3 - Towards a Moral Division of Labour between IHL and IHRL during the Conduct of Hostilities

Summary

When should international humanitarian law (IHL) apply? When should it prevail over, and when should it give way to international human rights law (IHRL) in regulating the conduct of hostilities during international and non-international armed conflicts? IHL and IHRL give diverging answers to the crucial question of when it is legally permissible to kill another person. Following the customary IHL principles of distinction, proportionality and necessity systematically leads to breaches of the legal provisions safeguarding the human right to life. Some legal scholars, notably Helen Duffy in this volume, do not acknowledge this norm conflict, but aver that the two bodies of law can be reconciled through interpretation. Those that reject a substantive convergence between IHL and IHRL tend to take one of three broad positions: many argue that the norm conflict can be resolved by reference to lex specialis; others suggest that, in each instance, the rule should prevail that affords greater protection; yet others cast the matter as depending on which rules states intended to apply in a given context.

I. Introduction

When should international humanitarian law (IHL) apply? When should it prevail over, and when should it give way to international human rights law (IHRL) in regulating the conduct of hostilities during international and non-international armed conflicts?¹ IHL and IHRL give diverging answers to the crucial question of when it is legally permissible to kill another person. Following the customary IHL principles of distinction, proportionality and necessity systematically leads to breaches of the legal provisions safeguarding the human right to life.² Some legal scholars, notably Helen Duffy in this volume, do not acknowledge this norm conflict, but aver that the two bodies of law can be reconciled through interpretation.³ Those that reject a substantive convergence between IHL and IHRL tend to take one of three broad positions: many argue that the norm conflict can be resolved by reference to lex specialis;⁴ others suggest that, in each instance, the rule should prevail that affords greater protection;⁵ yet others cast the matter as depending on which rules States intended to apply in a given context.⁶

This chapter, in contrast, treats the question of when IHL should prevail over IHRL as a moral question. That means a moral standard, based on a general theory of the moral purpose of law explained in Section III.A, should determine the applicability scope of a body of law. Specifically, I will argue that when two bodies of law make diverging substantive demands – as IHRL and IHL do – that which better discharges the law’s moral tasks should displace the other. The law’s two moral tasks, according to this theory, are to guide its subjects’ conduct, as often as possible, towards the course of action that conforms to their moral obligations (task one) and to secure the fullest feasible protection of rights in the outcome of conduct (task two). IHL and IHRL’s scope of application should be determined by their respective ability to guide individuals towards the fulfilment of their moral obligations and to secure morally desirable outcomes in war.

Both, our moral obligations on the battlefield and the morally desirable outcomes of warfare, centre on fighting without violating the rights of the persons against whom we fight.⁷ There are two scenarios in which it can be morally justified to infringe an individual’s right to life without violating it. First, if an individual A threatens an individual B with morally unjustified harm and B can only defend himself by harming A, then A has forfeited her moral right not to be intentionally harmed through non-performance of her duty not to harm B.⁸ A has made herself morally liable to B’s defensive harming. Individuals hence forfeit their moral right to life when they responsibly contribute to an unjustified threat and killing them is a necessary and proportionate defensive response. Secondly, besides killing individuals who have forfeited their right to life through their own conduct, it can sometimes be morally justified to override innocent bystanders’ moral right to life in order to prevent a greater moral evil, namely, a greater number of unjustified rights violations.⁹

During the conduct of hostilities, law then has the following two moral tasks: its first moral task is to guide soldiers towards directing fire only against individuals who have forfeited their individual moral right to life.¹⁰ If innocent bystanders are expected to be harmed in an attack, law must only permit the attack if overriding their moral right to life is a necessary and proportionate side-effect of the attack’s contribution to achieving a morally just war aim.¹¹ The law’s second moral task is to avoid and reduce as much as possible all morally unjustified infringements (i.e., violations) of individual rights in war. I will show that, if we compare IHL and IHRL, IHL’s provisions governing the conduct of hostilities further diverge from the principles setting out when it can be morally justified to infringe individual rights.¹² IHL permits a wider range of conduct that amounts to individual rights violations than IHRL. This makes IHRL the prima facie morally better law for governing hostilities.

However, even though IHRL more faithfully reflects fundamental moral principles, it is not necessarily better at discharging the law’s two moral tasks during armed conflict. In situations in which the morally right course of action is systematically difficult to discern (epistemic barriers), a legal rule that diverges from moral prescriptions may be better than one that mirrors these prescriptions at guiding the individual towards what is typically the morally right course of action (task one). In such situations, a rule that does not simply repeat moral principles may also secure a better protection of individual rights in the outcome of conduct (task two). Epistemic barriers to discerning the morally right conduct affect a rule’s ability to discharge both of law’s moral tasks. In addition, if in certain situations individuals cannot be moved to fully conform to their moral obligations (incurable volitional defects), a legal rule that asks for conduct other than the morally right course of action may generate better moral outcomes (task two). Incurable volitional defects affect a rule’s ability to discharge the law’s second moral task.

Whether IHRL or IHL should prevail depends on the epistemic and volitional context of decision-making during the conduct of hostilities. In other words, the empirical reality of armed conflict shapes the morally ideal scope of application of IHL and IHRL. I raise the question of whether IHL or IHRL should govern the conduct of hostilities for six types of non-international and international armed conflicts (NIACs and IACs). One of two characteristics distinguishes them from confrontations that do not count as armed conflicts: either the intensity of hostilities or a State’s (non-authorised) use of armed force outside its own territory. Armed confrontations count as law enforcement operations rather than armed conflicts if they are neither protracted, meaning hostilities remain below a threshold of intensity discussed in Section V.B, nor involve a State’s use of unauthorised force outside its own territory.¹³ If hostilities become protracted, but do not involve the unauthorised extraterritorial use of force by a State, they count as NIACs (types 1, 3 and 5a). If hostilities involve a State’s armed forces crossing international borders without the authorisation of the territorial State, they currently count as IACs, whether they are protracted or not (types 2, 4, 5b and 6).

I will argue that when hostilities become protracted or cross international borders, IHRL needs to be applied ‘symmetrically’. By this I mean that both sides should interpret their obligations under IHRL as if they faced unlawful threats and uses of violence from soldiers on the other side and as if they themselves had lawful aims, regardless of the legal status of their respective resorts to force. I will show in Section V.C that international law on the resort to force does not track moral principles. As a result, if we relied on general international law on the resort to force to determine the lawfulness of soldiers’ aims during organised armed violence that crosses international borders, the ius contra bellum would often empower the side without a morally just war aim and hamstring the side fighting for a just cause. A State’s armed forces crossing borders without the territorial State’s consent does not otherwise affect ‘symmetrical IHRL’s’ better ability to discharge the law’s two moral tasks. An increase in the intensity of hostilities, in contrast, raises the epistemic barriers to identifying the morally right course of action and it renders more acute incurable volitional defects in soldiers’ decision-making. Compared with IHL, ‘symmetrical IHRL’ nonetheless remains the better law for discharging the law’s first moral task. Its ability to discharge the law’s second moral task, however, declines as hostilities become protracted.

This analysis leads to the following proposal for a moral division of labour between IHRL and IHL: as the morally prima facie better law IHRL should govern the conduct of hostilities in law enforcement operations. In non-protracted IACs, ‘symmetrical IHRL’ should govern the permissibility of killing. Above the threshold of intensity at which hostilities count as protracted, hence during NIACs and protracted IACs we face a choice between affording individuals a guide towards what is typically the course of action that conforms to their moral obligations (task one) and reducing individual rights violations in the outcome of warfare (task two). ‘Symmetrical IHRL’ better discharges task one; IHL better discharges task two. It depends on the relative moral costs of prioritising one task over the other as to which body of law should prevail. I will argue that IHL currently offers a better, but far from morally ideal, law for governing the permissibility of killing during the conduct of hostilities, once these hostilities reach the crucial threshold of being protracted. IHL should therefore displace IHRL and govern, on its own, the conduct of hostilities during NIACs and protracted IACs.

An alternative way of thinking about the implications of this argument would be to assert that only protracted armed confrontations should count as armed conflicts in the first place. This would allow us to uphold the traditional understanding that all armed conflicts should be governed by IHL and all law enforcement operations by IHRL, but it would mean redefining IACs as only confrontations among States that are protracted. Instead, I accept the dominant classification of armed confrontations as IACs if they involve the non-consensual crossing of borders by a State’s armed forces even if hostilities are not protracted. The implication of the argument for a moral division of labour between IHL and IHRL is then that not all armed confrontations that count as IACs should be governed by IHL. Only those that are protracted should fall under its purview; non-protracted IACs should be governed by ‘symmetrical IHRL’. In turn, IACs that are protracted and all NIACs (which per definitionem are protracted) should be governed only by IHL. This chapter hence presents a moral argument for the displacement of IHRL during certain types of armed confrontations, rejecting the intellectual coherence of a convergence between IHRL and IHL and the moral desirability of a parallel application of both bodies of law.

The argument proceeds as follows. Section II outlines the substantive conflict between the two bodies of law regarding the conditions of permissible killing. Section III defines the law’s two moral tasks and thereby sets the moral standard that later determines which body of law should prevail. It then explores the extent to which IHRL and IHL, respectively, track moral principles about the permissibility of killing. Section IV offers a typology of armed conflicts, highlighting what characterises six different types of confrontations as armed conflicts. Section V systematises how these characteristics affect IHRL and IHL’s respective ability to discharge the law’s two moral tasks. Based on this, I propose a division of labour between the two bodies of law. The concluding section takes stock of the divergence between the current applicability scope of IHL and this morally better division of labour with IHRL.

II. The Human Right to Life and the Permissibility of Killing according to IHL

A. IHL and the Rights of Individuals in War

As argued by Helen Duffy in this volume, it is largely uncontroversial now that IHRL does not simply cease to apply during armed conflict, even if the conflict extends beyond the territory of a State.¹⁴ Indeed, the co-applicability of IHRL and IHL during armed conflict has become the ‘new orthodoxy’.¹⁵ Co-applicability raises two logically distinct questions.¹⁶ The first is whether the two bodies of law create a substantive conflict, in the sense that following one would systematically lead to or imply a breach of the other. Only if we answer this first question with yes, does the second question even arise: which body of law, or more specifically, which legal rule prevails?¹⁷ One of the most frequently quoted authoritative statements on the relationship between IHL and IHRL conflates these two questions. The ICJ holds that ‘[i]n 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, falls to be determined by the applicable lex specialis, namely the law applicable in armed conflict …’¹⁸

The ICJ’s statement suggests that IHL elucidates the meaning of IHRL in times of war, which would imply that following IHL does not lead to conduct that amounts to a breach of IHRL. At the same time, the statement refers to IHL as lex specialis, a tool for resolving a conflict between laws.¹⁹ In this section, I reject the ICJ’s position that IHL elucidates what it means to protect human rights in war.²⁰ I argue instead that IHL and IHRL give diverging answers to the crucial question of when and whom it is permissible to kill during the conduct of hostilities. Complying with IHL, namely, with its customary principles of distinction, necessity and proportionality, leads to conduct that will often violate IHRL. Attacks that are lawful under IHL will regularly deprive individuals of the legal right to life that they hold under IHRL.

Before we compare the substance of the protections afforded to the individual under IHL and IHRL respectively, we need to address the formal question of whether, like IHRL, IHL bestows rights directly onto the individual at all. This remains controversial because IHL, unlike IHRL, does not afford individuals standing before an international court or tribunal tasked with adjudicating violations of IHL.²¹ Crucially, the rise to prominence of international criminal law has largely settled any dispute over whether individuals incur duties directly by virtue of IHL. IHL imposes obligations on both the belligerent and the individual combatant.²² A body of law that looks past the State long enough to bestow duties on the individual certainly recognises the latter as a subject for its purposes and by implication an agent capable, in principle, of bearing rights as well.

A textual interpretation of the Geneva Conventions supports the position that IHL indeed bestows some rights directly onto the individual. The four Geneva Conventions feature a common provision that prohibits special agreements among belligerents that would ‘restrict the rights which [the Conventions] confer … upon’ protected persons.²³ All Conventions enjoin protected persons from renouncing ‘the rights secured to them’.²⁴ GCIII and GCIV, which are dedicated to the protection of prisoners and civilians, respectively, use the term ‘right’ pervasively and contain a long list of procedural²⁵ and substantive entitlements.²⁶ This leaves no room for doubt that, once in the power of an enemy belligerent, an individual, whether a civilian or combatant, bears legal rights. At the same time, it is arguably on the battlefield that individuals’ fundamental legal rights are most directly threatened.

The First Additional Protocol, the most elaborate and recent legal regime for the conduct of hostilities, recognises a right of combatants to participate in hostilities.²⁷ However, this cannot be understood as a claim right in the Hohfeldian sense because no other subject – a combatant’s own State, the enemy belligerent or other individuals – has a duty to not impede a combatant’s participation in war. Most recently, Adil Haque has convincingly argued that this ‘right’ is therefore best thought of as being an immunity from being prosecuted for engaging in hostile actions.²⁸ Otherwise, the chapters concerned with battlefield conduct are silent on the matter of individual rights either with reference to combatants or civilians. Juxtaposed with several references to the rights of individuals who are under the control of an adverse party to the conflict in the Protocol,²⁹ this omission of individual claim rights in the chapters on the conduct of hostilities has an air of purposefulness.

Of course, many of IHL’s restrictions on the conduct of hostilities evidently benefit individuals. If the substantive protections IHL affords soldiers and civilians were equivalent to those that individuals can claim as rights under IHRL, the formal question of whether or not IHL conceives of these protections as individual claim rights would be less important.³⁰ However, the observed disjuncture between provisions that regulate behaviour beyond the battlefield and the prescriptions concerned with the conduct of hostilities re-emerges when we inquire into the substance of IHL’s principles. The Protocol reiterates many of the substantive protections that the Conventions confer on individuals who are in the hands of the enemy. In contrast, the next two sub-sections will show that, during hostilities, neither civilians nor soldiers enjoy protections by virtue of IHL that safeguard the human right to life that they hold under IHRL.

B. IHL and Civilians’ Human Right to Life during Hostilities

During the conduct of hostilities civilians are generally immune from deliberate harming.³¹ They are legitimate targets of intentional attack only for such time as they directly participate in hostilities.³² This might suggest that civilians retain their right to life at least until they decide to directly contribute to overcoming the enemy. However, according to the principle of proportionality, it is permissible under IHL to injure or kill civilians as a foreseeable side-effect of an attack against a military objective, if the expected ‘incidental harm’ is not excessive in relation to the military advantage that is anticipated to arise from the attack.³³ Can killing civilians in accordance with the principle of proportionality amount to an unlawful deprivation of the right to life that they hold under IHRL? The following paragraphs show that IHL-compliant proportionate incidental civilian harm is not systematically covered by any of the exceptions to the right to life that IHRL recognises.

One exception to the prohibition on depriving individuals of their human right to life, contained in Article 2(2)(c) ECHR, permits the killing of individuals who threaten others with ‘unlawful violence’ or who resist or flee from ‘a lawful arrest’. The use of lethal force for the purposes of law enforcement is also widely recognised to give rise to an exception to Article 6 ICCPR. Depriving someone of their right to life is non-arbitrary, if it is necessary for the purposes of ‘self-defence or the defence of others against the imminent threat of death or serious injury; to prevent a particularly serious crime involving grave threat to life; to arrest a person presenting such danger and resisting their authority; or to prevent his or her escape’.³⁴ Crucially, these exceptions permit killing individuals who through their own unlawful conduct pose a threat to others. Civilians in war cannot be assumed to have broken the law, whether it is IHL, IHRL or domestic law. These exceptions cannot therefore explain why IHL-compliant incidental civilian harm would systematically be permissible under IHRL.

That civilians have not necessarily broken any laws also means that capital punishment, another recognised exception to the prohibition on depriving individuals of their right to life, cannot explain why incidental civilian harm would comply with IHRL.³⁵ Of course, it would be ludicrous to bring killing in war in connection with capital punishment, but a closer look at this exception is useful to elucidate what it means that a deprivation of the right to life is ‘non-arbitrary’ for the purposes of the ICCPR, in particular. The conditions that make a death sentence non-arbitrary are all oriented towards the goal of giving the defendant his or her individual legal due.³⁶ The Commentary to the ICCPR demands that ‘the personal circumstances of the offender and the particular circumstances of the offence, including its specific attenuating elements must be considered’.³⁷ Incidental civilian harm in war, in contrast, is not individuated to the circumstances of the civilian that is killed.³⁸

Killing civilians as a foreseen side-effect of an attack against a carefully selected and vetted military target may not seem arbitrary in the sense of being random, senseless or purposeless. Moreover, killing civilians in accordance with the principle of proportionality obviously has a basis in IHL and thus in law. However, the most recent UN Human Rights Committee’s Commentary on Article 6 ICCPR emphasises that the lawfulness of a rights deprivation is not the sole determinant of whether it is arbitrary. Rather arbitrariness ‘must be interpreted … to include elements of inappropriateness, injustice, lack of predictability, and due process of law’.³⁹ Appropriateness is often in the eye of the beholder. The next section will argue in detail that IHL does not systematically take account of the moral status of civilians and thus of whether or not their incidental killing is morally justified. Here it warrants reiterating that vetting a target for compliance with IHL does not amount to a process aimed at giving the individual his or her legal due.

Indeed, what makes IHL-compliant incidental civilian harm irrevocably arbitrary for the purposes of IHRL is its unpredictability. Besides the intention of the attacker, the following factors account for the legality of a civilian being deprived of her or his life under IHL: his or her physical proximity to a military target; the military value that the target has at that moment in the attacker’s campaign; the blast radius of the weapon the attacker happens to have at their disposal; and the absence of other civilians, who could render the expected incidental harm excessive. With the exception of the civilian’s physical proximity to the target, these factors are entirely beyond his or her control.⁴⁰ From the point of view of the individual that is deprived of his or her life, IHL’s permission to kill him or her incidentally is arbitrary in that the individual’s legally sanctioned fate is entirely disconnected from – and it cannot be influenced by – his or her legally required conduct.

In addition to these exceptions to the human right to life that are rooted in the individual’s own conduct, the ECHR also recognises an exceptional permission to kill individuals who are not themselves legally liable to harming, if this is ‘absolutely necessary … for the purpose of quelling a riot or insurrection’.⁴¹ A legal permission to override rights that is unrelated to the conduct of the affected individual, but instead hinges on necessity is the most likely point of convergence between IHRL and IHL.⁴² After all, the First Additional Protocol likewise stipulates that incidental civilian harm has to be necessary. Does the necessity to override an individual right to life in accordance with Article 2(2)(c) ECHR align with the necessity to kill civilians in order to pursue an anticipated military advantage under Article 57(2)(ii) API? No, two differences between necessity in IHL and necessity in IHRL account for why this exception to the human right to life does not systematically cover incidental civilian harm that complies with IHL.

The first difference between IHL and IHRL’s necessity exceptions concerns the aim with regard to which necessity has to obtain. Under IHRL necessity confers an exceptional empowerment to override a human right in order to achieve a legitimate aim such as the protection of human life or the protection of public order.⁴³ In contrast, under IHL killing civilians has to be necessary for the pursuit of a military advantage. This is not an emergency measure, but part and parcel of waging war. Notably, IHL does not require that a given military advantage is necessary to win a war. In turn, killing civilians under IHL does not have to be necessary for victory. Even if IHL did demand that incidental civilian harm was necessary also for victory, civilian harm would not automatically be necessary for the achievement of a legally recognised aim as required by the ECHR. To the contrary, at least on one side in each IAC, incidental civilian harm would be permissible because it was necessary for the furtherance of an aim that defies general international law.⁴⁴

The second difference between the meaning of necessity in IHL and the ECHR concerns the epistemic threshold at which we may consider a course of action necessary. In situations in which a reasonable observer perceives an imminent threat to human life, both IHRL and IHL deem overriding a right ‘necessary’. Where no such threat exists, the ECHR comes closer than IHL to still taking necessity literally to mean ‘lastness’. Under the ECHR, the course of action that overrides an individual right to life has to be the only, in the sense of the mildest, available path to the achievement of the recognised aim.⁴⁵ Establishing necessity in this sense means taking steps to ascertain the infeasibility of milder measures and the absence of alternative courses of action.⁴⁶ Crucially, if these steps cannot with reasonable certainty determine that overriding individual rights is necessary, the State would not be permitted to proceed.⁴⁷ Necessity under the ECHR implies an absolute standard of minimum reasonable knowledge about the alternatives to and consequences of conduct that overrides human rights.

Under IHL, the epistemic threshold at which a commander may assert that an attack is without alternative is much lower. Article 57(3) API demands that ‘[w]hen a choice is possible between several military objectives for obtaining a similar military advantage’ the attacker shall select the objective that can be attacked with the least expected incidental harm. This is not usually interpreted to mean that an attacker has to actively search as far and wide as reasonably possible for alternative targets with better expected incidental harm prognoses.⁴⁸ An attacker’s required knowledge about the consequences of an attack hinges on the feasibility of acquiring this knowledge. An attacker has to ‘take all feasible precautions in the choice of means and methods of attack with a view to avoiding, and in any event to minimizing, incidental loss of civilian life’.⁴⁹ That may often mean that an attacker explores the possibility of further reducing expected civilian casualties. However, equally as often, few or no measures to explore civilian harm mitigation may be feasible.

Crucially, an attacker is permitted to launch an attack even if all feasible verification measures were insufficient to establish with reasonable certainty that the expected incidental civilian harm could not have been further reduced while still achieving the military advantage at stake.⁵⁰ Although doing ‘everything feasible’ appears to be a demanding legal standard, it makes the level of knowledge required under IHL contingent on the circumstances of an attack.⁵¹ It follows that complying with IHL does not vouchsafe that a reasonable observer (with reasonably sufficient knowledge about the alternatives and consequences of an attack) deems the expected incidental civilian deaths the only possible path to the achievement of a given military advantage. This, however, is the connection between expected casualties and a military advantage that IHRL would require if achieving a military advantage was equivalent to a lawful aim recognised under IHRL.⁵²

The jurisprudence of the European Court of Human Rights (ECtHR) corroborates the argument that incidental civilian harm permitted under IHL is not systematically covered by the ECHR’s permission to exceptionally override an innocent bystander’s right to life.⁵³ The Court has ostensibly drawn on IHL’s requirement to take precautions in attack when elucidating the meaning of a necessary deprivation of the right to life in contexts of large-scale organised violence.⁵⁴ At the same time, the Court has mostly stayed faithful to the strict necessity standard of IHRL.⁵⁵ It has regularly enquired into the legitimacy of the aim behind the State’s use of violence.⁵⁶ Moreover, judgments feature not only questions about the availability of milder means, but also about the care devoted to exploring the latter.⁵⁷ A close reading of Article 57 API leaves little room for doubt that the Court has thereby asked for more care towards persons not directly participating in hostilities than would be required under IHL.

C. IHL and Soldiers’ Human Right to Life during Hostilities⁵⁸

IHL permits intentional lethal attacks against combatants and members of organised armed groups who have a continuous combat function,⁵⁹ hence against persons who regularly participate in hostilities.⁶⁰ Could their participation in hostilities, even if it conformed to the principles of IHL, be considered unlawful so that their threat or use of violence might trigger an exception to their human right to life? Individuals fighting for an organised armed group against their own State are indeed likely in breach of their domestic legal obligations.⁶¹ In an internal NIAC, when the State attacks enemy fighters, it thus deprives individuals of their right to life who engage in violence that is unlawful under domestic law. If an intervening State is authorised by the territorial State to use force against members of an organised armed group, it may also often be true that the latter’s fighters have defied the territorial State’s laws by taking up arms against a third State. We cannot assume this to be the case if an intervening State wages war against an organised armed group without the territorial State’s consent.⁶² Furthermore, we have no reason to believe that combatants fighting for a State party to a conflict have broken the domestic law applicable to them.

When it comes to the lawfulness of the use of force by combatants, rather than relying on domestic law, we could draw on general international law (namely, the ius contra bellum) to establish whether and when IHL-conforming participation in hostilities amounts to an unlawful threat or use of violence for the purposes of IHRL. Namely, we could argue that combatants who contribute to a State’s use of force in contravention of Article 2(4) UN Charter and corresponding customary law are using violence unlawfully, even if it accords with IHL. In this reading, participation in a legal (defensive or mandated) IAC would prima facie leave intact combatants’ right to life under IHRL. At the same time, none of the violence threatened or used by combatants on the side of the aggressor State would be lawful for the purposes of IHRL, even if it complied with the principles of IHL. Participation in hostilities on the side of an aggressor State would thus open a combatant to the permissible deprivation of their human right to life.

Of course, IHL permits attacks against combatants on all sides in an IAC, regardless of the status of a belligerent’s resort to force. More importantly, it is highly problematic to suggest that general international law could trigger an exception to the prohibition on depriving individuals of their human right to life. This would amount to anchoring an individual’s legal status under IHRL in the conduct of her or his State, over which she or he has likely little or no control.⁶³ Nonetheless, the United Nations Committee on Human Rights, in its recent commentary on Article 6, has endorsed a view that links the lawfulness of States’ use of violence on the battlefield to the status of the resort to force under general international law. Though not addressing the implications of this stipulation for the status of individual combatants, the Committee has argued that ‘[s]tates parties engaged in acts of aggression as defined in international law […] violate ipso facto Article 6 of the Covenant’.⁶⁴

Even if a State’s breach of the ius contra bellum could in principle trigger a permission to deprive an individual of his or her right to life under IHRL, even combatants whose IHL-compliant participation in hostilities contributed to waging an aggressive war would not necessarily have lost their human right to life. Depriving an individual who threatens another with unlawful violence of their right to life has to be necessary under IHRL.⁶⁵ Lethal attack has to be the last resort. IHL’s permission to intentionally kill combatants at all times, except when they are hors de combat, on the other hand, is not tied to combatants’ in fact using violence or posing a threat. The same is widely deemed to be true for members of organised armed groups with a continuous combat function.⁶⁶ Lethal attack against them can be the intended and planned first resort. Recent scholarly claims that IHL requires that killing soldiers is necessary⁶⁷ have been roundly rejected.⁶⁸

IHRL’s exceptional permission to override an individual’s human right to life for the purposes of ‘quelling an insurrection or emergency’ or ‘establishing public order’ does not cover IHL’s permission to intentionally kill soldiers either. IHRL again requires necessity, as discussed in the previous sub-section. Although combatants can through their conduct gain immunity from attack, namely, by surrendering,⁶⁹ no particular conduct on their part is required for IHL to endorse their killing in the first place.⁷⁰ IHL’s provision that is most directly geared towards benefiting combatants who are not already hors de combat is the prohibition on employing ‘weapons, projectiles and material and methods of warfare of a nature to cause superfluous injury or unnecessary suffering’.⁷¹ This provision shines a glaring light on IHL’s endorsement of virtually all intentional attacks against soldiers; its endorsement of conduct that will often amount to a deprivation of the soldiers’ right to life in contravention of IHRL.

D. IHL’s Authorisation of Conduct that Amounts to a Human Rights Violation

Some scholars argue that we can avoid the problematic conclusion that IHL authorises violations of IHRL by claiming that IHL is only ‘prohibitory’ and does not confer any authorisations at all. In this reading, the principle of proportionality is merely the absence of a prohibition on proportionate incidental civilian harm, not a legal liberty to cause foreseen non-excessive harm to civilians. Moreover, in this view, rather than a legally privileged course of action, killing combatants is conduct that IHL has simply omitted to outlaw. This interpretation is surprising because the Geneva Conventions as well as the Protocols contain a number of demands for positive causal interventions into the world, for instance, taking all feasible measures to minimise incidental harm or issuing a warning before an attack. Conduct meant to meet these requirements is surely, by logical implication, imbued with the authority of a legal permission. Nonetheless, this interpretation of the principles of distinction, proportionality and necessity as mere prohibitions is frequently stated rather than defended.⁷² The two most convincing explanations ultimately both fall short.

Adil Haque maintains that IHL affords combatants immunity from prosecution for participation in hostilities. However, it cannot grant combatants an authorisation to participate in hostilities because it does not ‘authorize organized armed groups to kill State armed forces or to harm civilians’.⁷³ He holds that this lack of authorisation for organised armed groups is evident because members of organised armed groups are subject to domestic criminal prosecution for participation in hostilities. IHL cannot then authorise combatants fighting for States, lest the principle of equal applicability is in jeopardy.⁷⁴ This reasoning rests on four claims which are unlikely to be simultaneously true: first, members of (non-State) organised armed groups are not authorised to use force because they are not immune from prosecution for participation in hostilities; secondly, combatants are not authorised to use force because a divergence in formal authorisations between combatants and members of non-State armed groups would challenge belligerent equality; thirdly, even though they are not authorised to participate in hostilities, combatants are immune from prosecution for participation in hostilities; fourthly, this divergence in immunities between combatants and members of non-State armed groups does not challenge belligerent equality.

Let us assume that members of non-State organised armed groups are indeed neither immune from prosecution nor authorised to use force (claim one), but combatants are both. Only in NIACs would this asymmetry in authorisations create an inequality among belligerents (claim two). In IACs, claim two would not hold. IHL allows States to incorporate or recognise paramilitaries or irregular armed groups.⁷⁵ This, in turn, confers the immunities and authorisations associated with combatant status on their members.⁷⁶ Individuals who fight in an IAC as members of an organised armed group without the backing of a State party share their lack of immunity from prosecution, and, in Haque’s view, the attending lack of authorisation to use force, with civilians who temporarily directly participate in hostilities, spies and mercenaries. As this fate befalls these groups of individuals on any side of an IAC, it does not affect the equal applicability of IHL.⁷⁷

In NIACs, on the other hand, if members of the State’s armed forces were authorised to use force and a lack of immunity from prosecution meant that members of (non-State) organised armed groups were not authorised to use force, an inequality among belligerents would indeed follow. However, the equality IHL affords belligerents in wars that pit a State against a non-State actor has arguably always been, and necessarily remains, qualified.⁷⁸ This qualification is evidenced not least in the asymmetry in immunities that Haque affirms in claim three: he holds that IHL affords immunity for combatants’ participation in hostilities (claim three), but his claim that IHL does not authorise participation in hostilities rests on the observation that IHL does not afford such immunity for participation to members of non-State organised armed groups. If in NIACs an asymmetry in immunities does not undermine belligerent equality (claim four), why should an asymmetry in formal authorisations to use force have this effect?

Even if we insisted on full belligerent equality also in NIACs, it would still not follow that IHL does not authorise killing in NIACs or indeed in IACs. In NIACs it is plausible to contest that IHL provides immunity from prosecution even for combatants fighting on behalf of the State. Article 6(5) APII urges the authorities in power at the end of hostilities to grant amnesty to persons who have participated in hostilities. Note that the provision refers to the ‘authorities in power’ at the end of hostilities, not the High Contracting Party. The Protocol evidently envisages the possibility that the regime challenger prevails and prosecutes soldiers who started out as combatants fighting on behalf of the State. The treaty hence acknowledges that in NIACs, IHL can afford immunity from prosecution to neither side. This is consistent with the absence in APII of a provision spelling out a ‘right’ of the State’s combatants to participate in hostilities. If we think of lack of immunity from prosecution as indicative of a lack of authorisation to use force, we could hence conclude that APII indeed does not authorise either side to use violence. Alternatively, the Protocols’ plea for immunity from prosecution under domestic law may simply not be indicative of whether IHL authorises participation in hostilities at all. In either case, nothing much follows for our interpretation of API and IHL’s permissions in IACs. The difference between the two treaties – APII lacks most of the provisions on the conduct of hostilities that I argue authorise attacks – means that something that is true for APII does not by logical implication hold for API.⁷⁹

Derek Jinks offers a different argument, unrelated to belligerents’ power to prosecute individual soldiers, for why IHL does not authorise the attacks that do not defy it. He maintains that the Geneva Conventions cannot be ‘read like domestic statutes’ because they do not ‘authorize states to engage in practices otherwise forbidden in law’.⁸⁰ It is not at all obvious, however, why international law should only be able to authorise conduct as an exception to its own prior prohibition. This evokes a bilateralist international system, in which States start out as utterly unfettered sovereigns whose actions are deemed acceptable unless expressly prohibited by international law. This view radically challenges dominant understandings of the contemporary international legal order.⁸¹ Crucially, even if we subscribed to the view that all State conduct not prohibited by international law is permissible, we would not necessarily have to endorse the position that international law is only prohibitory. As a matter of logic, nothing would prevent international law from endowing with the legitimacy of a legal permission conduct that a State could otherwise have engaged in without either violating or actualising a legal rule.

Jinks’ contention that IHL does not concern practices otherwise forbidden under international law is all the more astonishing because he is a proponent of the co-applicability of IHL and IHRL, but denies their convergence.⁸² Jinks endorses the view that before conflicts of laws are resolved or a State derogates, during war, IHRL prohibits any and all deprivations of the right to life that do not meet any of its own recognised exceptions (for co-applicability). He also correctly deems ‘perverse’ the ICJ’s view that IHL elucidates the meaning of such exceptions and that all killing in war permitted, or in his view not prohibited, by IHL does not violate IHRL (against convergence). Logically, IHL’s principles for the conduct of hostilities then govern practices that are already heavily regulated by international law, namely, by IHRL. The conduct of hostilities is rife with ‘prior prohibitions’,⁸³ to which IHL could be deemed to afford exceptions.

Even if the argument that killing in compliance with IHL is not formally authorised by IHL were compelling, we would still face a legal situation in which IHRL prohibits conduct that IHL does not prohibit. A denial of IHL’s authorisation of violence does not at all answer the question as to which body of law an individual on the battlefield ought to turn to for guidance. Moreover, regardless of whether IHL formally authorises the conduct that it does not prohibit, we have to grapple with the considerable expressive force of the uncontested legal situation that IHL does not prohibit intentional killings of soldiers and foreseen side-effect deaths of civilians.⁸⁴ The traditional reference to ‘belligerent privilege’ suggests that, scholarly debate notwithstanding, military practitioners widely assume that IHL endorses the attacks that do not defy it; attacks that will often breach IHRL, depriving civilians and sometimes soldiers of the right to life they enjoy under IHRL.

III. The Moral Right to Life and the Legal Permissibility of Killing in War

A. The Law’s Moral Tasks in War

When two simultaneously applicable bodies of law make diverging demands, the morally better law should prevail. How do we decide which one is the morally better law? Largely side-stepping the extraordinarily complex question of how law conceptually relates to morality, I make two assumptions: first, the content of law can be determined without reference to morality. It is hence possible that law makes demands on its subjects that diverge from their moral duties. Secondly, individuals have a moral interest in the rule of law ‘because human life goes better when subjected to governance by a (conscientious) authority’.⁸⁵ Legality therefore provides a moral reason for action. This reason is not decisive. It may sometimes be morally impermissible to obey a wicked law. At the same time, a morally wrongful act can be less wrongful because it is required by law.⁸⁶ With these assumptions, I rely on Raz’ conceptualisation of law as ‘a kind of complex social practice [that] can be put to moral use, and that, where it exists … has moral tasks to discharge’.⁸⁷

What are the law’s tasks through the performance of which it can morally improve human life? In Raz’ view, a law exercises legitimate authority over its subjects if the latter better conform to the moral reasons that apply to them when they act according to the guidance of the law.⁸⁸ Law can provide a moral service to the individual by assisting him or her in adopting the morally right conduct. This is the law’s first moral task. Crucially, this ability to guide the individual towards fulfilment of her or his moral obligations does not exhaust the law’s ability to morally improve human life. Law can also serve the individual as a guarantor of her or his moral rights. Law can be of moral use to each individual by ensuring that the conduct of legal subjects overall better conforms to the moral goal of preserving individual rights. Guiding behaviour towards the fullest possible protection of individual rights is the law’s second moral task.⁸⁹

When there is a conflict between simultaneously applicable legal rules, we should allow those rules to prevail that better accomplish these two moral tasks. The first step in determining whether IHL or IHRL ought to govern the conduct of hostilities is then to establish what moral obligations apply to the individual on the battlefield and what battlefield conduct leads to the best possible outcome in terms of protecting individual rights.⁹⁰ Ordinarily we forfeit our moral right to life only when we responsibly contribute to an unjustified threat against another person.⁹¹ The defensive harm we make ourselves liable to by virtue of this conduct has to be proportionate and necessary to avert the unjustified threat. Moreover, it can sometimes be morally justified to override innocent bystanders’ moral right to life in order to prevent a greater moral evil.⁹² Crucially, the only moral evil that warrants overriding an innocent bystander’s right to life is a great number of unjustified individual rights infringements.⁹³ Unintended, but foreseen harm against non-liable individuals has to be necessary for and proportionate to the moral aim of preventing a greater number of individual rights violations.

During the conduct of hostilities, law fulfils its first moral task if it guides soldiers towards directing fire against individuals who are morally liable to defensive harm or against targets that harm only individuals whose killing can be justified as a necessary and proportionate side-effect of the achievement of a morally just war aim. The law’s second moral task is to reduce as much as possible unjustified infringements (i.e., violations) of individual rights in a war. Guiding agents towards the course of action that fulfils their moral duties and that thus avoids individual rights violations in each instance is normally the best way to avoid such violations overall. Both moral tasks therefore prima facie point towards the same morally ideal legal rules: international law that discharges its two moral tasks in war prohibits all killing that amounts to, or results in, morally unjustified infringements of the individual right to life.

Is the law that better discharges these moral tasks in war, then, the law that more closely restates these fundamental moral prescriptions about the permissibility of killing? No, if there are genuine epistemic barriers to establishing which course of action avoids unjustified infringements of individual rights on the battlefield, a law that simply mirrors the moral demand will fail to discharge the law’s first moral task. By comparison, a law that accommodates systematic epistemic constraints on decision-making and that is clear, accessible and action-guiding in the context in which it actually addresses individuals is more effective in helping them conform to their moral obligations.⁹⁴ In general, a law that guides its addressees through the choice they typically encounter is better at discharging its first moral task than a law that addresses the individual as if he or she were omniscient. The epistemic context of decision-making on the battlefield is hence relevant to the question of whether IHL or IHRL better discharges the law’s first moral task.

Epistemic constraints account for why even the best possible law for discharging its first moral task may not always point towards the morally right course of action. The harder it is to morally parse the situations in which a law typically addresses the individual – due to time pressure on action, moral complexity or uncertain consequences – the more we should expect the morally best law for discharging its first task to diverge from underlying moral principles.⁹⁵ The more we should also expect that law may have to stipulate what is usually the morally right course of action, but not necessarily the right course of action in all situations. Law discharges its first moral task if an individual guided by law generally better conforms to her or his moral obligations than an individual not guided by law. The closer and more often law-directed conduct fulfils the guided individuals’ moral obligations, the better the law is at discharging its first moral task. However, compliance with law may not vouchsafe that conduct in all situations is morally right.

That law is action-guiding and sensitive to the epistemic context in which its subjects actually operate is also crucial for law’s ability to discharge its second moral task of securing the fullest feasible protection for individual moral rights. In addition, there is another reason why a law might better discharge its second moral task if it diverges from fundamental moral precepts: volitional defects, which are incentive structures, emotional or cognitive biases that account for why agents systematically fail to follow their moral obligations.⁹⁶ Law can cure some volitional defects by attaching sanctions to certain courses of action that subjects are tempted to take instead of the morally right course of action. Alternatively, law can counter some volitional defects by making detailed demands on how to act that ‘would be unintelligible as elements in moral rules or principles’.⁹⁷ Both sanctions and detailed action guidance can improve a law’s ability to secure a better moral outcome in terms of protecting individual rights in the face of volitional defects.

However, law cannot cure all volitional defects. Sometimes, outlawing morally wrongful actions creates incentives for further wrongdoing and thus morally worse results.⁹⁸ At other times, a law that prescribes the morally right conduct would be highly costly to obey, all while non-compliance is hard to sanction.⁹⁹ Such a law would likely be ignored and miss altogether the opportunity to morally improve the effects of conduct for the protection of individual rights.¹⁰⁰ By comparison, a law that accommodates the volitional defects that it cannot cure, by demanding less or other than the morally required course of action, may be able to secure better moral outcomes for rights. The volitional context of decision-making during armed conflict hence matters for the question of whether IHL or IHRL better discharges the law’s second moral task.

Incurable volitional defects can create a tension between the morally better law for discharging the law’s first task and the morally better law for discharging the law’s second task. If law demanded conduct that typically points individuals towards the conduct that fulfils their moral obligations, considering epistemic uncertainty and the need for law to guide action across a range of situations (task one), but this demand would, due to incurable volitional defects, be ignored or create worse moral results, law would fall short in its second task. On the other hand, if law demanded less than the morally right conduct in order to accommodate an incurable volitional defect and thereby morally improve the results of conduct (task two), it would fall short in discharging its first moral task. Both moral tasks are equally important. When they imply divergent rules, what is the solution?

If the demands associated with discharging the law’s two tasks diverge, a law that satisfices rather than maximises its ability to discharge either of its moral tasks may be morally preferable. Specifically, what such a law ought to look like depends on answers to two questions: first, how great a divergence from the morally right conduct would law have to prescribe in order to produce the fullest feasible protection of individual rights; and, secondly, how far from the morally best possible outcomes for the protection of individual rights would the results of conduct be, if law did not accommodate incurable volitional defects, but insisted on prescribing the typically morally right course of action? The law should strike the least morally costly compromise possible between its two tasks. In such a situation, the morally best possible law is then neither the law that, as often as possible, prescribes the typically morally right course of action (task one), nor the law that as much as possible protects rights in the outcome of conduct (task two).

There are two reasons then why even a law that discharges both its moral tasks (i.e., first, individuals guided by law, compared with those not guided by law, more often meet their moral obligations and, second, individual rights are better protected than in the absence of guidance by law) might permit or even ask for conduct that is morally wrongful in a particular situation: first, epistemic constraints mean law prescribes the course of action that is typically, but not always, the morally right course of action; and, second, law prescribes conduct other than the morally right course of action in order to accommodate incurable volitional defects. In situations in which law, as a result, demands conduct other than the morally right conduct, should an individual obey the law?

It can indeed be permissible or morally required to disobey a law that discharges its two moral tasks. However, such cases should be rare. As mentioned, law that discharges its moral tasks, but diverges from fundamental moral prescriptions, does so primarily because there are epistemic obstacles to determining the morally right course of action or volitional barriers to individuals’ conforming to their obligations. Knowing this, individuals would need to have particularly compelling reasons to trust their own judgement over the law’s guidance. Moreover, when deciding whether to obey a legal rule, besides the question of whether (non-)compliance would instantiate their moral obligations, individuals ought to consider how (non-)compliance would affect the rule of law. As alluded to above, that conduct is demanded by law adds a moral reason to the balance of reasons. This is not indefeasible, but it is a weighty reason, particularly if the law generally discharges its moral tasks of guiding individuals towards the fulfilment of their duties and of protecting their rights.

To recapitulate, when two bodies of law make diverging demands – as IHRL and IHL do regarding the permissibility of killing – that which better discharges the law’s two moral tasks should prevail. The law’s moral tasks are to guide individuals’ actions as often as possible towards meeting their moral obligations of not violating individual rights (task one), and to secure the morally best possible outcomes for rights, meaning minimising unjustified infringements of individual rights overall (task two). Which rules better accomplish these two tasks depends on the epistemic and volitional context of decision-making on the battlefield.¹⁰¹ If there is a tension between the demands of the law’s two moral tasks due to incurable volitional defects, the morally better law strikes the morally less costly compromise between its two tasks. The next two sub-sections explore to what extent IHL and IHRL, respectively, diverge from fundamental moral prescriptions about the permissibility of killing. Sections IV and V then systematically map which body of law better performs the law’s moral tasks in the epistemic and volitional context of real-world armed conflicts.

B. IHRL and the Moral Right to Life

Both moral justifications for infringing individuals’ moral right to life, that is, forfeiture and avoidance of a greater evil, have echoes in IHRL. A person who threatens another with unlawful violence, and who may hence be deprived of their human right to life under Article 2(2)(a) ECHR, may also be morally liable to necessary and proportionate defensive harm due to the forfeiture of her or his moral right. If a person poses a threat by evading lawful detention or arrest, just as they may have forfeited their legal right to life in accordance with Article 2(2)(b) ECHR, they may have forfeited their moral right to life. In addition, Article 2(2)(c) ECHR permits overriding individual rights with a view to allowing a State to avert an insurrection or riot, something that may well be a greater moral evil.¹⁰²

Although these exceptions to the prohibition on depriving individuals of their legal right to life under IHRL have similar structures to moral justifications for killing, the scope of the moral and legal permissions does not necessarily always align. Not every evasion even of a lawful arrest or detention warrants a lethal attack from a moral point of view, even if such an attack is ‘absolutely necessary’ and proportionate to the threat as demanded by the ECHR.¹⁰³ After all, the threat posed by the fleeing individual may be morally justified.¹⁰⁴ By the same token, we cannot be sure that in every instance in which State agents kill individuals for the purpose of quelling a riot or insurrection in accordance with IHRL, individual rights are really sacrificed for the avoidance of a greater moral evil.¹⁰⁵ Whether an insurrection or riots amount to a greater moral evil depends on the harm to innocent bystanders that quashing the insurrection likely necessitates as well as on the likely implications of the insurrection for citizens’ fundamental moral rights.¹⁰⁶

By hinging its permissions on the lawfulness of the threat or the use of violence rather than on its moral justification, IHRL opens the door to a possible lack of congruence between morally and legally permissible killing. At the same time, the more legitimate the authority of a State is and the more a State’s laws in general fulfil their moral tasks, the more likely it is that there will nonetheless often be an alignment between legally permissible human rights deprivations and morally justified killing. The reason is that overriding individual rights for the sake of establishing public order is more likely to be morally justified if the State institutions are morally legitimate, protecting individual rights in the first place. In turn, a riot or insurrection is more likely to be morally unjustified if the State exercises legitimate authority. Similarly, the morally better a State’s laws are at discharging their two moral tasks, in general, the more likely it is that an unlawful threat of violence or the evasion of a lawful arrest is also morally unjustified.

Moreover, as IHRL hinges the permissibility of individual rights infringements on the lawfulness of the arrest, the unlawfulness of the threatened violence and the preservation of the State’s legal authority, it generally permits the deprivation of individual rights in furtherance of the rule of law. As mentioned above, the latter is itself a morally important goal in as much as law performs its dual service to the individual of guiding him or her towards his or her moral duties and protecting a person’s moral rights. Some infringements of individual rights that IHRL authorises, which would on the balance of reasons be morally unjustified, might end up being morally permissible, or at least less morally wrongful, if we account for the moral importance of upholding the rule of law. Nonetheless IHRL’s focus on unlawful rather than on unjustified threats means that divergences between legally and morally permissible killing cannot be ruled out.

This very focus on unlawful rather than on unjustified threats, however, may improve IHRL’s ability to discharge its first moral task. When making what is potentially a split-second decision about whether it is necessary to shoot a fleeing individual, a law-enforcement officer may have neither the requisite information nor the cognitive capacity to make a judgement about whether the fugitive’s threat is morally justified. That lawfulness, not moral justification, is the reference point for the legal permissibility of rights deprivations according to IHRL is one way in which this law lightens the individual agent’s ‘burden of judgement’.¹⁰⁷ Similarly, by replacing ‘avoiding a greater moral evil in terms of individual rights violations’ with the morally potentially over- and under-inclusive, but much more concrete goal of ‘establishing public order in an emergency’, the ECHR better discharges its moral task of guiding action towards what is, in a State that exercises legitimate authority, typically the course of action that conforms to an individual’s moral obligations.

Of course, not all divergences between moral principles and IHRL’s rules on the protection of the individual right to life will necessarily serve the law’s ability to discharge its first moral task and guide its subjects towards the morally right course of action all while accommodating epistemic constraints. After all, drafting a treaty, such as the ICCPR or the ECHR, is a political process, which gives expression to the interests of the drafters. IHRL, for instance, allows deprivations of the right to life in the execution of a lawful death sentence. A convicted criminal in custody rarely poses an imminent threat. Whether killing a person because of their past culpable wrongdoing, hence for reasons of retribution and possibly in the uncertain hope of deterrence, can ever be morally justified is much more contestable than the justifiability of necessary lethal harm to avoid a future wrong.¹⁰⁸ Nonetheless, both mentioned treaties permit it.¹⁰⁹

At the same time, the permissibility of capital punishment may serve the law’s ability to discharge its second moral task. As mentioned above, considering the incentive structure of its addressees is one condition for law’s ability to morally improve the outcome of conduct. This not only means managing volitional defects of the individual whose conduct law seeks to guide, for instance, law-enforcement officers or soldiers. International law, unlike most domestic laws, also needs to accommodate the incentives of the agents who decide whether a treaty is ratified. It bears noting, however, that a treaty’s divergence from underlying moral prescriptions that are necessary to ensure ratification ultimately only helps law discharge its second moral task if, the pragmatic divergence from moral precepts notwithstanding, guidance of law still results in morally better outcomes for the protection of individual rights than conduct not guided by law.

From a moral point of view there is a limit then to how far international law should accommodate incurable volitional defects in order to secure treaty ratification and hence the law’s applicability. If law prescribes conduct that is not the morally right conduct in order to avoid being ignored, this has costs in terms of the law’s ability to discharge its first moral task. If the course of conduct that law prescribes in order to be deemed applicable also leads to outcomes that are overall morally worse or no better than those of conduct unguided by law, then accommodating incurable volitional defects also undermines law’s discharging of its second moral task. In this case, law should prioritise its first moral task and risk being ignored or not ratified, thereby failing to discharge its second task.

C. IHL and the Moral Right to Life

Both moral justifications for infringing the individual moral right to life, forfeiture and avoidance of a greater evil, have echoes also in IHL. IHL’s rules governing the conduct of hostilities, however, diverge significantly further than IHRL’s from the moral principles that determine the moral permissibility of killing. The general lack of congruence between IHL’s rules and moral principles centred on the protection of individual rights is well appreciated among just war theorists.¹¹⁰ This section briefly discusses the three main features of IHL that account for this further divergence: first, the legal permission to intentionally kill all soldiers, regardless of necessity and proportionality; secondly, IHL’s failure to demand actual necessity when permitting the deaths of innocent bystanders/civilians; and, thirdly, IHL’s equal empowerment of all parties to a war, regardless of the moral or legal status of their aims.

First, according to IHL, all combatants are always legally permissible targets of attack; civilians are prima facie immune from intentional harm. There may be a systematic coincidence between having combatant status and having the skills and intention to pose a threat to the enemy.¹¹¹ Ultimately, however, being a combatant is an assigned status, which first and foremost results from membership in a belligerent State’s organised armed forces. Military cooks, mechanics and logistics personnel are all permissible targets of lethal attack.¹¹² In fact, neither the ability nor the inclination to inflict harm on the enemy is a condition for combatant status under IHL. Even combatants actually trained for combat may lack the skills, courage or motivation to pose a threat. Finally, combatants retain their status as legal targets of intentional attack when their location or occupation makes them decidedly non-threatening. The latter is also true for members of non-State armed groups with a continuous combat function. They remain permissible targets of intentional attack regardless of their actual conduct, intention or location until they opt out of their combat function.

Moreover, even if all soldiers indeed posed a threat to the enemy, they would not all be morally liable to being killed. To be morally responsible for a threat they pose, individuals either have to be aware or reasonably should be aware of the moral status of this threat. In reality, soldiers often systematically lack the information necessary to make the determination of whether their aim in war is just.¹¹³ Moreover, some soldiers may have been coerced into fighting either by a State, an organised armed group or by circumstances. Soldiers who are coerced or who act in the reasonable, but mistaken, belief that their use of violence is justified may therefore be excused, in which case their moral liability to defensive harm is in question.¹¹⁴ Finally, even soldiers who are fully morally responsible for an unjustified threat they pose, are not necessarily morally liable to lethal attack if it is possible to avert the threat they pose by milder means.¹¹⁵ Just as many soldiers, who are permissible targets of lethal attack under IHL, retain their legal right to life under IHRL, many will not have forfeited their moral right to life.¹¹⁶

It is not hard to find reasons for why IHL should not attempt to restate the conditions of moral liability to harm in lieu of the principle of distinction. If it did, rather than a blanket permission to kill all soldiers and an obligation to spare all civilians, IHL would have to dictate that attackers direct harm towards individuals who responsibly contribute to the threat posed by a belligerent, but only if lethal attack is a necessary and proportionate response to that contribution. Such a law might fail in its first moral task of guiding soldiers who generally lack the information necessary to determine the moral status of the person they face on the battlefield.¹¹⁷ Indeed, such a law might fail to fulfil its second moral task as well. In the supremely stressful environment of battle, combatants who are subject to dehumanising narratives about and physical threats from the enemy may well perceive any individual ‘on the other side’ as morally liable to harm. A law that explicitly endorsed killing individuals based on their moral status, might exacerbate rather than cure this volitional defect, leading to more rather than less unjustified killing.

The second divergence of IHL’s prescriptions from moral principles concerns unintentional but foreseeable harm to innocent bystanders. When moral principles demand that such incidental harm has to be necessary, much like IHRL, that means harm ought to be the only and the mildest available course of action for the achievement of an aim. As outlined in the previous section, IHL in contrast only requires that an attacker does not eschew a target with a more favourable ‘collateral damage’ prognosis ‘when a choice is possible’ between similar targets. IHL further demands that the attacker takes all ‘feasible’ steps to reduce expected incidental civilian harm. Even if an attacker caused incidental civilian harm in order to achieve a morally just aim, IHL’s epistemic standard for when an attacker may deem necessary an attack that kills civilians is too low to vouchsafe that killing civilians is morally permissible.¹¹⁸

Thirdly, as already indicated, both moral permissions – intentionally killing an individual in defence against a threat and foreseeably killing an innocent bystander on grounds of necessity – depend on the attacker’s pursuing a morally just aim. In war that means that even if all soldiers responsibly contributed to the threat posed by their belligerent State, from a moral point of view only those on the unjust side would prima facie be liable to defensive harm. In contrast, the equal applicability of IHL, specifically the ‘right to participate in hostilities’ of combatants on both sides in an IAC without regard to ‘the causes espoused by … or attributed to the parties to the conflict’,¹¹⁹ means that IHL also endorses killing in defence against a morally justified threat. It is also due to the equal applicability of IHL that IHL’s principle of proportionality regularly permits killing civilians, hence potentially innocent bystanders, in pursuit of an aim that is morally unjust.¹²⁰

Again, it is not difficult to find epistemic constraints that account for why IHL needs to bracket the moral status of the resort to force when prescribing conduct in war if it is to fulfil its first moral task. Determining whether or not a belligerent’s resort to force is morally justified, and hence whether an attack contributes to the pursuit of a morally just aim, is a complex, future-oriented judgement. It involves estimating whether waging war is the lesser moral evil in terms of unjustified individual rights infringements compared with not resorting to force. I have argued elsewhere in detail that this question in reality regularly amounts to an ‘epistemically cloaked forced choice’.¹²¹ Even if we could overcome the inherent difficulty of judging whether the resort to force will be a moral lesser evil, this judgement is unlikely to be sound without precise information about what is at stake when a State resorts to force and how a war is meant to unfold from a military point of view. This is information that States typically do not share with their soldiers.¹²²

Tying the permissibility of conduct in war to the moral status of a war’s aim, would not only mean that compliance with law would require that combatants answer a question to which they almost certainly do not know the answer, undermining the law’s ability to discharge its first moral task. A law that tied the legal permissibility of killing to whether or not an attack was launched in pursuit of a just aim would also be undermined in the fulfilment of its second moral task of reducing individual rights violations as much as possible. In an international system of sovereign and formally equal States, who would hold to account a State that resorted to force for an unjust aim, but allowed combatants to avail themselves of the legal empowerments that IHL reserves for just combatants? An alternative IHL which differentiated between belligerents depending on the justice of their aim would, at best, secure both sides’ compliance with the law for the just side. Worse even, such an ‘asymmetrical IHL’ might simply miss the opportunity to better protect individual rights on at least one side in each war.

If we compare all three codes of conduct – IHRL’s prescriptions, IHL’s demands and moral obligations – side by side, it becomes evident that IHL diverges further from moral precepts than IHRL along the three lines mentioned: first, individuals are morally liable to lethal attack only if they pose or contribute to an unjustified threat for which they are responsible, and the threat means a lethal attack is necessary and proportionate. IHRL demands that intentional lethal attacks are directed against individuals who use or threaten unlawful violence which makes the use of defensive lethal force necessary and proportionate. In opposition, IHL permits targeting soldiers based on their status as combatants or because they assume a continuous combat function, regardless of the necessity or proportionality of the attack.

Secondly, it may be morally permissible to unintentionally kill an innocent bystander if this is necessary for and proportionate to preventing a greater number of individual rights violations. IHRL likewise demands that unintentional killing of innocent bystanders is strictly necessary for and proportionate to the achievement of a legally recognised aim, generally involving the protection of human life. In contrast, IHL only demands that attackers do everything feasible to minimise expected incidental civilian harm. And if two attacks are expected to yield the same military advantage and a choice is possible, they should attack the target that is expected to cause less incidental civilian harm. Civilian harm needs to be proportionate only to the achievement of a military advantage, not the achievement of a war’s aim.

Thirdly, from a moral point of view, individuals only forfeit their right to life if they contribute to an unjust threat. Similarly, it is only ever morally permissible to kill innocent bystanders in pursuit of a just aim – only then can their deaths potentially be justified as the lesser moral evil. Under IHRL, law enforcement officials may use violence only during a lawful arrest and in defence against unlawful violence. By the same token, the State needs a legally recognised aim to override the right to life of innocent bystanders, for instance, quelling an insurrection or averting ‘danger to life or limb’.¹²³ IHL, in contrast, is symmetrical. That means IHL affords the same permissions to both sides in a war, regardless of the legal or moral status of the resort to force or the aims that belligerents pursue in a war.

The first section of this chapter established that IHRL and IHL give diverging answers to the question of when it is permissible to use lethal force. This section showed that IHL’s answer diverges further than IHRL’s from the answer that moral principles give to the question of whether and when it is permissible to kill a person. This makes IHRL prima facie better suited to discharging the law’s two moral tasks. However, whether IHRL is actually better at guiding the individual soldier towards the course of action on the battlefield that typically conforms to her or his moral obligations (task one) and whether it really is better at securing the protection of individual rights in war (task two) depends on the epistemic constraints and the volitional defects of decision-making during the actual conduct of hostilities. The next section will delineate the empirical phenomenon IHL is meant to govern, identifying the characteristics of armed conflicts that could affect epistemic barriers to and volitional defects in soldiers’ decision-making.

IV. Six Types of Armed Conflict

A. When is a Violent Confrontation an Armed Conflict?

The preceding section showed that IHRL more closely than IHL tracks moral principles regarding the permissibility of killing. Although this makes IHRL the prima facie morally better law, which body of law should prevail in a given context depends on which law better discharges the law’s moral tasks in that context. Traditionally, we think of the division of labour between IHRL and IHL as the former governing the use of force during law enforcement operations and the latter governing armed conflicts.¹²⁴ This section brackets the question of whether IHRL should indeed govern law enforcement operations. Given its status as the prima facie morally better law for governing permissible killing, I assume the answer to this question is yes. Instead, I seek to establish whether IHL should govern the conduct of hostilities during violent confrontations that are currently deemed to constitute armed conflicts.¹²⁵ In order to establish this, we need to analyse what characterises an armed conflict. Or, put differently, what distinguishes a confrontation that counts as an armed conflict from one that does not?

Historically wars were declared. In order for a confrontation to be considered a war, it had to be recognised as such by the warring States. This recognition, in turn, triggered the applicability of IHL.¹²⁶ The Hague and the Geneva Conventions of 1929, as a result, did not define war or armed conflict. The Geneva Conventions of 1949 radically break with this understanding of war as depending on recognition. According to Common Article 2, the Geneva Conventions shall ‘apply to all cases of declared war or of any other armed conflict which may arise between two or more of the High Contracting Parties’.¹²⁷ The Commentary stresses that ‘the determination of the existence of an armed conflict … must be based solely on the prevailing facts demonstrating the de facto existence of hostilities between the belligerents’.¹²⁸ But what are those prevailing facts?¹²⁹ What we can infer from the wording of the provision is no more and no less than that an armed conflict is any situation that involves the use of force between States.¹³⁰

Are any and all uses of force by States against States armed conflicts? In 1949, State-on-State violence not governed by IHL would have been beyond the purview of international law. Drafters therefore deemed it appropriate to conceive of the applicability of the Geneva Conventions in the widest possible terms. The Commentary describes it as ‘in conformity with the humanitarian purpose of the Conventions that there be no requirement of a specific level of intensity of violence to trigger an international armed conflict’.¹³¹ Scholarly opinion is in almost total agreement that ‘[i]t makes no difference how long the conflict lasts, how much slaughter takes place, or how numerous are the participating forces’.¹³² Neither is the existence of an armed conflict between States contingent on the purpose for which they use force.¹³³ The phenomenon that IHL is meant to govern then includes all situations in which one State uses armed force against another.

The identification of armed conflicts between a State and a non-State actor is by comparison more complicated. Even in 1949, the use of force by a State on its own territory would have been regulated by domestic law. An intensity threshold for the applicability of IHL to internal armed conflicts did therefore not create a legal black hole.¹³⁴ NIACs, first mentioned in Common Article 3 of the Geneva Conventions, are correspondingly widely deemed to be defined by a threshold of minimum intensity.¹³⁵ For the Second Additional Protocol of 1977 to apply to an armed confrontation the organised armed group has to be ‘under responsible command, exercise such control over a part of its territory as to enable them to carry out sustained and concerted military operations and to implement this Protocol’. Crucially, these criteria trigger the applicability of the treaty. They are not constitutive of an armed conflict because Common Article 3 is applicable to ‘armed conflict[s] not of an international character’, but it makes no such demands.

Where then is the threshold of minimum intensity for a NIAC more generally? One of the most widely reproduced concretisations of the required threshold of minimum intensity was articulated by the Appeals Chamber of the ICTY. It defines a NIAC as ‘protracted armed violence between governmental authorities and organised armed groups or between such groups within a State’.¹³⁶ Though widely accepted, this definition does not necessarily provide a definitive and incontestable test for each empirical case. Specifically, what criteria make armed violence ‘protracted’ is far from obvious.¹³⁷ Indicators drawn on in international jurisprudence include the frequency of hostile confrontations, the type and range of weapons and the calibre of munitions used, the number of persons participating in combat, wounded or killed as a result of hostilities, the severity and extent of the physical destruction and the number of displaced persons.¹³⁸

Assuming we can tell when these indicators of intensity point towards the existence of an armed conflict, is killing in all such situations really meant to be governed by IHL’s principles of distinction, proportionality and necessity? Common Article 3 does not itself concern the conduct of hostilities and not all NIACs also trigger the applicability of APII. The most recent authoritative commentary to Common Article 3, however, states that ‘when common Article 3 is applicable, other rules, especially those on the conduct of hostilities, with different restraints on the way force may be used compared to peacetime law, may also apply’.¹³⁹ This statement does not rule out that IHRL is simultaneously applicable during NIACs, but the contrast to ‘peacetime law’ is a gesture towards the traditional view that IHL’s rules for the conduct of hostilities displace otherwise applicable more stringent rules when armed violence meets the threshold of a NIAC.¹⁴⁰

In sum, IHL’s principles for the conduct of hostilities are meant to govern the permissibility of killing in all situations of armed force used between States and in all situations of ‘protracted armed violence between governmental authorities and organised armed groups or between such groups within a State’.¹⁴¹ It is, hence, either a particular configuration of belligerents, namely States on both sides, or the intensity of a confrontation between a State and non-State challenger on the former’s territory that defines the real-world armed confrontations that count as armed conflicts under the purview of IHL. The former characterises and differentiates an IAC, the latter a NIAC from armed confrontations presumed to be governed by IHRL.

B. When does Intensity Matter?

In recent years instances of armed violence that seem to fit neither the definition of IACs as violence purely between States, nor that of NIACs as protracted violence on the territory of one High Contracting Party have become more frequent. Instead, these conflicts cross international borders while also involving non-State organised armed groups. As we have already established the applicability of the rules for the conduct of hostilities to all NIACs and IACs, we might be tempted to bracket the contested question of how to classify armed confrontations that appear to have elements of both types. However, to delineate the universe of real-world confrontations that IHL is meant to govern, and to understand the epistemic and volitional context of decision-making in these situations, we have to know whether we need to enquire into the intensity of a particular armed confrontation to assert the applicability of IHL (NIAC) or not (IAC).

Beyond classic NIACs (1) and IACs (2), we can distinguish four types of armed confrontation that cross international borders while also involving non-State organised armed groups:¹⁴² (3) internal confrontations in which the territorial State is supported by a third State; (4) internal confrontations in which the non-State actor is supported by a third State; (5) internal confrontations that involve only one State, but spill over onto the territory of a neighbouring State; and (6) transnational confrontations between a State and a non-State actor entirely conducted on a third State’s territory, but without the consent of that territorial State. Which of these four types of armed confrontations are NIACs and which IACs? Or, put differently, which of these confrontations only fall under the purview of IHL if hostilities are ‘protracted’?

When another State or coalition of States intervenes in an internal confrontation on the side of the territorial State, we have an ‘internationally-supported internal confrontation’ (3). While the intervention lends the confrontation an international element, hostilities remain confined to one territory and there is no violence of one State against another, so that such internationally supported internal confrontations should be deemed NIACs. They fall under the purview of IHL only if they cross the threshold of being protracted.¹⁴³ The outside intervention on the side of the territorial State will often intensify hostilities that were previously below that threshold. Outside intervention can thus turn a situation from a law enforcement operation into a NIAC by intensifying hostilities, but this is not automatically the case and has to be separately established.

In contrast, if a State or coalition of States intervenes on the side of the non-State actor in an internal confrontation (4), the use of force between States ensues. Such an ‘internationalised internal confrontation’ is hence an IAC.¹⁴⁴ Even an internal confrontation that did not rise to the intensity of a NIAC before the intervention becomes an IAC once internationalised. Internationalised internal confrontations are therefore not subject to a requirement of minimum intensity in order to come under the purview of IHL. The difference between an ‘internationally supported internal confrontation’ and an ‘internationalised internal confrontation’ shows that it is not the involvement of more than one State in an armed confrontation per se that constitutes it as an IAC, but the crossing of borders by a State’s armed forces into another State’s territory without the latter’s consent or authorisation.

What about the use of force by a State on another State’s territory without the latter’s consent, but against a non-State actor rather than against the territorial State? Both ‘spill-over internal confrontations’ (5) and ‘transnational confrontations’ (6) raise this question. Dapo Akande has convincingly argued that in such situations, two armed conflicts exist in parallel: a NIAC between the State and the non-State actor; and an IAC between the intervening and the territorial State.¹⁴⁵ He has further stressed that it is practically impossible to establish whether an attack by the intervening State, meant to weaken the non-State actor but carried out on the territory of another State, is part of the NIAC or the IAC. Whether or not their intensity means that they meet the threshold of a NIAC, all attacks by a State against a non-State actor on another State’s territory are hence also part of an IAC.¹⁴⁶

In the case of a transnational confrontation (6), the cross-border scope of hostilities in effect moots the intensity requirement for the applicability of IHL. In the case of spill-over internal confrontations (5), attacks outside the State’s territory are likewise inevitably part of an IAC. However, this is not true for attacks carried out on the State’s own territory. We will have to separately establish that these internal hostilities cross the required threshold of intensity of being protracted to count as a NIAC. Spill-over internal confrontations might hence present the odd situation in which hostilities between the State and the organised armed group on the State’s own territory remain below the threshold of a NIAC, but their hostilities on another State’s territory are inextricably intertwined with and therefore part of an IAC, even if they are no more intense.¹⁴⁷

In sum, IHL is meant to govern the permissibility of killing in purely internal confrontations or classic NIACs with the involvement of no more than one State on that same State’s territory (1), internationally supported internal confrontations (3), and the internal part of a spill-over confrontation (5a), but only if hostilities cross a threshold of minimum intensity at which they count as ‘protracted’. In addition, IHL is meant to govern any armed force used by a State on the territory of another State, without the consent and not on behalf of the territorial State, regardless of its intensity. This category includes classic State-on-State IACs (2), internationalised internal confrontations (4), the part of an internal confrontation that spills-over into another State’s territory (5b), and transnational confrontations (6). It is thus one of two features that distinguish an armed confrontation that counts as an armed conflict under the purview of IHL from a law enforcement operation governed by IHRL: either the intensity of hostilities, that is, their protracted nature, or a State’s (non-authorised) use of armed force outside its own territory.

V. Discharging the Law’s Moral Tasks in Armed Conflicts

Section III showed that IHRL is prima facie better than IHL at guiding soldiers towards the conduct that conforms to their moral obligations (task one) and better at securing the protection of individual rights in the outcome of conduct (task two). However, epistemic constraints and volitional defects might make a law that further diverges from underlying moral principles, such as IHL, better at discharging one or both of the law’s moral tasks. Section IV suggested that an armed confrontation qualifies as an IAC that comes under the purview of IHL when a State is using force outside its own borders on the territory of another State without the territorial State’s authorisation or consent, regardless of the intensity of hostilities.¹⁴⁸ Moreover, armed confrontations count as NIACs that are governed by IHL when they do not involve a State’s using unauthorised force outside its own borders, but hostilities reach a threshold of intensity at which they count as ‘protracted’. In this section, we seek to answer the question as to whether either of these two features should trigger the applicability of IHL because they create epistemic barriers or volitional defects that undercut IHRL’s ability to better discharge the law’s two moral tasks.

Section A will show that the use of force beyond a State’s own territory does not per se create epistemic barriers that could affect IHRL’s ability to guide soldiers towards the fulfilment of their moral obligations (task one). Section B asserts that hostilities becoming more intense, in contrast, does create such epistemic barriers. Still, IHRL remains the better law for discharging its first moral task compared with IHL. Section C shows that it is not only the reality of armed conflict, but also the legal context that structures the epistemic environment in which law addresses the soldier on the battlefield. Due to the open-endedness of the ius contra bellum and its divergence from moral principles, IHRL only retains its ability to better discharge the law’s first moral task as hostilities become protracted or cross international borders if it is applied ‘symmetrically’, meaning as if both parties to an armed confrontation had a lawful aim. Finally, Section D turns towards the law’s second moral task and the implications of extraterritoriality and the intensity of hostilities for the volitional context of decision-making on the battlefield. It again finds that extraterritoriality does not per se create volitional defects. When hostilities reach the threshold of becoming protracted, however, IHL is better than ‘symmetrical IHRL’ at discharging the law’s second moral task.

A. The Use of Force across International Borders and the Law’s First Moral Task

Does the unauthorised use of force by a State outside its own borders create epistemic constraints that could interfere with IHRL’s ability to discharge the law’s first moral task? Let us imagine an internal confrontation between State A and an organised armed group, which spills-over into neighbouring State B’s territory (type 5). In A’s territory hostilities are not protracted and therefore governed by IHRL alone. However, on B’s territory, hostilities are part of an IAC and thus under the purview of IHL. It is important to stress that the question at hand is not whether State A would be able to guarantee the full panoply of human rights just as easily on State B’s territory as it would be able to do so on its own territory. Instead, the question is whether soldiers on the battlefield are less able to discern the implications of the more complex demands of IHRL regarding the permissibility of killing when they operate in neighbouring State B compared with when they use armed force on their own territory. Traditionally, State A would have been less familiar with the terrain across the border, which could reduce the situational awareness of its soldiers operating in B’s territory compared to those fighting at home. In the twenty-first century, however, satellite imagery mostly makes up for any such shortfall.

In other than spill-over confrontations, a State may face a less familiar enemy when fighting beyond its own borders. For instance, a State opposing a non-State armed group in a transnational IAC (type 6) or another State in a traditional IAC (type 2), might have less insight into the opponent’s organisation and conduct, compared with a State fighting an organised armed group that operates on its territory. However, ultimately the level of intelligence that belligerent C has about belligerent D has much more to do with the sophistication and length of C’s intelligence-gathering than with where D operates. Moreover, in internationalised internal confrontations (type 4), which involve the use of extraterritorial unauthorised force, just like in internationally supported internal confrontations (type 3), which do not, the intervening State can benefit from the intelligence of the belligerent it supports about the belligerent it opposes. These examples suggest that a State’s unauthorised crossing of international borders does not necessarily change the epistemic context of battlefield decision-making, though sometimes it can.

Familiarity with the terrain and with the enemy, which are contingently linked to where a belligerent is conducting hostilities, may speak to what human rights jurisprudence conceives of as a State’s ‘effective control’. Although the ECHR and the ICCPR both define their jurisdictional reach primarily with reference to a State’s own territory,¹⁴⁹ domestic as well as international case law has over the last decades converged on the interpretation that IHRL applies extraterritorially if a State exercises effective control.¹⁵⁰ Effective control, like familiarity with an enemy and terrain, is systematically linked to, but not necessarily only a function of whether the State operates on its own territory. The ECtHR and the Human Rights Committee indicate that just as a State does not necessarily lack control outside its borders, it does not always have control over all individuals on its own territory. Organised armed violence on a State’s territory is one possible reason for why a State might lack internal control.¹⁵¹ At the same time, a State can have extra-territorial control ‘as a consequence of … military action’.¹⁵² It follows that crossing international borders normally reverses the presumption from a State having effective control to a State not having effective control. On a State’s own territory, armed violence may signal the absence of control. Beyond a State’s territories it may be an indicator of the opposite.

Does IHRL require that a State has effective control in order to discharge the law’s first moral task? Or, put differently, is the loss of effective control associated with the emergence of epistemic barriers to discerning the morally right course of action which would undercut IHRL’s ability to guide soldiers’ actions? If this were true, then the morally right point for IHL to displace IHRL would be the (now widely accepted) de iure limit of IHRL’s extraterritorial applicability: where the State lacks effective control. This raises the question of what it looks like when a State exercises extraterritorial effective control in the context of the use of force. The ECtHR’s jurisprudence asserts that a belligerent can exercise ‘temporarily, effective overall control of a particular portion’ of another State’s territory,¹⁵³ in which case IHL’s rules for occupation, enshrined in GCIV, will likely apply. Alternatively, a belligerent State can exercise authority and control over persons without controlling territory.¹⁵⁴ Although this latter type of extraterritorial control over a person has mostly been found to obtain in the context of detention,¹⁵⁵ the Court has also asserted that a State can have control over an individual passing through a checkpoint¹⁵⁶ and even one affected by the use of force.¹⁵⁷

When it comes to defining the factual indicators of such extraterritorial effective control over persons during the conduct of hostilities, the literature unfortunately often verges on tautology. Murray, for instance, mentions ‘factors such as, troop density, effective command and control, control of the skies, control of the sea lines of communication, a robust intelligence picture, suitable military hardware, control of cyberspace, or control of infrastructure and logistic support’.¹⁵⁸ The case law meanwhile overwhelmingly focuses on the parameters of a State exercising effective control over individuals in detention.¹⁵⁹ ‘What is decisive in such cases is the exercise of physical power and control over the person in question’.¹⁶⁰ This leaves unclear what it looks like when a State has effective control over persons affected by the use of force, that is, during the conduct of hostilities. Actual physical control, the literal power to handle a person’s body and command them to be in one place rather than another, is a feature of detention. The use of force against a person during the conduct of hostilities, however, will often signal the absence of this power.

Indeed, if we home in on the ordinary meaning of the word ‘control’ as ‘the power to influence or direct people’s behaviour or the course of events’,¹⁶¹ it becomes evident that the use of violence is either itself an exercise of such control or it is a sign of its absence. If an attacker subdues a challenger through the use of armed force, this is an expression and indeed conclusive evidence of their effective control over the challenger. Exchanges of armed force that go on, in contrast, suggest that neither side has the power to subdue, that is, to control, the other. During an ongoing violent confrontation neither side then has effective control. It follows that it is the intensity of hostilities, particularly their quality of being ‘protracted’ that determines whether a party to an armed confrontation has effective control and whether IHRL can discharge the law’s first moral task. We can thus conclude that crossing international borders does not itself undercut the ability of IHRL to discharge the law’s first moral task. In peacetime, extraterritoriality may reverse the presumption of a State’s effective control, but whether the use of force is an exercise of effective control or the expression of its absence, depends on the intensity of hostilities not on where the confrontation takes place.

B. The Intensity of Hostilities and the Law’s First Moral Task

How does an increase in the intensity of hostilities affect IHRL’s ability to guide soldiers towards fulfilling their moral obligations? The more intense hostilities are, the more difficult it is to determine whether, at any given moment, IHRL indeed exceptionally permits the use of intentional lethal force. If IHRL governed hostilities, it would address the soldier on the battlefield with the prescription to kill combatants or enemy fighters only if they in fact used or threatened violence, and a lethal attack was necessary to defend themselves or a third party against this threat.¹⁶² Soldiers would not be allowed to plan on using force as a first resort against enemy fighters as soon as they present themselves. Capture would have to be impossible.¹⁶³ Implementation of this provision requires a judgement about an individual’s likely conduct and about their intention. Two recognised indicators for the intensity of hostilities, in particular, diminish the attacker’s capacity to divine a potentially hostile individual’s state of mind and predict their behaviour: the type of weapons, namely, their range, and the number of persons involved in hostilities. The latter accounts for how many different individuals’ conduct and state of mind a soldier likely has to evaluate at the same time, potentially straining her or his cognitive capacity beyond the humanly possible.

An intensification in hostilities similarly increases the epistemic barriers to discerning when IHRL exceptionally permits overriding an innocent bystander’s right to life on grounds of necessity. IHRL either requires an immediate threat to human life or it imposes on the attacker a duty of care. A duty of care means an attacker is only authorised to launch the attack if a reasonable observer would with reasonable certainty affirm that this attack was the last and mildest means of achieving the aim which is legally recognised to unlock this exception. It may not be possible to establish true ‘lastness’ even in an environment in which the attacker has relatively solid knowledge about how reality will likely unfold in the near future. As the adversary gets a vote in the consequences of an attack, and success in war partly depends on not being predictable to the enemy, achieving reasonable certainty about the consequences of one’s conduct during an armed confrontation is likely rare. The more numerous, complex and fast-paced exchanges of violence are, the rarer will be moments in which the attacker can fulfil a duty of care towards innocent bystanders.¹⁶⁴

That it becomes harder to discern when IHRL exceptionally permits an attack on the battlefield does not mean IHRL ceases to have implications for action. Both Conventions discussed in Section II start out with a blanket prohibition on deprivations of the right to life. The use of lethal force is an exceptional empowerment. If the conditions of an exception cannot be established with reasonable certainty, the obligation not to deprive individuals of their human right to life remains intact. In the midst of hostilities, if it is impossible to establish who is threatening human life by using the kind of unlawful violence that can only be neutralised through lethal attack, then not carrying out a lethal attack is the reaction IHRL demands. Similarly, if it is impossible to establish with reasonable certainty the necessity of overriding a bystander’s right to life, IHRL does not grant such a permission.

As hostilities intensify, IHL’s rules of distinction, proportionality and necessity do not become harder to apply in the same measure as do IHRL’s rules. Neither does IHL revert to a default of no permission to use lethal force as does IHRL. Combatants do not necessarily become harder to distinguish from the civilian population in all-out war. Members of organised armed groups with a continuous combat function may even be easier to tell apart from the general population the less sporadic and more protracted hostilities are. Moreover, the implications of IHL’s principles of proportionality and necessity depend on an attacker’s expectations. The care in formulating these expectations that is owed to civilians under Article 57 API diminishes with intensifying hostilities, as the range of verification measures that are ‘feasible’ shrinks. As hostilities intensify it may also be less often true that ‘a choice is possible’ among several targets, which are anticipated to yield the same military advantage.¹⁶⁵ As its permissions do not hinge on an absolute threshold of minimum knowledge about the status of a target or the consequences of an attack, IHL is no less likely to permit attacks as intensifying hostilities decrease the knowability of an attack’s alternatives and consequences.¹⁶⁶

Which body of law then guides the soldier towards what is typically the morally right course of action as hostilities intensify? Moral principles, like IHRL, start with a presumption against killing and only allow infringing or overriding an individual’s moral right to life in exceptional circumstances. If these circumstances cannot be established with reasonable certainty, the morally right course of action is typically not to kill another person or launch an attack that is expected to kill an innocent bystander. Past a certain point of doubt, I would not be permitted to defend myself by killing an individual that may or may not threaten me. Similarly, past a certain point of doubt, rescuing innocent bystanders by killing an individual who I thought was attacking them would not be the morally right course of action. Inaction, as demanded by IHRL, is prima facie the morally appropriate reaction to uncertainty. That IHRL, on epistemic grounds, permits fewer attacks as hostilities intensify hence means it continues to guide the individual towards what is typically the course of action that conforms to his or her moral obligations.

One may reasonably interject here that compliance with IHRL would surely hamstring a belligerent in prosecuting a war. Not winning or not even fighting a war, in turn, can carry a moral cost if the war has a just aim, such as repelling a brutal aggressor or preventing a genocide and thereby protecting individual rights overall. Here we need to note what we have so far bracketed: IHRL requires a lawful aim for the exceptional deprivation of an individual right to life. Of course, moral principles also only permit the use of lethal force in pursuit of a just aim and in defence against an unjustified threat. If having a lawful aim for the purposes of IHRL was the same as having a morally just aim, IHRL would guide soldiers on both sides of an armed conflict towards what is typically the morally right course of action in their respective situations: inaction on the side in want of a lawful/just aim and exceptional permissions to infringe individual rights on the side pursuing a lawful/just aim. This would make the belligerent with the just/lawful aim likely to win, the constraining effect of IHRL notwithstanding.

We can conclude that intensifying hostilities change the epistemic context of decision-making on the battlefield, making it more difficult to discern the implications of IHRL’s demands and reducing the instances in which IHRL affords an exceptional permission to deprive an individual of their right to life. In contrast, it does not become more difficult to follow the prescriptions of IHL as hostilities become protracted. IHRL nonetheless continues to be a better guide to soldiers meeting their moral obligations on the battlefield. An intensification of hostilities does not undermine its better ability to discharge the law’s first moral task.

C. The Legal Context and the Law’s First Moral Task

The above assertion that IHRL better discharges the law’s first moral task even in the context of protracted armed hostilities and when States cross borders to use unauthorised force on other States’ territory rests on the assumption that having a lawful aim for the purposes of IHRL coincides with having a morally just aim. If this were not the case, IHRL would risk guiding soldiers with an unjust aim towards the perpetration of individual rights violations. It would also systematically prohibit soldiers with a morally just aim from committing infringements of individual rights that are necessary and justified in order achieve this aim. The above assertion further rests on the second assumption that soldiers are generally able to determine whether they have a lawful aim. Otherwise they would not be able to establish the implications of IHRL for their actions and IHRL would in fact fail to guide them towards meeting their moral obligations. Let us focus on the first assumption for now: is it systematically the case that in each armed conflict, the side that uses force lawfully also has a morally just aim and the side that has broken the law uses morally unjustified violence?

Section II.C touched on the difficult question of what it means to use or threaten unlawful violence for the purposes of IHRL during an organised armed confrontation that counts either as a NIAC or an IAC. For armed conflicts that pit non-State organised armed groups against State belligerents, we can sometimes rely on domestic law to establish that members of the organised armed group, specifically if they challenge their territorial State, use force unlawfully.¹⁶⁷ For conflicts among States, we could draw on general international law, namely, the ius contra bellum, to determine which side has a lawful aim.¹⁶⁸ We need to establish then whether the domestic law-based differentiation between State and non-State belligerents and the ius contra bellum track the moral principles that determine when it is morally justified to resort to armed force.

According to the moral standard outlined in Section III.A, the sole locus of moral value is the individual and resorting to force is justified only in defence of individual rights. States’ and other political communities’ rights are derivative of the rights of the individuals that constitute them. A community’s moral right to resort to force in self-defence is contingent on its being a vehicle for the protection of individual rights.¹⁶⁹ Crucially, this justification for resorting to force is the same for non-State organised armed groups as it is for States. There is no reason to assume that a non-State actor that challenges the territorial State has necessarily resorted to force unjustifiably and that the territorial State is always morally justified in fighting back. In contrast, as mentioned above, if we rely on domestic law to determine the lawfulness of a soldier’s use of violence during a NIAC, we will often find that the belligerent State’s soldiers have a lawful aim, whereas members of non-State organised armed groups threaten and use unlawful violence. This domestic law-based differentiation between State and non-State belligerents’ use of force does not track moral principles.

Having a morally just war aim means fighting to preserve individual rights or to prevent their violation. Past a certain level of intensity, armed confrontations inevitably involve the killing of innocent bystanders. As warfare thus always leads to infringements of individual rights, establishing a just cause means determining whether a resort to force is the lesser moral evil. When facing an outside aggressor or when deciding whether to intervene to halt a genocide, the State or non-State actor resorting to force has to ask: what is the lesser evil in terms of morally unjustified infringements of individual rights, resorting to war or refraining from using force? If the individual rights infringements that a resort to force will inevitably cause are the lesser evil, are they also proportionate to the aim of avoiding the greater number of rights violations?¹⁷⁰ If, but only if, the answers to both questions are yes, is a resort to force morally justified?¹⁷¹

Calculating whether a resort to force would entail fewer or more unjustified infringements of individual rights than would not responding militarily to an aggression or a humanitarian catastrophe, is an extraordinarily difficult, future-oriented judgement. It not merely requires divining the intended and unintended consequences of one’s own actions. The extent to which a defensive war jeopardises individual rights – which rights, for how many individuals and how severely – also depends on the reactions of the adversary.¹⁷² From a moral point of view, the decision as to whether or not to resort to armed force therefore often takes the form of what I have termed an ‘epistemically cloaked forced choice’¹⁷³ between allowing individual rights violations to occur and committing potentially unjustified individual rights infringements by resorting to force. Given this high epistemic barrier to determining whether and when resort to armed force is morally justified, it is not surprising that general international law diverges from moral principles. Just restating that a resort to armed force is lawful if it is a lesser evil in terms of individual rights infringements would be rather unhelpful. At the same time, if general international law on the resort to force discharges the law’s first moral task, it will typically both permit morally justified resorts to force and prohibit those that are not. Does it?

The resort to force is prohibited according to Article 2(4) of the UN Charter and under customary international law.¹⁷⁴ Article 51 of the UN Charter recognises individual and collective self-defence as an exception to this prohibition. It is the only such exception where the unilateral resort to force by States is concerned. The continued lack of a legal permission to resort to force for the purposes of humanitarian intervention means that a class of cases of potentially morally justified resort to force – necessary intervention to rescue individuals from their own State’s egregious individual rights violations – does not qualify as lawful.¹⁷⁵ There may be good moral reasons for the continued legal prohibition on humanitarian intervention,¹⁷⁶ but the absence of a clearly delineated international legal empowerment to use force as a means of rescue is one source of a likely divergence between a morally justified and a legally permissible resort to armed force.

International law empowers the Security Council to authorise the use of force. However, Article 39 UNC does not ask the Security Council to authorise specifically measures that are necessary to protect individual rights or even measures that are necessary to defend a State’s right to territorial integrity, which could be a vehicle for protecting individual rights. Rather, it empowers the Security Council to authorise measures that are necessary ‘to maintain or restore international peace and security’. Some scholars argue that the Security Council’s mandate of maintaining peace and security coincides with the broader goal of enforcing international law.¹⁷⁷ Individual rights are protected under international law, but is the protection of individual rights therefore co-extensive with the preservation of international peace and security? That may sometimes be the case. However, that a resort to force can be in defence of both international peace and security and individual rights does not mean that these goals never conflict and that measures necessary for the achievement of the former are also necessary (or sufficient) for securing the latter. Whether or not Security Council-mandated resorts to force have a morally just cause is therefore entirely contingent.

Does Article 51 UNC at least permit, more often than not, resorts to force that are also morally justified while excluding those that are not? The right to use force in self-defence is triggered ‘if an armed attack occurs against a Member of the United Nations’. Scholarly opinion broadly coalesces around the understanding that not just any use of force on another State’s territory amounts to an armed attack,¹⁷⁸ and that a mere threat of force is not enough to warrant forcible self-defence.¹⁷⁹ A State that is subject to the use of force below the threshold of an armed attack ‘is bound, if not exactly to endure the violation, at least to respond only by means falling short of the use of cross-border force’.¹⁸⁰ Any threat or use of force clearly jeopardises the State’s right to territorial integrity, but from a moral point of view, this is not a sufficient just cause for war. If a ‘bloodless invasion’ only violated individuals’ political rights, but an armed response would also endanger their right to life, not resorting to war in self-defence could be the lesser moral evil.

For us to understand whether armed attacks in the meaning of international law usually create a cause for morally justified resort to force, we need to know whether a use of force that rises to this threshold typically threatens individual rights, while a use of force below this threshold typically does not. The first obstacle to answering this question is the contestability of the minimum threshold that the use of force has to meet in order to fall in the category of an armed attack.¹⁸¹ As the ICJ in the Nicaragua case assumed there was a ‘general agreement on the nature of the acts which can be treated as constituting armed attacks’,¹⁸² the Court expended little ink on discussing the minimum intensity or scale of violence required to meet the threshold.¹⁸³ Every time the Court has returned to the concept of an armed attack, its application has been highly context-specific. Scholars have correspondingly found the concept ‘too vague to be useful’¹⁸⁴ in adjudicating real-world cases.

An alternative way of gauging whether the kind of war that responds to an armed attack is typically a morally justified resort to force is to delineate the goals that Article 51 UNC envisages as a lawful purpose of self-defence. However, like the minimum threshold of intensity that defines an armed attack, the horizon towards which defensive force has to be directed is subject to uncertainty.¹⁸⁵ Some scholars argue that it is ‘halting and repelling’ the armed attack.¹⁸⁶ Other scholars maintain ‘that the legitimate ends of using force in self-defence may differ, depending, inter alia, on the nature and scale of the armed attack, the identity of those who carried it out, and the preceding relationship between the aggressors and the victim state’.¹⁸⁷ Crucially, even scholars who agree on the interpretation that a State must seek to halt and repel the attack admit that what that means is controvertible. Many argue that it must be more than an empowerment to end an ongoing aggression.¹⁸⁸ As it is not clear what counts as a lawful aim for the resort to force in self-defence, there is no reason to assume that it is necessarily a morally just aim.

In sum, we cannot assume that soldiers using unlawful violence for the purposes of IHRL necessarily lack a just aim while those who use lawful violence fight in pursuit of a morally just aim. International law does not regulate the resort to force by non-State armed groups. If we rely on domestic law, a non-State actor that rises up against the territorial State in order to prevent the latter’s individual rights violations, likely uses violence unlawfully even if it is in pursuit of a morally just aim. In addition, we uncovered three major sources of divergence between moral principles and the ius contra bellum. First, the absence of a humanitarian intervention exception accounts for why international law may prohibit morally justified resorts to force. Second, the Security Council’s focus on international peace and security rather than on individual rights means mandated resorts to force may or may not be morally justified.¹⁸⁹ Third, it is very likely that international law systematically authorises the use of force by States in self-defence when this is not morally justified in cases in which the territorial integrity of a State is threatened by an armed attack, but resorting to force would be a greater evil in terms of individual rights infringements.

Moreover, the ius contra bellum does not afford a straightforward guide for soldiers according to which they can easily determine the lawfulness of their cause. Specifically, the threshold for and permissible aims of self-defence are contestable. Our second assumption made above – soldiers on the battlefield are generally able to determine the lawfulness of their aims – is therefore not warranted either. The combination of a significant substantive divergence from moral principles and a high degree of contestability means that general international law on the resort to force poorly discharges its first moral task. As it partly determines what individuals on the battlefield know about the lawfulness of their aims, its limitations affect the epistemic context of soldiers’ decision-making. The law that determines the permissibility of resort to armed force, when force is protracted or crosses international borders, thereby undermines IHRL’s ability to discharge its first moral task during hostilities. Even if individuals knew whether their aims were (un)lawful, and they acted accordingly, IHRL might fail to guide them towards the course of action that fulfils their moral obligations because the ius contra bellum sometimes empowers and hamstrings the wrong sides, respectively.

Given these limitations of general international law on the resort to force, IHRL may better discharge the law’s first moral task if it is applied ‘symmetrically’, meaning as if both sides in a war had a lawful aim and as if the threats and uses of violence that soldiers on both sides encountered from their opponents on the other side were unlawful. As mentioned, IHRL will regularly empower what is from a moral point of view ‘the wrong side’ and undermine the pursuit of a just aim on ‘the right side’. ‘Symmetrical IHRL’ would still empower rights violations on the part of soldiers who lack a morally just war aim. It would, however, avoid undercutting soldiers’ pursuit of morally just war aims not recognised as lawful under international law. Whereas regular IHRL might vouchsafe that the side with the morally unjust aim secures victory, ‘symmetrical IHRL’ would merely fail to prejudge which side prevails.

Would IHRL that is applied as if both sides had a lawful aim still better discharge the law’s first moral task than IHL? The answer is certainly yes. If soldiers on both sides in a war ended up applying IHRL as if their war aim was lawful and the violence they encountered from soldiers on the other side amounted to a threat or use of unlawful violence, IHRL would still guide soldiers on the just side towards the typically morally right course of action. IHL, in contrast, would not. Its status-based distinction and laxer standard of care towards civilians would license additional unnecessary (and therefore unjustified) infringements of individual rights. On the unjust side, even if soldiers were permitted to act as if their aim was lawful, IHRL would guide soldiers towards courses of action that are less morally wrongful than IHL. A ‘symmetrical IHRL’, one in which soldiers on both sides act as if the threat or use of violence on the other side was unlawful, still better discharges the law’s first moral task than IHL.

In sum, given that the ius contra bellum diverges from moral principles, lacks determinacy and leaves questions unanswered where non-State actors are concerned, ‘symmetrical IHRL’ will more often guide individuals towards the conduct that conforms to their moral obligation than asymmetrical, that is, regular IHRL. ‘Symmetrical IHRL’ will typically guide soldiers fighting for a just aim towards the course of action that conforms to their moral obligations, and soldiers fighting for an unjust aim towards conduct that is less morally wrongful than the conduct allowed by IHL, given the latter’s greater divergence from moral principles. Before the next section turns to the moral implications of compliance with law for the protection of individual rights (task two); it is worth emphasising that, if general international law on the resort to force were clarified or changed, if it were to systematically regulate the resort to force by and against non-State actors across international borders, we would have to revisit the question of whether IHRL should be symmetrical or not with a view to fulfilling the law’s first moral task.

D. The Use of Force across International Borders, the Intensity of Hostilities and the Law’s Second Moral Task

We have assumed that in typical law enforcement contexts, in which hostilities are neither protracted nor do they cross international borders, IHRL discharges both of the law’s moral tasks better than IHL. The analysis in the preceding sections then revealed that neither a state using unauthorised force beyond its own borders nor an increase in the intensity of hostilities undercuts IHRL’s better ability to discharge the law’s first moral task. However, given the shortcomings of the international legal regulation of the resort to force, ‘symmetrical IHRL’ discharges the law’s first moral task better than regular IHRL when hostilities cross borders or become protracted. A law that better discharges the law’s first moral task also prima facie better discharges its second moral task of securing the fullest feasible protection of individual rights. That is unless incurable volitional defects mean that a law that reminds soldiers of their moral obligations and guides them towards the course of action that typically fulfils these obligations fails to attract compliance or leads to morally worse outcomes.

In this final sub-section, we turn to the law’s second moral task. We have to answer two questions. First, does a State’s use of unauthorised force beyond its own territory create volitional defects that would undermine ‘symmetrical IHRL’s’ ability to discharge its second moral task in IACs that are not protracted?¹⁹⁰ Secondly, does an increase in the intensity of hostilities mark the emergence of such defects in NIACs and protracted IACs? I will discuss these two questions in turn.

Our assumption that in regular law enforcement contexts IHRL better than IHL fulfils both of the law’s moral tasks implies that States and their law enforcement officials are by and large willing to obey IHRL in such situations. Again, a spill-over armed confrontation (type 5) that remains below the threshold of being protracted in a neighbouring State’s territory proves instructive in showing that extraterritoriality does not on its own undercut this ability of IHRL to attract compliance. If hostilities are not protracted, the State has the capacity to subdue a non-State challenger. There is no reason then why soldiers should be less willing to follow the guidance of IHRL simply because they operate in a neighbouring State’s territory. But what if the opponent is not a non-State actor on another State’s territory, but really another State, such as in a classic IAC (type 2) or an internationalised internal confrontation (type 4)? If hostilities are not protracted, a State has effective control as a result of the extraterritorial use of force and is, by logical implication, able to subdue the opposing State’s soldiers. In other words, if the use of force is itself an exercise of control rather than a struggle signalling the absence of such control, the volitional context is conducive to compliance with IHRL.¹⁹¹ Extraterritoriality alone does not create volitional defects.

Here it becomes obvious that, when States use force beyond their own borders, IHRL needs to be applied as if both sides had a lawful aim not only for the purpose of discharging its first moral task, as discussed above, but also in order to fulfil its second moral task. Let us again imagine a non-protracted classic IAC (type 2). It is highly unlikely that combatants on the side that lacks a lawful aim would comply with regular IHRL. Assuming that State A’s combatants knew their State’s resort to force violated the ius contra bellum, the only guidance IHRL would have to offer them would be to hold still, to cease threatening or using violence like a criminal in a domestic law enforcement context. Section III.C has already highlighted that this could violate the spirit of IHRL if A’s combatants were conscripted or had little choice but to participate in hostilities. Even if A’s combatants fought voluntarily in a war which they knew to lack a lawful aim, it would not necessarily be reasonable to expect them to hold still while according the authority to enforce international law to State B’s combatants. Even if this were a reasonable expectation, the international legal order, in which soldiers are primarily and much more directly subject to the authority of their own State rather than any international institution, provides soldiers with few incentives to follow the demand of IHRL to hold still over their State’s command to participate in hostilities. Regular IHRL would likely miss the chance to guide the actions of soldiers on one side in each war, forgoing the opportunity to morally improve the outcome of warfare and to discharge the law’s second moral task.

If ‘symmetrical IHRL’ is meant to attract compliance from both sides in non-protracted IACs, is there not a more fundamental challenge to its discharging its second moral task? Contrary to IHL, IHRL does not bind non-State armed groups.¹⁹² IHRL is addressed to the State; it concerns the State’s obligations vis-à-vis its subjects. The horizontal implications of IHRL are traditionally fairly weak.¹⁹³ However, three out of the four types of IACs discussed here involve non-State actors (all except type 2). How can a body of law hope to better secure the protection of individual rights in the outcome of warfare if it does not even address all belligerents in each of these confrontations? For an analysis that sought to establish lex lata, this would be a crucial limitation. However, this section provides an answer to the question of when and to whom IHRL should apply from a moral point of view. Nothing prevents us from stipulating that in instances of organised armed violence in which ‘symmetrical IHRL’ ought to govern the permissibility of killing by a State, that is, in non-protracted IACs (types 2, 4, 5b and 6), it ought to also govern the permissibility of killing by a non-State organised armed group. After all, the applicability of IHL to non-State actors is also stipulative. Non-State armed groups are not parties to the pertinent IHL treaties. ‘Symmetrical IHRL’ is hence no less procedurally legitimate than IHL as a framework for non-State actors’ conduct, nor is ‘symmetrical IHRL’ less likely than IHL to in fact attract compliance by non-State armed groups.¹⁹⁴

What about armed confrontations that are protracted, either NIACs or protracted IACs? Does crossing this threshold of intensity create volitional defects that could undermine the ability of ‘symmetrical IHRL’ to secure the protection of individual rights in the outcome of warfare? Could IHL potentially better discharge the law’s second moral task? Volitional defects are in the first instance created by compliance costs, that is, incentives to ignore the demands of moral principles. Moral principles that demand that soldiers fight wars without violating the rights of the individuals against whom they fight have two major compliance costs: first, diminished military effectiveness, that is, a reduced likelihood of prevailing in the armed confrontation; and, secondly, diminished survival chances, that is, a reduced likelihood of escaping the confrontation unharmed and alive. These compliance costs increase radically as hostilities become more intense and they account for why individuals left to their own devices often fail to meet their moral obligations in protracted armed confrontations.

Compliance with IHL and ‘symmetrical IHRL’ likewise bears a cost of reduced military effectiveness. Crucially, ‘symmetrical IHRL’s’ compliance costs rise much faster than IHL’s as hostilities intensify. As previously noted, under IHRL, during protracted hostilities, a soldier will often have to forgo attacks because it is unclear whether the targeted individual really does present a threat that makes force necessary. Under IHL, in contrast, an attacker may kill opposing soldiers to further military progress, even if it is not strictly necessary. The duty of care towards innocent bystanders, imposed by IHRL, likewise means that an attacker who cannot establish with reasonable certainty the consequences of their attack on a military objective – and the strict necessity of the expected incidental deaths – would be enjoined not to attack. IHL’s verification demands, in opposition, shrink with the increasing intensity of hostilities, as described above.

The other type of compliance costs, diminished survival chances, likewise expand more quickly for ‘symmetrical IHRL’ than for IHL as hostilities intensify. The faster-paced, more numerous and more violent confrontations are, the more likely it is that any additional effort devoted to establishing that killing a prima facie hostile person is indeed strictly necessary, reduces the attacker’s own chances of survival. IHRL would not demand that a soldier sacrifice him- or herself on the battlefield, but if he or she was averse to taking the increased risk involved in establishing the strict necessity of infringing the human rights of innocent bystanders, this would diminish the soldier’s ability to launch attacks. Resisting the rise in one type of IHRL’s compliance costs (diminished survival chances) could thus further accelerate the rise in the other type of compliance costs (diminished military effectiveness). Under IHL, in contrast, a soldier may attack all other soldiers even before they pose a threat to him or her. Although contestation persists about exactly what makes a verification measure ‘infeasible’ for the purpose of Article 57 API, many militaries limit the extent to which they put their own forces at risk in order to verify the necessity of expected incidental civilian harm.¹⁹⁵

Compliance costs incentivise non-compliance, a volitional defect that law would normally cure by making non-compliance costly. This can be achieved by making visible unreasonable interpretations or outright violations of law, by opening them up to social opprobrium, or by attaching sanctions to compliance failures. However, there is a general lack of oversight over the conduct of hostilities that may allow soldiers to obscure an unreasonable prioritisation of the attacker’s safety over that of civilians. Although lack of oversight is a challenge in all armed confrontations, it is exacerbated as the fog of war thickens during more protracted hostilities. Indeed, the more intense hostilities are, the less likely it becomes that an investigation and assessment after the fact would even uncover intentional attacks against non-threatening individuals or incidental harm that was not strictly speaking necessary. As hostilities become protracted, law cannot easily cure the volitional defects associated with IHRL’s higher compliance costs. These compliance costs hence create the kind of volitional defect that means law risks being ignored.

A different type of volitional defect – affective and cognitive bias – risks that law is systematically misinterpreted leading to morally worse rather than better outcomes. In an armed confrontation in which one side is strong enough to subdue the other before hostilities become protracted, we may be able to count on soldiers’ being no more biased than most of us are in our day-to-day lives. That means soldiers likely have an affective preference for their compatriots and fellow soldiers over members of the ‘outgroup’. And they likely display varying degrees of outgroup hostility.¹⁹⁶ In all-out war, in contrast, the exhaustion of battle, the stress of being under lethal threat and seeing comrades die likely amplify these biases against individuals ‘on the other side’.¹⁹⁷ Protracted hostilities breed bias which, in turn, creates biased interpretations of the law.

The more uncertain a law’s implications are in the situation in which it typically addresses the individual, the more room there is for the individual’s bias to distort these implications. As previously indicated, the implications for actions of ‘symmetrical IHRL’ become harder to discern the more intense hostilities become. Under an IHRL paradigm, all individuals ‘on the other side’, including civilians, are potentially lawful targets of intentional attack, depending on their conduct. For a soldier during protracted hostilities, every person ‘on the other side’, whether they wear a uniform or not, may appear threatening. Neutralising that threat may appear of paramount necessity the more often a soldier has seen a comrade die or the further out of reach military victory appears. If a soldier in that situation draws on ‘symmetrical IHRL’, the law risks empowering morally unjustified individual rights infringements that the soldier would not have committed if they had been left to their own devices. ‘Symmetrical IHRL’ might not merely miss the opportunity to secure a better protection of individual rights in the outcome of warfare, it might make these outcomes worse compared with soldiers’ following their own judgement.

Bias against individuals on the other side, like high compliance costs, creates volitional defects during protracted hostilities that law cannot easily cure. IHL, however, accommodates these volitional defects. It forestalls, as much as possible, the effect of bias against individuals ‘on the other side’ by reducing the individual soldier’s burden of moral judgement. Indeed, IHL demands that soldiers suspend their moral judgement: civilians are immune from attack and combatants are legitimate targets of lethal attack regardless of necessity, regardless of their conduct,¹⁹⁸ their mental state or their cause, hence irrespective of any parameter that requires a moral judgement or that is open to biased interpretation. IHL severs the rules for the conduct of war as much as possible from moral judgements. In circumstances in which bias is pervasive, this means IHL can secure better outcomes for individual rights. Of course, it is exactly this disconnect from underlying moral principles that enhances IHL’s ability to discharge the law’s second moral task that also makes IHL so bad at guiding soldiers towards the fulfilment of their moral duties.

If we focus on the law’s second moral task of securing the fullest feasible protection of individual rights in the outcome of warfare – that is, minimise as much as possible unjustified infringements of individual rights – IHL’s principles for the conduct of hostilities perform better than IHRL even if IHRL is applied as if both sides had lawful aims. IHL accommodates the volitional defects that it cannot cure; volitional defects that stem from higher costs of compliance and soldiers’ more acute bias during protracted hostilities. That does not mean IHL is the morally ideal law for the performance of law’s second moral task during the conduct of protracted hostilities. It may well be possible that law could demand the necessity of lethal attacks against enemy soldiers and still attract reciprocal and good faith compliance even during highly intense hostilities. Such a law would reduce the number of unjustified infringements of individual moral rights compared with IHL as it currently stands. However, if the choice is between IHL as it stands and ‘symmetrical IHRL’, then from the point of view of the law’s second moral task, IHL should govern the conduct of hostilities above the threshold of intensity at which the discussed incurable volitional defects kick in, meaning when hostilities become protracted.

VI. Conclusion

When should IHL apply? When should it prevail over, when give way to IHRL in regulating the conduct of hostilities during IACs and NIACs? The argument presented in this chapter leads to the following proposal for a moral division of labour between IHL and IHRL. IHRL better discharges both of the law’s moral tasks during law enforcement operations, when organised armed violence neither involves the use of unauthorised force by a State beyond its own borders nor rises to the threshold of being protracted. When a State uses unauthorised extraterritorial violence, IHRL needs to be applied as if both sides were facing unlawful threats and uses of violence from the other side. Applying IHRL ‘symmetrically’ is necessary due to the failure of general international law to discharge its first moral task. It also helps IHRL attract compliance from soldiers on both sides of a non-protracted IAC. As long as hostilities do not become protracted, ‘symmetrical IHRL’ discharges the law’s two moral tasks better than IHL. A moral division of labour between IHRL and IHL, then clearly requires that non-protracted hostilities during armed conflicts are governed by ‘symmetrical IHRL’ rather than by IHL.

Among the six types of armed conflicts we distinguished, currently only NIACs – that is, purely internal NIACs (type 1), internationally supported NIACs (type 3) and the internal part of spill-over confrontations (type 5a) – are subject to an intensity requirement. That means that if conflicts with these belligerent configurations are not protracted, they do not count as armed conflicts at all and they are governed by regular IHRL. Classic IACs (type 1), internationalised internal conflicts (type 4), the extraterritorial part of a spill-over confrontation (type 5b) and transnational confrontations (type 6), in contrast, fall under the purview of IHL regardless of their intensity. From a moral point of view, if conflict types 1, 4, 5b and 6 are not protracted, they should instead be governed by ‘symmetrical IHRL’. Conflicts with States on both sides may not seem amenable to governance by IHRL. However, this intuition is likely rooted in the association of these conflicts with intense hostilities. In reality, extraterritorial spill-over conflicts and transnational conflicts, in particular, are often characterised by sporadic rather than protracted hostilities.¹⁹⁹ ‘Symmetrical IHRL’ should prevail over IHL during such non-protracted IACs.

Above the threshold of intensity at which hostilities count as protracted, meaning during protracted IACs and during NIACs, ‘symmetrical IHRL’ remains the better law for guiding individuals towards conformity with the moral reasons that apply to their conduct (task one). However, IHL generates morally better outcomes for individual rights (task two). Insisting that ‘symmetrical IHRL’ should prevail over IHL would have moral costs. ‘Symmetrical IHRL’ has compliance costs that are almost as high as those of moral principles. It thus risks being ignored. Moreover, due to the acute bias associated with intense hostilities, those soldiers who do recur to the guidance of ‘symmetrical IHRL’ might mistakenly feel empowered to commit more individual rights violations than if they had been left to their own judgement. Allowing IHL to displace ‘symmetrical IHRL’ during protracted hostilities would, however, also have moral costs. IHL operates as if soldiers on the battlefield inevitably ceased to be active centres of moral intelligence. It effectively asks soldiers to suspend their moral judgement. In reality, soldiers might be better served by law reminding them of the morally right course of action even if taking this course of action is costly and recognising it requires that soldiers transcend their bias against the enemy.

This chapter nonetheless makes a moral case for the displacement of ‘symmetrical IHRL’ and for IHL alone to govern the conduct of hostilities during protracted IACs and during NIACs. This is not because securing individual rights in the outcome of war (task two) is more important than guiding the individual towards the typically morally right course of action (task one). Both of the law’s moral tasks are equally important. Rather, it is because IHL falls short of discharging the law’s first task by a metre, whereas IHRL falls short of discharging its second moral task by a mile. IHL still guides soldiers towards a course of action that is less morally wrongful than the course of action they would likely take if left to their own devices. IHRL, even if applied symmetrically, in contrast, either fails to improve outcomes altogether or risks making them worse. As a result, the moral costs of ‘symmetrical IHRL’s’ failing to fulfil task two are higher than the moral costs of IHL’s not fully discharging task one. Put differently, allowing IHL to prevail for the sake of securing individual rights in the outcome of warfare is less morally costly than risking that protracted hostilities are unguided by law. The staggering moral catastrophe presented by armed conflicts waged beyond the international community’s gaze and thus seemingly out of reach of international law serve as a reminder of how important it is for international law to stand a chance at attracting the good faith compliance of individuals on the battlefield.

It warrants restating though that allowing IHL to prevail in governing the conduct of protracted hostilities during NIACs and intense IACs means failing the soldier as a moral agent. We have a legitimate expectation that law, at least to some extent, helps us to meet our moral obligations. That IHL fails in this crucial task makes the ICJ’s claim that compliance with IHL actualises what it means to uphold human rights in war not only false as a matter of legal exegesis, but morally problematic. The least societies can do when sending soldiers into protracted battle with the demand that they obey IHL, is not pretend that this is the be-all and end-all of a moral conversation about battlefield conduct.