Law and Morality
Law and Morality
Abstract and Keywords
Conventionalists argue that combatants are legally and morally permitted to fight irrespective of their cause. Revisionists argue that combatants are not morally permitted to fight for an unjust cause and conclude that the law of war sharply diverges from the deep morality of war. Many revisionists conclude that the law of war should simply aim to reduce suffering in war to the greatest extent practically possible. This chapter argues that the law of war does not permit or authorize combatants to fight for an unjust cause. The law of war is prohibitive, not permissive. The law of war does not confer symmetrical legal permissions but instead imposes symmetrical legal prohibitions and grants symmetrical legal immunities. The law of war applies alongside other legal and moral norms, including human rights law. Finally, this chapter argues that the law of war should aim to help combatants better conform to their moral obligations.
International law prohibits the use of military force by states on the territory of other states except with the consent of the territorial state, with the authorization of the United Nations Security Council, or in individual or collective self-defense against an armed attack.1 Importantly, the law governing the conduct of hostilities (the jus in bello) applies independently of the law governing the resort to military force (the jus ad bellum or jus contra bellum) and applies symmetrically to all sides irrespective of the legality of their respective war aims.
In international armed conflicts, international law grants combatants “the right to participate directly in hostilities,” irrespective of the party for which they fight, so long as they distinguish themselves from the civilian population while they are engaged in an attack or in military operations preparatory to an attack.2 Similarly, international law protects civilians, as well as combatants who surrender, are taken prisoner, or are wounded, sick, or shipwrecked, “without any adverse distinction based on the nature or origin of the armed conflict or on the causes espoused by or attributed to the Parties to the conflict.”3 In short, international law accepts both the legal equality of combatants and the legal equality of noncombatants.
For some time, just war theory was dominated by the conventionalist view according to which the morality of war tracks the law of war in most respects.4 On the conventionalist view, combatants on all sides enjoy symmetrical moral permissions while noncombatants on all sides enjoy symmetrical moral protections. All combatants are morally permitted to intentionally harm opposing combatants and to inflict necessary and proportionate collateral harm on noncombatants. All noncombatants are morally protected from intentional, unnecessary, or disproportionate harm. In (p.20) short, the conventionalist view accepts both the moral equality of combatants and the moral equality of noncombatants.
Since roughly the turn of the century, a number of revisionist scholars have systematically attacked the conventionalist view and with it the moral foundations of the law of war. Revisionists deny that the moral norms governing armed conflict are symmetrical in the conventionalist sense. Most revisionists deny the moral equality of combatants and many deny the moral equality of noncombatants as well. On the revisionist view, only combatants pursuing a just cause are morally permitted to intentionally harm opposing combatants or to inflict necessary and proportionate collateral harm on noncombatants. Some revisionists also argue that noncombatants who indirectly contribute to an unjust war effort are morally liable to intentional or collateral harm.
Importantly, most revisionists agree that the legal norms governing armed conflict should remain symmetrical, not as a matter of moral principle but for pragmatic and epistemic reasons. On the revisionist view, the law of war must diverge from the deep morality of war. The law must permit what morality forbids—killing in pursuit of an unjust cause—and forbid what some revisionists believe that morality permits—killing civilians who support an unjust war effort.
Many revisionists, convinced that law and morality must diverge, are driven to embrace the humanitarian view that the law of war should aim to reduce unnecessary suffering in war to the greatest extent practically possible. According to the humanitarian view, while the deep morality of war may protect the moral rights of individuals, the law of war can at best reduce aggregate harm. On this view, armed forces convinced (or at least claiming) that they fight for a just cause will simply ignore legal constraints that make military victory impossible. It follows that the law of war can only effectively prohibit militarily unnecessary harm.
In this chapter, I argue that the debate between conventionalists and revisionists regarding the morality of war rests, in part, on a mistake. The legal equality of combatants does not consist in symmetrical legal permissions to fight but in symmetrical legal prohibitions on certain means and methods of fighting as well as symmetrical legal immunities from criminal prosecution for acts that do not violate those prohibitions. The law does not permit what morality forbids because the law does not, in fact, permit combatants to fight for an unjust cause. On my view, the law of targeting and attack contains no permissions but instead contains only prohibitions. Conventionalists committed to the convergence of law and morality need not defend symmetrical moral permissions to fight irrespective of one’s war aims. At the same time, revisionists who reject such symmetrical moral permissions need not insist on a sharp divergence between law and morality.
Finally, I reject the humanitarian view and defend the service view, according to which the law of war should aim to help combatants better conform to their moral obligations. On the service view, individuals generally have decisive moral (p.21) reason to obey the law only if they will better conform to their moral obligations by following the law than by following their own moral judgment. The law of war achieves its aim to the extent that just combatants are more likely to avoid acting wrongfully if they obey the law while unjust combatants will act less wrongfully if they obey the law.
The service view can accommodate the revisionist insight that the deep morality of war is asymmetrical while providing a non-consequentialist defense of a symmetrical law of war. The contingent aim of many states in developing the law of war may be to reduce unnecessary suffering in war, but the constitutive aim of the law of war itself is to provide a moral service to combatants. We should defend, interpret, and develop the law of war by reference to the service view.
Conventionalists and Revisionists
The conventionalist view, that combatants are both legally and morally permitted to fight irrespective of their war aims, is most closely associated with Michael Walzer. In Just and Unjust Wars, Walzer offers a critical interpretation and qualified defense of “the war convention,” which Walzer describes as “the set of articulated norms, customs, professional codes, legal precepts, religious and philosophical principles, and reciprocal arrangements that shape our judgments of military conduct.”5 According to Walzer, “the task of the moral theorist is to study the pattern as a whole, reaching for its deepest reasons.”6
Since Walzer views the law of war as an integral part of the war convention, it is no surprise that his interpretation and defense of the latter reads like an interpretation and defense of the former. Famously, Walzer argues that the moral principles governing the resort to war and the moral principles governing the conduct of war are “logically independent. It is perfectly possible for a just war to be fought unjustly and for an unjust war to be fought in strict accordance with the rules.”7 In particular, Walzer argues that combatants on all sides of a conflict possess “an equal right to kill” opposing combatants as well as an equal (though limited) right to collaterally kill noncombatants.8 Finally, Walzer argues that all noncombatants retain an equal right not to be intentionally killed or to be collaterally killed unnecessarily or disproportionately.
(p.22) On Walzer’s view, combatants are morally equal because they pose threats to others, while noncombatants are morally equal because they pose no threats to anyone. According to Walzer, the “right not to be attacked … is lost by those who bear arms ‘effectively’ because they pose a danger to other people.”9 It follows that opposing combatants have “an equal right to kill” one another because they pose lethal threats to one other (as well as to noncombatants) and thereby lose their right not to be killed by one another. Importantly, combatants lose their right not to be killed even if they do not act wrongfully. “Simply by fighting,” combatants lose “their title to life and liberty, … even though, unlike aggressor states, they have committed no crime.”10 Conversely, since noncombatants pose no threats, they retain their ordinary rights not to be intentionally or unnecessarily killed.
The revisionist view, that combatants are only morally permitted to fight for a just cause, is most closely associated with Jeff McMahan. In Killing in War, McMahan argues that human beings lose their moral right not to be killed by posing unjust threats but not by posing just threats. Combatants who fight for an unjust cause (such as territorial conquest) typically pose unjust threats to opposing combatants and thereby lose their moral right not to be killed. In contrast, combatants who fight for a just cause (such as national self-defense) typically pose just threats to opposing combatants and therefore retain their moral right not to be killed. On the revisionist view, the soldier fighting in national self-defense and the soldier fighting for territorial conquest do not have an “equal right to kill” one another. The former has a right to kill the latter and the latter has no right to kill the former.11
McMahan grants that combatants are legally permitted to inflict collateral harm on civilians irrespective of their war aims, but argues that combatants are only morally permitted to inflict collateral harm on civilians in pursuit of a just cause. Soldiers fighting in national self-defense can sometimes morally justify collaterally harming civilians as a necessary and proportionate side effect of preventing a greater evil. In contrast, soldiers fighting for territorial conquest can never morally justify collaterally harming civilians as a side effect of achieving an even greater evil.
While McMahan rejects the moral equality of combatants on deontological grounds, he accepts the legal equality of combatants on broadly consequentialist (p.23) grounds. As McMahan observes, a legal prohibition on killing in pursuit of an unjust cause will do little good and may do great harm. Most combatants, who believe that they fight for a just cause, will simply ignore such a prohibition. Worse, if combatants fear that they will be punished merely for fighting, then they may be less willing to surrender and more willing to win at all costs.12 As Vattel observed, “each party asserting that they have justice on their own side, will arrogate to themselves all the rights of war, and maintain that their enemy has none.”13 It follows that, perhaps counter-intuitively, the legal equality of combatants reduces wrongful suffering in war.
For McMahan, the deep morality of war reflects the moral rights and duties of each affected individual. In contrast, McMahan writes that “the law of war is designed not to protect rights but to prevent harm.”14 On this view, “the laws of war are conventions that we design for … mitigating the savagery of war, seeking to bring about outcomes that are more rather than less just or morally desirable.”15 In such passages, McMahan suggests that the legal equality of combatants rests not on the deep morality of war but on the humanitarian view of the law of war.
I will have more to say about the humanitarian view shortly, but first things first. In the next section, I reject the premise that the legal equality of combatants consists in symmetrical legal permissions to fight irrespective of one’s war aims. If I am right then the dispute between conventionalists and revisionists, as well as the appeal of the humanitarian view, partly rests on a mistake.
The Right to Fight: An Immunity, Not a Privilege
As noted at the outset, international law grants lawful combatants “the right to participate directly in hostilities” irrespective of the party for which they fight.16 Importantly, this legal right is granted to lawful combatants but denied to civilians. Indeed, international law positively defines combatants as those with a right to fight and negatively defines civilians as those who are not combatants. In this respect, the right to participate directly in hostilities contrasts with the right to be treated as a prisoner of war (p.24) upon capture, which is granted both to lawful combatants and to certain categories of civilians.17
The right to participate directly in hostilities is often referred to as “the combatant’s privilege” or as “the privilege of belligerency.” As David Rodin writes:
The legal privilege functions as a positive right to kill, much in the way that the liberty to kill in self-defence functions as a right within domestic criminal law. It functions as a codified exception to an established prohibition … .18
Similarly, Jeremy Waldron writes that
The default position, apart from any convention, is that intentionally killing or attacking any human being is prohibited as murder. The laws of armed conflict provide an exception to that; they establish what we call in the trade a Hohfeldian privilege in relation to what is otherwise forbidden. And the rule about civilians is to be understood as a limitation on the scope of that privilege.19
On this view, international law grants all combatants a limited legal permission to target opposing combatants and to inflict necessary and proportionate collateral harm on noncombatants. Simply put, international law grants lawful combatants a “license to kill.”20
In sharp contrast, on my view, international law does not grant lawful combatants a symmetrical legal privilege to fight but instead grants lawful combatants a symmetrical legal immunity from criminal prosecution. To see this, recall that international law grants “the right to participate directly in hostilities” exclusively to combatants and not to civilians. If this exclusive legal right involves a legal permission, then it should follow that combatants are legally permitted to fight while civilians are legally forbidden to fight. However, international law nowhere forbids civilians from participating directly in hostilities. Civilians do not violate international law simply by taking up arms and joining the fight.21
(p.25) It follows that the legal “right to participate directly in hostilities” conferred on combatants but denied to civilians cannot be a privilege or liberty-right to participate directly in hostilities, that is, the absence of a legal duty not to fight.22 After all, neither combatants nor civilians have a legal duty under international law not to fight. Of course, this legal right also cannot refer to a claim-right that imposes on others legal duties of non-interference (let alone positive assistance). Obviously, opposing combatants have no legal duty not to prevent each other from fighting.
Instead, international law confers a limited legal immunity on lawful combatants, but not on civilians, from subsequent criminal prosecution.23 International law prohibits states from criminally prosecuting combatants for acts that may violate national law but do not violate international law. In contrast, international law does not prohibit states from criminally prosecuting civilians under national law merely for participating directly in hostilities. Civilians may be criminally prosecuted for acts that do not violate international law but that violate national law, including killing opposing combatants. The right of combatants, but not civilians, “to participate directly in hostilities” is simply the right not to be criminally prosecuted merely for participating directly in hostilities.24
As Seth Lazar observes, “immunity from prosecution can be justified on its own terms; it need not be grounded in a right to fight.”25 Lazar writes that disconnecting combatant immunity from the right to fight “would not need too radical a change in the laws of war as they currently stand.” On my view, no change in the laws of war is necessary because the right to fight just is a limited legal immunity from prosecution for fighting.
(p.26) Importantly, this legal immunity is limited in several ways. First, combatants forfeit their legal immunity by failing to distinguish themselves from civilians.26 Second, this legal immunity only exists in international armed conflicts between state armed forces, not in non-international armed conflicts between states and organized armed groups or between such groups.27 Most importantly, combatants are only legally immune from criminal prosecution by foreign states under their domestic criminal law. Conversely, combatants may be criminally prosecuted by their own state under its domestic criminal law.28 Paradigmatically, lawful combatants fighting for their own state may be criminally prosecuted by their own state for violating rules of engagement that are more restrictive than international law. In addition, combatants fighting against their own state may be criminally prosecuted—most notably for treason—merely for directly participating in hostilities.29
As we have seen, combatants have no general “right to fight.” Instead, lawful combatants in international armed conflicts have specific rights against foreign governments not to prosecute them for acts that violate domestic criminal law but do not violate international law. Vattel seemed to express this view when he wrote that the law of nations
does not, to him who takes up arms in an unjust cause, give any real right that is capable of justifying his conduct or acquitting his conscience, but merely entitles [an unjust combatant] to the benefit of the external effect of the law, and to impunity among mankind.30
It seems that, according to Vattel, only the jus ad bellum promises a justification for fighting, while the jus in bello offers only immunity from punishment.
States and scholars often use the terms “combatant’s privilege” and “combatant immunity” interchangeably. As should now be clear, this lack of precision is apt to mislead. Strictly speaking, international law confers no legal privilege on (p.27) combatants to fight irrespective of their war aims. International law does not permit what a combatant’s own state forbids. States are free to punish their own citizens for acts that violate national law but not international law. On the contrary, international law prohibits what a combatant’s own state might permit. States are free to punish combatants of any nationality for acts that violate international law even if those acts were authorized under national law. International law only prohibits states from punishing foreign lawful combatants for acts that violate national law but do not violate international law.
This dual function of international law—imposing legal prohibitions on all participants in hostilities while granting limited legal protections to lawful combatants—belies the notion that killing in war is presumptively lawful,31 or, as Jeremy Waldron describes it, “a model that assumes that the default position is that you can kill anyone you like in wartime.”32 On the contrary, killing in war, as elsewhere, is presumptively criminal. International law bars the prosecution of foreign combatants for such presumptively criminal acts only when the latter kill in conformity with international law. In contrast, the ordinary law of criminal homicide remains in effect during war and may be applied to lawful combatants who kill in violation of international law, to unlawful combatants, to combatants who fight against their own state, and to civilians who directly participate in hostilities.
Importantly, a legal immunity is not a legal permission or an exemption from a legal prohibition. A legal immunity is not a justification (like self-defense), an excuse (like duress), or a denial of responsibility (like insanity). On the contrary, a legal immunity is a bar to criminal prosecution that applies irrespective of the wrongdoing and blameworthiness of the actor.
To illustrate, consider that diplomats are immune from criminal prosecution even if they satisfy all the elements of a criminal offense and can offer no exculpatory defense on their behalf. Of course, diplomatic immunity does not rest on a special legal permission conferred on diplomats but denied to ordinary citizens.33 Instead, diplomatic immunity is a non-exculpatory public policy defense that allows an individual to “escape conviction in spite of  culpability.”34 As Antony Duff explains
someone who claims diplomatic immunity when charged with an offense is not claiming that she was authorized to engage in that conduct, or that her conduct was legally (p.28) permissible—indeed, she might admit that what she did was culpably criminal: but she is denying that this court has the authority to call her to account for her conduct.35
Put another way, justifications and excuses answer a criminal charge, while immunities entail that the accused need not answer the charge. Accordingly, legal immunities are typically raised and adjudicated on prior to trial and, if sustained, result in dismissal of charges. In contrast, justifications and excuses are typically raised and adjudicated on during trial and, if sustained, result in a verdict of acquittal.
Although justifications, excuses, denials of responsibility, and bars to prosecution allow the accused to avoid conviction, they each express a distinct moral message. Justifications express the message that the accused acted permissibly, all things considered, and that no one has a legitimate grievance or complaint against her. Excuses express the message that the accused acted impermissibly but she does not deserve blame. Bars to prosecution express the message that the accused may have acted impermissibly and may deserve blame but that there are other reasons why the state should not punish her. Less is required to morally justify bars to prosecution because bars to prosecution make relatively modest moral claims.
On my view, we should understand lawful combatant immunity as a bar to prosecution that rests on prosaic considerations of treaty and custom, reciprocity and impartiality, marginal incentives and aggregate consequences. No state wants its own soldiers prosecuted by its adversaries, and so all states agree not to prosecute the soldiers of their adversaries. In addition, such a legal immunity gives soldiers determined to fight an incentive to fight within the constraints of international law. If combatants will act less wrongfully if they obey the law than if they violate the law, then the law should create such incentives.
Revisionists may be right that such prosaic considerations are too weak to morally justify a legal permission to kill in pursuit of an unjust cause. Perhaps, as Rodin argues, a legal system that “creates a legal right for certain people to violate the moral rights of others, as a means to achieving a broader desirable end” thereby violates rights itself.36 However, such prosaic considerations may be strong enough to ground a prohibition on criminally prosecuting foreign combatants who kill in pursuit of an unjust cause but who do not violate international law. Put another way, if the law claims that combatants are morally permitted to kill in pursuit of an unjust cause, then law’s claim is false. If law claims only that states are morally prohibited from prosecuting combatants for fighting for an unjust cause, then law’s claim may be redeemed. Those killed in pursuit of an unjust cause have a legitimate moral grievance or complaint—one that a legal permission would deny—but not necessarily one that their state may vindicate through domestic criminal prosecution.
(p.29) On many jurisprudential views, including my own, law necessarily makes moral claims. On such views, a legal right is a (putative) moral right claimed by law.37 In some cases, the law purports to identify and protect a pre-legal moral right, such as the right not to be tortured. In other cases, the law purports to create a new moral right, such as the right to vote, by changing the moral reasons that apply to those subject to the law. On such views, if the law permits some actor to perform some act, then the law claims that actor is morally permitted to perform that act. In particular, if international law confers a symmetrical legal permission to fight then it asserts a symmetrical moral permission to fight. Indeed, if international law did not make such moral claims, then conventionalists would not consider it part of the war convention that conventionalists seek to interpret and defend.38
Thankfully, on my view, international law does not claim that killing in pursuit of an unjust cause is morally permissible, that combatants are exempt from ordinary moral constraints, or that combatants are always or even typically justified or excused. International law claims only that states are morally prohibited from criminally prosecuting lawful combatants for acts that conform to international law. Put another way, international law does not purport to change the moral reasons that apply to combatants but rather to change the moral reasons that apply to states. Accordingly, Walzer was incorrect when he wrote that combatants who kill in pursuit of an unjust cause “have committed no crime.” Often they have committed crimes, albeit crimes for which they may not be prosecuted.
No doubt, the legal immunity of combatants carries a moral cost. Many combatants know or should know that they fight for an unjust cause. They may believe that their war is unlawful or their war may be manifestly unlawful. There may be retributive reasons to allow states to punish such combatants in proportion to the wrongs they commit and their degree of moral fault. However, there are also retributive reasons against allowing such prosecutions. Such prosecutions would often be misdirected or abused—particularly by states prosecuting enemy soldiers—resulting in the unjust punishment of combatants who fought for a just cause or who reasonably believed that they fought for a just cause. In addition, such prosecutions would often result in disproportionate punishment of combatants who unreasonably but sincerely believed that they fought for a just cause. In (p.30) any event, even if the retributive reasons against combatant immunity outweigh the retributive reasons supporting combatant immunity, it is hard to believe that the former outweigh both the latter and the instrumental reasons supporting combatant immunity.
Importantly, combatant immunity does not entail general impunity for harms inflicted in pursuit of an unjust cause. International law recognizes the crime of aggression, which imposes criminal liability on political and military leaders for the use of force in manifest violation of the United Nations Charter.39 Such political leaders do not enjoy combatant immunity and such military leaders may not claim combatant immunity with respect to the crime of aggression. If the crime of aggression is indeed “the supreme international crime differing only from other war crimes in that it contains within itself the accumulated evil of the whole” then by punishing the whole we punish its constituent parts, at least to a degree.40 We will not thereby punish all those directly responsible for each killing in an unjust war, but we will punish those ultimately responsible for all killing in that unjust war.41
To conclude, international law does not prohibit lawful combatants from participating in aggression, and immunizes them from prosecution for acts that international law does not prohibit. The non-prohibition and the immunity are conceptually distinct but inextricably linked. If international law prohibited lawful combatants from participating in aggression, then foreign states would be free to prosecute them on that basis. No doubt, such a prohibition would have considerable expressive value. However, there is no combatant immunity for acts prohibited by international law. We must choose between the proposed prohibition and the existing immunity and, as we have seen, the reasons for the immunity are weighty indeed.
Prohibitions and Permissions
In my view, “the law relating to the conduct of hostilities is primarily a law of prohibition: it does not authorize, but prohibits certain things.”42 While the jus ad (p.31) bellum sometimes authorizes the use of force by states, the jus in bello never authorizes acts of violence by armed forces. Under international law, authorization for acts of violence, if any, must come from the jus ad bellum.
As we have seen, international law applies the jus in bello equally to parties conforming to the jus ad bellum and to parties violating the jus ad bellum. Yet it would be illogical for international law to prohibit a use of force under the jus ad bellum while authorizing the acts of violence that make up that use of force under the jus in bello. The prohibition of aggressive force under the jus ad bellum, the equal application of the jus in bello, and the notion that the jus in bello authorizes acts of violence form a logically inconsistent set. In my view, we should reject the final proposition.
Similarly, international law applies the jus in bello equally to state armed forces and to non-state armed groups. Yet international law hardly gives non-state armed groups a legal right to wage war against their governments.
The law of armed conflict (LOAC) does not tell combatants what they may do, only what they may not do.43 The LOAC prohibits the intentional, unnecessary, or disproportionate killing of civilians. However, no rule of the LOAC authorizes or justifies the unintentional, unavoidable, and proportionate killing of civilians. Similarly, the LOAC prohibits the intentional killing of combatants who have surrendered or are incapacitated by injury or illness. However, no rule of the LOAC authorizes or justifies the intentional killing of combatants who have not surrendered or been incapacitated. When we say that the LOAC “permits” such killings we refer either to the absence of a prohibition in international law or to lawful combatant immunity from prosecution for violations of domestic law. Indeed, it would be more precise to describe acts not prohibited by the LOAC as merely “tolerated” rather than as “permitted.”44
Along similar lines, John Westlake wrote of the laws of war that
These rules are always restrictive, never permissive in any other sense than that of the absence of prohibition, for law can give no positive sanction to any act of force of which it cannot secure the employment on the side of justice alone, even if the particular act be not one which the law would prohibit both to the just and to the unjust if it could. Whenever (p.32) therefore in speaking of the laws of war it is said that a belligerent may do this or that, it is always only the absence of prohibition that must be understood.45
Put another way, the law of targeting and attack confers no strong permissions. As Joseph Raz explains, “an act is strongly permitted only if its being permitted is entailed by a norm. It is permitted in the weak sense if the permission … is simply a consequence of there being no norms prohibiting the performance of the action.”46 For example, the oft-cited Lotus principle—that states are permitted to do what international law does not prohibit—clearly refers to weak permissions rather than to strong permissions.47
Infelicitously, some legal prohibitions are phrased as legal requirements, which may in turn suggest implicit legal permissions. As we will see in Chapters 5 and 7, international law requires attackers to take feasible precautions to avoid mistakenly targeting civilians, unnecessarily harming civilians, or disproportionately harming civilians. However, combatants are not authorized to attack so long as they take these precautions. Instead, combatants are prohibited from attacking without taking these precautions. Put another way, these legal requirements are conditional—if you attack, then you must take these precautions—and entail no unconditional permissions to attack.48 Similarly, combatants “shall at all times distinguish between [civilians or] civilian objects and military objectives and accordingly shall direct their operations only against military objectives.”49 Obviously, this basic rule means that combatants shall not direct their operations against civilians or civilian objects. The rule was hardly intended to encourage combatants to kill each other.
It is sometimes claimed that “the principle of military necessity in the customary law of war may be viewed as justifying or permitting certain acts.”50 Since military necessity is not defined in any treaty, proponents of this claim typically draw on two sources. In my view, neither source supports this claim.
First, the so-called Lieber Code states that
Military necessity, as understood by modern civilized nations, consists in the necessity of those measures which are indispensable for securing the ends of the war, and which are lawful according to the modern law and usages of war.51
(p.33) By the terms of this definition, measures are militarily necessary only if they are lawful. However, measures are not lawful in virtue of military necessity. Measures derive their lawfulness not from military necessity but “according to the laws and usages of war,” that is, according to the specific legal rules that other provisions of the Code purport to list. Military necessity is not a source of legal authority or even part of the law of war.
The Lieber Code states that “Military necessity admits of all direct destruction of life or limb of armed enemies, and of other persons whose destruction is incidentally unavoidable in the armed contests of the war.”52 However, these acts are not lawful because they are “admitted” by military necessity. On the contrary, military necessity “admits” these acts because they are lawful (that is, not prohibited) according to the law and usages of war. Similarly, the Lieber Code states that military necessity “does not admit” of cruelty, unnecessary suffering, torture, poison, or perfidy.53 However, these tactics are not unlawful because military necessity does not “admit” them. On the contrary, military necessity “does not admit” of these tactics because these tactics are not “lawful according to the modern law and usages of war.” Each of these tactics is specifically prohibited in other articles of the Code, by reference to the laws of war rather than to military necessity.54
Almost a century later, the American Military Tribunal at Nuremberg wrote that
Military necessity permits a belligerent, subject to the laws of war, to apply any amount and kind of force to compel the complete submission of the enemy with the least possible expenditure of time, life, and money.55
As before, by the terms of this definition, military necessity permits force only if that force is lawful. However, force is not lawful because it is militarily necessary. Instead, every amount and kind of force is “subject to the laws of war,” with which military necessity is partially contrasted.
Importantly, the Tribunal repeatedly states that “international law is prohibitive law.”56 On this view, international law is a source of legal prohibitions, not a source of legal authority.57 Accordingly, states are always (weakly) permitted to do what (p.34) international law does not prohibit them from doing. It is in this sense that belligerent states are (weakly) permitted to use force subject to the laws of war. We do not need to invoke military necessity to explain this result.
As the Tribunal observed, the laws of war prohibit the destruction or seizure of property “unless such destruction or seizure be imperatively demanded by the necessities of war.”58 Other legal rules “make no such exceptions to [their] enforcement.”59 Today, specific legal prohibitions contain exceptions for cases of military necessity, public necessity, medical and investigative necessity, or the necessity of providing for the civilian population.60 However, “The[se] prohibitions … control, and are superior to military [or other] necessities of the most urgent nature except where the [prohibitions] themselves specifically provide the contrary.”61 Accordingly, military necessity, public necessity, medical and investigative necessity, and so forth are not free-standing sources of legal authority. These are non-legal concepts that must be incorporated into specific legal rules in order to have any legal effect.
Since the jus in bello applies equally to opposing states, military necessity cannot legally authorize the acts of violence that make up an unlawful act of aggression.62 Since the jus in bello applies equally to state armed forces and non-state armed groups, military necessity cannot grant non-state armed groups a legal right to wage war against their governments. Accordingly, lawful combatants charged with murder in the criminal courts of an adversary would hardly claim that their killings were justified by military necessity. Instead, such lawful combatants would claim that they are immune from prosecution for acts not prohibited by the law of armed conflict.
To conclude, international law does not guarantee the equal legal status of combatants but only the equal application of the jus in bello.63 Since the jus in bello is primarily prohibitive rather than permissive, combatants may conform to the jus in bello yet remain legally unequal under other branches of international or national law. Accordingly, common article 3 of the 1949 Geneva Conventions, as well as article 4 of Protocol I, underscore that their equal application “shall not affect the legal status of the Parties to the conflict.”64 The prohibitions contained in the LOAC (p.35) apply alongside other applicable legal rules, working together “to ensure a better protection for the victims of those armed conflicts.”65
Human Rights in War
Since the LOAC does not authorize or justify the killings that it does not specifically prohibit, it cannot permit killings that are prohibited by other bodies of law. Most importantly, international human rights law prohibits the arbitrary deprivation of life.66 In armed conflict, this general prohibition on arbitrary killing applies alongside the specific prohibitions of the LOAC. As the Inter-American Commission on Human Rights observes, “humanitarian law generally afford[s] victims of armed conflicts greater or more specific protections than do the more generally phrased guarantees in … human rights instruments.”67 Importantly, these specific protections were designed to guide combatants and protect civilians in the unique circumstances of armed conflict.
Similarly, Protocol II states that “international instruments relating to human rights offer a basic protection to the human person” while the LOAC aims “to ensure a better protection for the victims of those armed conflicts.”68 As Protocol II makes clear, human rights law and the LOAC do not conflict, as both offer protections from violence rather than licenses to commit violence. Accordingly, “when Protocol II in its more detailed provisions establishes a higher standard than the Covenant [on Civil and Political Rights], this higher standard prevails,” while “provisions of the Covenant … which provide for a higher standard of protection than the protocol should be regarded as applicable” in appropriate cases.69 On this view, the LOAC can only add to, but can never subtract from, the protection that individuals enjoy under human rights law.
It is true that, in its Nuclear Weapons advisory opinion, the International Court of Justice (ICJ) famously wrote that
In principle, the right not arbitrarily to be deprived of one’s life applies also in hostilities. The test of what is an arbitrary deprivation of life, however, then falls to be determined by (p.36) the applicable lex specialis, namely, the law applicable in armed conflict which is designed to regulate the conduct of hostilities.70
This passage suggests that, in the context of armed conflict, deprivations of life that do not violate the LOAC are necessarily non-arbitrary under human rights law. On this view, the LOAC does not need to permit what human rights law prohibits because the LOAC determines what human rights law prohibits in armed conflict. We should reject this view.
In its advisory opinion, the ICJ took for granted the contingent content of the LOAC, namely that its “cardinal principles” prohibit weapons “that are incapable of distinguishing between civilian and military targets” or that “cause unnecessary suffering to combatants.”71 Understandably, the ICJ considered these cardinal principles sufficient to adjudicate the legality of the threat or use of nuclear weapons. However, if these cardinal principles never entered the LOAC then clearly the LOAC would not prohibit all or even most arbitrary deprivations of life in armed conflict.
The extent to which the LOAC prohibits arbitrary killing in armed conflict is contingent on the content of the LOAC, which is itself contingent on treaty and custom, which is in turn contingent on what states agree to, how states behave, and what states believe. There is no reason why such contingencies should determine the content of human rights law.
Of course, it is hard to imagine the LOAC without the prohibitions on indiscriminate weapons and unnecessary suffering. A different example may help illustrate that the content of the LOAC is a contingent matter. Protocol I provides that
When a choice is possible between several military objectives for obtaining a similar military advantage, the objective to be selected shall be that the attack on which may be expected to cause the least danger to civilian lives and to civilian objects.72
Strikingly, the US Department of Defense denies that this rule reflects customary international law.73 This view seems clearly mistaken.74 But imagine if this view were true. Imagine that attacking forces could cut off an enemy supply route by destroying either of two bridges, the first with no civilian traffic and the second full of civilians on their way to work, school, or their homes. Assume that the military advantage of destroying the second bridge would render the collateral harm to civilians proportionate. However, such collateral harm would be unnecessary, since destroying the first bridge would yield the same advantage. Clearly, if attacking forces strike the second bridge rather than the first, then they would kill the civilian travelers arbitrarily whether or not they would thereby violate the LOAC.
(p.37) Alternatively, assume for the sake of argument that Protocol I crystallized the precautions rule for the first time.75 Now imagine that the Diplomatic Conference that produced Protocol I never convened. If wartime killings are arbitrary only if they are prohibited by the LOAC, then careless and easily avoidable killings would not be arbitrary. Put another way, on this view, Protocol I did nothing to make the LOAC more respectful of human rights or to limit arbitrary killing in armed conflict. Similarly, if Protocol I had omitted the precautions rule, then this omission would not have affected Protocol I’s human rights credentials. This seems impossible to believe.
When applying human rights law, we should ask whether a particular deprivation of life was arbitrary given the circumstances. While we may certainly look to the LOAC to inform our interpretation of which deprivations of life are arbitrary in armed conflict, there is no reason to assume that the LOAC is so perfect that it effectively prohibits all arbitrary deprivation of life in armed conflict. In the end, whether a particular deprivation of life is arbitrary remains a question of human rights law, not of the LOAC. Our best interpretation of human rights law, informed but not determined by looking to the LOAC, should prevail.
During armed conflict, the LOAC may or may not prove sufficient to prohibit arbitrary killing. That will depend on the contingent content of the LOAC, as well as on the factual circumstances, not on some a priori relationship between the LOAC and human rights law. Outside of armed conflict, killing is almost always arbitrary unless it follows strict rules governing self-defense or law enforcement. During armed conflict, killing opposing combatants may seldom prove arbitrary. However, as we shall see in Chapter 4, killing a combatant whom one could safely capture may not violate the LOAC but may fail to respect the human right to life.
The relationship between the LOAC and human rights is teleological, not constitutive. The LOAC aims to protect human rights to the greatest extent possible in armed conflict, not to define what human rights mean in armed conflict.76 It is not an a priori or conceptual truth that the LOAC, whatever its content, prohibits arbitrary killing. If the LOAC prohibits even most arbitrary killing in armed conflict, then this is a contingent and reversible human achievement. Accordingly, we may draw on the LOAC to interpret human rights law, or draw on human rights law to interpret the LOAC, but neither body of law determines the content of the other.77
As we have seen, revisionists assume that combatants are legally permitted to fight for an unjust cause, argue that combatants are not morally permitted to fight for an unjust cause, and conclude that law and morality sharply diverge. Moreover, revisionists argue that law and morality must diverge, since a legal prohibition on fighting for an unjust cause will either be ignored or create perverse incentives. These revisionists conclude that, while the deep morality of war is concerned with individual rights and directed duties, the law of war should content itself with the consequentialist aim of reducing wrongful suffering in war to the greatest extent practically possible.
As we have seen, this revisionist argument rests on a false premise. Combatants are not legally permitted to fight for an unjust cause, though lawful combatants are legally immune from criminal prosecution so long as they fight according to the rules. It follows that revisionists should not feel compelled to embrace the humanitarian view. At the same time, the humanitarian view can seem plausible—even attractive—on its face. As a general matter, the aim of reducing wrongful suffering in war is one that we should all share. We should therefore evaluate the humanitarian view on its own merits.
According to the humanitarian view, the guiding aim of international humanitarian law (IHL) should be to reduce wrongful suffering in war by prohibiting militarily unnecessary killing, maiming, and destruction. More precisely, IHL should aim to prohibit types of harmful acts that are generally or typically militarily unnecessary. For example, on this view, IHL rightly prohibits targeting civilians because generally it is unnecessary to target civilians in order to defeat opposing armed forces. The humanitarian view is sometimes expressed in terms of striking a balance between military and humanitarian considerations. Unfortunately, the rhetoric of balancing often proves empty. When military and humanitarian considerations directly and broadly conflict, military considerations always seem to prevail.
The humanitarian view was elegantly expressed by Vattel, who held that
All acts of hostility which injure the enemy without necessity, or which do not tend to procure victory and bring about the end of the war, are unjustifiable, and as such condemned by the natural law.78
(p.39) At the same time, Vattel saw that warring parties should not be left free to judge the necessity or utility of their own military operations on a case-by-case basis, since such an open-ended principle would invite self-serving judgments and endless recriminations. Instead,
as between Nation and Nation, we must lay down general rules, independent of circumstances and of certain and easy application … . Hence, … the voluntary Law of Nations limits itself to forbidding acts that are essentially unlawful and obnoxious … . On the other hand, it permits or tolerates every act which in its essential nature is adapted to attaining the end of the war; and it does not stop to consider whether the act was unnecessary, useless or superfluous in a given case unless there is the clearest evidence that an exception should have been made in that instance; for where the evidence is clear freedom of judgment cannot be exercised.79
On this view, international law should prohibit acts of war that are typically militarily unnecessary while tolerating acts of war that are typically militarily necessary.
The humanitarian view later found expression in several foundational treaties. Most notably, the St Petersburg Declaration states that “the progress of civilization should have the effect of alleviating as much as possible the calamities of war,” that law must fix “the technical limits at which the necessities of war ought to yield to the requirements of humanity,” and that the parties will continue to work “to conciliate the necessities of war with the laws of humanity.”80 Similarly, the fourth Hague Convention was “inspired by the desire to diminish the evils of war, as far as military requirements permit.”81
Recently, Janina Dill and Henry Shue have offered a vigorous philosophical defense of the humanitarian view. On their account, the aim of IHL should be “to limit all killing [in armed conflict] as much as possible.”82 It is not possible to limit killing in ways that make winning impossible, because parties determined to win will simply ignore such limitations. However, it is possible to limit killing in ways that make winning more difficult. Dill and Shue reason that the law is right to categorically prohibit targeting civilians because this prohibition limits the killing of a large category of individuals and generally it is possible to win without targeting civilians. In some cases, it may be easier to win by targeting civilians, but this is an option that the law can foreclose without being systematically ignored. The law is (p.40) also right not to prohibit targeting combatants because generally it is not possible to win without targeting combatants. According to Dill and Shue, “the rules for the conduct of war cannot in general restrict the killing of combatants” or they will be ignored.83 Since all individuals are either civilians or combatants, it follows that the legal restrictions on killing civilians limit lawful killing in armed conflict as much as possible.84
Importantly, Dill and Shue argue that most combatants—just and unjust alike—are not morally liable to be killed. It seems to follows that, on their view, law and morality sharply conflict, since intentionally killing individuals who retain their moral right not to be killed is almost always morally wrong. Nevertheless, Dill and Shue argue that, by limiting lawful killing as much as possible, IHL indirectly reduces wrongful killing as much as possible.85 Thus, rather than aiming to prohibit wrongful killing, IHL aims to minimize wrongful killing in part by permitting some wrongful killing.
As Dill and Shue observe, “even if we all agree on noncombatant immunity, it matters what underlies the application of discrimination.”86 As it happens, Dill and Shue provide the wrong explanation for the right result. After all, if many soldiers retain their moral rights then in some cases soldiers may violate fewer rights by targeting civilians than by targeting combatants. In these cases, by limiting lawful killing in war IHL may increase actual killing. When means subvert ends in this way, rule-consequentialism seems like rule-fetishism. It is not clear why combatants who internalize the aim of reducing wrongful suffering in war will follow a legal rule prohibiting attacks on civilians when attacking civilians will in fact reduce wrongful suffering. Indeed, while the humanitarian view aims to reduce wrongful suffering in war, IHL in fact concentrates suffering in war on one category of individuals rather than another.
The deeper problem with the humanitarian view is that while the utility of targeting civilians varies, the morality of targeting civilians remains constant. To take an extreme example, suppose that if IHL does not prohibit the intentional killing of young children, then this will in fact reduce wrongful suffering in war. Such a position might deter states from initiating armed conflict and encourage belligerents to surrender more quickly when victory is in doubt. On the humanitarian view, (p.41) IHL should adopt this position and not prohibit the intentional killing of young children. If this implication seems hard to accept, then civilian immunity likely does not rest on empirical contingencies in the way the humanitarian view suggests. I will have more to say about civilian immunity in Chapter 3.
Similarly, there is only an empirically contingent connection between the voluntary conduct of combatants and the utility of killing them, but there are morally necessary connections between the voluntary conduct of combatants and the morality of killing them. Many combatants pose unjust threats to civilians who retain their basic rights, while most civilians make indirect and superfluous contributions to their armed forces. In addition, combatants can avoid being eliminatively harmed more easily than civilians can avoid being opportunistically harmed, in part because combatants choose to make themselves lawful targets. I will say more about these considerations at the end of Chapter 3.87
These moral considerations systematically favor legal rules that prohibit targeting civilians but do not prohibit targeting combatants. Killing civilians is, at a minimum, intrinsically morally worse than killing combatants. It follows that IHL should not aim to limit all killing as much as possible and distribute lawful killing in whatever way will in fact minimize wrongful killing in war. At the very least, IHL should skew the distribution of actual killings by strictly limiting the more wrongful killings (of civilians) but not comparatively less wrongful killings (of combatants).
Indeed, in subsequently published work, Dill shows that law might limit killing as much as militarily possible in at least two very different ways.88 According to the “logic of sufficiency,” law should tolerate intentionally and collaterally killing combatants, as this is both necessary and sufficient for military victory. At the same time, the law should prohibit intentionally killing civilians, as this is generally unnecessary for military victory. In contrast, according to the “logic of efficiency,” the law should allow each party to achieve its war aims in the quickest and least costly way possible, even if this means targeting civilians and civilian objects. Importantly, Dill argues that it is impossible to empirically determine which “logic” of warfare will best reduce killing in war, that is, whether “contained wars are the least destructive” or whether “sharp wars are brief.”
Accordingly, Dill argues that we should prefer the logic of sufficiency to the logic of efficiency because generally it is morally worse to kill civilians than to kill combatants. As Dill observes, civilians are often vulnerable, defenseless, and non-threatening. Moreover, a legal rule prohibiting attacks on civilians allows civilians to be secure in their expectations that they will not be attacked unless they exercise their agency in defined ways. This refinement of Dill’s view is both significant and welcome. However, the “logic of sufficiency” remains grounded in rule-consequentialism. (p.42) Non-consequentialist considerations merely support one set of rules over another when we cannot tell which set of rules would yield better consequences.
It is not clear why individual soldiers would follow rules justified on rule-consequentialist grounds rather than follow their own moral judgment. After all, Dill and Shue do not argue that soldiers will wrongfully kill fewer people by following IHL than by violating IHL, or even that a smaller proportion of their killings will be wrongful killings if they follow IHL than if they violate IHL. They argue that soldiers who follow IHL will wrongfully kill one category of people but not another category of people. It is hard to see why a rational, moral soldier would abstain from militarily efficacious actions on such grounds. Presumably, such soldiers will be concerned primarily with the rights and duties they infringe and only secondarily with the aggregate consequences of general rules.
Soldiers are not computers, and legal norms are not software programs. Soldiers are human beings who bear moral obligations and exercise moral judgment. The law must address them as such. Soldiers need a reason to obey the law, and the most obvious reason to do so is that they will better fulfill their moral obligations by following the law than by following their independent moral judgment. As Seth Lazar nicely observes, rule-consequentialism is a third-person view of moral justification, while soldiers need a first-person view that addresses them as moral agents.89
None of this is to say that legal rules should never aim at producing good aggregate consequences, or that individuals never have reasons to obey such rules. Often, we have moral reasons to cooperate with others in order to achieve shared aims together that we could not achieve individually (think of making small contributions to the same charity). Similarly, we may have moral reasons to coordinate with others so that we can pursue our individual aims more effectively or more safely (think of driving on the same side of the road). In contrast, in war we compete with our adversary, and our moral reasons do not depend on their behavior. We can avoid targeting civilians, and have moral reasons to do so, irrespective of how our adversary behaves.
Of course, many scholars believe that rule-consequentialist sensibilities, whatever their normative credentials, drove the historical development of IHL. For example, David Luban argues that “humanitarianism in war is plainly a form of negative benthamism” aimed at reducing aggregate suffering.90 Luban cites the preamble to the St Petersburg Declaration as evidence that the historical aim of IHL was “alleviating as much as possible the calamities of war.”91
(p.43) In fact, the St Petersburg Declaration prohibited the use of a particular weapon—exploding bullets—thought to inflict suffering on each individual combatant unnecessary to incapacitate him. This suggests that the Declaration’s purpose was to prevent unnecessary individual suffering, not to prevent unnecessary aggregate suffering. To see this, suppose that the drafters received credible reports that the use of exploding bullets actually reduces aggregate suffering, because the extreme suffering that they inflict on individual combatants encourages other combatants to surrender more quickly. It is far from clear that the drafters would have abandoned their project, allowing the infliction of unnecessary individual suffering in order to reduce aggregate suffering.
While the humanitarian view may or may not provide a historical explanation of how IHL developed, it certainly does not provide a moral explanation for why soldiers should follow IHL. The contingent aim of some states in developing IHL may have been to reduce aggregate suffering. However, I argue below that the constitutive aim of IHL itself is to provide a moral service to combatants.
The Service View
At times, McMahan suggests an alternative to the humanitarian view, writing that “Ideally we should establish laws of war best suited to get combatants on both sides to conform their action as closely as possible to the constraints imposed by the deep morality of war.”92 Importantly, McMahan observes that “combatants should be reluctant to give their individual judgment priority over the law, for the law has been designed in part precisely to obviate the need for resort to individual moral judgment in conditions that are highly unconducive to rational reflection.”93 These passages suggest that the aim of the law of war should not be to minimize wrongful harm overall. Instead, the law should aim to help combatants to conform to their moral obligations more closely than they would by relying on their individual moral judgment. This seems to me the better view.
Why should any individual follow the law rather than follow his or her own moral judgment? As moral agents, our ultimate aim should be to fulfill our moral obligations and conform to the strongest moral reasons that apply to us. We should follow our own moral judgment if and only if this is the best available means of achieving our ultimate aim. In some cases, we are more likely to fulfill our moral obligations if we defer to the moral judgment of another person who is better informed, more perceptive, more logical, or more virtuous. Similarly, sometimes we are more likely to fulfill our moral obligations if we follow a rule that directs the morally correct action more often than will a (p.44) series of case-by-case judgments. Such a rule provides a service to moral agents by helping them to conform to their moral obligations better than they could on their own. According to the service view, the LOAC should aim to provide such a service to combatants.
Joseph Raz famously argues that law necessarily claims legitimate authority, that is, that those subject to the law morally ought to follow the law rather than their own moral judgment when the two conflict.94 Of course, law’s necessary claim is not necessarily true. Law has legitimate authority over an individual only if that individual “would better conform to reasons that apply to him anyway (that is, to reasons other than the directives of the authority) if he intends [or tries] to be guided by the authority’s directives than if he does not.”95 When this normal justification condition is met, law’s claim to legitimate authority is vindicated by the service that it provides those subject to the law by helping them conform to the reasons that apply to them independently of the law. Accordingly, Raz names his account of legitimate authority the service conception.
The service conception eases the tension between legal authority and individual autonomy. First, we must exercise our own moral judgment in order to determine that the normal justification condition is satisfied and only then accept the law as a legitimate authority over us. Second, if the normal justification condition is satisfied, then the law is “simply one device, one method, through the use of which people can achieve the goal (telos) of their capacity for rational action, albeit not through its direct use.”96 In other words, we can choose to comply with the law as an indirect strategy for pursuing our ultimate aim of fulfilling our moral obligations.
Let me elaborate on the idea of an indirect strategy. As moral agents, our ultimate aim is to act permissibly in the objective or fact-relative sense, that is, to conform to the moral reasons that objectively apply to us given all the morally relevant facts. Since we do not have unmediated access to the moral reasons that objectively apply to us, we can only pursue our ultimate aim indirectly, by way of our beliefs, the evidence available to us, or the rules applicable to us. We might pursue our ultimate aim by acting permissibly in the subjective or belief-relative sense, that is, by doing what would conform to the moral reasons that would apply to us if the morally relevant (p.45) facts were as we believe. We might pursue our ultimate aim by acting permissibly in the epistemic or evidence-relative sense, that is, by doing what would conform to the moral reasons that would apply to us if the morally relevant facts were as our evidence suggests.97
However, often our subjective beliefs are unreliable, the available evidence is incomplete or misleading, or our moral reasoning is distorted by emotion or bias. In such cases, we may better pursue our ultimate aim not by following our beliefs or our evidence but by following a rule. Following a rule is just another indirect strategy for conforming to the moral reasons that objectively apply to us.
Put another way, individuals always have decisive reason to do what is objectively morally required but sometimes have decisive reason not to try to do what is objectively morally required. In some cases, if they try to do what is objectively morally required, then they are likely to fail, but if they try to follow a rule, then they are more likely to succeed in doing what is objectively morally required. In other words, often we are more likely to satisfy the moral standards that apply to us, not by applying those standards, but instead by following a well-designed decision procedure. In such cases, we have decisive reason to adopt that decision procedure rather than exercise our (more fallible) moral judgment on a case-by-case basis. We may say that an act that is permissible under a rule or decision procedure that we have decisive reason to adopt is permissible in the instrumental sense.
Importantly, in exceptional cases, generally reliable rules may provide no service to their subjects and obedience would only lead them astray.98 If a generally reliable rule prohibits an act that is clearly morally required, or requires an act that is clearly morally prohibited, then individuals have no reason to defer to that rule rather than follow their own moral judgment. Individuals should neither blindly follow generally reliable rules in morally clear cases nor second-guess generally reliable rules in morally unclear cases. Instead, individuals should exercise their moral judgment when they adopt generally reliable rules, trust in those rules in morally unclear cases, and trust in their moral judgment when those rules yield clearly wrong results.99
Like all law, the LOAC claims legitimate authority over those subject to it. The LOAC claims that combatants should follow the law rather than their own moral judgment when the two conflict. According to the service conception, this claim is vindicated only to the extent that combatants will better conform to the moral reasons that apply to them by following the law than by trying to act directly on (p.46) those reasons. Put another way, combatants have decisive reason to follow the law only if doing so will help them fulfill their moral obligations. If combatants will make morally better decisions by ignoring the law and relying on their own moral judgment, then they have decisive reason to do just that.
In war, moral judgment is easily misled by limited or unreliable information; clouded by stress, fatigue, anger, and fear; and distorted by all-too-familiar psychological dynamics. Combatants are human beings and, like the rest of us, too readily obey authority; conform to group behavior; and rationalize wrongdoing, for example, by adverting to the wrongful conduct of our adversaries, to the moral guilt of our victims, or to our own feelings of victimization. As one important study concluded:
While attempts at justification such as those referred to can enable combatants to switch off guilt feelings in the face of inhuman acts and to stretch moral values by legitimizing such acts, they cannot confer legality on such behaviour. The [law] draws an easily identifiable red line, whereas [moral] values represent a broader spectrum which is less focused and more relative.100
In other words, rather than attempt sound moral reasoning in such adverse circumstances, combatants may better conform to their moral obligations by following the law.
Throughout this book, I appeal to the service view to evaluate apparent gaps between the law of war and the morality of war. For example, in Chapter 3, I defend the legal prohibition of direct attacks on civilians not directly participating in hostilities. I will argue, against many revisionists, that almost all such civilians retain their ordinary moral rights. I then appeal to the service view to show that combatants will best conform to their moral obligations by following the general legal prohibition rather than by trying to identify and target civilians who have forfeited their ordinary moral rights. Very few civilians forfeit their moral rights, these civilians are very difficult to distinguish from the majority of civilians who retain their moral rights, and killing these civilians is seldom necessary to prevent harm to others. If soldiers try to target morally liable civilians, then they are much more likely to wrongfully kill than to prevent wrongful killing. It follows that soldiers will better conform to their moral obligations by following the law than by following their own moral judgment.
In contrast, in Chapter 4, I propose a limited legal prohibition of directing attacks at members of armed forces or armed groups whom the attacker knows or has decisive reason to believe perform noncombat functions. Often it is objectively wrong to kill noncombat personnel. At the same time, if combatants must positively (p.47) identify opposing combat personnel, distinguishing them from opposing noncombat personnel, then they will often hesitate to use objectively permissible force and allow their fellow soldiers or civilians to be harmed. Moreover, in general, noncombat personnel can avoid being mistaken for combat personnel more easily than can civilians, simply by conforming to more stringent moral duties than civilians bear. It follows that attackers need not take the same precautions to avoid mistakenly killing such individuals that they must take to avoid mistakenly killing civilians. However, if an attacker knows or has decisive reason to believe that a person performs a noncombat function, then the moral costs of restraint are low relative to the moral costs of attack. I argue that such a limited legal prohibition will help combatants better conform to their moral obligations.
It might appear that on these issues the service view simply requires epistemic permissibility. If most civilians are not morally liable, then perhaps it is presumptively unreasonable to believe that any particular civilian is morally liable. Similarly, if most unjust combatants are morally liable, then perhaps it is presumptively reasonable to believe that each particular combatant is morally liable.101 However, this is not quite correct. Objective permissibility depends on intrinsically morally relevant facts. Epistemic permissibility depends on one’s evidence regarding such intrinsically morally relevant facts. In contrast, the rule prohibiting attacks on civilians makes no reference to intrinsically morally relevant facts, such as the moral rights or liabilities of particular individuals. Instead, the rule refers to facts that reliably indicate that such intrinsically morally relevant facts exist or do not exist.102 The point of relying on IHL is to allow combatants to bypass individualized assessments of intrinsically morally relevant facts when the moral stakes are high but directly relevant evidence is weak. Moreover, combatants should not trust the reasonableness of their beliefs regarding the moral liability of civilians, which are susceptible to distortion by non-rational factors. Combatants would better avoid killing morally protected civilians by following IHL than by following their own judgment. This is the service that IHL should aim to provide combatants.
Importantly, some rules of IHL invite combatants to exercise moral judgment. Most notably, the proportionality rule invites attackers to judge whether the collateral harm they expect their attack will inflict on civilians would be excessive in relation to the concrete and direct military advantage they anticipate. The proportionality rule nevertheless performs an important guidance function by framing the relevant moral issue and focusing moral deliberation. The proportionality rule directs attackers to consider only two morally relevant variables—expected civilian losses and anticipated military advantage—while excluding all others. Moreover, (p.48) the proportionality rule directs attackers to consider only military advantages that are both concrete and direct, and to resist wishful thinking. Finally, combatants may receive moral guidance from the rule itself, from authoritative interpretations of the rule, and from rules of engagement that implement the rule. In these ways, moral guidance and moral judgment may be mutually reinforcing rather than mutually exclusive.
Revisionists might argue that the LOAC can only help just combatants to conform to their moral obligations. On their view, unjust combatants cannot conform to their moral obligations except by laying down their arms, which the law does not require. Yet the law does not claim legitimate authority only over just combatants or only over combatants who believe—rightly or wrongly—that their cause is just. The law also claims legitimate authority over combatants who are not sure whether their cause is just or unjust and seek to hedge against the moral risk that they fight for an unjust cause. Finally, the law claims legitimate authority over combatants who believe that their cause is unjust, elect to fight under duress or out of excessive partiality, yet wish to limit the kind and degree of their own wrongdoing.
On one view, the LOAC should prohibit those acts that would be morally impermissible if committed by combatants fighting for a just cause. The law should then simply apply those prohibitions to the just and the unjust alike, on the assumption that all combatants believe that they fight for a just cause.103 Though attractive, this view seems too elegant to cope with a messy reality. The truth is that many combatants fight despite grave doubts regarding the war as a whole. They may doubt that they fight for a just cause, or that war was necessary or proportionate. They may suspect that justice favors their adversary, or that justice condemns both parties. Nevertheless, they may choose to fight in order to protect their friends, families, and communities from the consequences of their government’s decision to go to war. The LOAC must speak to these combatants as well, by showing them that, even in an unjust war, there are rules worth following.
On the service view, the legitimate authority of law does not require that individuals who conform to the law will never perform objectively impermissible acts, only that they will perform fewer or less wrongful objectively impermissible acts than they would by relying on their own moral judgment. As noted in Chapter 1, while permissibility is all-or-nothing, wrongfulness is a matter of degree. An act is objectively impermissible only if the moral reasons against its performance outweigh the moral reasons in favor of its performance. In contrast, the wrongfulness of an impermissible act depends on the degree to which the moral reasons against its performance outweigh the moral reasons in favor of its performance. Although the balance of reasons matters most, it still matters whether the balance of reasons is (p.49) tipped by a penny or by a pound. Even when we fall short of our moral obligations, it matters whether we fall short by an inch or by a mile.
To vindicate the legitimate authority of the LOAC, it is enough to show that combatants fighting for an unjust cause will act less wrongfully if they follow the law than if they follow their own moral judgment and violate the law. In other words, it is enough to show that killings in pursuit of an unjust cause are morally worse if they also violate the law than if they at least conform to the law.
As we will see in Chapter 3, targeting civilians is morally worse than targeting combatants even when neither has forfeited their moral right not to be killed. Just combatants often pose unjust threats to civilians who have a moral right not to be killed. It follows that there are moral reasons to kill just combatants that offset the moral reasons not to kill just combatants. In addition, combatants can avoid eliminative harm more easily than civilians can avoid opportunistic harm and, on a related though distinct point, forgo fair opportunities to avoid making themselves lawful targets. It follows that the moral reasons not to kill just combatants are weaker than the moral reasons not to kill civilians. In these ways, among others, it is less wrongful to kill just combatants than to kill civilians.104 It follows that unjust combatants will act less wrongfully by following the law than by targeting combatants and civilians without distinction.
As we will see in Chapter 8, jus in bello proportionality is best understood to prohibit attacks that inflict greater collateral harm on civilians than they prevent the opposing force from inflicting on civilians and combatants in current or future military operations. Since there are strong moral reasons to prevent harm to civilians as well as significant moral reasons to prevent harm to most unjust combatants, an attack that satisfies jus in bello proportionality is generally less wrongful than one that does not. Such attacks may remain all-things-considered impermissible, but the balance of reasons is hardly one-sided. The moral reasons to launch such attacks partially offset the moral reasons not to launch such attacks. It follows that unjust combatants will act less wrongfully by following the law than by inflicting disproportionate harm on civilians.
Revisionists may be right that it is objectively morally impermissible to fight for a party pursuing an unjust cause. Nevertheless, ordinary combatants cannot force their political leaders to abandon their war aims and make peace. Instead, ordinary combatants must choose between killing opposing combatants and foreign civilians or allowing opposing combatants to kill their fellow soldiers and their own civilians. Over the remainder of the book I hope to demonstrate what I have already suggested: that by following the LOAC (suitably interpreted and developed) all combatants can ensure that they will fight, if not permissibly, then less wrongfully than they would otherwise. In other words, I hope to vindicate the LOAC’s claim to legitimate authority over just and unjust combatants alike.
As we have seen, Raz’s service conception provides an analysis of law’s claim to legitimate authority as well as a standard by which to evaluate that claim. My own view is that we should interpret and develop the LOAC so as to vindicate the law’s claim of legitimate authority. We should interpret and develop the law such that the law satisfies the normal justification condition with respect to as many combatants as possible.
The constitutive aim of all law is to guide human conduct through positive normative standards. The positive nature of law distinguishes law from morality while the normative nature of law distinguishes law from a system of mere incentives. For law to succeed in its constitutive aim, it must provide those subject to the law with decisive reasons to accept its guidance. On my view, the basic task of legal interpretation is to help law to achieve its constitutive aim by casting positive legal materials in their morally best light, that is, in the light that gives those subject to the law the strongest moral reasons to obey the law.
Of course, some laws are both legally unambiguous and morally indefensible. In such cases, positive legal materials make a morally attractive interpretation unsustainable. The failure of such laws to achieve their constitutive aim is so complete that the failure cannot be cured through interpretation alone. But if the content of a particular law is ambiguous then interpretation can serve the constitutive aim of all law without directly contradicting the contingent aim of the law in question. Put another way, if the specific aim of the drafters of a law is indeterminate then it is appropriate to revert to the constitutive aim that the drafters implicitly adopted as their own general aim when they chose to pursue their specific aims through law.
The states that create international law through treaty and custom have a variety of contingent aims, including the humanitarian aim of reducing wrongful suffering in war. However, these states chose to pursue their contingent aims through law, and the constitutive aim of law is to give those subject to the law decisive reason to follow the law rather than their own moral judgment. Our aim in interpreting the LOAC should be to ensure that the law achieves its constitutive aim.
It might seem that mine is a heterodox approach to the interpretation of international law. According to the orthodox approach, the terms of a treaty should be “interpreted in good faith in accordance with the[ir] ordinary meaning … in their context and in the light of its object and purpose.”105 If this (p.51) textual–purposive approach “[l]eaves the meaning” of these terms “ambiguous or obscure” or “[l]eads to a result which is manifestly absurd or unreasonable,” then their meaning may be determined by “supplementary means of interpretation” such as preparatory materials and concluding commentary.106 This textual–purposive approach is appropriate in most cases, since lawmakers can only exercise legitimate practical authority if they can anticipate how the laws they make will be interpreted.107
As we shall see in later chapters, the drafters of Protocol I deployed evaluative terms such as “feasible,” “reasonable,” and “excessive”—and introduced such terms into customary law—without explicitly defining these terms or coming to a consensus regarding their intended meaning. Neither ordinary meaning nor context provides these terms with determinate content. Relevant preparatory materials and concluding commentary shed little additional light. In each case, we must interpret these terms in light of the object and purpose of Protocol I, namely “protecting the victims of armed conflicts,”108 as well as the purpose of the LOAC as a whole, namely “to protect combatants and non-combatants from unnecessary suffering and to safeguard the fundamental human rights of persons who are not, or are no longer, taking part in the conflict … and of civilians.”109
As David Luban observes, the difficulty with purposive interpretation “lies in what you take the purpose of laws of war to be. Is it to protect civilians, even at cost to military effectiveness, or is it to give full sway to military necessity and protect civilians (only) against military excess?”110
Endorsing the former view, the International Criminal Tribunal for the former Yugoslavia (ICTY) controversially held that “a rule of international humanitarian law [that] is not sufficiently rigorous or precise … must be interpreted so as to construe as narrowly as possible the discretionary power [of] belligerents and, by the same token, so as to expand the protection accorded to civilians.”111
Endorsing the latter view, Michael Schmitt posits that states, as authors of international law, would never agree to legal norms that “unduly restrict their freedom of action on the battlefield, such that national interests might be affected.”112 Schmitt concludes that sound interpretation of international law should reflect “a (p.52) reasonable balance between military necessity and humanity,” that is, a balance that leaves “states … reasonably free to conduct their military operations effectively.”113
These dueling approaches may seem worlds apart both from each other and from the service view. But things are not as they appear. Perhaps surprisingly, my apparently heterodox approach to purposive interpretation both captures and reconciles what is attractive in these dueling mainstream approaches while leaving behind their excesses.
According to the service view, we should indeed interpret imprecise legal norms so as to legally protect civilians—from objectively impermissible harm, epistemically impermissible harm, and instrumentally impermissible harm. At the same time, according to the service view, we should indeed interpret imprecise legal norms so as to balance military and humanitarian considerations—by reference to their respective moral weight.
On my view, military considerations support a permissive interpretation to the extent that they reflect moral reasons to perform the type of act in question, such as the losses that such acts might prevent to the attacking force or to its civilian population. Humanitarian considerations support a restrictive interpretation to the extent that they reflect moral reasons to refrain from the type of act in question, such as the losses that such acts might inflict on the civilian population of the opposing party. We balance military and humanitarian considerations by balancing the moral reasons for and against performing the act in question. When the former outweigh the latter we should adopt a permissive interpretation; when the latter outweigh the former we should adopt a restrictive interpretation.
Crucially, when adopting either a permissive or a restrictive interpretation we should consider not only whether the act in question is morally permissible in the objective and epistemic senses but also whether the act in question is morally permissible in the instrumental sense. In other words, we should adopt the interpretation that will best help combatants to conform to their objective moral obligations. Sometimes that will mean tracking the conditions of objective or epistemic permissibility; often it will mean departing from those conditions. As we have seen, combatants are sometimes more likely to act objectively permissibly if they follow the law rather than their own moral judgment. We should interpret imprecise laws so that they provide combatants with the best possible moral guidance, understood in this instrumental sense.
There is another reason to interpret the LOAC in its morally best light, namely to narrow the gap between the LOAC and human rights law. Since human rights are moral rights that we all posses in virtue of our humanity, morally unjustified killings violate our right not to be arbitrarily deprived of our lives. Accordingly, we should interpret the LOAC so as to prohibit morally unjustified killings. Evidently, a killing that is morally permissible in the objective, epistemic, or instrumental sense (p.53) is not morally unjustified and should not be considered arbitrary. For example, a wartime killing that is morally permissible in the instrumental sense should not be considered arbitrary because, by hypothesis, following the relevant rule is the best way to avoid violating the human rights of others, such that one could not better respect human rights by relying on one’s own moral judgment. Accordingly, we should interpret the LOAC so that it helps combatants avoid violating human rights better than they could on their own.
International Law and National Law
The discussion so far raises important questions regarding the relationship between international law and national law. Does national law permit combatants to fight, or merely bar their prosecution for fighting? When international law and national law diverge, which should combatants follow? Let us take each question in turn.
Under the national law of many states, soldiers are justified in fighting to the extent that they exercise lawful authority or obey lawful orders. For example, according to the United States Manual for Courts-Martial, “A death, injury, or other act caused or done in the proper performance of a legal duty is justified and not unlawful.”114 In particular, “killing an enemy combatant in battle is justified.”115 In addition, the US Manual provides that “An act performed pursuant to a lawful order is justified.”116 Accordingly, under national law, US soldiers who kill enemy combatants in conformity with the LOAC are typically justified and not merely immune from prosecution. In contrast, under national law, foreign combatants who kill US soldiers in conformity with the LOAC are typically not justified but merely immune from prosecution.117 While the LOAC applies symmetrically, national law typically applies asymmetrically.
As we have seen, law necessarily makes moral claims. Accordingly, a legal justification is a moral justification claimed by law. It is hardly surprising that states claim that their soldiers are justified in fighting. Nor is it surprising that states do not claim that opposing soldiers are justified in fighting but instead claim only that opposing soldiers are immune from prosecution. After all, states invariably claim that they are waging a just war while their adversaries are waging an unjust war. Indeed, a state that orders its soldiers to war, authorizes them to use lethal force, and obligates them (p.54) to follow orders necessarily claims that its soldiers are morally permitted (indeed, morally required) to fight. Of course, most such claims turn out to be false.118
Under national law, soldiers have a legal duty to obey general orders, including rules of engagement, as well as specific orders. According to the service conception of legitimate authority, soldiers morally should defer to their superiors only if they are more likely to conform to the moral reasons that apply to them by obeying their orders than by relying on their own moral judgment. Conversely, soldiers should not defer to their superiors if ordered to commit acts that are clearly immoral, or if important new information or circumstances arise that their superiors did not anticipate, or if they know that their superiors issued their orders arbitrarily or in bad faith.119
If a soldier’s orders prohibit what international law tolerates, then there is no true conflict. As we have seen, international law does not authorize or entitle soldiers to kill in ways that it does not prohibit, or give soldiers a right to kill that national law might infringe. Typically, soldiers morally ought to follow orders that are more restrictive than international law. Following such orders will help soldiers avoid wrongful killing and, of course, satisfy their oath. Moreover, since armed forces typically have an interest in using as much force as law and morality permit, a decision by military commanders to issue such restrictive orders warrants significant deference.
In contrast, if a soldier’s orders require what international law prohibits, then there is a true conflict. As Raz observes, “When several authorities pronounce on the same matter and their directives conflict, we must decide, to the best of our ability, which is more reliable as a guide.”120 Typically, soldiers morally ought to refuse to obey orders that violate international law. As we will see throughout this book, almost all killings that violate the LOAC are seriously wrongful, such that a soldier’s promissory obligation to obey her superiors cannot render these killings morally permissible. Moreover, since armed forces typically have an interest in using more force than morality permits, a decision by military commanders to issue such permissive orders warrants little deference.
Fortunately, under the national law of most states, soldiers have no duty to obey orders that violates international law. Unfortunately, in many states, soldiers may be punished for disobeying orders that they believe (rightly or wrongly) violate international law but that a court-martial later decides (rightly or wrongly) do not violate international law.121 Accordingly, a soldier may be punished under national (p.55) law if she disobeys an order to perform an act that violates international law but that conforms to her state’s mistaken interpretation of international law. Conversely, that same soldier may be punished under international law if she obeys that same order, provided that she believes that the ordered act is unlawful or the ordered act is manifestly unlawful. For this reason, among many others, it is imperative that states adopt reasonable interpretations of international law that bring national law, international law, and morality into alignment.
International law does not confer on combatants a legal permission to fight for an unjust cause and therefore does not claim that combatants are morally permitted to fight for an unjust cause. Conventionalists need not defend such a moral permission in order to defend international law. Revisionists need not insist on a sharp divergence between law and morality or feel compelled to embrace the humanitarian view that the law should aim to reduce wrongful suffering in war to the greatest extent practically possible. The law of armed conflict is prohibitive, not permissive, and applies alongside other applicable moral and legal norms. Accordingly, acts not prohibited by the law of armed conflict may be prohibited by human rights law or by moral rules.
Though initially attractive, the humanitarian view does not stand up to critical scrutiny. Instead, we should accept the service view that the law should aim to provide moral guidance to combatants, helping them to conform as closely as possible to their moral obligations. Ideally, combatants on all sides will commit fewer and less serious moral wrongs by following the law than they would by violating the law and following their own moral judgment. Finally, wherever possible, we should interpret and develop the law so as to vindicate the law’s claim to legitimate authority and provide the best possible moral guidance to combatants.
(1) United Nations Charter arts 2(4), 42, 51.
(2) Protocol Additional to the Geneva Conventions of 12 August 1949, and relating to the Protection of Victims of International Armed Conflicts (Protocol I) (adopted 8 June 1977, entered into force 7 December 1978) 1125 UNTS 3, arts 43(2), 44; International Committee of the Red Cross, Customary International Humanitarian Law, vol 1 (CUP 2009) 11, 384.
(3) Protocol I preamble.
(4) In important respects, the conventionalist view itself departed from traditional just war theory. See Gregory M Reichberg, “Just War and Regular War: Competing Paradigms,” in David Rodin & Henry Shue (eds), Just and Unjust Warriors: The Moral and Legal Status of Soldiers (OUP 2008) 193.
(5) Michael Walzer, Just and Unjust Wars (2nd edn Basic Books 1992) 44.
(7) Walzer, Just and Unjust Wars, 21; see also 136 (“The rules of war apply with equal force to aggressors and their adversaries … . Soldiers fighting for an aggressor state are not themselves criminals: hence their war rights are the same as those of their opponents”).
(8) Walzer, Just and Unjust Wars, 41; see also 36 (“Though there is no license for war-makers, there is a license for soldiers, and they hold it without regard to which side they are on; it is the first and most important of their war rights. They are entitled to kill”).
(11) Jeff McMahan, Killing in War (OUP 2009). McMahan allows that combatants fighting for a just cause may pose unjust threats in particular cases, for example by threatening to intentionally, unnecessarily, or disproportionately harm innocent civilians. Conversely, combatants fighting for an unjust cause may pose just threats in particular cases, for example in the course of protecting innocent civilians from intentional, unnecessary, or disproportionate harm. However, McMahan argues that such cases do not amount to a general moral equality of combatants.
(13) Emer de Vattel, The Law of Nations (1758) (Thomas Nugent trans, Liberty Fund 2008) book III, §188.
(15) Jeff McMahan, “The Morality of War and the Laws of War,” in David Rodin and Henry Shue (eds), Just and Unjust Warriors (OUP 2008) 19, 34–5 (“My suggestion, then, is that we distinguish sharply and explicitly between the morality of war and the law of war”). See also Jeff McMahan, “The Ethics of Killing in War” (2004) 114 Ethics 693, 730.
(16) Protocol I art 43(2).
(17) See Geneva Convention (III) relative to the Treatment of Prisoners of War (opened for signature August 12, 1949, entered into force October 21, 1950) 75 UNTS 135. Protocol I clearly distinguishes between the right to be a combatant and the right to be a prisoner of war. See Protocol I, art 43(2) (stating that members of armed forces “are combatants, that is to say, they have the right to participate directly in hostilities”), art 44(1) (“Any combatant, as defined in Article 43, who falls into the power of an adverse Party shall be a prisoner of war”), arts 44(2), 44(5), and 47(1) (distinguishing the right to be a combatant from the right to be a prisoner of war).
(18) David Rodin, “Morality and Law in War,” in H Strachan & S Scheipers (eds), The Changing Character of War (OUP 2011) 446, 455.
(19) Jeremy Waldron, Torture, Terror, and Trade-offs (OUP 2010) 107. But see Jeremy Waldron, “Responses to Zedner, Haque and Mendus” (2014) 8 Criminal Law and Philosophy 137 (writing that “Haque’s account of this in terms of a ban on prosecutions is more accurate and more precise than my characterization in terms of a Hohfeldian privilege”).
(20) Yoram Dinstein, The Conduct of Hostilities under the Law of International Armed Conflict (CUP 2010) 33.
(21) See, eg, International Committee of the Red Cross, Interpretive Guidance on the Notion of Direct Participation in Hostilities Under International Humanitarian Law (ICRC 2009) 83–4 (stating that “civilian direct participation in hostilities is neither prohibited by IHL nor criminalized under the statutes of any prior or current international criminal tribunal or court”); Philip Alston, Report of the Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions, Study on Targeted Killings, Human Rights Council, UN Doc A/HRC/14124/Add.6 (May 28, 2010), para 71 (“Under IHL, civilians … are not prohibited from participating in hostilities”).
(22) See generally Wesley Hohfeld, Fundamental Legal Conceptions (1919).
(23) See, eg, ICRC, Interpretive Guidance 83 (“This right [to directly participate in hostilities] … merely provides combatants with immunity from domestic prosecution for acts which, although in accordance with IHL, may constitute crimes under the national criminal law of the parties to the conflict (the so-called combatant privilege)”); Alston, Study on Targeted Killings, para 71 (“the consequence of participation is two-fold. First, … [civilian participants] may themselves be targeted and killed. Second, [civilian participants] do not have immunity from prosecution under domestic law for their conduct”)).
(24) Notice that civilians may be criminally prosecuted for directly participating in hostilities irrespective of jus ad bellum considerations, while lawful combatants may not be criminally prosecuted for directly participating in hostilities irrespective of jus ad bellum considerations.
(25) Seth Lazar, “The Morality and Law of War,” in Andrei Marmor (ed), Routledge Companion to Philosophy of Law (Routledge 2012) 376. See also Adil Ahmad Haque, “International Crime: In Context and in Contrast,” in RA Duff, Lindsay Farmer, SE Marshall, Massimo Renzo, & Victor Tadros (eds), The Structures of Criminal Law (OUP 2011) 106; Adil Ahmad Haque, “Law and Morality at War” (2014) 8 Criminal Law & Philosophy 79.
(26) Protocol I, art 44(3).
(27) See ICRC, Customary IHL Study, 12 (“Persons taking a direct part in hostilities in non-international armed conflicts are sometimes labelled “combatants” … but [this] does not imply a right to combatant status or prisoner-of-war status, as applicable in international armed conflicts … . The lawfulness of direct participation in hostilities in non-international armed conflicts is governed by national law”).
(28) See, eg, Lassa Oppenheim, International Law, Volume II: Disputes, War and Neutrality (Hersch Lauterpacht ed, 7th edn 1952) 115 (“The privileges of members of armed forces cannot be claimed by members of the armed forces of a belligerent who go over to the forces of the enemy and are afterwards captured by the former. They may be, and always are, treated as criminals”).
(29) See Waldemar A Solf, “The Status of Combatants in Non-International Armed Conflicts under Domestic Law and Transnational Practice” (1983) 33 American University Law Review 53, 59.
(30) Vattel, Law of Nations, book III, §192. According to an alternative translation, the law of war “does not confer upon him whose cause is unjust any true rights capable of justifying his conduct and appeasing his conscience, but merely makes his conduct legal in the sight of men, and exempts him from punishment” (Emer de Vattel, The Law of Nations (1758) (Charles G Fenwick trans, Carnegie 1916) book III, §192).
(31) For an example of this view, see George P Fletcher & Jens David Ohlin, Defending Humanity (OUP 2008) 100.
(33) See, eg, Vienna Convention on Consular Relations (opened for signature April 18, 1961, April 24, 1964) 500 UNTS 95, art 55(1) (“Without prejudice to their privileges and immunities, it is the duty of all persons enjoying such privileges and immunities to respect the laws and regulations of the receiving State”).
(34) Paul H Robinson, “Criminal Law Defenses: A Systematic Analysis” (1982) 82 Columbia Law Review 199, 229–32. Prior acquittal is another example of a bar to prosecution that does not reflect on the legal or moral guilt of the accused.
(35) RA Duff, “‘I Might Be Guilty, But You Can’t Try Me’: Estoppel and Other Bars to Trial” (2003) 1 Ohio State Journal of Criminal Law 245, 247.
(37) See, eg, John Gardner, “How Law Claims, What Law Claims,” in Law as a Leap of Faith 133 (OUP 2012); Joseph Raz, “Law, Morality and Authority,” in Ethics in the Public Domain (OUP 1994).
(38) But see McMahan, Killing in War, 105 (“The law of war does not assert the moral equality of combatants but it does assert the legal equality of combatants”). On the contrary, by asserting the legal equality of combatants the law of war necessarily asserts the moral equality of combatants. It is just that the legal equality of combatants consists in immunities from prosecution rather than in permissions to fight. Accordingly, the law of war necessarily asserts that states have decisive moral reasons not to criminally prosecute opposing combatants for acts not prohibited by the law of war.
(39) See, eg, Rome Statute of the International Criminal Court (opened for signature July 17, 1998, entered into force July 1, 2002) 2187 UNTS 3, art 8 bis.
(41) While threatening leaders with prosecution for the crime of aggression carries costs and risks—including creating incentives to win by any means necessary and thereby avoid trial and punishment—the strong retributive reasons to punish such leaders may justify such costs and risks. Thanks to Yitzhak Benbaji for pressing this point.
(42) Protocol I Commentary para 2238. See also Richard R Baxter, “So-Called ‘Unprivileged Belligerency’: Spies, Guerillas, and Saboteurs” (1951) 28 British Yearbook of International Law 323, 324 (“The law of war is … ‘prohibitive law’ in the sense that it forbids rather than authorizes certain manifestations of force”); Derek Jinks, “International Human Rights Law in Time of Armed Conflict,” in Andrew Clapham et al, Oxford Handbook of International Law in Armed Conflict 656 (OUP 2014).
(43) The law of detention and occupation may both authorize and constrain. On this view, detaining and occupying powers temporarily assume responsibility for the wellbeing of those detained or occupied, and therefore must have the legal power to issue authoritative directives until the end of detention or occupation. See, eg, Geneva Convention (III), art 21; Geneva Convention (IV) relative to the Protection of Civilian Persons in Time of War (opened for signature August 12, 1949, entered into force October 21, 1950), art 42. But see Jinks, “International Human Rights Law in Time of Armed Conflict,” 666–7.
(44) See Accordance with International Law of the Unilateral Declaration of Independence in Respect of Kosovo, Advisory Opinion, 2010 ICJ Rep 403, paras 8–9 (declaration of Judge Simma) (criticizing the view that “everything which is not expressly prohibited carries with it the same colour of legality; [this view] ignores the possible degrees of non-prohibition, ranging from ‘tolerated’ to ‘permissible’ to ‘desirable’ … That an act might be ‘tolerated’ would not necessarily mean that it is ‘legal,’ but rather that it is ‘not illegal’ ”).
(45) John Westlake, II International Law (1907) 52.
(46) Joseph Raz, Practical Reason and Norms 86 (OUP 1999).
(47) See SS “Lotus” (France v Turkey) (Judgment)  ICGJ 248.
(48) See, eg, Protocol I art 57(5) (“No provision of this Article may be construed as authorizing any attacks against the civilian population, civilians or civilian objects”).
(49) Protocol I art 48.
(50) US Department of Defense, Law of War Manual 220.127.116.11.
(51) Instructions for the Government of Armies of the United States in the Field, General Order No 100, art. 14 (April 24, 1863).
(52) Instructions for the Government of Armies of the United States in the Field, General Order No 100, art 15.
(53) Instructions for the Government of Armies of the United States in the Field, General Order No 100, art 16.
(54) Instructions for the Government of Armies of the United States in the Field, General Order No 100, arts 44, 56, 65, 70.
(55) US v List (American Military Tribunal, Nuremberg, 1948), 11 NMT 1230, at 1253.
(56) US v List (American Military Tribunal, Nuremberg, 1948), at 1247, 1252, 1256. See also at 1236 (“acts done in time of war … cannot involve any criminal liability … if the acts are not prohibited by the conventional or customary rules of war”).
(57) The Tribunal did not mistake the absence of a legal prohibition for an affirmative authorization or endorsement. The Tribunal condemned reprisals against civilian hostages as “a barbarous relic of ancient times” and bemoaned the “complete failure on the part of the nations of the world to limit or mitigate the practice by conventional rule” (at 1249, 1251).
(58) US v List (American Military Tribunal, Nuremberg, 1948), at 1296.
(59) US v List (American Military Tribunal, Nuremberg, 1948), at 1256.
(60) See, eg, Protocol I, arts 34(4)(b), 63(5), 70(3)(c).
(61) US v List (American Military Tribunal, Nuremberg, 1948), at 1296.
(62) Note that the Lieber Code long predates the legal prohibition of aggression. See General Order No 100, art. 67 (“The law of nations allows every sovereign government to make war upon another sovereign state”).
(63) I owe this crisp formulation to Dapo Akande.
(64) See, eg, Geneva Convention (I) for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field (opened for signature August 12, 1949, entered into force October 21, 1950), art 3; Protocol I, art 4.
(65) Protocol Additional to the Geneva Conventions of August 12, 1949, and relating to the Protection of Victims of Non-International Armed Conflicts (Protocol II) (adopted June 8, 1977, entered into force December 7, 1978) 1125 UNTS 609, preamble.
(66) See, eg, International Covenant on Civil and Political Rights (ICCPR) (opened for signature December 19, 1966, entered into force March 23, 1976) 999 UNTS 171, art 6(1). Note that an act may be otherwise lawful yet arbitrary. See ICCPR, art 17(1) (“No one shall be subjected to arbitrary or unlawful interference with his privacy, family, home or correspondence”).
(67) Juan Carlos Abella v Argentina (Case 11.137) Report No 55/97 [18 Nov 1997] OEA/Ser L/V/II.98, paras 159–60 (“It is, moreover, during situations of internal armed conflict that these two branches of international law most converge and reinforce each other”).
(68) Protocol II, preamble.
(69) Michael Bothe, Karl Josef Partsch, and Waldemar A Solf, New Rules for Victims of Armed Conflicts: Commentary on the Two 1977 Protocols Additional to the Geneva Conventions of 1949 (Martinus Nijhoff 1982) 636.
(70) Legality of the Threat or Use of Nuclear Weapons, Advisory Opinion, 1996 ICJ Rep 226, para 25.
(71) Legality of the Threat or Use of Nuclear Weapons, Advisory Opinion, para 78.
(72) Protocol I, art 57(3).
(73) US Department of Defense, Law of War Manual (2015) 5.11.5.
(74) See Adil Ahmad Haque, “Off Target: Selection, Precaution, and Proportionality in the DoD Manual” (2016) 92 International Law Studies 31.
(75) This assumption is probably correct, but I do not wish to argue for it here. A very limited version of the precautions rule is found in Hague Convention (IX) concerning Bombardment by Naval Forces in Time of War (opened for signature October 18, 1907, entered into force January 26, 1910), art 2.
(76) See, eg, UK Ministry of Defense, Law of Armed Conflict Manual (OUP 2005) 1.8 (one purpose of the law of armed conflict is “to safeguard the fundamental human rights of persons who are not, or are no longer, taking part in the conflict … and of civilians”).
(77) The European Convention on Human Rights prohibits intentional deprivation of life, with narrow exceptions, but permits measures derogating from that obligation “in respect of deaths resulting from lawful acts of war.” Convention for the Protection of Human Rights and Fundamental Freedoms, 213 UNTS 222, entered into force September 3, 1953, art 15. Importantly, such measures are permitted only “to the extent strictly required by the exigencies of the situation” (art 15). On my view, a court applying the Convention may find that intentional killings that conform to the law of war nevertheless violate the Convention because the measures taken were not strictly necessary. For example, a court may find that measures derogating from the Convention are strictly required only in certain parts of a state’s territory, or only in certain situations. In other areas or situations, the Convention may very well apply with full force.
(80) Declaration Renouncing the Use, in Time of War, of Explosive Projectiles Under 400 Grammes Weight [St Petersburg Declaration], Nov 29/Dec 11, 1868, 138 Consol TS 297, 18 Martens Nouveau Recueil (ser 1) 474, preamble.
(81) Hague Convention (IV) Respecting the Laws and Customs of War on Land (opened for signature October 18, 1907, entered into force January 26, 1910) 3 Martens Nouveau Recueil (ser 3) 461, preamble.
(82) Janina Dill & Henry Shue, “Limiting the Killing in War: Military Necessity and the St. Petersburg Assumption,” (2012) 26 Ethics & International Affairs 311, 319.
(84) Interestingly, Dill and Shue write that “proportionality … prohibits conduct that, even though it might be necessary, can be expected to cause unintended but foreseeable civilian damage that is excessive” (“Limiting the Killing in War,” 320). Presumably they believe that attacks that are tactically necessary to achieve concrete and direct military advantages are seldom strategically necessary to defeat opposing armed forces. Otherwise, on their assumptions, jus in bello proportionality would fail to constrain the conduct of parties determined to win.
(87) See also Seth Lazar, Sparing Civilians (OUP 2016).
(88) Janina Dill, Legitimate Targets (CUP 2015) 262–3.
(90) David Luban, “Human Rights Thinking and the Laws of War,” in Jens David Ohlin (ed), Theoretical Boundaries of Human Rights and Armed Conflict (OUP 2016) 52.
(91) St Petersburg Declaration, preamble.
(94) Joseph Raz, The Morality of Freedom (OUP 1986).
(95) Joseph Raz, Between Authority and Interpretation (OUP 2009) 136–7. Raz also describes an independence condition according to which authority is not legitimate if it encroaches on a domain in which the value of your actions depends on your reasons for acting. For example, the value of proposing marriage depends on your subjective judgment that your partner is right for you. In contrast, the value of avoiding unjustified harm does not depend on your subjective judgment that the harm is unjustified. It is much more important that combatants avoid inflicting morally unjustified harm than that combatants decide for themselves which harms are morally unjustified.
(97) See Derek Parfit, On What Matters, vol 1 (OUP 2011) 151.
(99) Lazar is therefore incorrect when he writes that “if other moral reasons could justify disobedience to the law, then we would have to consult those reasons in any situation to determine whether it is exceptional” (“Morality and Law of War,” 368). In morally unclear cases, we should not trust own judgment that a given situation is exceptional.
(100) Daniel Muñoz-Rojas & Jean-Jacques Frésard, “The Roots of Behaviour in War: Understanding and Preventing IHL Violations” (2004) 86 International Review of the Red Cross 189, 203.
(101) See, eg, Jeff McMahan, “Innocence, Self-Defense, and Killing in War” (1994) 2 Journal of Political Philosophy 193, 218.
(102) cf Donald H. Regan, “Authority and Value: Reflections on Raz’s Morality of Freedom” (1989) 62 Southern California Law Review 1003, 1007.
(103) See, eg, Vattel, Law of Nations §191.
(105) Vienna Convention on the Law of Treaties (opened for signature May 23, 1969, entered into force January 27, 1980) 1155 UNTS 331, art 31. Since the customary law that I will discuss in this book tracks treaty law, particularly Protocol I, I will only discuss treaty interpretation here. For a view similar to mine regarding the identification and interpretation of customary international law, see John Tasioulas, “Custom, Jus Cogens, and Human Rights,” in Curtis Bradley (ed), Custom’s Future: International Law in a Changing World (CUP 2016).
(106) Vienna Convention art 32.
(107) See, eg, Raz, Between Authority and Interpretation, ch 11.
(108) Protocol I Preamble.
(109) UK Ministry of Defense, Law of Armed Conflict Manual (OUP 2005) 1.8. Note that the first clause refers to protecting each individual combatant and non-combatant from unnecessary suffering, not to minimizing aggregate suffering.
(110) David Luban, “Military Necessity and the Cultures of Military Law” (2013) 26 Leiden Journal of International Law 315, 323.
(111) Prosecutor v Kupreškić (Judgment) IT-95-16-T (Jan 14, 2000), para 525.
(112) Michael N Schmitt, “Military Necessity and Humanity in International Humanitarian Law: Preserving the Delicate Balance” (2010) 50 Virginia Journal of International Law 795, 801.
(114) United States Manual for Courts-Martial §916(c) (2012).
(115) United States Manual for Courts-Martial §916(c) Discussion.
(116) United States Manual for Courts-Martial §916(d) Discussion (in contrast, “An act performed pursuant to an unlawful order is excused unless the accused knew it to be unlawful or a person of ordinary sense and understanding would have known it to be unlawful”).
(117) For a recent US case in which an individual fighting for a foreign armed force asserted a justification defense, which the court denied, see United States v Hamidullin, 114 F Supp 3d 365, 382 (ED Va 2015).
(118) As McMahan observes, all wars are unjust on at least one side, and some wars are unjust on all sides.
(121) See, eg, United States Manual for Courts-Martial §14(c)(2)(a)(i) (“An order requiring the performance of a military duty or act may be inferred to be lawful and it is disobeyed at the peril of the subordinate”).