Pro patria mori: When States encourage civilian involvement in armed conflict

Abstract

Contemporary armed conflicts have increasingly been accompanied by belligerents’ calls for civilians to support their military efforts. This article investigates the legal consequences of civilians taking up arms provided by or with the tacit support of the State. It first looks at the implications of civilian involvement from the perspective of a State's international humanitarian law (IHL) and international human rights law obligations, focusing on removing civilians from the vicinity of hostilities, informing and training civilians on the implications of directly participating in hostilities, and respecting and ensuring respect for the law. It then demonstrates that the broader fabric of public international law is tested when civilians are encouraged to engage in hostilities, through a close analysis of the challenge of attributing civilian acts to the State. The article closes with practical recommendations for States to ensure that they uphold their humanitarian and human rights obligations, and to render the law of international responsibility effective when civilians commit systemic violations of IHL.

Introduction

Much ink has been spilled over what Wilfred Owen termed the “old lie”: dulce et decorum est pro patria mori.¹ Originally intended as an exhortation for Romans to embrace the young Empire's martial tradition, the phrase is perhaps better known today for Owen's poem Dulce et Decorum Est, a ballade against the glorification of war. As societal understanding of the phrase has shifted over time, so too has the way that international society conceptualizes war: in contemporary international law the use of force is strictly curtailed, and armed conflict is governed in order to protect civilians, and the wounded and sick, from war's worst effects. Modern international humanitarian law (IHL) relies, at its core, on the principle of distinction: in order to protect the vulnerable, belligerents must at all times distinguish between civilians and combatants.² Death for one's country is not, necessarily, considered “sweet and fitting”.

Despite this change in the way we understand war, contemporary armed conflicts have increasingly been marked by belligerents’ calls for civilians to support the military effort. Examples of this phenomenon have been reported in media pertaining to international armed conflicts (IACs) and non-international armed conflicts (NIACs), as well as other situations of violence, occurring in all parts of the world.³ The exact modalities of States’ calls to civilians to participate in conflict vary across contexts – in certain cases, for example, States have encouraged or facilitated direct participation in hostilities (DPH) by dramatically liberalizing gun licensing and ownership laws.⁴ In others, States have encouraged the formation of civilian brigades but have explicitly stated that these will be placed under the supervision of the military,⁵ or they have facilitated the uptake of arms by civilians as a way to police restive territories.⁶

The involvement of civilians in the war effort is not new; indeed, it is most likely as old as the existence of armed conflict itself. However, it has become a more pressing and visible issue in recent decades, as States have begun to rethink their military doctrines and to rely more heavily on civilian capabilities.⁷ In addition to the comparatively straightforward instances of fighting, civilians have provided logistical support, including by transporting weapons and fighters, contributing or moving food items, and providing media coverage.⁸ Not all of these forms of support amount to acts constituting DPH,⁹ but many are “harmful”, broadly speaking, to an opposing belligerent.¹⁰ In the legal realm, various authors have underlined how conflict civilianization leads to an erosion of the principle of distinction and thus an increased exposure of civilians to harm.¹¹ Nonetheless, debates have largely been centred around two issues: determining the threshold at which the conduct of a civilian renders them a lawful target, and examining the roles and responsibilities of contractors, including private military and security companies.¹² Comparatively less attention has been given to identifying the full scope of implications of any involvement in an armed conflict (including acts below the DPH threshold), and the legal issues which may arise for States from encouraging such conduct.

In this article, we examine the legal consequences arising from encouraging civilian involvement in armed conflict.¹³ We first analyze the different forms that civilian support may take, and reaffirm that while not all of them amount to DPH, they nonetheless may undermine the protections that civilians enjoy under both IHL and international human rights law (IHRL). We observe this issue both from the perspective of the civilian and from that of the State. We then highlight that encouraging civilian involvement in conflict has little operational, tactical or humanitarian benefit overall. We proceed to consider in depth a specific issue raised by civilian involvement: whether the law of international responsibility is sufficiently capable of responding to acts perpetrated by civilians directly participating in hostilities, if such acts would violate IHL when committed by members of the armed forces. We close by providing a set of recommendations for States on mitigating the risks identified, and suggesting avenues for further research.

Facilitating civilian involvement in military activities: A humanitarian and human rights law perspective

A bottom-up approach: The civilian's perspective

Elements comprising the legal test for determining when a person directly participates in hostilities have been extensively studied and examined in the literature.¹⁴ While it is the most broadly accepted, the definition set out in the International Committee of the Red Cross (ICRC) Interpretive Guidance on the Notion of Direct Participation in Hostilities under International Humanitarian Law (ICRC Interpretive Guidance)¹⁵ is not agreed upon by all States. Some adopt a broader understanding both on the types of conduct which could qualify as direct participation¹⁶ and on the time frame during which protection from being targeted is lost because of such participation.¹⁷ Incidental evidence of States endorsing broader understandings of the DPH notion might not bear too much influence on how it is ultimately defined as a matter of international law. Still, from a humanitarian perspective, one would probably not advise a civilian to rely on this legal nuance and expose themselves to the risk of being targeted by engaging in an activity for which it is known that it might fall within the scope of such expanded definitions. Regardless of the legal interpretation of the notion, bearing in mind the pace at which modern warfare is developing, an increasing number of acts invite opening a conversation on whether they amount to DPH.¹⁸ However, a soldier making split-second decisions cannot be expected to carry out such highly sophisticated legal analyses (indeed, IHL rules were precisely conceived with the aim of being straightforward and readily applicable by combatants on the ground). This phenomenon exacerbates the already existing difficulties in distinguishing combatants from civilians in the “fog of war”.¹⁹

A civilian becomes a lawful target under IHL for such time as they take direct part in hostilities.²⁰ During that time, they are also removed from the calculation of collateral damage resulting from an attack, and are not “beneficiaries” of precautions in attack. In addition to the risk posed to themselves, given that civilians directly participating in hostilities become lawful targets, they put civilians who are not directly participating in hostilities and civilian objects in their surroundings in jeopardy of suffering the consequences of attacks.²¹ These injuries or destructions would not necessarily be unlawful, seeing that the proportionality rule allows “collateral damage” to result from attacks on lawful military targets.

A final factor in the equation is the absolute nature of the principle of distinction. Among the three cardinal principles governing the conduct of hostilities, the principle of distinction could be deemed as the most straightforward and strict, given that it is not defined as a balancing exercise (unlike proportionality) and is an obligation of result rather than means (unlike precaution). In addition to the obligation to take all feasible measures to verify the nature of the target,²² IHL imposes a duty to presume a person's civilian status in the event of doubt.²³ Nonetheless, the practice of several States reveals that this presumption in cases of doubt “does not override commanders’ duty to protect the safety of troops under their command or to preserve their military situation”,²⁴ thus relativizing the automatic nature with which the principle should be applied. Consequently, in cases where uncertainties may arise as to whether a civilian is directly participating in hostilities or not, scenarios where their civilian nature is contestable might lead them to be targeted under the premise that military considerations warranted doing so. Scholars have increasingly highlighted the reality of such a risk. For instance, Mačák explains that the trend of targeting civilians based on being suspected of DPH is on the rise, and necessarily includes cases where such suspicions turn out to be misguided.²⁵ Similarly, legal experts have drawn attention to an additional risk factor stemming from a blanket call for civilians to take up arms. Such an act would affect not only the safety of those who decide to follow that appeal but all civilians, because it creates the belligerent's expectation that every civilian could potentially carry out harmful acts.²⁶

While it is only conduct that amounts to DPH that legally deprives the civilian of protection from being targeted, this is not the only source of increased risk of harm. Being involved in the conflict without directly participating in hostilities still has implications in the context of the proportionality rule – namely, such involvement often entails coming into the vicinity of persons or objects that are lawful targets. The direct and concrete military advantage expected to result from an attack on a military objective could possibly outweigh the anticipated incidental loss of life or injury of the close-by civilian who supports the war effort, even though this support does not amount to DPH (and thus does not render the person, in and of themselves, targetable). In other words, even if the person's conduct does not amount to DPH, an increase in proximity to lawful targets leads to an increase in the chance that they will be counted as lawful “collateral damage”. The same reasoning would apply in relation to precautions in attack. The spectrum of what would (factually, not as a matter of legal test) be considered “feasible” could be influenced by the behaviour of the civilian themselves. Therefore, increasing the proximity between lawful military objectives and civilians could mean that the attacking party would objectively be left with less means at its disposal to decrease civilian damage.

Even more than the risk of harm from hostilities as such, civilians involved in activities related to the conflict bear certain risks as a result of being perceived as cooperating with the opposing side.²⁷ The brutal implications of being considered a collaborator of the enemy have been well documented by international law scholars.²⁸ IHL itself provides the possibility of adopting several measures with regard to individuals acting in support of a belligerent. In addition to the straightforward case of the capture of combatants,²⁹ IHL envisages a basis for the administrative detention of (protected) civilians³⁰ for imperative security reasons.³¹ Such security reasons could be identified if a person engages in subversive activities or directly assists the enemy, including by providing logistical support.³² IHL also provides States with the possibility of applying derogation measures in relation to a person suspected of engaging in activities affecting State security, including limiting their communication rights.³³ Therefore, while being subjected to any form of ill-treatment would certainly be unlawful, involvement in the conflict might expose the civilian in question to security measures and result in rights restrictions.

A top-down approach: The State's perspective on civilian involvement in military activities

IHL provides neither for an explicit right for civilians to directly participate in hostilities, nor for a prohibition against it; rather, it deals with the factual consequences arising from such participation.³⁴ The majority of legal scholarship argues that a State does not violate an explicit provision of IHL by encouraging civilians to support the military effort.³⁵ Nonetheless, as Mačák underlines in the context of civilian involvement in cyber warfare, a requirement for States to refrain from doing so can be deduced from other provisions of IHL, most notably precautionary measures. Several due diligence obligations stemming from IHL come into play: the duty to take precautions against the effects of hostilities, and the obligation to respect and ensure respect for IHL found in Article 1 common to the four Geneva Conventions (common Article 1).³⁶

Duty to take precautions against the effects of attacks

Encouragement of civilian involvement in a conflict potentially raises several questions when observed from the perspective of the parties’ obligation to take, to the maximum extent feasible, precautions against the effects of attacks. Firstly, this obligation could be read to incorporate the duty not to prompt civilians to come closer to the battlefield area or to lawful targets in general, since this increases their exposure to a risk of harm stemming from the military operations. Secondly, it could require belligerents not to invite doubts as to whether a civilian constitutes a lawful military target or not, because of uncertainties as to whether their actions, at that moment, amount to DPH. Seeing that doing so would increase the risk that those civilians taking part in military activities (but not directly participating in hostilities) are exposed to the dangers of military operations, such conduct could be seen to counter the obligation to strive towards removing civilians from the vicinity of military objectives. Moreover, extending this argument, IHL could be seen as obliging belligerents to not encourage civilians to directly participate in hostilities, as this increases the risk that other civilians in their vicinity would be exposed to higher levels of danger arising from military operations. Exacerbating this risk could arguably be seen to also go against the obligation to take efforts not to locate military objectives near or within densely populated areas.³⁷ For all these components, the assessment needs to be made on a case-by-case basis, in light of the factual circumstances and relatedly the scope of what would be expected to be feasible on the part of the State. Nonetheless, these multiple facets of how civilian engagement stands in tension with precautionary measures showcase that the threshold for proving that such conduct goes against the elements comprising passive precautions would not be that high.

Duty to respect and ensure respect for IHL

States’ obligations stemming from common Article 1 encapsulate the duty to ensure respect for IHL. This is evidently a due diligence obligation and does not require the State to do the impossible. While the precise scope of measures expected of the State will vary in each context, the duty does require States to refrain from encouraging the commission of IHL violations, including by civilians.³⁸

What is less settled is whether, as a preliminary point, common Article 1 also has any bearing on States’ encouragement of civilians to participate in hostilities. This is not explicitly provided in the article itself, but such an interpretation can be deduced from the due diligence obligation to prevent IHL violations – namely, in cases where there is a reasonably foreseeable risk that an IHL violation will be committed, all States have an obligation to take measures to prevent such an occurrence.³⁹ This obligation is indifferent to the status of a State in relation to the conflict in question; it applies to neutral, ally and enemy States.⁴⁰ In the present context, such a duty would arise in two dimensions. Firstly, blurring the lines between civilians and lawful targets (civilians directly participating in hostilities and combatants), and the consequent erosion of the principle of distinction,⁴¹ increases the likelihood of violations being committed by belligerents. Therefore, an obligation to prevent IHL violations would arise even if the participation itself entailed IHL-compliant acts. Secondly, IHL violations might be committed by participating civilians themselves, depending on the extent of their military training and knowledge of IHL.⁴²

In other words, this duty could be read as entailing the obligation to put efforts into not making it more difficult for the enemy to act in compliance with IHL. This interpretation is bolstered by another element of common Article 1: the duty to refrain from carrying out acts which contribute to IHL violations.⁴³ Arguing to the contrary would lead to a paradoxical result where the State would be required to deal with the consequences of such a risk (i.e., do everything feasible to counter the risk) but not to refrain from creating the risk in the first place.

The above proposal might seem to suggest that the party in question would be expected to make waging war easy for its enemy. This may appear counterintuitive, but numerous examples of IHL norms demonstrate that such an interpretation would not be so shocking. For example, IHL requires combatants to distinguish themselves in order to benefit from prisoner of war status upon capture – a duty which precisely aims to facilitate compliance with the principle of distinction. Similarly, several obligations of means are imposed on the defender to mitigate the effects of military operations on its own civilian population, including by separating civilians from military objectives. Lastly, IHL already provides that parties are prohibited from involving children in hostilities. The latter provision encompasses not only instances of direct participation, but also other forms of association and provision of support to the armed forces. Interestingly, it is precisely the exposure to the danger of being targeted that was identified as a compass in determining whether a certain act amounted to a prohibited form of participation or not by the International Criminal Court (ICC) Trial Chamber in the Lubanga case.⁴⁴ The cardinal nature of the principle of distinction is therefore reflected in the logic underpinning several other norms of IHL. Interpreting the duty to ensure respect with IHL as comprising a due diligence obligation to refrain from conduct which contributes to this principle's erosion would therefore be in line with the rationale of IHL.

In addition to IHL, IHRL brings important perspectives in relation to both the rights of civilians involved in the conflict and those of other civilians affected by their conduct. The first aspect is examined in the present section, while the latter is appraised in the context of the duty to provide information and training in the following section. Specifically in the context of the duty to prevent violations of the right to life, the added value of IHRL in this regard lies in the fact that it would not only protect the person in cases where they are mistaken for a lawful target, but also against the risk of becoming such a target in the first place. Along those lines, Mačák invokes States’ obligation to minimize the risks placed on the right to life of civilians in cases where they are engaging in dangerous activities.⁴⁵ He puts forward that States should “prioritise those forms of civilian involvement in the war effort that do not place these individuals in harm's way”, equating the forms coming under this scope to conduct constituting DPH.⁴⁶ Nonetheless, the previous analysis has showcased why the forms coming under this notion should be understood more broadly than the legal threshold for DPH (and therefore temporary loss of protection from being targeted). More specifically, conduct placing individuals at risk should include (1) an increase in the likelihood of being targeted by mistake and (2) an increase in the risk of being considered as part of lawful collateral damage because of proximity to lawful military objectives.

Beyond the preliminary question: The duty to provide information and training

The analysis provided in the section above has shown why it would not be unreasonable to assume that the obligation to take passive precautions would entail not encouraging civilians to engage in military activities. Even moving past the preliminary question related to the legality of whether a State could encourage civilian involvement in an armed conflict, IHL and IHRL norms bring in important elements relating to the obligations that a State would still need to abide by if it were to do so. Both legal frameworks bring in components of a potential duty to inform civilians about the implications of being involved in the conflict, including of directly participating in hostilities, and an obligation to provide them with appropriate training in this regard. This section examines these two components in turn.

Jurists seem to agree that IHL does not explicitly require the defending party to warn civilians about the dangers arising from military activities.⁴⁷ Nonetheless, they identify several norms of IHL that such an obligation could be seen to stem from. These include adopting a progressive approach to interpreting the scope of “other necessary precautions” required of the defender, the duty of constant care to spare civilians,⁴⁸ and the obligation to disseminate IHL among the civilian population.⁴⁹ Therefore, these components of IHL norms could all be seen to require informing civilians, but would operate at different levels – namely, obligations related to the precautions would only be breached in cases where a State is involved in encouraging or prompting civilian involvement;⁵⁰ the degree of the State's involvement would dictate the implications stemming from the obligation to ensure respect for IHL; and the obligation to disseminate IHL would stand at all times.

Such a duty is reinforced by IHRL. For instance, under the European jurisprudence, (real or constructive) knowledge about a “real and immediate risk to the life” would be a trigger for the State's positive obligation to take feasible measures to counter that risk.⁵¹ Dangers arising from the proximity of civilians to the battlefield area, from being converted into a lawful military objective in the context of an armed conflict or from increasing the likelihood of mistakes being made in this regard, would seem to meet this risk threshold.⁵² Outside the realm of conduct of hostilities, an argument could be made that part and parcel of the State's duty to protect individuals against being deprived of their liberty by armed groups (in the NIAC context) or other States (in the IAC context)⁵³ would be not to knowingly place them, or encourage carrying out acts which would place them, at risk of being subjected to detention.

Precisely in this regard, Mačák draws an analogy between IHL and the human rights jurisprudence dealing with States’ preventive obligations in relation to the right to life, as concerns the risk of environmental and industrial disasters.⁵⁴ He highlights the population's right to information about the potential risks to the right to life arising from dangerous activities. In keeping with the protection of personal autonomy, Mačák puts forward that the State would not be acting in violation of its right to life obligations if a civilian were to decide to engage in DPH despite having been fully informed by the State authorities of the implications of such actions.⁵⁵ Nonetheless, it is interesting to note that, in the context of examining a civilian's collaboration with the Occupying Power, an Israeli Supreme Court decision limited the weight given to the civilian's consent to be placed in a zone of hostilities.⁵⁶ In addition to the IHL prohibition on using protected residents as part of the Occupying Power's war effort (for which an explicit parallel does not exist in the law of NIACs),⁵⁷ this conclusion was reached by the Court based on IHL's requirement to keep civilians removed from hostilities, but also the factual assessment of the risks of physical harm arising from the proximity to the hostilities and the risk of ill-treatment that a person suspected of collaboration would face.

The obligation to disseminate IHL is important in one additional context: civilians’ compliance with IHL throughout their involvement in the armed conflict. The State's conduct in this regard could be observed from the perspective of the duty to respect IHL in relation to the conduct attributable to the State, and the duty to ensure respect by individuals under its authority.⁵⁸ In relation to the former, having IHL norms embedded in military training which soldiers are required to undergo is usually considered part of this analysis. With regard to the latter, the scope of measures expected of the State in discharging this obligation varies based on factors such as foreseeability, the State's knowledge about a breach, and the gravity of the breach, as well as the means reasonably available to the State and the influence exercised over the private person.⁵⁹ Arguably the breadth of the measures would increase with greater involvement of the State. Moreover, several factors could inform the foreseeability of IHL breaches by civilians engaged in armed conflict. These include the requirements set out in the State's call for civilians to take up arms, the plan envisaging how civilians would be engaged, and subsequent training. In fact, most examples referred to above do suggest certain forms of supervision over civilians involved in military activities and the training they would undergo. Examining the adequacy of such training in relation to the risk that they are placed at would necessarily be part of this picture.

Moreover, analogies can be drawn from the realm of IHRL in relation to a State's obligations towards its own soldiers in order to grasp what would be expected of the State in relation to civilians called upon to take part in military activities. Namely, while the scope of States’ duty to protect the lives of its own soldiers is not clear-cut, it would put forward an obligation for States to refrain from placing their own soldiers in situations where the risk to their lives is disregarded, or where there is no clear and legitimate military purpose behind imposing such risks upon them. One of the manifestations of States’ duty to protect the lives of their own soldiers relates to obtaining adequate military training. One argument put forward in the context of the UK Supreme Court's decision in the Smith case was that a State could be found responsible for a violation of a soldier's right to life when such a violation would stem from a lack of adequate military training.⁶⁰ A similar argument could be made when civilians become incorporated in the State's armed forces. Even more broadly, it would be reasonable to espouse that, once it has already called upon them to join the military effort, the State has an obligation to provide civilians with the means to do so, including by providing them with training.

This obligation is reinforced by other IHRL norms – this time, in relation to persons subjected to the conduct of those civilians becoming involved in the conflict. These duties would feature most prominently in the context of States’ obligations with regard to the rights to life and liberty. The duty to protect the individual's right to life would require States to take positive steps to counter reasonably foreseeable threats to life, including those emanating from private individuals or entities, as well as other States.⁶¹ Further, a State would remain bound by a duty of care towards persons deprived of liberty by private actors who perform such actions or operate such facilities pursuant to a State's authorization. Particularly where such a State authorization to civilian formations were to be identified, one could argue that the State would have the obligation to prevent arbitrary or unlawful arrests or detentions. It would also need to provide effective remedies to victims in such cases, and maintain control over the exercise of such powers.⁶²

A tactical and humanitarian appraisal

Seeing that IHL is built on the balance between humanity and military necessity, inquiring into the possible tactical benefit of civilian involvement in conflicts is pertinent to the topic. Here, however, even the argument about the increase in military capacity stemming from civilian involvement in conflict does not hold, and in fact such involvement tends to create more problems than benefits. In relation to civilian hackers, their encouragement to become involved in conflict is based precisely on their expertise and level of knowledge in cyber and defence capacities.⁶³ However, outside of this specific case, the overview of practice seems to suggest that looser requirements are envisaged in general calls for civilian involvement in traditional forms of kinetic warfare, casting doubts on the extent to which regular civilians could truly increase the State's defence capacity.

Even the short-term benefits of increasing force capacity and presenting challenges to the opposing belligerent's ability to lawfully conduct hostilities are not worth the long-term cost of inevitable violations of IHL and IHRL which encourage a race to the bottom and an abandonment of the principle of distinction. Moreover, eroding the principle of distinction creates further challenges not only from the perspective of the application of IHL on the ground, but also to efforts to ensure accountability for violations of the principle post facto – an already existing challenge for all norms relating to the conduct of hostilities. Accountability challenges, as part of the broader inefficacy of IHL's implementation mechanisms, further feed into the narratives undermining IHL's value.

Responsibility for acts committed by civilians directly participating in hostilities: A public international law perspective

This article has so far demonstrated that encouragement of civilian engagement in armed conflict creates a host of legal risks, both to the individual and to the State. A State may find itself in violation of its basic IHL or IHRL obligations, a civilian who directly participates in hostilities loses (for such time as they directly participate) their protection under IHL, and the more that the lines between civilians and combatants are blurred, the more challenging it is for belligerents to effectively respect the principle of distinction. Furthermore, with the possible exception of cyber operations, assumption of these risks does not necessarily correlate with increased tactical or operational utility for the State.⁶⁴

In the present section we argue that one specific phenomenon, encouragement of DPH, is also troubling from a general international law perspective. This is because in contemporary international law, encouraging direct participation creates a profound risk of impunity and consequently risks eroding IHL's normative core. The legal mechanisms designed to secure accountability – in particular, the law of international responsibility – are not, in their present state, effective deterrents.

In this section, we first (re)familiarize readers with the law of international responsibility as it applies to States. We then consider each step in the process of invoking a claim of responsibility. We identify that within the established framework of the law of responsibility, civilian conduct is difficult to attribute; more often than not, the “umbrella” obligation to respect and ensure respect for IHL is the most viable route for asserting, attributing and invoking a wrongful act. Invocation also presents profound challenges, especially given that most armed conflicts today are non-international. Taking these points into consideration, the law of responsibility is not appropriately developed or equipped to deter States from encouraging DPH. We close the article by considering avenues for strengthening compliance with IHL, both by individuals and by States.

Restating the law of international responsibility

The law of international responsibility is a system of secondary rules which activate when a primary rule is violated.⁶⁵ It is a long-standing system – elements of the law of responsibility are apparent in early jurisprudence of the Permanent Court of International Justice⁶⁶ – and retains a special importance in contemporary international law. This is because the law of responsibility also contains the primary avenues for securing a remedy: countermeasures. The general prohibition against intervention and the prohibition of the threat or use of force effectively render unlawful the two avenues by which States have historically sought to enforce their claims. Responsibility can attach to the actions of any subject of international law, although its modalities vary somewhat.⁶⁷ For the purposes of this article, State responsibility is the relevant mode. It is difficult to conceive of a situation in which an international organization has ever encouraged, or ever would encourage, civilians to directly participate in hostilities, and the responsibility of individuals is contained within international criminal law. The core elements of a claim of State responsibility are well established, and are accepted as customary international law:

1. 1. An action or omission constituting a breach of a State's international obligations must occur. Such an act or omission may be composite.

2. 2. That action or omission must be attributable to the State, through one of the avenues detailed in the Draft Articles on Responsibility of States for Internationally Wrongful Acts (DARSIWA).

3. 3. There must be no circumstances precluding wrongfulness – namely consent, self-defence, a situation of force majeure, distress, necessity, or establishing that a wrongful act is in fact a lawful countermeasure.⁶⁸

4. 4. An injured State must invoke the law of responsibility. In some situations, a State other than an injured State may be capable of invoking responsibility, particularly where an obligation is owed erga omnes or erga omnes partes (and, in the latter case, protects a collective interest of the group).⁶⁹

5. 5. When an internationally wrongful act has been established, the perpetrator has a subsequent obligation to cease the wrongful act and provide guarantees or assurances of non-repetition.

6. 6. Reparation for an injury can take the form of restitution, compensation or satisfaction; in almost all cases, reparation is a combination of all three. However, erga omnes violations do not necessarily trigger the obligation of reparation; Article 48(2)(b) of the DARSIWA has not yet achieved customary status. Rather, they trigger the subsequent obligation of cessation and non-repetition.

Elements 1 and 2 are preliminary and substantive; if either is not proven, a wrongful act has not been evidenced. Element 3 is contingent and substantive; if a circumstance precluding wrongfulness applies, an otherwise wrongful act does not result in an obligation to remedy the breach. Element 4 is procedural; a procedural defect does not preclude an act's wrongfulness, but it may preclude claiming reparation.

Identifying a wrongful act

The two core elements of an internationally wrongful act, as described in the DARSIWA,⁷⁰ are clear: (1) an action or omission must be attributable to the State, under international law, and (2) that action or omission must violate an international obligation of the State. However, the application of these elements is not clear, for three reasons. First, it is not clear whether the elements must be assessed in a particular order: this is primarily a concern for obligations arising from treaty law. Second, attribution is exceptionally complex and strictly constrained by international law. Third, much of an act or omission's legality depends on how it is characterized.

In the following subsections we consider how these three concerns might interact with the encouragement of DPH. We argue that the first concern – the order of the elements – is largely immaterial when it comes to IHL questions since almost all obligations are both universal and customary. The questions of attribution and act characterization, however, both present serious challenges. It is possible – perhaps even likely – that in most cases of misconduct perpetrated by civilians directly participating in hostilities, only a minimum core of internationally wrongful acts will be attributable to the State.

Ordering the elements?

It is not altogether clear, either in doctrine or practice, whether the elements must be examined in a prescribed order. The commentaries to the DARSIWA suggest that they should, and that attribution should always be considered first.⁷¹ Stern argues the same, because in her view it is not possible to assess whether an act violates an obligation without attributing it – otherwise, the relevant obligations cannot be identified.⁷² Stern acknowledges, however, the counter-argument that in some circumstances – particularly those concerning violations of IHL – it may very well be possible, and desirable, to consider whether an act or omission violates an obligation before answering the question of attribution.⁷³ As far as practice is concerned, the International Court of Justice (ICJ) has not adopted a consistent methodology. In the Nicaragua case, the Court considered whether the conduct of the Contras was – in part or in full – attributable to the United States before it considered whether that conduct breached an obligation owed.⁷⁴ It took the same approach in Diplomatic and Consular Staff in Tehran. ⁷⁵ In Armed Activities on the Territory of the Congo, however, the Court reversed its approach: it first considered whether the relevant act or omission was capable of violating an international obligation of the State.⁷⁶ It did the same in the Bosnian Genocide case, where the entire judgment was structured around assessing conduct before attribution, and followed this same methodology in Croatia v. Serbia. ⁷⁷ In Oil Platforms, the Court took both approaches at once: it focused on attribution as far as the missile attack on the Sea Isle City was concerned,⁷⁸ but when considering the question of cumulative acts amounting to an armed attack it set aside the question of attribution entirely, preferring to consider whether an obligation was breached.⁷⁹ In that same case, when considering the United States’ counter-claim, the Court chose to analyze the relevant conduct first and to consider attribution second.⁸⁰

This inconsistent doctrine and practice cannot be meaningfully resolved here. For the purposes of this article, it suffices to say that the structural concern raised by Stern does not arise with regard to IHL, except with regard to certain obligations contained exclusively in treaty law and the rare case of the persistent objector. This is because the vast majority of IHL is universally ratified or otherwise customary in nature. An attribution–acts order may be necessary when considering alleged breaches of the Additional Protocols, particularly Additional Protocol II. However, a responsibility claim, including relating to the conduct of civilians, is unlikely to otherwise be impeded by what is ultimately a matter of form.

Attribution

Attribution is essential to the identification of an internationally wrongful act. After all, an act may be immoral, inhumane or unlawful, but if it cannot be attributed to a State, then State responsibility is not the appropriate mode of international responsibility.⁸¹ There are three broad avenues of attribution:

1. 1. An actor is part of what James Crawford described as the “hard core” of the State: an organ or agency “exercising sovereign authority”.

2. 2. An actor is acting under the instructions, or direction and control, of the State.

3. 3. An actor falls within what Crawford describes as the “exceptional categories”: either it fulfils State functions in the absence of official authorities, or it is an insurrectionist movement that subsequently becomes the government of the State in question (or a new State), or its conduct is adopted by the State as its own.⁸²

The “core” links

The attribution links contained within the “hard core” are unlikely to capture the majority of private conduct. Article 4 of the DARSIWA – where the author of an act is an organ of the State – may be relevant when State employees directly participate in hostilities and, in doing so, are acting in their official capacity. It could cover, for example, acts committed by members of a police agency, provided those are sufficiently connected with their official functions.⁸³ Organs are identified with reference to the “internal law of the State”.⁸⁴ This ordinary link is uncontroversial, but is only capable of capturing the acts of a limited set of civilians, in a very limited set of circumstances. However, in exceptional cases it may also be possible to attribute acts committed by a “de facto organ” on the basis of internal practice – as Crawford noted, a State “cannot evade responsibility … merely by denying [a de facto organ's] status as such under internal law”.⁸⁵ Such situations are exceptionally rare: in the Bosnian Genocide case, the ICJ suggested they may arise where an actor is “so closely attached as to appear to be nothing more than [a State's] agent”, and where failure to acknowledge the reality of the situation would allow States to escape responsibility “by choosing to act through persons or entities whose supposed independence would be wholly fictitious”.⁸⁶ This modified link – one of complete dependence – is an exceptionally high bar, and in Genocide the existence of “some qualified, but real, margin of independence” was sufficient to disconnect the State from the actions of the Republika Srpska and the VRS.⁸⁷ With such strict prior practice, it is nearly impossible to see how the encouraged, but uncontrolled, conduct of civilians could possibly render them a State organ within the meaning of Article 4.

Article 5 of the DARSIWA, which deals with entities empowered by domestic law “to exercise elements of the governmental authority”, is somewhat different, since it applies to private actors exercising public function. This could include privatized but empowered regulatory bodies or private security firms, or even in some situations autonomous but State-established charitable foundations.⁸⁸ The category may in some circumstances include mercenary organizations,⁸⁹ and likely also captures certain conduct of “community defence groups” when they are empowered by the law of the State.⁹⁰ It does not, however, extend to authorization of conduct by “citizens or residents generally”, because this is not an exercise of governmental authority.⁹¹ Acts such as providing small arms or blanket firearms licenses, or conferring a right to carry arms in anticipation of exposure to conflict, may have the effect of encouraging participation, but they do not confer a right to exercise governmental authority. Neither does explicitly encouraging DPH. Arguably, even a public statement that direct participation and its consequent acts – including killing – will not be prosecuted may not amount to “empowerment”.

Instructions, direction or control

The “hard core” of responsibility is, we suggest, not well suited for attributing the acts of civilians directly participating in hostilities, including when those acts may amount to serious violations of IHL or even grave breaches. A natural next step is to consider whether Article 8 of the DARSIWA, relating to actions committed under the instruction of, or direction and control of, a State, provides a more promising link. Article 8 reflects the customary rule that, in certain situations, a State can be responsible for the conduct of private actors.⁹² Here we consider only the question of “instruction”: this is because “direction or control” – typically interpreted by international courts as a unified concept⁹³ – has most often been understood as a reference to the “effective control” test outlined in the Nicaragua case.⁹⁴ That standard is so high as to almost never be met. We suggest that despite appearances, Article 8 is also not a promising avenue for attributing the conduct of civilians who directly participate in hostilities. This is because it is not evident what, precisely, constitutes an instruction, nor is it clear how precise an instruction must be.

It is unclear when encouragement or exhortations become an “instruction”; all that is generally agreed is that there must be a factual relationship between the actor and the State. In Diplomatic and Consular Staff in Tehran, a general call for attacks against other States was not considered to be an “instruction”.⁹⁵ In a recent investment arbitration, the tribunal held that the takeover of a foreign-owned plant by union members, allegedly in pursuance of an instruction by a head of State to “let the company … become state controlled and transfer it to [the State's] hands”, was not an act on the “instruction” of the State.⁹⁶ By contrast, a different arbitral tribunal considered that the acts of a State-owned enterprise, directed by a minister,⁹⁷ were attributable under Article 8 of the DARSIWA as acts under the “instruction” of a State.⁹⁸ In contemporary scholarship, it has been argued in relation to armed conflict that “a general ‘rallying call’ … encouraging likeminded but unspecified ‘patriotic’ hackers to engage in offensive action” is not sufficient to constitute an instruction under Article 8, and nor is mere encouragement of conduct, even where that conduct may be unlawful or morally reprehensible.⁹⁹

In addition, an instruction must have some undefined minimum level of specificity. The approach of the ICJ suggests that instruction must, at least in armed conflicts, specifically relate to “each operation in which the alleged violations occur”.¹⁰⁰ Crawford suggests that within a sufficiently delineated “instruction”, a “general direction which leaves [open the] method of fulfilling the directive” may suffice.¹⁰¹ Tonkin agrees, and suggests that instructions which authorize, but do not oblige, unlawful behaviour – for example, instructing a private military or security company to extract information “by any means necessary”, or to “shoot anyone who comes near [a] protected object” – may trigger Article 8.¹⁰² As Tonkin also notes, however, the question is markedly more complicated when the instructions given are vague or ambiguous.¹⁰³ Tonkin's concern is with vague but constrained instructions, such as to “protect this place”; she identifies that there are fringe cases where an “instruction” within the meaning of Article 8 may, or may not, exist. This is not in contradiction to either Crawford's suggestion that instructions may be open-ended, or to the ICJ's position that instructions must be at least somewhat situationally specific.

Taken together, these factors suggest that an instruction must not only demonstrate some form of factual relationship: it must also be at least situationally specific, even if the means of achieving the instruction are left open-ended. Encouraging civilians to take up arms is not, bearing these points in mind, enough to constitute an “instruction”. There would have to be something more: a minimum level of specificity about the associated operation or acts commanded, and at least a plausible inference that the instruction authorizes or encourages unlawful conduct. While international law may be evolving towards a more permissive understanding of Article 8 of the DARSIWA,¹⁰⁴ the law as it stands today nonetheless remains strict, making Article 8 attribution of civilian conduct very challenging.

The exceptional cases

Articles 9 and 10 of the DARSIWA – respectively covering the exercise of governmental authority in the absence of the official authorities, and the conduct of a successful insurrectionist movement – do not provide any further clarity.¹⁰⁵ This leaves one final avenue for attribution: Article 11, which applies to conduct that is, ex post facto, both acknowledged and adopted by a State as its own. Much like Article 8, Article 11 acknowledges that, in some limited situations, private conduct may acquire a public character and therefore be attributable.¹⁰⁶ The leading jurisprudence remains Diplomatic and Consular Staff in Tehran, which concerned the express adoption of private conduct as a matter of State policy, thereby transforming it into an attributable, public act.¹⁰⁷ The DARSIWA commentaries confirm that, while acknowledgement and adoption can be either express or inferred, a State may nonetheless limit its acquired responsibility: “the act of acknowledgement and adoption, whether it takes the form of words or conduct, must be clear and unequivocal”.¹⁰⁸ In practice, States very rarely acknowledge and adopt the conduct of third parties; they more often simply “tolerate” it.¹⁰⁹ This, in itself, is not sufficient to trigger the application of Article 11. As far as the conduct of individual civilians directly participating in hostilities is concerned, it seems that Article 11 could only apply in the most extreme circumstances: for example, endorsement of grave breaches committed by civilians.

Bearing the above in mind, attribution is a serious challenge. The core rules of attribution are of little use, and Article 11 likely only applies to the most extreme situations. Article 8 appears promising at first glance, but in practice is very difficult to trigger. This leads to a situation where even if the acts act perpetrated by civilians are clearly wrongful – the deliberate targeting of other civilians, for example, or attacks on civilian infrastructure, or the wilful killing of captured combatants – the State may not bear responsibility for them. Instead, it would bear responsibility only for the high-level acts or omissions from which further wrongful conduct flows: for example, failing to “ensure” respect for the Geneva Conventions or Additional Protocol I (AP I). This is, in effect, an obligation of conduct to be measured against a standard of due diligence.¹¹⁰

Act characterization

Placing attribution aside, the identification of the relevant wrongful act is also a challenge. Acts which violate IHL are inherently capable of being construed as violations of a State's international obligations (with the exception of conduct by opposing non-State armed groups (NSAGs), unless those groups then either overthrow the government or establish a new State).¹¹¹ However, it is not always clear whether an apparent violation of IHL is in fact a violation of it. An act may be internationally lawful or outside of the law, or it may be governed by a separate legal regime. Three examples serve to illustrate this point:

1. 1. A civilian opens fire on, and kills, another civilian.

2. 2. A civilian or group of civilians destroys a nominally civilian object, such as a town hall.

3. 3. A civilian, dressed in civilian attire, enters a building under special protection and kills an individual located within.

All of these acts suggest violations of some legal obligation, but not all obligations are internationally owed, nor are all obligations necessarily governed by IHL.¹¹² A civilian may kill another in an act of murder, or banditry, unrelated to an armed conflict; they may target a civilian who is also directly participating in hostilities; they may be factually mistaken, and believe they are targeting a combatant; or they may be violating the principle of distinction intentionally; or they may be committing an act of persecution within an occupied territory; or they may simply be acting in lawful self-defence, entirely unrelated to an armed conflict. Destruction of a civilian object may be a violation of the principle of distinction, or it may be a lawful attack on a military objective, or it may be a violation of the law of occupation – or, again, it may simply be criminal in nature. A killing in a protected building may be perfidious, or may be an extrajudicial killing, or may simply be an act of murder. In short, even where conduct can be attributed to the State, it may be challenging to establish that said conduct actually breaches one or more of the State's international obligations.

Which State(s) can invoke responsibility?

Assuming that a wrongful act can be identified, this is only the first part of the law of responsibility. Responsibility is not autonomous; it must be invoked. There are two avenues for invocation: either it must be invoked by an injured State, or the obligation breached must be of an erga omnes or erga omnes partes nature. The invocation of responsibility also presents challenges when considering the conduct of civilians who directly participate, for three reasons:

1. 1. If the “injured State” approach is taken, only the State(s) actually injured may invoke responsibility or lawfully implement countermeasures in order to incentivize returning to compliance. However, the forms of reparation available are wider.

2. 2. If the “erga omnes (partes)” approach is taken, then while any State may invoke responsibility and enact countermeasures, the forms of reparation available amount to, in effect, primarily declaratory relief.

3. 3. The route by which responsibility is invoked also depends in no small part on the conduct which is alleged to constitute a violation. In a NIAC, for example, it is difficult to envisage conduct that specifically “injures” another State, within the meaning of Article 42 of the DARSIWA. This presents an additional complication because depending on the context, it may not be possible to invoke responsibility for anything except perhaps the most high-level failures relating to the conduct and due diligence obligations of the State.

Invocation by an injured State

“Injury” is defined in Article 31 of the DARSIWA as referring to “any damage, whether material or moral, caused by the internationally wrongful act of a State”.¹¹³ The breached obligation constituting the wrongful act must either have been owed to the injured State directly, or to a group of States (including the invoking State), or to the international community as a whole.¹¹⁴ An obligation may be simultaneously bilateral and multilateral. This is the most traditional avenue by which responsibility is invoked. Sassòli has previously convincingly argued that the only “injured” State(s) in relation to IAC are those which either experience violations of IHL on their territory or which can exercise diplomatic protection over victims of violations.¹¹⁵ In this view, only States to which an obligation is bilaterally owed, or specially affected States, may be considered “injured”. This is because, as Sassòli notes, the criterion contained in Article 42(b)(ii) of the DARSIWA is related to a very specific category of obligations¹¹⁶ – specifically, those where “each party's performance is effectively conditioned on and requires the performance of each of the others”.¹¹⁷ This includes, for example, disarmament treaties;¹¹⁸ it does not apply to IHL, where both the core treaties and the customary law are explicitly non-reciprocal.¹¹⁹ Arguments that violations of common Article 1 somehow extend the notion of “injury” to all States party to the Geneva Conventions and/or their Additional Protocols are inconsistent with the drafting decisions, including State input, that led to the DARSIWA.¹²⁰ Consequently, invocations of responsibility via injury, relating to obligations owed to a group or the international community as a whole, should only be considered, at most, as valid for specially affected States.

Overall, the category of States actually “injured” by most violations of IHL during an IAC is rather narrow. For most violations, only a small number of States – often only one State, when a conflict is strictly between two belligerents – can viably claim to be “injured”. Injured status relies on a breached obligation, which, in turn, means that it is inextricably linked to the question of whether conduct can be attributed. As previously discussed, this is exceptionally challenging with regard to encouraged direct participation. The challenge of identifying an injured State is even more sharply pronounced in NIAC contexts, particularly those occurring within a single State rather than across borders. In such situations, both the author of the alleged wrongful act and the victim of the injury may well be the same State. This risk is particularly pertinent given that the vast majority of contemporary armed conflicts are non-international in nature.¹²¹

If an injured State (or group of States) can be identified, the full range of reparation within the law of international responsibility is available – restitution, compensation and satisfaction.¹²² Identification of an injured State is also a necessary preliminary step for the lawful implementation of countermeasures.

Invocation erga omnes (partes)

Obligations erga omnes and erga omnes partes are somewhat different. Rather than being owed bilaterally, they are owed to an entire group, whether that be the international community as a whole (erga omnes) or an identifiable subgroup (erga omnes partes).¹²³ The two forms are addressed in tandem here for two reasons. First, the vast majority of IHL treaty rules are also customary in nature. If their treaty form is an obligation inter partes (as opposed to a purely bilateral obligation), their customary form is almost certainly also erga omnes. Secondly, although the two concepts bind different groups, the process for identifying such obligations does not vary. The identification process has two limbs:

1. 1. An international obligation exists which is binding upon the State.

2. 2. That international obligation is one in which “all states [or States Parties, for obligations erga omnes partes] can be held to have a legal interest”;¹²⁴ that is, it protects a common community interest.

As far as IHL is concerned, the first element is generally uncontroversial (with the limited exception of certain obligations contained within the Additional Protocols). It is less certain precisely which rules of IHL are owed to the community as a whole. It is generally agreed that common Article 1 has an erga omnes character, both erga omnes partes – as concerns its enforcement between signatories to the Geneva Conventions and their Additional Protocols – but also as a “pure” obligation erga omnes, because a free-standing dyad of customary obligations, namely to respect and ensure respect for IHL, is merely given “specific expression” by the article.¹²⁵

It is not clear, however, which other rules of IHL are of an erga omnes character. The International Criminal Tribunal for the former Yugoslavia (ICTY), for its part, has stated that IHL's non-reciprocal nature means that the obligations contained within it are necessarily obligations erga omnes. ¹²⁶ The fundamental principle of distinction and the prohibition against unnecessary suffering have been recognized by the ICJ in its Advisory Opinion on the Legality of the Threat or Use of Nuclear Weapons as “cardinal” in nature – itself not a controversial statement, as these rules are indeed the core of IHL, and compliance with them is in the interest of all States.¹²⁷ The ICJ then proceeded to state that “a great many rules of humanitarian law applicable in armed conflict are so fundamental to the respect of the human person and ‘elementary considerations of humanity’ … because they contain intransgressible principles of international customary law”.¹²⁸ In Legal Consequences of the Construction of a Wall in the Occupied Palestinian Territory, the Court repeated this position before then commenting solely on common Article 1.¹²⁹ In Legal Consequences Arising from the Policies and Practices of Israel in the Occupied Palestinian Territory, the ICJ repeated this language, and stated that it should be interpreted as describing “obligations which are essentially of an erga omnes character”.¹³⁰ In the same Advisory Opinion, it referred to “certain … obligations under international humanitarian law” as being of an erga omnes character, but did not mention any specific obligations enjoying this character.¹³¹

The advisory jurisprudence of the ICJ appears to suggest that many rules of IHL are, at the least, obligations erga omnes, if not obligations jus cogens; it is difficult to disentangle the Court's treatment of the two concepts. The ICJ has in any event not elaborated on precisely which particular rules are of, at the least, an erga omnes character. This approach resonates with that taken by the ICTY, but is distinct; the ICTY jurisprudence on the matter would have the logical end effect of rendering all rules not dependent on reciprocity obligations erga omnes, whereas the ICJ's jurisprudence suggests many, but not necessarily all, such obligations are of an erga omnes character.

Scholarly opinion on the matter is varied. The most progressive position is that all customary international law is, by its very nature, erga omnes – accordingly, all rules of IHL would also be of an erga omnes character.¹³² Some IHL scholars consider that by virtue of common Article 1's erga omnes nature, all rules of IHL are also obligations erga omnes; they flow from the overarching rule.¹³³ Others temper their approach somewhat, instead referring to “most norms of IHL” as erga omnes;¹³⁴ others are more cautious still, suggesting that only a “general obligation of guarantee” is, with certainty, an erga omnes obligation.¹³⁵ The ICRC position is also cautious; it holds that the Geneva Conventions, through common Article 1, create obligations of an erga omnes partes nature.¹³⁶ In the ICRC Customary Law Study, the ICRC references the claim that all of IHL is comprised of obligations erga omnes only in passing.¹³⁷

All told, the farther that one strays from the very core of IHL, the more difficult it is to convincingly contend that a rule is erga omnes in character. The basic humanitarian obligations of respect and ensuring respect, and the core humanitarian principles of distinction and the prohibition of unnecessary suffering,¹³⁸ certainly are; beyond this, courts and scholarly authorities alike have generally refrained from identifying specific erga omnes norms, instead asserting that an as-yet-undefined portion of the law has this character. Attempts to trigger State responsibility through Article 48 are therefore most likely to succeed where they appeal to IHL's core obligations and principles – more progressive claims are less likely to be accepted. This again speaks to the difficulty of holding States to account for the actions of individuals who directly participate in hostilities: even if conduct can be attributed to the State, and may be clearly wrongful or reprehensible, if the relevant obligation is not one erga omnes, then only an injured State may lawfully invoke responsibility. This issue is particularly salient in NIAC contexts.

Where an obligation is indeed owed erga omnes, the issue of remedy becomes pertinent. Breaches of obligations erga omnes do not trigger the same obligation for a breaching State to make reparation as breaches of bilateral obligations; instead, they are described as triggering an obligation to make reparation “in the interest of the injured State or of the beneficiaries of the obligation breached”.¹³⁹ The DARSIWA commentaries are clear that this is a “progressive development”, insofar as it concerns the obligation for States to act on behalf of the injured party (whether that be a State, or a beneficiary of the obligation).¹⁴⁰ It is not clear whether State practice and opinio juris have, as yet, transformed this act of progressive development into a customary rule. Consequently, if an obligation of this type is breached, in principle only Article 48(2)(a) of the DARSIWA, which incorporates Article 30, is certain to be triggered – the breaching State must cease the wrongful act and offer, if appropriate, assurances and guarantees of non-repetition.¹⁴¹

It is possible that some – but not all – violations of obligations erga omnes will also trigger the rules contained within Articles 40 and 41 of the DARSIWA, relating to jus cogens norms. This is almost certainly the case for violations of the core principles or “basic rules” of IHL¹⁴² – in particular the principle of distinction – but again, as one moves away from the core of IHL, rules become less likely to have a jus cogens character.¹⁴³

Conclusions on invocation

It is clear that even beyond the question of attribution and identification, the invocation of State responsibility is also a challenge. A civilian who takes up arms may commit acts that are morally or ethically repugnant, but those acts may nonetheless not be internationally wrongful. Even if they are, and even where they can be attributed to a State, invocation of responsibility may be exceptionally challenging – especially in NIACs. While obligations erga omnes may appear to be an appropriate solution to this difficulty, the precise scope of erga omnes obligations under IHL remains unclear; the farther a claim strays from the core rules, the higher the likelihood that it will fail. Even if a claim succeeds, erga omnes invocations trigger only a limited obligation of cessation and non-repetition, and do not make good the wrong.

The limits of responsibility

The above discussion has demonstrated that the law of international responsibility is limited in its ability to respond to the acts of private individuals, even where such acts are encouraged or abetted by the State. This is because:

• • not every act can be attributed to the State;

• • not every attributable act is actually wrongful;

• • even where an act is wrongful, it does not necessarily follow that a State has the right to invoke responsibility; it must either be injured, or the obligation must be owed erga omnes; and

• • if the obligation is owed erga omnes, only a limited obligation of cessation and non-repetition is certain to arise. It is not altogether settled whether a State can demand reparation on behalf of another injured State or the beneficiary of an obligation.

In the context of IHL, these limitations on responsibility are concerning. Scholarship has previously identified, for example, that the actions of hackers may not be easily attributable;¹⁴⁴ consequently, efforts to constrain or regulate cyber warfare have primarily focused on ensuring that States appropriately inform and train those individuals who choose to support hostilities in this manner.¹⁴⁵ Focus has been placed not on State responsibility for the harmful acts actually perpetrated by civilian hackers, but instead on the obligation of the State to prevent violations of the law – that is, the overarching obligation to “ensure respect” for the law.

This concern holds true also for physical direct participation by civilians, even where participation is encouraged or facilitated by the State. Take, for example, a NIAC wherein the State encourages individuals to take up arms in defence of their communities. Such civilians, acting outside of a responsible command structure, typically with minimal or no training, are at markedly increased risk of violating IHL – but their acts are challenging to attribute to the State. The same is also true in IAC, for example where a State liberalizes gun ownership laws or encourages civilians to defend their homes with force. Consider also the example of a State which, following allegations that its civilians who are directly participating in hostilities have violated IHL, declines to investigate or prosecute their conduct. In such cases, the State's action encourages direct participation, but this does not, in itself, mean that the State bears legal responsibility for such conduct. The only wrong for which it bears responsibility, with certainty, would be the act or omission from which the other wrongs flow – that is, its possible failure to “ensure respect” for IHL. Because this is an obligation erga omnes, depending on the factual situation in which responsibility is invoked it may also be that States are strictly limited in their ability to seek a return to compliance with IHL: this is because obligations erga omnes do not, as a matter of customary international law, trigger a right to implement countermeasures, or the obligation of reparation enforceable by States other than an injured State.¹⁴⁶ The law of State responsibility was simply not designed to deal with the diffusion of an essentially sovereign act – the waging of war – among the civilian population.

Concluding remarks

Recommendations

Throughout this article we have identified the following risks arising from encouraged civilian participation in hostilities:

1. 1. In modern warfare, the distinction between civilian and combatant is already being blurred in practice. Calls for civilians to engage in acts which support the armed conflict erode this distinction further, weakening what is arguably the core principle of IHL.

2. 2. The more that civilians are engaged in the conflict, the greater the chance that their legal rights are affected. Not only do they risk being (lawfully) targeted while directly participating in hostilities, but carrying out acts in support of one belligerent, even acts falling short of direct participation, also exposes them to possible internment on imperative security grounds.

3. 3. Encouraging civilian participation creates a risk of the defender violating their passive precautionary obligations.

4. 4. More generally, encouragement of participation may violate common Article 1 and its customary equivalent, particularly where civilians are not informed of their obligations when participating in hostilities. Similarly, encouraged participation may result in a State breaching the rights of its own nationals.

5. 5. With rare exceptions, there is little (lawful) tactical advantage to be gained from encouraging civilians to engage in armed conflict. This minimal advantage is accompanied by a severe human cost and an erosion of the principle of distinction.

6. 6. Encouraged direct participation, in particular, increases the risk of IHL violations. At the international level, the law of State responsibility should be the preferred tool for responding to such violations; however, it is not adequately equipped to do so for various reasons:

   • • The law of responsibility establishes a high threshold for the attribution of private acts to a State.

   • • Invocation of responsibility relies on either the existence of an injured State or violation of an obligation erga omnes.

   • • The challenges associated with attribution mean that more likely than not, the obligation erga omnes to “ensure respect” for IHL is the avenue by which responsibility could be invoked. It is not settled whether invocation of responsibility erga omnes, as a matter of law, entitles States to implement countermeasures; it also remains unclear whether, as a matter of customary law, such invocation triggers an obligation of reparation enforceable by non-injured States. States therefore have a limited range of means available to induce (re)compliance with the violated obligation.

These risks are, sadly, not hypothetical. Civilian involvement – encouraged, tolerated or otherwise – is a fact of modern warfare, and the core principles of IHL are already under constant strain; it is a question of when, not if, civilian actors will be accused of perpetrating widespread violations of IHL. One may argue that international criminal law is the appropriate response to this concern, and in part, one would be correct.¹⁴⁷ However, State responsibility serves the essential purpose of reaffirming that States – not only the individuals who comprise them – are governed by and accountable to the law. International law's integrity rests just as much on States being held responsible for their breaches of obligation as it does on effective and visible administration of justice for individual criminal conduct.

What, then, should States do? We close by suggesting a small number of practical, realistic steps that can be taken to manage the risks associated with civilian participation, and also suggest that structural shifts may be needed in the law of responsibility.

1. 1. As a matter of principle, States should not encourage civilians to become involved in armed conflict, except as lawful combatants. If a State finds itself in need of combatants, it has avenues available to it; there is no general prohibition in international law, for example, against conscription

2. 2. When States do encourage or otherwise facilitate civilian involvement in hostilities, they must do so in a manner that complies with their international obligations. Civilians should be made fully aware of the risks they are assuming through their actions, and should also be made fully aware of their legal obligations – in particular, respect for the principle of distinction and the basic regimes of protection in IHL. This necessarily requires States to fulfil their obligations to disseminate IHL.

3. 3. Where a State encourages DPH, it has an ongoing obligation to ensure respect for IHL. To meet its due diligence obligations, it should implement ongoing measures to deter violations of the law by civilians, and to encourage civilians who directly participate to distinguish themselves from the wider civilian population. It necessarily must also investigate and, where appropriate, prosecute alleged breaches of IHL.¹⁴⁸

4. 4. At a structural level, States and others who contribute to the development and adaptation of international law should consider whether the attributive modes of State responsibility are sufficiently flexible in situations of armed conflict. State legal offices would benefit from taking a clear position on Article 48(2)(b) of the DARSIWA, both in general and in the specific context of armed conflict; its uncertain legal effect contributes to a genuine risk of impunity for acts committed at arm's length from the State, because it renders meaningful reparation challenging at best.

5. 5. Until structural change is achieved, States should exercise the measures available to them to ensure collective respect for IHL. States may, for example, wish to publicly condemn the involvement of civilians in armed conflict, or to pass domestic legislation prohibiting the taking up of arms by civilians in conflict situations except in cases of levée en masse. They may also resort to measures of retorsion, or use other established legal mechanisms to ensure that the involvement of civilians in hostilities is neither accepted nor tolerated, neither overtly nor tacitly.

Avenues for future research

The impact of civilian involvement on post-conflict justice

While this article has focused on the implications of civilian involvement in conflict throughout the duration of the conflict, the consequences of such involvement extend beyond this period. In many of the previously mentioned contexts, authors have specifically highlighted the implications that civilian involvement in conflict has on subsequent efforts at reconciliation. These include a strengthening not only of the ethnic and religious dimensions of the conflict¹⁴⁹ but also of tribal rivalries,¹⁵⁰ deepening societal divides in general. Furthermore, in addition to its implications on the security of the society, arms proliferation has an important legal dimension as well – in fact, control over the availability of arms within a society is one component of the State's duty to protect the right to life.¹⁵¹ Complying with this duty would require the State to introduce measures addressing whether civilians who had previously taken up arms during the conflict could unconditionally retain the weapons in question, and if not, organizing their return.

Belligerent equality and obligations of non-State armed groups

This question has an NSAG perspective as well – namely, the variety seen in the examples discussed in this article showcases that civilian involvement in armed conflict is a phenomenon that spans across both international and non-international armed conflicts. In fact, many of the examples involving NIACs also encompass the practice of both States and NSAGs.¹⁵² This article has primarily examined the obligations incumbent upon States, in large part because of its analysis of the law of international responsibility – a regime which is subject to debates in the context of NSAGs.¹⁵³

There are reasons why certain aspects of this analysis would differ in the case of NSAGs. A question that is part of this picture, but falls outside the scope of this article, is the meaning of belligerent equality in this regard. It should be examined whether belligerent equality would prompt us to transpose this article's conclusions in relation to States’ obligations to NSAGs. In other words, encouraging engagement of civilians in military activities could then also be seen as irreconcilable with NSAGs’ obligations to take precautions against the effects of attacks and to ensure respect for IHL. Nonetheless, such a conclusion could be at odds with obligations that could realistically be met by NSAGs. NSAGs are characterized by very fluid concepts of membership, and often arise out of the civilian population or are structured in a way that makes them very tied to the communities in which they are based, precisely due to the support they draw from the local population.¹⁵⁴ Therefore, perhaps a more nuanced analysis of NSAGs’ obligations would be more appropriate, including by adopting the sliding scale of obligations approach.¹⁵⁵