International Law and Governance by Armed Groups: Caught in the Legitimacy Trap?

ABSTRACT

Empirical research shows that armed groups may provide governance services to civilians where they have established some form of control over a territory or population. At the same time, armed groups often pursue policies designed to enhance legitimacy in order to strengthen their claim to power. This gives rise to the question: Does the virtuous governance circle, to which this special volume is dedicated, also apply to governance efforts on the part of armed groups and which conditions must be fulfilled in order for the virtuous circle to operate? The article highlights that the contradictory demands of the multiple audiences of legitimacy with which armed groups need to communicate, create a dilemma that may interrupt the virtuous circle. Bringing together international legal and political science discourses on armed group governance, the article argues that the state-centric fabric of the international legal framework contributes significantly to this process.

KEYWORDS:

• Armed groups
• legitimacy
• international humanitarian law
• non-international armed conflict
• rebel governance

1. Introduction

Political science literature has identified a so-called virtuous circle of governance: the more legitimate a political actor or institution is, the more effective it becomes, and the more effective it is, the more legitimate it becomes (Levi and Sacks 2009; McLoughlin 2015). In cases of governance in areas of limited statehood, to which this special issue is devoted, the question arises whether such a virtuous circle also applies to non-state actors, such as armed groups. In areas of limited statehood, the state lacks the ability to make and implement decisions and/or lacks the monopoly over the use of force. However, these areas are in general not ungoverned. Instead of the state, non-state actors, including violent ones, often provide a panoply of goods and services to the local population (Risse 2011; Risse and Stollenwerk 2018); in the case of armed groups, frequently, but not exclusively, in the realm of security governance. Forms of security governance magnify the complexities of the interactions between effectiveness and legitimacy (on these concepts see Schmelzle and Stollenwerk 2018). International law contributes to these complexities in an important yet underresearched manner: security governance may, on the one hand, include elements of law-making and law enforcement. On the other hand, international law, in particular, the laws of war, provide a legal framework which affects and purports to direct, the behaviour of all actors involved in security governance. This article aims to bring together the discourses on armed group governance in political and legal science and thereby explain situations in which the virtuous circle of governance may be interrupted due to the state-centred fabric of international law.

The debate about new wars and asymmetric warfare (Kaldor 2013; Münkler 2005) created the impression, in particular, amongst international lawyers, that conflicts with or between armed groups are dominated by random violence exerted by irrational actors or actors which use violence against civilians as an end in itself (Bassiouni 2008, 765; Lamp 2011, 242). Taking this perspective, armed groups are not likely to provide security governance, including engagement with international humanitarian law, because the laws of war aim to direct the conduct of war by states and not the military strategies of armed groups (Coggins 2015, 104; Lamp 2011). However, in international law and political science, the debate has moved on. International legal scholars have increasingly focussed on the applicability, adaptability and compliance-pull of international humanitarian law in non-international armed conflicts (Bellal 2016; Fortein 2017; Krieger 2015; Rodenhäuser 2018). Political science research surrounding rebel governance has demonstrated that armed groups may provide governance structures where they have established some form of control over a territory or population (Mampilly 2011; Mampilly 2015a, 46; Weinstein 2007; Wood 2015, 21–24). In political science debates, the importance of legitimacy for rebel governance (inter alia: Arjona 2017; Mampilly 2015b; Podder 2017; Schlichte and Schneckener 2015), including rebel diplomacy employed to bolster legitimacy claims, (Coggins 2015; Huang 2016b) has been emphasized. Thereby, the debate identifies a puzzle. While, on the one hand, rebel diplomacy may boost an armed group’s capability to provide governance, on the other hand, there are concerns that ‘outward looking groups tend to invest less energy in developing governance foundations [and] are likely to be more abusive in their relationship with civilians’ (Podder 2017, 700; see also Coggins 2015, 115).

While these contributions provide important insights into the dynamics of armed group governance, they do not dig deeper into the impact of international law on the interplay between effectiveness and legitimacy of armed group governance. The present article aims to fill this gap. By drawing on both disciplinary discourses, it examines the ambiguous influence which international law exerts on the effectiveness and legitimacy of armed group governance efforts. The article aims to rebut the assumption that the impact of international humanitarian law on ‘rebels’ participation in diplomacy [… is] limited and rarely realized in practice’ (Coggins 2015, 102) and offers an explanation as to why the effects of armed group engagement with international humanitarian law, in regards to governance, vary. The article thus contributes to an explanation as to why, under certain circumstances, armed groups that have externally engaged with the international community and provided internal governance structures, may at a certain point in time nonetheless turn against the civilian population.

From the perspective of the state, armed groups which contest the ruling government and its legal monopoly over the legitimate use of force, are an illegal and illegitimate phenomenon. Armed groups cannot rely on formal sources of legitimation available for state actors. Instead, they need to utilize other strategies aimed at various audiences in order to consolidate their legitimacy. Since armed groups operate under a constant threat of violence, further conditions need to be fulfilled so that their governance efforts are perceived as effective and therefore increase their legitimacy in the eyes of their supporters, local and national as well as international audiences. International humanitarian law is one instrument which armed groups can utilize in this process. However, even if the most important conditions for the virtuous governance circle to operate are fulfilled, in the case of certain types of armed groups, the causal link between legitimacy and effectiveness may still be interrupted by the contradictory demands of the multiple audiences of legitimacy. International audiences, such as third states and international organisations, are legally bound by international law. The state-centred fabric of the international legal order opposes efforts to support armed groups outside the context of decolonization and aims to obstruct efforts to enhance their legitimacy. As a consequence, any possible gain in empirical legitimacy amongst international audiences remains legally contested and thus precarious. This precariousness increases the likelihood that armed groups lose their interest in effective governance efforts to increase their legitimacy, even in the eyes of local audiences under their control and turn to other violent strategies. Changes in the União Nacional para a Independência Total de Angola’s (UNITA) governance efforts in the Angolan civil wars exemplify this.

In order to explore these interactions, the article will first interrogate the prominent role legitimacy plays for armed groups (1). It will then examine how the virtuous circle could work in principle in relation to armed group governance and which conditions need to be fulfilled in order for it to operate (2). The article then turns to the challenges which disrupt the causal pathway from legitimacy to effectiveness (3) and ends with a number of policy considerations (4).

2. Armed groups’ strategic approach to legitimacy

Legitimacy is an essential prerequisite for effective governance in areas of limited statehood: ‘Without empirical legitimacy, there is no successful governance’ (Risse and Stollenwerk 2018, 403; Krasner and Risse 2014). In the case of armed groups, the significance of legitimacy strategies results from two dilemmas: the need to distinguish between legality and legitimacy (1.1) and the need to satisfy different audiences of legitimacy with diverging normative expectations (1.2).

2.1. The need to distinguish between legality and legitimacy

The first dilemma arises from the very nature of armed groups as violent non-state actors. Armed groups challenge the state and its institutions through an illegal use of force. Such groups contribute to dissolving the public order and to diminishing the state’s capabilities. Therefore, governments consider armed groups as illegal as well as illegitimate. The characteristics of armed groups exclude them from any formal source of legitimation available for state actors, in fact, from any form of legitimacy based on legality. International law does not grant armed groups formal legal status unless states have legally recognized them as belligerents or national liberation movements. International humanitarian law does not privilege fighters of armed groups as it refuses to grant them the status of a combatant. Unlike combatants in international armed conflict, members of armed groups can be criminally prosecuted by the de iure government based on their simple participation in a non-international armed conflict (Sassòli and Shany 2011).

Since justificatory discourses based on legality are not available for armed groups, legitimacy as a category distinct from legality gains strong significance for them (Schlichte and Schneckener 2015, 414):

In the international realm, only state actors have the power to define what is legal and what is illegal. From their point of view, legitimacy should first of all be derived from legal recognition. In other words, state actors have an interest in leaving no room in between legality and legitimacy, but armed groups have to argue that there is a crucial difference between these concepts.

Of course, the characteristics, capabilities and (political) goals of armed groups vary considerably. From an empirical perspective, there is a wide range of groups which can be qualified as violent non-state actors. Likewise, international law does not offer a generally agreed upon definition. The detailed categorization is often disputed (Müller, forthcoming), not least because categories will overlap. If one applies an understanding which combines empirical findings with normative-legal qualifications, the following entities, amongst others, will fall under the concept of ‘armed group’: de facto regimes/de facto governing authorities, i.e. territorial entities which are not recognized as states but exercise effective governmental authority and have thus acquired a limited legal status; national liberation movements as a historical, legally recognized category; armed opposition groups and warlord configurations with or without emancipatory social, political or ethnical goals; paramilitary groups or militias acting as irregular combat units which have some support of a government but are not integrated into the state’s security structures; vigilante or self-defence groups with a low degree of organisation and the primary goal of temporarily defending themselves; territorial gangs without political purpose which strive to control a territory for pursuing criminal activities (Bellal 2016, 45).

Not all of these entities will provide security governance nor strive for legitimacy. Groups which do not hold territory or lack any stable structures (Jo and Bryant 2013, 242; Kasfir 2015, 23) will, in general, not benefit from the virtuous circle of governance because they lack the necessary capacity. Militias may provide basic governance services to the local population, but their politics of legitimacy may differ from those of other armed groups since they may enjoy some form of ‘borrowed legitimacy’ from the government which sponsors them (Schneckener 2017, 799). Groups which merely exist for criminal purposes are often not interested in legitimacy since they do not aim to acquire any political, let alone legally conferred, status. This includes territorial gangs particularly those which are short-lived because they do not intend to provide any form of governance for the population (Bellal 2016, 52).

Primarily, for those armed groups which consider themselves as actors with long-term political goals, legitimacy is a strategic resource necessary to the pursuit of their political cause (Bellal 2016, 135 et seqq.; Seymour 2017, 818; Stewart 2018, 210). Political science debate has labelled such groups as rebel groups, i.e. ‘those who try to overthrow the state, insurgents who fight for secession, or those who force the government of an incumbent state to reform’ (Kasfir 2015, 24; see also Huang 2016b; Podder 2017, 699). While rebel groups represent the main type of group which falls under the present analysis, the term ‘armed groups’ is preferable, on the basis that it is normatively more neutral than the term ‘rebel’. Since international humanitarian law deliberately refrains from evaluating the political cause pursued by these entities (except in the case of national liberation movements in the context of decolonization), it adopts the term ‘armed groups’ in the pertinent legal texts, such as Article 1 (1) of Additional Protocol II to the 1949 Geneva Conventions (AP II). Moreover, there is evidence that other types of armed group may provide security governance and engage in diplomacy efforts (Kasfir, Frerks, and Terpstra 2017, 259). Examples include Afghan warlords (Malejacq 2017) or fragmented armed groups, such as in the 2016 civil war in South Sudan (Kindersley and Rolandsen 2017) or in Syria. Finally, efforts to enhance claims to legitimacy are relevant for a broader array of violent non-state actors than those captured by the term ‘rebel groups’. Most armed groups qualify as political actors and, therefore, have to pursue politics of legitimacy. This requires them to build up arguments and to engage with norms and values to support their claims, even if these claims are self-centred and partial (Schlichte and Schneckener 2015, 412).

2.2. The need to reconcile different audiences of legitimacy

As a strategic resource, legitimacy requires armed groups to communicate their agenda to a variety of different audiences (Huang 2016a; Mampilly 2015b, 84–87; Podder 2017, 687). In the case of armed groups, three different types of audience can be distinguished: Armed groups must address their active followers as well as local, regional or national communities and the international community, in particular, third states, international organisations, and NGOs. In order to maintain stable control and achieve political goals within their territory, support and obedience on the part of their followers as well as local, regional or national communities are of utmost importance (Metelits 2010; Schlichte and Schneckener 2015, 419).

At the international level armed groups aim to achieve social validation and legal recognition which allows them to participate in the social and legal practices of the international community (cf. Zaum 2016, 1115–1116). Here, international law creates obstacles. In principle, armed groups, as violent, non-state actors do not enjoy full legal personality. They are not entitled to conclude treaties with states or international organisations. As a general rule, they neither have standing before international courts nor observer status within international organisations. Thus, their ability to communicate with international audiences is restricted. Still, under international humanitarian law, armed groups become bearers of international obligations based on their effective territorial control. Where a state has lost control over territory and people, Common Article 3 of the Geneva Conventions I-IV (GC I-IV) and AP II foresee that armed groups are legally responsible for actions pertaining to non-international armed conflict within their area of control. As bearers of duties under international humanitarian law, armed groups can seize the opportunity to utilize this position as a starting point for conducting what political science literature has called ‘rebel diplomacy’ (Coggins 2015, 105): ‘a rebel group’s conduct of foreign affairs during civil war for the purpose of advancing its military and political objectives’ (Huang 2016b, 90). Since the international community has, inter alia, a legally induced interest in armed group compliance with international law, a willingness to engage with them may arise, offering armed groups an opportunity to communicate with international audiences – an engagement from which armed groups may benefit in terms of enhanced legitimacy (Coggins 2015, 106; Mampilly 2015b, 94).

Armed groups are mindful of their need for multilevel legitimacy (Mampilly 2015b, 84–86) although the extent to which a group will strategically communicate with international audiences apparently still depends, to some extent, on the existence of a political wing (Jo and Bryant 2013, 245; Podder 2017, 693) and the overall political objectives of a group. Political wings were, for instance, instrumental in the cases of the United People’s Front in Nepal and the Sudan People’s Liberation Movement in Sudan (SPLM/A) (Coggins 2015, 112). As a means of communication, the names of armed groups often refer to international norms. A case in point is the Forces pour la Défense des Droits Humains, an armed group in the Democratic Republic of Congo (DRC). Some armed groups have engaged with regional or international powers as well as international organisations or NGOs on matters of compliance with international humanitarian law (Jo 2015; Podder 2017, 699–700; Schlichte and Schneckener 2015, 419–420). For instance, the Rassemblement Congolais pour la Démocratie in the DRC, the United Somali Congress faction in Somalia, or the SPLM/A in Sudan have all collaborated with the International Committee of the Red Cross (ICRC) (Coggins 2015, 115; Jo and Bryant 2013, 244; Mampilly 2015b, 75).

However, the need to communicate with different audiences gives rise to a second dilemma: The causal pathway from governance effectiveness to legitimacy demands that ‘the conditions for linking effective governance with more legitimacy have to be met by as many legitimacy audiences as possible’ (Schmelzle and Stollenwerk 2018). Legitimacy amongst one audience may contribute to increasing effectiveness and thus legitimacy amongst another audience. However, since legitimacy is based on the collective beliefs of a particular audience and its normative points of reference (Zaum 2016, 1109), the diversity of audiences of legitimacy and their normative beliefs entails the risk that perceptions surrounding legitimacy may differ or even be diametrically opposed. Where targeted audiences are characterized by considerable differences in terms of interests, expectations and normative reference points, reconciling claims to legitimacy addressed to a variety of audiences is difficult to achieve. The more expectations of audiences of legitimacy conflict with one another, the more they limit an armed group’s ability to manoeuvre politically, as well as its chance to gain stability. Failure to address these rifts will contribute to a constant challenge to the group’s legitimacy and its claim to power (cf. Zaum 2016, 1120–1121; Gerdes 2015).

Both dilemmas described above impact the virtuous governance circle. The ambiguous effect the use of violence has on legitimacy reveals that certain conditions need to be fulfilled in order for the virtuous circle to operate (2). The legitimacy dilemma arising from the need to address audiences with conflicting normative expectations may nonetheless interrupt the virtuous circle (3).

3. Operating the virtuous circle

In order to analyze the impact of international law on the effectiveness and legitimacy of armed group governance, it is particularly important to take into consideration the following aspects of the virtuous circle: rule quality, provision capacity, and performance-based legitimacy beliefs and shared values (3.1). Engaging with international humanitarian law may be an important instrument in this process (3.2).

3.1. From effectiveness to legitimacy?

Literature about areas of limited statehood defines governance as institutionalized modes of social coordination to produce and implement collectively binding rules or to provide collective goods (Risse 2011; Risse and Stollenwerk 2018). These modes of social coordination are also employed by armed groups where they provide for ‘the organization of civilians within rebel-held territory for a public purpose [such as] rebel encouragement of civilian participation, provision of civilian administration, or organization of civilians for significant material gain’ (Kasfir 2015, 24). Providing for security and public order is a central element of armed group governance (Jentzsch 2017, 326; Kasfir 2015, 36). The Liberation Tigers of Tamil Eelam (LTTE), for example, established a police force and a judiciary which enjoyed legitimacy among local constituencies (Mampilly 2011, 116; Terpstra and Frerks 2017, 280). The National Resistance Army in Uganda held, inter alia, elections for village officials during the civil war it had started in 1981 (Kasfir 2005).

Even where armed groups provide governance infrastructure they constantly operate under the threat of violence (Kasfir 2015, 25). Thus, effective governance by armed groups only contributes to their legitimacy if the ambiguous effect which violence exerts on legitimacy is taken into account. While effective control by armed groups entails a demonstration of power, their legitimacy may be weakened if civilians within territory they control are exposed to indiscriminate violence over a long period of time. There is, further, an inherent danger of violent escalation which may simultaneously reveal a certain lack of control on the part of the armed group evident to its constituency:

Mass killings, atrocities, sexual violence, human rights abuses, bloody internal rivalries and massive counter-violence by other actors (in particular the state) all seriously undermine leaders’ credibility and diminish the chances that their claims will be seen as legitimate. One can often observe a downward spiral which is characterized by intensified violence and radicalization on the one hand and by a rapid process of delegitimization on the other hand. More violence leads then to less legitimacy, less legitimacy leads to even more violence and so on. (Schlichte and Schneckener 2015, 416)

In order to consolidate a claim to power then, it becomes necessary to translate the military control of an armed group into a more institutionalized form of exercising power. For this purpose, the virtuous circle model demonstrates that additional conditions need to be fulfilled in order for the circle to operate. Several conditions related to the ‘virtuous circle’ are particularly important in order to analyze the impact of international law on the effectiveness and legitimacy of armed group governance: rule quality and provision capacity, i.e. the capacity of armed groups to provide governance functions; performance-based legitimacy beliefs among the governance audience; and the extent to which goals and values are shared between rulers and ruled.

In their model of the virtuous circle, Schmelzle and Stollenwerk (2018) have highlighted that rule quality and governance provision capacity comprise,

… a condition necessary for the linkage between compliance and effectiveness. In order for compliance and cooperation to have the assumed positive impact on governance effectiveness, governance actors need to codify sensible rules, which, if complied with, can actually achieve their intended purpose. Additionally, governance actors must possess sufficient capacity to actually provide goods and services, if a satisfying level of cooperation is met.

The second condition concerns the existence of performance-based legitimacy beliefs (see Schmelzle and Stollwerk 2018):

This condition states that the effectiveness of a governance institution will only contribute to its legitimacy, if the governance audience at least partly bases its legitimacy beliefs on assessments of effectiveness. In other words, governance effectiveness must be a part of the audience’s conception of legitimacy, in order to establish a causal pathway from effectiveness to legitimacy.

Rule quality and provision capacity, which correlate with performance-based legitimacy beliefs among the local communities under the control of an armed group, contribute to the stability and coherence of the group’s governance. Law and law enforcement can be instrumental in this respect since the provision of order and security, as well as foreseeability and preventing discretionary decision-making are the predominant tasks of legal regulation. Such qualities allow for predictable and regular governance activities and are therefore important goods for the daily life of civilians (Kasfir 2015, 39), appealing to their legitimacy beliefs.

In particular, international humanitarian law offers sensible rules which armed groups can utilize for their security governance because these rules are explicitly tailored to protect the most basic security needs of civilians in times of protracted armed conflict. Common Article 3 of Geneva Conventions I-IV (Art 3 GC I-IV) and AP II to the 1949 Geneva Conventions oblige armed groups to respect and ensure certain rules of international humanitarian law. Art 3 GC I-IV contains a condensed version of the Geneva Conventions outlining rules regarding the humane treatment of all persons in enemy hands. The article explicitly forbids murder, mutilation, torture, cruel, humiliating and degrading treatment, the taking of hostages, and being submitted to an unfair trial. It obliges parties to a non-international conflict to collect and care for the wounded, sick and shipwrecked and grants the ICRC the right to offer its services. Obligations under AP II are more demanding. They include, inter alia, more detailed provisions surrounding the internment of persons whose liberty has been restricted (Art. 5 AP II) and in relation to criminal prosecutions (Art. 6 AP II).

Thirdly, in order to understand the impact of international law on the virtuous governance circle as it applies to armed groups, ‘shared social goals and values’ must be considered. According to Schmelzle and Stollenwerk (2018) ‘the effectiveness of a governance regime will only increase its legitimacy, if the rulers and the ruled share the same governance goals and social values, at least to a certain extent.’ International humanitarian law can claim to represent fundamental universal norms regarding warfare and human life, such as the principle of distinction between combatants or fighters and civilians, as well as the specific protection of certain vulnerable groups. Members of armed groups sometimes explicitly demonstrate their willingness to respect fundamental rules of humanity (Jo 2015, 49). Bellal articulates this point as follows:

According to a recent Geneva Call study on the perceptions of ANSAs [armed non-state actors] on humanitarian action, many ANSAs view themselves as being responsible for the fate of the civilian population under their control as exemplified by this statement of a KNU [Karen National Union] representative: ‘Although I don’t know all the rules, I do think we should take part and fulfil our responsibilities’. (Bellal 2016, 48 quoting Geneva Call 2016, 22)

Numerous studies show that local norms often reflect international humanitarian law. Some armed group codes of conduct include rules related to the protection of civilians, in particular, woman and children, which are rooted in local traditions, such as the norms adopted by the Farabundo Martí National Liberation Front (Jo 2016, 71) or the codes of conduct of the Dinka and the Nuer communities of South Sudan (ICRC 2018, 55).

Reference to universal norms, such as those contained within international humanitarian law, may even serve as a strategy where armed groups are confronted with internal ethnic or religious differences (cf. Mampilly 2015b, 92) since these rules appear to be intuitive and reflect a certain general moral agreement (Krieger 2015, 514). It may even be preferable for armed groups to pledge compliance to international humanitarian law, rather than international human rights law on the basis that the values contained in the former body of law are less likely to be seen as alien and therefore be rejected by local communities, whereas those contained within the latter are often perceived as being culturally more contested.

3.2. Engaging with international humanitarian law as a means to increase legitimacy?

Effective governance based on engagement with international humanitarian law may increase a group’s legitimacy in various ways according to different audiences concerned. Armed groups have utilized numerous instruments to signal their capability and willingness to comply with the laws of war. Instruments and measures employed are directed at the armed group’s followers, local audiences in the conflict region, as well as at international audiences, with legitimizing effects being intertwined.

At the local level, rebel courts act as a prominent tool for resolving disputes (Sivakumaran 2009; Willms 2015). For instance, in Afghanistan, the Taliban operate a court system with stationary and mobile judges and two stages of appeal (Jackson 2018, 19). Such courts may act as an alternative and more efficient dispute settlement system for civilians than that which was provided by the state. In addition, the establishment of courts by armed groups may be designed to maintain order amongst its ranks and impose penal sanctions, in particular, for war crimes perpetrated by its members, thereby aiming to strengthen the group’s accountability. Given that courts are an important symbol of state order, armed groups raise a stringent claim to legitimacy by effectively providing for criminal accountability (Jo 2016, 77). The creation of rebel courts is part of a strategy to use symbolic processes to bolster a group’s legitimacy. ‘By mimicking the behaviour of the modern state, rebels seek to discursively construct a political authority imbued with a comparable legitimacy enjoyed by national governments’ (Huang 2016b, 101; see also Mampilly 2015b, 77–78; Podder 2017, 692). Armed groups can thus reproduce and reinforce state symbols and practices which the population under their control is accustomed to, thereby relying on, and benefiting from, pre-existing legitimacy beliefs (Kasfir 2015, 40; Mampilly 2015b, 83).

Other important instruments, such as military codes of conduct issued by armed groups follow a comparable legitimacy strategy and address the relevant audiences by imitating the practices of state armed forces. The LTTE, for instance, established an armed force which closely resembled a regular state army, with a strong chain of command and a strictly enforced code of conduct. Breaches of the LTTE’s code of conduct were regularly seen to be punished (Terpstra and Frerks 2017, 287). Other pertinent examples include the 1947 Three Main Rules of Discipline and the Eight Points for Attention of the Chinese People’s Liberation Army, the 2009 Rules of Conduct by The Revolutionary Armed Forces of Colombia – People’s Army, otherwise known as FARC, and the 2011 Rules issued by the Libyan National Transitional Council (Amstel 2011). The Taliban code of conduct, the layha, developed from a list of rules on military discipline into an even broader code for accountable governance (Jackson 2018, 8). These codes enable armed groups to contain the use of force within their ranks, to improve security for local communities, and to communicate to the international community that they aim to abide by the law and to differentiate between legal and illegal use of force.

On the international level, armed groups can also issue or sign a variety of documents in order to demonstrate their commitment to international humanitarian law. Various forms of unilateral declarations – some carrying legal weight, others designed more to enhance legitimacy – also exist. For example, Art. 96 (3) Additional Protocol I to the Geneva Conventions (AP I) allows for the possibility of national liberation movements to issue unilateral declarations whereby the Geneva Conventions become applicable to the parties to the conflict. The 2015 unilateral declaration by the Frente Popular de Liberación de Saguía el Hamra y Río de Oro (Popular Front for the Liberation of Saguia el-Hamra and Río de Oro, known as POLISARIO) was the first declaration of this kind ever accepted by the Swiss government, as the depositary of the Geneva Conventions, and notified to member states. Furthermore, Deeds of Commitment established by Geneva Call, a non-governmental organisation, are a specific form of a unilateral declaration which armed groups can sign in order to demonstrate their intention to comply with certain norms embodied by the Deeds. Though the signing of Deeds does not constitute an international legal act, the process nevertheless, embodies high symbolic significance. Currently, three different Deeds of Commitment exist: the 2000 Deed of Commitment for Adherence to a Total Ban on Anti-Personnel Mines and for Cooperation in Mine Action signed by 49 armed groups; the 2010 Deed of Commitment for the Protection of Children from the Effects of Armed Conflict signed by 19 armed groups; and the 2012 Deed of Commitment for the Prohibition of Sexual Violence in Situations of Armed Conflict and towards the Elimination of Gender Discrimination signed by 16 armed groups (Bellal 2016, 50). In order to increase their legitimacy in the eyes of the international community, armed groups may wish to appear as being more law-abiding than the state whose power they contest. Thus, some armed groups, which have signed Geneva Call’s Deed of Commitment against the use of anti-personnel mines, are engaged in conflicts with states which have not ratified the 1997 Anti-Personnel Mine Ban Convention. Such states include India, Iran, Myanmar, and Somalia (Bellal 2016, 47, note 39). Deeds of Commitment are signed by armed groups in proceedings which imitate states’ signing of international treaties. Their representatives travel to Geneva and sign the Deed in a ceremony which takes place in the Alabama Room of Geneva’s Town Hall where the 1864 Geneva Convention was adopted. Deeds are countersigned by Geneva Call as well as by the ‘Government of the Republic and Canton of Geneva’ and deposited with the Canton which acts in a manner similar to a depositary of international treaties (Bongard and Somer 2011, 686). Lastly, armed groups may also conclude so-called special agreements which are foreseen by Art 3(3) GC I-IV (Bellal 2016, 50). Since the term ‘agreement’ is also used for formal treaties between states, armed groups may see these agreements as another tool to mimic state actions at the international level and to gain legitimacy by behaving as the relevant representatives of their constituency.

Such forms of diplomacy, in particular formal ceremonies as described above, are attractive for armed groups strategically seeking legitimacy, since any form of social validation embodies symbolic significance – allowing armed groups to mimic the state not only through the adoption of internal governance functions, but also by imitating a state’s external relations (Huang 2016b, 101; Mampilly 2015b). Diplomacy based on engagement with international humanitarian law may, therefore, be seen as a tool to further long-term policy goals, such as legal recognition, as well as short-term goals, such as gaining material support for governance structures or military advantage (Coggins 2015, 115). Insofar as long-term policy goals are concerned, armed groups may hope to achieve de facto or – in the long run – even de iure recognition. In particular, those armed groups which strive for secession or to replace governments in power, will ultimately need international legal recognition. The success of political aspirations which fall short of these far-reaching goals will, however, also hinge on obtaining some form of status, for instance through participating in ceasefire agreements, truces, peace talks or other types of political negotiation (Müller forthcoming). While it is legally disputed whether ceasefire agreements or truces are at all covered by international law (Müller forthcoming), participation in such agreements, at least in part, validates armed groups as relevant political actors. Sometimes these forms of participation may even imply a qualified form of legal recognition which confers a more stable formal status on armed groups (Mampilly 2015b, 86). Political support may also turn into material aid for sustaining armed groups’ military objectives. Benefits resulting from conducting diplomacy based on an engagement with international humanitarian law may range from recognition by or observer status within international organisations or expressions of political sympathy from states, to the supply of material support, for example in the form of money or weapons (Huang 2016b, 99). Gaining some level of legitimacy from the international community may thereby contribute to increasing effectiveness and thus legitimacy amongst local audiences because it allows groups to demonstrate at least some political relevance at the international level (Huang 2016b, 102).

Finally, engaging with international humanitarian law may be a starting point for armed groups to widen their diplomatic efforts with third states and international organisations in order to gain support for broader political goals. It may be a way to ‘to push their claims onto the agenda’ at the international level (Seymour 2017, 822). The likelihood of succeeding in this respect is considerably higher if the group complies with the laws of war from the very onset of a civil war. Compliance can demonstrate a willingness to behave according to the rules of the international community and signal that an armed group is ‘more than a violent military organisation, is politically organized and willing to co-operate’ (Huang 2016b, 100). By engaging in negotiations with established procedures, armed groups can ‘align [their] claims with wider standards of legitimacy in international society’ (Seymour 2017, 820 and 823). International humanitarian law can be instrumental in this respect because it applies even if more complex governance structures are not established. This means that even armed groups with fragmented structures will be engaged by external actors, for example, the ICRC. Due to the highly diverse character of armed groups, the law’s normative preconditions aim to adequately account for different degrees of armed group capacity (Krieger 2015, 510). International humanitarian law provides for a two-tiered system: AP II only applies to armed groups which meet the threshold of its Article 1(1). The article makes the application of the Protocol dependent on the existence of the ‘exercise [of] such control over a part of its territory as to enable [armed groups] to carry out sustained and concerted military operations and to implement this Protocol’. In all other cases of non-international armed conflict, Art 3 GC I-IV provides for a minimum standard. Correspondingly, in cases of Art 3 GC I-IV, a considerably lower degree of organisation and control is required (Krieger 2015, 510; Zegveld 2002, 135/140). The Syrian civil war offers numerous examples of fragmented armed groups which have attempted to increase their legitimacy in the eyes of international audiences by pledging compliance with international humanitarian law: For example, in 2017 several factions signed various Geneva Call Deeds of Commitment: the Free Syrian Army, including Brigade Al Motasem, Division Hamza, Brigade 51, the Al Watan Liberation Movement, the 21st Force Union, the 23^rd Division, Central Division, First Costal Division and the Democratic Self-Administration in Rojava, People’s Protection Units, and Women’s Protection Units.

If, on the basis of the signing of such instruments, compliance with international humanitarian law is furthered, the increased protection of civilians, restrictions on excessive uses of violence, and diplomacy based on engagement with international humanitarian law, may create or increase an armed group’s legitimacy from the perspective of its constituency, the local community and also the international community. However, even if an armed group complies with international law and negotiates with international actors, thereby benefiting in terms of legitimacy, this will not necessarily increase the effectiveness of their governance efforts. Instead, the virtuous circle may be interrupted on the basis that armed groups are caught in what may be called a ‘legitimacy trap’.

4. Caught in the legitimacy trap

Given the multiplicity of factors that may influence the course of a civil war, one needs to be cautious when making clear-cut cause–effect statements (cf. Terpstra and Frerks 2017). However, there are certain indications that a legitimacy trap may arise from the conflicting perspectives of different audiences of legitimacy: Governance by armed groups exclusively contributes to their legitimacy whilst simultaneously corroding the legitimacy of the state on whose territory they perform governance services. From the perspective of the state then, a need will arise to contest an armed group’s legitimacy. The ensuing struggles, which both sides may lead, have aptly been labelled a ‘legitimacy war’ (Jo 2015, 257). Effective and legitimate governance by armed groups may appeal to the legitimacy beliefs of local communities. However, the international legal framework promotes the state’s claims to legitimacy and creates an obligation for third states and international organisations to contest armed groups’ claims to legitimacy.

4.1. International law opposes the legitimizing effects of armed group diplomacy

To understand how and why armed groups may be caught in a legitimacy trap, it is necessary to shift our perspective slightly and analyze how the international community deals with security governance by armed groups and their related diplomacy efforts. States have always been aware that engaging with armed groups may enhance the group’s legitimacy and have therefore developed strategies to reduce such effects. This is reflected within the applicable legal framework – international law favours existing governments and state structures. International law comprehensively prohibits support for governance by armed groups and in fact, backs efforts to weaken them. International humanitarian law explicitly does not grant any form of legal status to armed groups or their fighters, with the specific aim of preventing an accrual of legitimacy. Furthermore, through national legislation and or practice, some states even try to prevent NGOs from providing humanitarian support to armed groups.

International law comprehensively protects the state against claims of armed groups: State sovereignty is protected by the principle of non-intervention. Outside the context of decolonization, international law does not in principle foresee a right to secession. Some cases often cited as examples of secessionist rebel governance (e.g. Podder 2017, 698), such as East Timor, Namibia or Western Sahara, actually legally qualify as cases of decolonization. In other cases, where armed groups strive for independence, such as in Somaliland or South Sudan, self-determination was denied on the basis that it would privilege support of the armed groups involved. Therefore, while support for a people exercising their right to self-determination is in theory legally permissible according to UNGA Resolution 3314 (XXIX) (14 December 1974), support for armed groups in contemporary conflicts which take place outside the context of decolonization, in principle, violates international law. Many present-day conflicts are instead designated as cases of terrorism and thus criminalized (Krieger 2015, 529 et seq.). These rules entail adverse consequences for social validation and legal recognition in relation to armed groups. Premature recognition of an entity as a state constitutes a violation of the principle of non-intervention. Even where armed groups have established a more permanent form of territorial control, variances in the legal degree of recognition conferred by other states allow the state on whose territory an armed group operates to maintain its sovereignty. Law distinguishes between de iure and de facto recognition, recognition as a national liberation movement in the context of decolonization and recognition as belligerents, the latter being presently rarely used by third states (Azarova and Blum 2015, para. 13). Military intervention in support of insurgents within a non-international armed conflict, including the provision of material assistance, infringes the principle of non-intervention (Kunig 2008, para. 24) whereas military support for the ruling government upon invitation is permissible. Military intervention in support of armed groups also violates the prohibition against the use of force. According to the ICJ’s judgments in the Nicaragua Case (ICJ 1986, 101 para. 191, 104 para. 195; 109 para. 209) and the Armed Activities Case (ICJ 2005, 226 paras. 161 et seqq.) as well as the Friendly Relations Declaration (UNGA Resolution 2625 (XXV) (24 October 1970) the prohibition on the use of force and the duty of non-intervention prohibit third states from supplying arms to rebel groups (Ruys 2014, 32). Since the training of armed group personnel is in principle not allowed, some states hold that supporting armed groups through international humanitarian law training is also illegal (Krieger 2015, 529). In an outspoken manner, UN Security Council Resolution 2078 demonstrated the international community’s principled rejection of governance by armed groups when the Council stated that it, ‘condemns the attempts by the M23 to establish an illegitimate parallel administration and to undermine state authority of the Government of the DRC’ (UNSC Resolution 2078 (28 November 2012), para. 6).

Admittedly, states have not always complied with this legal framework. National interests or policy preferences have sometimes outweighed considerations of international legality (cf. Krasner 1999). The conflicts in Libya and Syria, for instance, have raised concerns among international legal scholars that the prohibition against providing material assistance to armed groups, in particular weapons, is currently being eroded (Krieger 2018, 459 et seq.). In these conflicts, the US and European states have supported armed groups opposing governments which were considered by some within the international community as illegitimate. In the Libyan conflict, France recognized the National Transitional Council as the ‘legitimate representative’ of the Libyan people (Corten and Koutroulis 2013, 65) and France, Qatar and the UK sent weapons to armed opposition groups (Corten and Koutroulis 2013, 90 et seq.). In the Syrian conflict, France, Qatar, Saudi Arabia, the UK, and the US supplied weapons to various armed groups (Ruys 2014, 14–16). Even more significantly, the EU reviewed its position and took note of Member states’ commitments to export arms to the Syrian Opposition Council (Council of the EU 2013). It is, however, too early to assume that the prohibition has been entirely abandoned (Krieger 2018, 461). The ratio of the prohibition to render material support to armed groups still claims relevance: support for all sides of an armed conflict risks perpetuating the conflict and to further undermine the existence of the state (Krieger 2018, 460; Regan 2014, 320). The state-centred nature of international law, therefore, sides with the de jure government.

Also, international humanitarian law aims to contain any legitimization which engagement with armed groups may produce. Although international humanitarian law obliges armed groups to comply with its rules, Art 3(4) GC I-IV explicitly states that the application of its rules does not affect the legal status of parties to the conflict. The travaux préparatoires of the Geneva Conventions, in fact, show that the substantive provisions of Art 3 GC I-IV could only be agreed upon because of the caveat included in para. 4 (ICRC Commentary 2016, 862). The provision makes clear that the application of Art 3 GC I-IV does not confer any status on an armed group nor imply any recognition by the de iure government. The provision does not restrict a government’s right to militarily oppose armed groups and does not exclude the right to prosecute members of armed groups for participation in armed conflict. ‘Indeed, the application of Art 3 GC I-IV to a non-international armed conflict does not confer belligerent status or increased authority on the non-state armed group’ (ICRC Commentary 2016, para. 864). International law needed to reconcile the tensions which arise from the gap between the illegal position of armed groups and the necessity to engage with these actors in the interest of humanitarian concerns. Therefore, Art 3(4) GC I-IV was specifically introduced into the Geneva Conventions in order to meet states’ concerns that the application of these rules to non-international armed conflicts could somehow hinder the de jure government’s lawful armed opposition against military activities of armed groups:

Nothing since the introduction of Common Article 3 in 1949 has altered the fact that the applicability of humanitarian law to situations of non-international armed conflicts does not affect the legal status or enhance the legitimacy of non-state armed groups. This remains as essential today as it was at that time, as any other interpretation will almost inevitably lead states to deny the applicability of Common Article 3 and thereby undermine its humanitarian objective. (ICRC Commentary 2016, para. 869)

Accordingly, the 2006 UN Guidelines on Humanitarian Negotiations with Armed Groups stress that ‘because of their exclusively humanitarian character, humanitarian negotiations do not in any way confer legitimacy or recognition upon armed groups’ and warn humanitarian organisations that their efforts may be abused by armed groups as a means to gain legitimacy. Therefore, the ICRC’s strategy exclusively focuses on the humanitarian objectives of the Geneva Conventions:

This clause makes absolutely clear that the object of the Convention is purely humanitarian, that it is in no way concerned with the internal affairs of states, and that it merely ensures respect for the essential rules of humanity which all nations consider as valid everywhere, in all circumstances. (ICRC Commentary 2016, para. 861)

Moreover, some states have even enacted national legislation enabling the criminalization of armed groups through the paradigm of terrorism. This includes US legislation according to which ‘it is a federal crime to knowingly provid[e] material support or resources to a foreign terrorist organization’ (US Supreme Court Holder v. Humanitarian Law Project, 130 S Ct 2705 (2010); 18 U.S.C. § 2339 B(a)(1), USA Patriot Act.). Since material support includes training and expert advice or assistance, activities of NGOs, for example, efforts of the ICRC or Geneva Call to disseminate international humanitarian law or to provide legal advice or assistance in regard to the disposal of landmines fall under the provision. Groups classified as terrorists according to this regulation have included the Kurdistan Workers’ Party and LTTE (Krieger 2015, 548–549). Concerns surrounding the potential for anti-terrorism legislation to restrict the provision of medical care have also been raised. In his 2017 report, the UN Secretary-General stressed that in some states anti-terrorism legislation even ‘criminalizes the provision of medical care to members of some non-state armed groups, such as those designated as terrorist by the Security Council or under national law’ (UN Secretary-General 2017, para. 43). Already in 2012, the UN Secretary-General had explicitly criticized such practices (UN Secretary-General 2012, para 76). Nonetheless, the US legislation stands as of 2018.

Thus, while armed groups may attempt to benefit from the effects of the virtuous circle of governance, the international legal framework within which they are forced to conduct their diplomacy hinders these efforts and supports those actors who seek specifically to weaken armed groups.

4.2. Breaking the virtuous circle – UNITA’s changing governance attitudes in the Angolan civil war

Through its paradoxical implications, international law creates a dilemma for armed groups: While it encourages armed groups to provide security governance for local communities under their control, it continues to conceive armed groups as illegal actors and therefore simultaneously delegitimizes them, irrespective of their governance efforts. Although these efforts may appeal to the legitimacy beliefs of local communities, considerations of international legality promote states’ claims to legitimacy in the eyes of international audiences. As a consequence, any possible gain in empirical legitimacy for armed groups often remains legally contested and thus precarious. De facto recognition may easily be withdrawn, supply of money and weapons may be frozen, political sympathy lost. Armed groups’ legitimacy gains among international audiences are therefore neither permanent nor stable. The legal framework makes it more difficult to maintain legitimacy based on performance over time. As a consequence, armed groups will seldom reach the point where compliance ‘allows for rule at lower costs’ (cf. Schmelzle and Stollenwerk 2018). Maintaining international legitimacy, which, for armed groups means struggling against the grain of the dominant legal framework, therefore requires constant efforts. Where armed groups come under international pressure to comply with humanitarian norms, yet their legitimacy at the international level is simultaneously contested, the likelihood increases that armed groups will lose their interest in governance and communication activities and turn to other violent short-term strategies to the detriment of the civilian population. The changing behaviour of UNITA in the Angolan civil wars between 1975 and 2002 exemplifies the backlash that might result from an internationally contested legitimacy.

In the first war following Angola’s independence from Portugal, between the MPLA government (Movimento Popular de Libertação de Angola) and UNITA, the latter engaged in governance efforts within the territory under its control and pursued intensive diplomacy efforts to increase its legitimacy. Part of these diplomacy efforts were directed at engaging with the ICRC. In 1980, UNITA issued a unilateral declaration to respect the basic principles of international humanitarian law and, in 1988, allowed the ICRC to visit detained persons. The ICRC and UNITA leader Jonas Savimbi concluded special agreements concerning humanitarian assistance and the release of prisoners of war in the territory controlled by UNITA. This resulted in the release of numerous Angolan government prisoners of war and foreign abductees between 1982 and 1984. The ICRC also operated a centre for contacts with the armed group in areas under UNITA control. Further, the 1991 ceasefire agreement, concluded between the government of Angola and UNITA, provided for verification of the release of prisoners of war by the ICRC (Ewumbue-Monono 2006, 908, 912, 916, 919). These efforts contributed to UNITA’s success in gaining international as well as domestic legitimacy (Huang 2016b, 118–119).

However, the 1992 elections in Angola and shifting geopolitics following the end of the cold war led to a push for the political isolation of UNITA. The international political and legal framework increasingly delegitimized UNITA, manifesting in the UN Security Council sanctions regime (UN SC Res 864 (1993)). Correspondingly, UNITA ceased to employ diplomacy instruments and no longer invested resources into governance structures for the local population. Instead, it reversed its strategy. Following the recognition of the MPLA government, UNITA’s opponent, by the US, UNITA had lost its legitimacy war. In order to sustain its position, UNITA relied on diamond mining and trade and turned to a strategy of terrorizing the civilian population through use of indiscriminate violence. Apparently UNITA’s reliance on the diamond trade and its loss of international legitimacy eventually also destroyed its strategic interest in maintaining legitimacy with local audiences (Huang 2016b, 114–123; Jo 2016, 242).

5. Concluding policy considerations

The risk that, in the case of armed groups, the virtuous circle of governance may be interrupted raises the question as to whether international actors involved in humanitarian efforts and statebuilding initiatives in areas of limited statehood should re-evaluate and adjust their policy approaches to armed groups. Humanitarian efforts to engage with armed groups within the context of on-going civil wars, of course, differ from state-building initiatives. Efforts do not seek to achieve long-term goals of consolidating governance, rather they strive to prevent, from a short-term perspective, the deterioration of civilian protection. Therefore, one could argue that international humanitarian actors need not take into account whether their activities influence armed groups’ governance efforts and the virtuous circle. In any case, international humanitarian lawyers and organisations have long been aware of the ambiguous consequences engagement with armed groups entails. As has been shown above, the 1949 Geneva Conventions explicitly aim to contain any potential legitimizing effect, while some states are still reluctant to classify internal wars as a non-international armed conflicts ‘often due to a concern that such classification somehow confers a certain status or legitimacy on non-state armed groups, be it legal or political’ (ICRC Commentary 2016, para. 865). From this perspective, however, the caveat included in Art 3 GC I-IV should be seen as a sufficient safeguard against any ‘side-effect legitimacy’ diplomacy with armed groups might have.

To some extent, this seems to be the currently prevailing approach. The UN Secretary-General, for instance, insists that ‘focus on recognition and legitimacy is problematic in that it detracts from the more serious issue of the consequences for civilians when engagement does not take place’ (UN Secretary-General 2017, para. 45). The perceived increase in non-international armed conflicts following the end of the cold war pushed the international community towards acknowledging that armed groups are ‘key-actors in today’s conflicts’ (UN Secretary-General 2017, para. 30). Simultaneously, these conflicts have given rise to the impression that the international community faces a ‘global protection crisis’, according to the UN Secretary General (2017, para. 3). Any attempt to improve the protection of civilians necessitates engagement with armed groups. Therefore, in his report for the 2016 World Humanitarian Summit, the UN Secretary General emphasized the importance of the involvement of impartial humanitarian actors (UN General Assembly 2016, para. 51), accepted the need for directly engaging with armed groups and ‘emphasized repeatedly the need for consistent engagement with those groups to seek improved compliance’ (UN General Assembly 2016, para. 41). A certain change in the rejectionist approach by states when it comes to engaging with armed groups can also be observed. Switzerland and Germany, for example, support initiatives to engage with armed groups, as long as ‘engagement does not constitute political recognition of these groups’ (UNSC, 6790th Meeting (25 June 2012) UN Doc. S/PV 6790, Statement of the Representative of Germany, p. 23.).

However, while the law can prevent armed groups from obtaining legally recognised legitimacy, it cannot so easily contain the development of legitimacy among local or regional communities based on effective governance efforts by armed groups. Humanitarian actors should not lose sight of the fact that even short-term governance initiatives may have long-term effects on post-conflict peace-building processes (Huang 2016a, 184). Eventually, however, the old dilemma remains and will continue to persist. The delicate balance between inducing compliance with international humanitarian law in the interest of protecting civilians and states’ interests in containing armed group legitimacy, which may flow from compliance, is a constant challenge. As long as legitimacy remains a political resource for armed groups, efforts on the part of the international community and third states to build upon armed groups’ security governance initiatives, in the interest of humanitarian concerns, tread a thin line. In particular where efforts are not embedded in long-term strategies for ending a conflict.

International actors should, therefore, take the effects of the governance circle into account. Humanitarian actors should not ignore the fact that they contribute nolens volens to the legitimacy of armed groups and that armed groups use this interaction for political leverage. International actors must recognize that part of the incentive for armed groups to accept international humanitarian law, lies in the potential for compliance to enhance the group’s legitimacy. This applies, above all, to all instances where an armed group can mimic state behaviour at the international level. Less public efforts to engage with armed groups could, for instance, contain some of these effects. International actors should be mindful that shifting approaches towards engagement with armed groups may impact the situation of civilians. A change in policies towards a certain armed group should always take into account the consequences for the governance circle. In particular, the international community should be prepared to deal with a deterioration of the civilian security situation.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributor

Heike Krieger is Professor of Public Law and International Law at Freie Universität Berlin and Max Planck Fellow at the Max Planck Institute for Comparative Public Law and International Law in Heidelberg. She is Co-Chair of the Berlin Potsdam Research Group on „The International Rule of Law – Rise or Decline?“ and Co-Editor of the Yearbook of International Humanitarian Law (T.M.C. Asser Press and Springer) and on the Editorial Board of the Journal of Conflict and Security Law. Her recent publications include the edited volumes “Inducing Compliance with International Humanitarian Law: Lessons from the African Great Lakes Region” (Cambridge University Press, 2015, 576pp.) and “The Legal Pluriverse Surrounding Multinational Military Operations” (Oxford University Press; ed. together with R. Geiß, forthcoming).

Additional information

Funding

This work was supported by Deutsche Forschungsgemeinschaft through the Collaborative Research Center (SFB) 700: Governance in Areas of Limited Statehood at the Freie Universität Berlin, Germany.