civilians represented approximately 15% of deaths. In World War II that percentage dramatically increased to 65%, and in recent conflicts the proportion has risen yet again to over 84%. See Douglas H. Fischer, Comment, Human Shields, Homicides, and House Fires: How a Domestic Law Analogy Can Guide International Law Regarding Human Shields Tactics in Armed Conflict, 57 Am. U. L. Rev. 479, 484 n.30 (2007) (citing Edmund Cairns, A Safer Future: Reducing the Human Cost of War 17 (1997).
• 18George P. Fletcher & Jens David Ohlin, Reclaiming Fundamental Principles of Criminal Law in the Darfur Case, 3 J. Int’l Crim. Just. 539, 542 (2005).
• 19See Judith Gail Gardham, Proportionality and Force in International Law, 87 Am. J. Int’l L. 391, 394–403; see also A.P.V. Rogers, Law on the Battlefield 17 (2d ed., 2004).
• 20See Rogers, supra note 19 at 208–09.
• 21Gardham, supra note 19 at 400 (“With aerial warfare, civilians became extremely vulnerable and were inevitably collateral targets, potentially on a much larger scale than previously. Henceforth, the primary focus of proportionality was to be in relation to civilian losses.”).
• 22See Rogers, supra note 19.
• 23Id.
• 24Collateral damage definition, Merriam-Webster Online Dictionary. 2016, http://www.merriam-webster.com/dictionary/collateral%20damage (last visited Feb. 16, 2016).
• 25Though this statement is qualified by the fact that many AP I signatories included qualifying signing statements, as discussed infra in note 36.
• 26See Samuel Estreicher, Privileging Asymmetric Warfare? Part I: Defender Duties under International Humanitarian Law, 11 Chi. J. Int’l L. 425, 428 (2011).
• 27Customary IHL, International Committee of the Red Cross, https://www.icrc.org/customary-ihl/eng/docs/v1_cha_chapter4_rule14 (last visited Feb. 16, 2016).
• 28Parks, supra note 14, at 173 n.526 (pointing out that the United States has not conceded that proportionality as defined in AP I is customary international law).
• 29See John Finnis, Natural Law and Natural Right 236–48 (2d ed., 1982).
• 30Fenrick, supra note 13, at 98.
• 31Rome Statute, supra note 4, art. 8(2)(e)(i).
• 32Gary D. Solis, The Law of Armed Conflict: International Humanitarian Law in War 120–21 (2010).
• 33See Restatement (Third) of Foreign Relations Law of the United States § 102(2) cmt. j (1986).
• 34See the hypothetical explored in Section I.
• 35Customary IHL, supra note 27.
• 36International Committee of the Red Cross, Australia’s Reservation to the 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 1977, https://www.icrc.org/applic/ihl/ihl.nsf/Notification.xsp?action=openDocument&documentId=10312B4E9047086EC1256402003FB253 (Australia’s signing statement insisting that military advantage in art. 57 refer to military advantage “as a whole” and that the judgment of military commanders is necessarily ex ante); International Committee of the Red Cross, United Kingdom of Great Britain and Northern Ireland's Declaration to the 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 1977, https://www.icrc.org/applic/ihl/ihl.nsf/Notification.xsp?action=openDocument&documentId=0A9E03F0F2EE757CC1256402003FB6D2 (the United Kingdom expressing the same reservations); consider, for example, the cumulative effect of statements (c) and (i) made by the UK on ratification. See also Parks, supra note 14, at 174 (stating that, in reviewing a draft of AP I during negotiations, the U.S. Department of Defense concluded that the concept of proportionality was not a customary rule of law as presented in the draft).
• 37Estreicher, supra note 26, at 428.
• 38Aaron Fellmeth, Proportionality Principle in Operation: Methodological Limitations of Empirical Research and the Need for Transparency, 45 Isr. L. Rev. 131–33 (2012).
• 39 As a consequence, the 123 signatories to the statute are committed unequivocally to the definition provided therein.Id.
• 41 Therefore, the Rome Statute serves as a genuine common denominator between signatories to both.

  d) None of the above. While it is possible to imagine a CIL proportionality regime that prohibits more military conduct than the AP I (strict liability for civilian casualties, for instance) or less military conduct than the Rome Statute, these alternatives are significantly less plausible as candidates for CIL than those outlined above.For an implicit rejection of both options, see Section IV infra.

• 43It is considered settled that the rules of International Humanitarian Law have been incorporated as a part of CIL. U.N. Secretary General, Report of the Secretary-General Pursuant to Paragraph 2 of Security Council Resolution 808, ¶ 34, U.N. Doc. S/25704 (May 3, 1993).
• 44AP I, supra note 5, art. 51(5)(b).
• 45Rome Statute, supra note 4, art. 8(2)(b)(iv).
• 46Id.
• 47Sloane, supra note 8, at 309.
• 48Rome Statute, supra note 4, art. 5 (a)–(d).
• 49Adil Ahmad Haque, Protecting and Respecting Civilians: Correcting the Substantive and Structural Defects of the Rome Statute, 14 New Crim. L. Rev. 519, 523 (2011).
• 50See AP I, supra note 5, arts. 85(3)(b) and 85(5).
• 51AP I, supra note 5, art. 85(3)(b).
• 52Haque, supra note 49, at 525–26.
• 53AP I, supra note 5, art. 85(3)(b). (“[T]he following acts shall be regarded as grave breaches of this Protocol, when committed willfully, in violation of the relevant provisions of this Protocol, and causing death or serious injury to body or health . . . (b) launching an indiscriminate attack affecting the civilian population or civilian objects in the knowledge that such attack will cause excessive loss of life, injury to civilians or damage to civilian objects, as defined in Article 57, paragraph 2(a)(iii).)(emphasis added).
• 54Prosecutor v. Stanislav Gali, Case No. IT-98-29, Judgement and Opinion, ¶ 54 (Int’l Crim. Trib. for the Former Yugoslavia Dec. 5, 2003), http://www.icty.org/x/cases/galic/tjug/en/gal-tj031205e.pdf.
• 55U.N. Preparatory Commission for the Int’l Crim. Ct., Finalized Draft Text of the Elements of Crimes, Art. 8(2)(b)(i), U.N. Doc. PCNICC/2000/I/ADD.2 (Nov. 2, 2000).
• 56Antonio Cassese, International Criminal Law 290 (2d ed. 2008) (“This erroneous belief about factual circumstances must be based on reasonable grounds or in other words not be specious or far-fetched. More specifically, the mistake must not result from negligence.”).
• 57Rome Statute, supra note 4, art. 30; see also Haque, supra note 49, at 532–33 (concluding, after reviewing several attempts to loosen the mens rea requirement of the Statute with respect to proportionality that, “The Rome Statute does not prohibit reckless or negligent attacks on civilians, only intentional and knowing attacks on civilians, and this remains the case even if negligence and recklessness are relied upon as evidence of intent”).
• 58See, for example, HCJ 769/02, The Public Committee Against Torture in Israel v. Gov’t of Israel et al., ¶ 46, 54 [2006] (Isr.) (Justice Aaron Barak of Israel’s Supreme Court arguing that targeted killing operations should be subject to both ex ante and ex post investigation, citing to the precautionary obligation introduced by the AP I definition of proportionality).
• 59Lest there be skepticism that military commanders are concerned with legal standards in the heat of war, it should be mentioned that in situations of targeting, military lawyers are normally involved in the target selection process. See, for example, Leonardo Tricarico, Identification of Targets and Precautions in Attacks in Air Warfare: Operation Allied Force as a Case Study, in Protecting Civilians in 21st Century Warfare: Target Selection, Proportionality, and Precautionary Measures in Law and Practice 39­–44 (Mireille Hector & Marine Jellma eds., 2001).
• 60A.P.V. Rogers, The Principle of Proportionality, in The Legitimate Use of Military Force 189, 208–09 (Howard M. Hansel ed., 2008) (by [the Rome Statute’s] use of the words “clearly excessive in relation to the concrete and direct overall military advantage anticipated,” the drafters of this article took into account the various statements made on ratification of Protocol I and, by adopting a middle way, have tried to accommodate the requirements of military necessity without abandoning humanity, by allowing one to look at the bigger operation picture) (emphasis added).
• 61William H. Boothby, The Law of Targeting 97 (2012).
• 62See Human Rights in Palestine and Other Occupied Arab Territories: Rep. of the U.N. Fact-Finding Mission on the Gaza Conflict, 12th Sess., Sept. 25, 2009, U.N. Doc A/HRC/12/48 (2009) [hereinafter Goldstone Report].
• 63ICC Office of the Prosecutor, Letter to Senders Concerning the Situation in Iraq, 5–6 (Feb. 9, 2006), available at https://www.icc-cpi.int/NR/rdonlyres/04D143C8-19FB-466C-AB77-4CDB2FDEBEF7/
  143682/OTP_letter_to_senders_re_Iraq_9_February_2006.pdf (last visited Feb. 16, 2016).
• 64Goldstone Report, supra note 62, ¶¶ 436, 629.
• 65Id.
• 66Amnon Rubinstein & Yaniv Roznian, Human Shields in Modern Armed Conflicts: The Need for a Proportionate Proportionality, 22 Stan. L. & Pol'y Rev. 93, 126–127 (2011) (arguing that proportionality should be read more expansively to permit, under certain circumstances, strikes against targets protected by human shields).
• 67Rome Statute, supra note 4, art. 8(2)(e)(i).
• 68I do not argue for such a conclusion here. Any argument that did attempt to reach a conclusion along these lines would require an intensely empirical demonstration of how the incentives of insurgents would change in light of greater proportionality latitude in attacking human shields.
• 69See Goldstone Report, supra note 62, ¶ 435.
• 70See, for example, Implementing Limitations on the Use of Force: The Doctrine of Proportionality and Necessity, 86 Am. Soc’y Int’l L. Proc. 39, 42 (1992) (remarks of Fred Green, Counsel of the Joint Chiefs of Staff) (Stating that the principle of proportionality is “well-understood and play[s] a very real role in decision making within out government generally and within the Department of Defense—the military establishment, specifically”). But cf. II Int’l Comm. of the Red Cross, Customary International Humanitarian Law, Volume II: Practice (Jean-Mariet Henckaerts & Louise Doswald-Beck eds., 2005) (quoting Russian’s view that proportionality is the “weakest point of IHL” and that states do not, in fact, comply with it in any meaningful sense).
• 71Ariane L. DeSaussure, The Role of the Law of Armed Conflict During the Persian Gulf War: An Overview, 37 A. F. L. Rev. 41, 59–61 (1994).
• 72U.S. Dept. of Defense, Conduct of the Persian Gulf War: Final Report to Congress, (Wash. D.C., GPO, Apr. 1992), http://www.ssi.army.mil/!Library/Desert%20Shield-Desert%20Storm%20Battle%20Analysis/Conduct%20of%20the%20Persian%20Gulf%20War%20-%20Final%20Rpt%20to%20Congress.pdf.
• 73Id.
• 74Which perhaps encourages a preexisting tendency in the media to claim war crime violations; see id. at 730–32.
• 75Oppenheim, International Law, vol. I, § 17 (8th ed. 1955); see also North Sea Continental Shelf, 1969 I.C.J. 44 (Feb. 20).
• 76See Rome Statute, supra note 4, art. 38(1)(b).
• 77See Finnis, supra note 29, at 238.
• 78See Ian Brownlie, Principles of Public International Law 7–9 (4th ed. 1990).
• 79See, for example, The Paquete Habana, 175 U.S. 677 (1990) (“By an ancient usage among civilized nations, beginning centuries ago, and gradually ripening into a rule of international law, cost fishing vessels . . . have been recognized as exempt . . . from capture as a prize of war”).
• 80This is a difficulty, in particular, for those who argue that prohibitions against torture is a rule of CIL, see Mark Weisburd, Customary International Law and Torture: The Case of India, 2 Chi. J. Int’l L. 81 (2001).
• 81See Filartiga v. Pena-Irala, 630 F.2d 876, 822 (2d Cir. 1980).
• 82See Brownlie, supra note 78, at 5–6.
• 83Filartiga v. Pena-Irala, supra note 81, at 822.
• 84See Brownlie, supra note 78, at 7.
• 85Restatement (Third) of the Foreign Relations Law of the United States, supra note 33, § 102(2).
• 86International Criminal Court, The States Parties to the Rome Statute, https://www.icc-cpi.int/en_menus/asp/states%20parties/Pages/the%20states%20parties%20to%20the%20rome%20statute.aspx (last visited Feb. 16, 2016).
• 87International Committee of the Red Cross, Treaties and States Parties to Such Treaties, https://www.icrc.org/applic/ihl/ihl.nsf/States.xsp?xp_viewStates=XPages_NORMStatesParties&xp_treatySelected=470 (last visited Feb. 16, 2016).
• 88International Committee of the Red Cross, Canada’s Reservation to the 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 1977, https://www.icrc.org/applic/ihl/ihl.nsf/Notification.xsp?action=openDocument&documentId=172FFEC04ADC80F2C1256402003FB314 (last visited Feb. 16, 2016).
• 89See North Sea Continental Shelf Cases, supra note 75.
• 90Fellmeth, supra note 38.
• 91Jean-Marie Henckaerts & Louise Doswald-Beck, supra note 70, at 48-49, 58-59.
• 92Fellmeth, supra note 38.
• 93Rogers, supra note 60, at 147.
• 94Id. at 148.
• 95See J. Patrick Kelly, The Twilight of Customary International Law, 40 Va. J. Int’l L. 449, 469 (2000) (“[T]he . . . [ICJ], in most cases, declares rules of law without investigating the attitude of states on the legal character of a customary norm or undertaking an investigation of the actual practices of the majority of states.”).
• 96Id. at 469–79.
• 97Id.
• 98Id.
• 99Id.
• 100Id.
• 101Id.
• 102A sentiment echoed by other skeptics of CIL; see generally Samuel Estreicher, Rethinking the Binding Effect of Customary International Law, 44 Va. J. Int’l L. 5 (2003).
• 103Eric Posner & Jack L. Goldsmith, A Theory of Customary International Law, 66 U. Chi. L. Rev. 1113, 1132 (1999).
• 104Id. at 1175.
• 105Estreicher, supra note 102, at 11. An acknowledgement of the effect that moral appeals in domestic politics can have in the formation of international law is not contrary to realism. See Stephen D. Krasner, Realist Views of International Law, 96 Proceedings of the Annual Meeting (Am. Soc. of Int’l L.) 265, 266 (2002).
• 106Estreicher, supra note 102, at 11–12.
• 107For this Comment’s purposes, an extended treatment of “realism” is not necessary. But a realist account means roughly an approach that explains and predicts state behavior based solely on the distribution of power. See Kenneth Waltz, Theory of International Politics (1979).
• 108The term “moral” here is used in a limited and provisional sense. I do not want to imply that Western Europe—to generalize—acts morally in the context of CIL, while “Westphalian” state actors act amorally or immorally. Rather, I mean to say that something like a moral motivation is probably the best explanation those countries give for their desire to create standards of conduct that can be used to brand outlier nations. It’s possible, however, to explain the difference in approach to CIL through the cost/benefit analysis favored by Posner & Goldsmith above.
• 109Sweden, for instance, which made no reservations regarding proportionality in its signing statement. International Committee of the Red Cross, Sweden’s Signature to the 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 1977, https://www.icrc.org/applic/ihl/ihl.nsf/Notification.xsp?action=openDocument&documentId=343E4A648EC4DF2DC1256402003FB674 (last visited Feb. 16, 2016).
• 110Kelly, supra note 95, at 472.
• 111See Ronald Reagan, Agora: The U.S. Decision Not to Ratify Protocol I to the Geneva Conventions on the Protection of War Victims: Letter of Transmittal, 81 Am. J. Int’l L. 910, 911 (1987).
• 112Amichay Ayalon & Brian Michael Jenkins, War by What Means, According to Whose Rules?, Proceedings of a RAND–Israel Democracy Institute Workshop (2014), available at http://www.rand.org/pubs/conf_proceedings/CF334.html.
• 113See Rubinstein and Roznai, supra note 66, at 95–96.
• 114See id.; see also Samuel Estreicher, Privileging Asymmetric Warfare (Part III): The Intentional Killing of Civilians under International Humanitarian Law, 12 Chi. J. Int’l L. 589 (2012).
• 115At least for countries that are partially democratic. See Beth Simmons, Mobilizing for Human Rights: International Law in Domestic Politics 15–16 (2009).
• 116See generally Eric Posner, The Twilight of Human Rights Law (2014).
• 117Lest the United States’ refusal to sign the Rome Treaty be thought of as evidence that the definition of proportionality offered there is too constrictive, the stated reasons given by the United States for rejecting the treaty made no mention of proportionality. See Reagan, supra note 112; see also Abraham D. Sofaer, The Rationale for the United States Decision, 82 Am. J. Int’l L. 784, 785 (1988).
• 118Luke N. Condra et al., The Effect of Civilian Casualties in Afghanistan and Iraq, Nat’l Bureau of Econ. Res., Working Paper No. 16152, (2010) (arguing that tightened rules of engagement in Afghanistan resulted in reduced American casualties).
• 119North Atlantic Treaty Organization, Tactical Directive (July 6, 2009), http://www.nato.int/isaf/docu/official_texts/Tactical_Directive_090706.pdf.
• 120David Luban, Opting out of the Law of War: Comments on Withdrawing from International Custom, 120 Yale. L.J. Online 151 (2010), available at http://yalelawjournal.org/forum/opting-out-of-the-law-of-war-comments-on-withdrawing-from-international-custom (“A significant consequence of the rules of CIL is that states train their militaries in them. The United States, which did not ratify Additional Protocol I, nevertheless accepts portions of it as legitimate statements of customary international law and build these rules into its law-of-war training and JAG practice.”).
• 121Id.
• 122Against the notion that military commanders will always stress prefer the most expansive proportionality regime, Gen. McChrystal’s restrictive rules of engagement make clear that in modern asymmetric conflicts, commanders understand the strategic value of minimizing civilian casualties, see supra note ; see also International Security Assistance Force, Counter-insurgency Guidance Document, http://www.nato.int/isaf/docu/official_texts/counterinsurgency_guidance.pdf (last visited Feb. 16, 2016).
• 123Amos Guiora & David Luban, Was the Gaza Campaign Legal?, 31 Am. Bar Ass’n Nat’l Sec. L. Rep. 1, 11 (2009) (describing Hamas’ theft of humanitarian aid and use of human shields, including children, to immunize an area from attack).
• 124See, for example, press reports on the discovery of Hamas’ urban warfare manual which describes the benefits of civilian deaths reported by Bob Fredericks, Hamas’ Disturbing ‘Human Shields’ Manual, N.Y. Post (Aug. 5, 2014, 9:12PM), http://nypost.com/2014/08/05/hamas-manual-details-civilian-death-plan-israel/.
• 125Central Intelligence Agency, Guide to the Analysis of Insurgency, 1 (2012), http://www.mccdc.marines.mil/Portals/172/Docs/SWCIWID/COIN/Doctrine/Guide%20to%20the%20Analysis%20of%20Counterinsurgency.pdf (defining insurgency as a “protracted political-military activity directed toward completely or partially controlling the resources of the country through the use of irregular military forces and illegal political mobilization. Insurgent activity . . . is designed to weaken government control and legitimacy while increasing insurgent control and legitimacy”).
• 126Indeed, scholarship around military commanders in recent conflicts has detailing the excessive amount of intelligence relied upon by military commanders. See Peter W. Singer, Tactical Generals: Leaders, Technology, and the Perils, Brookings Air & Space Power J. (2009).
• 127See supra note 132.
• 128See supra Part III.
• 129See Prosecutor v. Ante Gotovina & Mladen Markač, IT-06-90-A, Judgement (Int’l Crim. Trib. for the Former Yugoslavia Nov. 16, 2012), http://www.icty.org/x/cases/gotovina/acjug/en/121116_judgement.pdf (criticizing the Trial Court’s use of an ad hoc judicial device, the “200 meter rule” to determine whether a military commander had committed a war crime).
• 130See, for example, Antonin Scalia, Judicial Deference to Administrative Interpretations of Law, 1989 Duke L. J. 511, 512–14 (1989).
• 131See supra Part IV(B).
• 132See International Committee of the Red Cross, supra note 27. (Australia signing statement, etc.).
• 133Id.
• 134See Rome Statute, supra note 4 (Rome Statute has no qualification).