Introduction

The ICRC interacts with State and non-State actors that play a role in contemporary armed conflicts and other situations of violence. Engaging armed groups more specifically represents a substantial part of the work of the institution considering their pervasiveness in the non-international armed conflicts and in other situations of violence in which the ICRC operates.1 Although the ICRC needs to preserve the confidentiality of its engagement with armed groups, this article aims to present general information for the public interest about contemporary dynamics of armed conflicts and their consequences on the work of the institution.

As an independent and neutral humanitarian institution, the main purposes of the ICRC's interactions with armed groups are to protect and to assist civilian populations affected by armed conflicts and to promote international humanitarian law (IHL). This mission is legally grounded in the Geneva Conventions of 1949, their Additional Protocols, the ICRC's own Statutes and those of the International Red Cross and Red Crescent Movement.2 In practice, the ICRC interacts with more armed groups than any other humanitarian organization considering the range of countries and activities in which it is involved. Interacting with armed groups is necessary to gain safe access to vulnerable populations that live under their direct control or influence. It also helps the ICRC to act as a neutral intermediary3 and promote its modus operandi based on independence and neutrality. Last, engaging armed groups seeks to ensure better respect for the laws of armed conflict.

In 2021, the ICRC identified 605 armed groups whose capacity to cause violence is of humanitarian concern in the contexts where the ICRC is operating.4 The ICRC interacts with 418 of them worldwide.5 Armed groups range from State-like territorial bureaucracies that rule populations larger than Switzerland (population: about 8.7 million) to networks partly embedded in communities that exist only in political and security vacuums when the State is no longer able to govern locally. The ICRC has developed an array of tools to analyse and synthesize its patterns of interaction and humanitarian dialogue with these groups.6 These analyses help us better understand the functioning of armed groups worldwide, including the nature of their activities, objectives, relations to the population and other States, and patterns of violence. Since early 2020, these analyses also have guided our response to COVID-19 by situating armed groups’ reactions to the crisis in a consistent analytical framework.

The outbreak of COVID-19 presented several challenges in humanitarian contexts. The challenges were particularly severe for civilian populations, since the pandemic has exacerbated the humanitarian situation in countries that were already experiencing armed conflicts before the pandemic. The presence of an array of armed groups that sometimes replace existing governmental institutions was a notable predicament. Any local response to the pandemic was affected by the proliferation of a growing number of non-State parties.7 The in-depth analysis of armed groups’ characteristics across cases and of our patterns of interactions with them were therefore important to the ICRC to situate those groups’ positions on the crisis and better calibrate our response. This understanding helped us define the feasibility, the practicalities of our engagement and potential response to the pandemic.

This article builds upon internal yearly surveys undertaken since 2018 and our work engaging armed groups. The specific details of the surveys, especially country-specific information, cannot be published to protect the ICRC's operational dialogue. Our work relies on the maintenance of confidentiality and the protection of the ICRC's interactions with State and non-State interlocutors alike. Nonetheless, we have decided to release general background information that illustrates important aspects of our work while ensuring full respect of the Fundamental Principles of the International Red Cross and Red Crescent Movement. In terms of methodology, the data are extracted from a comprehensive survey sent out yearly to every ICRC delegation to collect exhaustive data about the number, type and characteristics of the ICRC's interlocutors on the ground. The statistics provide a general overview of some of these groups’ most important features.

Our estimate (see below) of the population affected by the control of armed groups includes civilians that live under the exclusive political and security control of armed groups, although the central government might still maintain some public services (e.g. in the fields of education or health), and people living in areas where this control is disputed. These estimates are notoriously difficult to assess since frontlines can change quickly, and the number of residents in some of these places is at best an approximate assessment. These numbers nonetheless provide an important snapshot on this phenomenon. We do not claim that governance by armed groups is a new phenomenon, but we believe that it is critical to evaluate its contemporary magnitude and humanitarian consequences.

The paper is structured in three parts. First, the article presents the most important trends concerning armed groups’ presence worldwide. Two notable characteristics of contemporary armed conflicts are the widespread external State support for armed groups and the extent of armed groups’ areas of control. Almost half of the armed groups receive external support. In contexts where control is full, armed groups often try to replicate traditional State structures to rule the population after the elimination of the political and security presence of internationally recognized governments. The second part of this article examines the impact of these two characteristics on our engagement with armed groups. It contends that, regardless of their ideological leanings, the type of non-military activities performed by armed groups informs the type of humanitarian objectives that the ICRC can achieve as well as the quality of dialogue with these groups. This finding confirms that facts on the ground are more determinative than the narratives built around them. Finally, the third section builds upon these findings to illustrate our analysis of the impact of COVID-19 and the range of response that can possibly be provided. It exposes armed groups’ enduring and contextual characteristics and their impact on the ICRC's ability to engage these actors.

Two dominating trends: External support and governance

Armed groups’ activities are primarily local (55%), sub-national (19%) or national (10%) in nature. They range from low-intensity armed actions perpetrated by local groups structured around their communities to significant nationwide campaigns put together by groups with national ambitions effectively in charge of large proportions of territories. Most groups remain primarily engaged for national or sub-national purposes, as direct challengers to central State authorities or seekers of some sort of alternative social order.

However, even national conflicts rarely remain only driven by national dynamics. The first notable trend of contemporary armed conflicts is the importance of external support for armed groups, which is a major vector through which conflicts can spill over across the border. Nearly half of armed groups worldwide are supported by one or more foreign States.8 External State support is particularly preponderant in countries like Iraq, Libya and Syria. Most State support is provided by States from the same region, while large-scale contribution from Western countries tends, with limited exceptions, to assist the territorial State or to take the form of a direct intervention with the projection of their own military forces.

The diversity of armed groups, their localism, and the existence of multiple contending alliances involving States, questions the common understanding of external support as a mechanical relationship whereby armed groups would be a mere tool for foreign States. External support is too often understood as a rather unilateral patron–client relationship where power and agency primarily lie in foreign States. However, the local embeddedness of many groups also acts as a real constraint on supporting States, which must concede the limits of their actions. Local groups are occasionally reluctant to be responsive to the demands of their external supporters when local communities oppose foreign States’ specific demands. Moreover, being able to switch alliance in a crowded marketplace reinforces the bargaining power of local groups who can play contending coalitions against each other. According to our research, State support is particularly important in favour of armed groups that only conduct armed activities or engage in full-scale governance, while it is substantially weaker for the groups that only provide some type of social services or security.

The geopolitical capacity to form coalitions is not limited to States. Armed groups also establish cross-group alliances where their level of cooperation ranges from transactional exchanges of material support, to tactical military coordination, strategic partnerships and, finally, full mergers.9 Cross-group alliances can be short-lived. They also compete with one another in most conflicts in a dizzying number of combinations that forms the kaleidoscope of today's armed conflicts. Cross-group alliances are important to examine carefully since they complexify the legal determination of IHL's applicability.10

The second notable feature of contemporary armed conflicts is the widespread phenomenon of armed groups’ governance. For the sake of our internal study, governance refers to the organization of civilian life such as policing the population, administration of some mechanisms of dispute resolution (i.e. justice), and sometimes the provision of public goods related to health or education, or the imposition of taxes. Governance can be exercised more or less directly in informal or bureaucratic patterns. Armed groups whose local governance is legally recognized domestically, for instance as part of a peace agreement regulating power sharing, are not included. Moreover, this definition includes contested areas where State and non-State authorities compete to impose themselves on local communities. These areas are potentially the most problematic from a humanitarian point of view considering the challenges posed by the presence of contending local authorities. In these areas, the State and local armed groups fight over control and strive to dissuade civilians from collaborating with the other side, including with violence. Last, our definition is agnostic on the type of group controlling the population.

With this definition of “full territorial control”, our internal study estimates that, at the time of writing11 between 50 to 60 million individuals live under the full control of armed groups worldwide while an approximate 100 million live in areas where this control is contested.12

Armed groups’ governance varies substantially across cases. Some armed groups compete with territorial States by intentionally replicating State authorities. They set up formal governing bodies in charge of the population that increasingly cooperate with international institutions on cooperation and humanitarian assistance. In some cases, armed groups rely on technocratic figures to create cooperation channels with international organizations and maintain some level of plausible deniability regarding their connection to the alternative authority that they support locally. In other cases, they directly rule the population through their own organizational structures. Many direct and indirect forms of governance are comparable to established governments, although they typically lack international recognition. Armed groups’ governance arrangements can, at times, be effectively as sophisticated as the governance of the internationally recognized governments that they oppose. In other cases, loosely structured armed groups can only establish rudimentary forms of governance tending to be more tightly embedded in their local communities. These cases expose the inherent organizational strength of these groups as much as the weakness of domestic governments and highlight that armed groups governance is a critical feature of contemporary armed conflicts.

Differences in organizational structures and capacities of armed groups are important for the debate on the applicable legal framework in which humanitarian engagement with armed groups takes place. For instance, to be party to an armed conflict and bound by IHL, the latter legal framework (an exhaustive presentation of which is outside the scope of this paper) requires an armed group to show a certain degree of internal organization and be involved in a sufficiently intense level of violence. This is particularly challenging to establish for de-centrally structured groups with an opaque chain of command. On the other extreme, if NSAGs exercise stable control over territory and are, de facto, able to act like a State authority, their interaction with and impact on the life and wellbeing of persons living under their control is significant. In these situations, it may become relevant for humanitarian organizations to also refer to human rights norms to ensure the protection of affected populations, in addition to these armed groups’ IHL obligations as applicable to non-international armed conflicts.13 The question of the applicable international legal framework has immediate consequences, for instance on individuals detained by armed groups worldwide. Beyond detention, we notice that armed groups’ involvement in governance largely determines their decisions regarding the humanitarian situation of the communities under control. Imperatives linked to local governance, including the need to be responsive to the local needs and expectations of the population, play a vital role in armed groups leaderships’ decisions to adopt certain policies and engage external actors.

External State support and local governance are two key challenges for armed groups and international organizations’ interactions with them. They shape these groups’ interests and perceptions. Moreover, they situate the broader contexts in which they can be engaged regardless of their ideological leanings and organizational affiliations.

Engaging armed groups

Armed groups vary according to different factors including their ideological leanings, areas of influence, type of non-military activities they are engaged in, and external ties. These dimensions inform the type of engagement that humanitarian organizations have with them since armed groups that share important characteristics often behave in a congruent manner. The second factor concerns armed groups’ own organizational structures, especially when they control territory. These two features are strongly predictive of the type of engagement and dialogue with armed groups that can be pursued beyond anecdotal evidence and assumptions, such as that more radical groups would systematically refuse to acknowledge or follow IHL while in practice it is not necessarily the case.

For the sake of our survey we defined engagement with armed groups along five categories. The first category refers to direct protection dialogue with armed groups to which IHL applies in view of these groups qualifying as a party to a non-international armed conflict in the sense and for the purpose of IHL. That type of dialogue raises concrete protection concerns and points NSAGs to their legal obligations. In the second category, the ICRC does not refer to IHL directly, notably where an armed group is not party to an armed conflict or if the group is not willing to recognize this body of international law. In such cases, the ICRC relies rather on other legal norms or general principles of humanity, such as respect for human dignity and the protection of human life, which are common to all legal traditions and stem from religious or cultural sources. The third category of engagement is more geared towards practically enabling humanitarian assistance: dialogue can include direct discussions on operational concerns, including access to territories where armed groups are active, to negotiate acceptance and security, but without going into the issue of how hostilities should be conducted. The fourth category is when the ICRC entertains dialogue with NSAGs that references IHL and other legal frameworks in a non-contextual manner. This comprises, for example, the provision of IHL training. The last category is simply an absence of dialogue. Differentiating these types of dialogue across the spectrum of armed groups helps to better understand the nature of our engagement with them, and the level of mutual understanding and trust that consolidates over time.14 There is a relatively strong level of engagement across the spectrum of armed groups. Most of them, regardless of their ideological leanings and positions, can be engaged at least on operational matters. Engaging armed groups on IHL is generally more difficult as it requires a stronger degree of awareness and understanding of the importance of engagement with external actors, which only materializes over time. Engagement is weaker for the groups situated on the more criminal side of the spectrum, as they are not usually willing to engage external organizations. Engagement is associated with a stronger measurement of armed groups’ perception of the ICRC. Engagement improves mutual understanding of one another. It also lowers misperceptions.

Although some armed groups have specific ideological leanings that might be antagonistic to international engagement with external organizations, the type of activities conducted by armed groups locally has a critical impact on their willingness to collaborate with humanitarian and other actors. There is a systematic relationship between the level of armed groups’ engagement with our organization and perception of our work and the type of activities conducted by armed groups. Regardless of their ideological commitments, armed groups providing a range of social services, or fully governing parts of their territories have a much stronger engagement with the ICRC, including on IHL, where applicable. This finding suggests that they are responsive to different types of incentives, including the necessity to access their local communities and establish new external relations in the current international environment, when their range of activities widens. It also exposes their more advanced organizational features that facilitate that type of engagement. Understanding these features is critical to foster dialogue instead of merely assuming that certain types of groups are simply not responsive to external engagement, for ideological reasons for instance. It also helps better understand how to tackle new crises, such as pandemics, and the evolution of armed groups’ long-term trajectories more generally.

The systematic analysis of armed groups’ comparative features and engagement over the years has well positioned our organization to emerging challenges. These included, in the past decade, adapting to new networks of armed groups and global franchises. A major challenge, since 2020, has been the COVID-19 pandemic against the backdrop of substantial territorial control by armed groups worldwide. Understanding the combination of several key structural and contextual factors underpinning armed groups’ responsiveness was important to understand their positioning and suggest possible avenues for dialogue.

The COVID-19 pandemic

The COVID-19 pandemic illustrates important issues when engaging armed groups. As previously indicated,15 the authors of this article estimate that, as of July 2021, 150 to 160 million individuals currently live under the direct control of armed groups, or in areas where this control is fluid. Groups active in these grey areas can be vying to take over central governments, achieve territorial autonomy, or be simply behaving with short-term economic rationale. These differences are important to define the type of engagement that is possible after the outbreak of pandemics like COVID-19.

Armed groups worldwide have adopted three main positions on COVID-19 (vertical axis in Figure 1). (1) On the lower end of the spectrum, armed groups or influential figures close to them have simply denied its existence, often blaming their global enemies. Others have turned responsibility to deliver public health responses back to the State, while refusing to engage in dialogue of substance on the subject matter. (2) In the intermediary category, armed groups may have engaged in public campaigns for precautionary measures showcasing their State-like capabilities but without strong commitment on addressing the crisis. (3) The last category includes the groups that have taken proactive measures and/or been willing to actively engage with international partners to join international efforts against the pandemic. Figure 1 represents the different trajectories, as well as two turning points (A and B) in which armed groups’ decisions can potentially change over time (horizontal axis).

Figure 1. Potential trajectories followed by armed groups.

Regarding the reasons explaining changes of trajectories on COVID-19, we differentiate two types of factors. The first are relatively structural. They are situated on the top-row boxes of Figure 2, which schematize from left to right a group's potential trajectory from denial to engagement, with A and B symbolizing the turning points between different attitudes. These structural factors include armed groups’ lasting characteristics that external actors cannot directly influence. For example, armed groups that are less embroiled in cross-group competition within the same social movement,16 are aligned with the State, and are strongly embedded in their communities are more likely to recognize the severity of COVID-19 and try to play a relatively active role. They are also more inclined to collaborate with international organizations when they seek to nurture international ties, or when health conditions deteriorate quickly. The second type of factors, in the lower-row boxes, are more conjunctural. They can be used by international and local organizations to encourage immediate action on pandemics. Engaging in direct dialogue with armed groups or specific figures of influence can help reduce the amount of denial and lead to more proactive engagement.

Figure 2. Structural and conjectural factors.

This understanding has informed the range of engagement with armed groups regardless of their official ideological positioning. For example, many Islamist armed actors expressed relatively strong denials in the beginning of the crisis by invoking religious scriptures and anti-Western feelings. However, only referring to these positions was not a satisfactory way to account for their leadership decisions, although religious explanations are important for the community to make sense of these decisions. Our comparison of different types of groups suggests that what mattered was not merely their reading of religious texts but the structure of their broader social movement and these groups’ relations with State authorities. Armed groups that are aligned to domestic State authorities have attempted to position themselves as credible providers to their constituencies. Public responses, including communiqués and street spreading of disinfectants, were therefore particularly common in some countries. This situation contrasted substantially with armed groups whose social movement was very divided over religious authority. Internal divisions over religious legitimacy and credibility were reflected in their divergences on COVID-19. Divergences on COVID-19 were often, in these cases, the outcome of internal competition between different religious authorities. Internal competition incentivizes individual actors to develop a stronger theological reading to appear more religiously rigorous, and therefore legitimate. Differences of positioning between Islamist armed groups therefore do not stem from essentialist readings of their theological methodologies but from broader political variables that we include in our analyses.

The same arguments apply to the global and trans-border franchises. Groups’ affiliates have been largely driven by their own position in armed conflicts rather than the position of their patron. Franchises in charge of populations have been more willing to take some measures. This was not the case for the franchise groups that barely control territories in their areas of influence, since they were not pressured to develop a position in the absence of territorial control. This position could have been substantially different had the group been in charge of populations.

Armed groups’ positioning must be understood more systematically, beyond the individual positions of their leaders and ideological readings. Situating these groups in their contexts and relations to one another and to the civilian population reveals underlying factors that might not seem to be evident in the first place, especially in the case of arguably “ideological” groups that appear to be more strongly committed to certain principles antagonistic to IHL and its implementation (such as access to health care in the case of COVID).

Conclusion

Humanitarian crises present important specificities that shape the possibility of engagement with armed groups and respond to crises like the COVID-19 pandemic. Many features of contemporary non-international armed conflicts mentioned in this article are not necessarily new, especially foreign support for armed groups, but their relevance has increased over the past decade. More ideologically committed groups should not be essentialized. These groups are also responsive to their environments, which shape their behaviour and political positions beyond stated ideological commitments. This approach maintains a healthy distance from the labels attached to these groups by their opponents or the narratives actively promoted by themselves, which act as framing devices obfuscating any attempt at engagement, be it for humanitarian purposes, or for larger conflict transformation ambitions. The attempt to build a more systematic understanding of armed groups in armed conflicts that the ICRC has developed notably suggests that, regardless of ideological commitments and leanings, establishing structures of governance is a major factor influencing these groups’ trajectories and informing avenues for engagement. First, these features play a particularly prominent role when new crises occur, and armed groups must respond quickly to new needs that they are not always equipped to cater for. Second, by focusing on behaviour rather than supposed intentions, it puts people's need firmly at the centre of a neutral humanitarian planning process before any other policy consideration. Although limited exceptions exist, especially in the case of specific global franchises, these exceptions should not conceal a much more frequent case in favour of calibrated dialogue and collaboration in which respect for humanitarian considerations constitute the minimum common ground.

• 1The ICRC used an internal working definition of armed groups as “broad range of groups with varying goals, structures, doctrines, funding sources, military capacity and degree of territorial control. It denotes a group that is not recognized as a State, but has the capacity to cause violence that is of humanitarian concern. Included in this broad operational category are ‘non-State armed groups’ (NSAGs) that qualify as a party to a non-international armed conflict and are therefore bound by international humanitarian law.”
• 2For a legal discussion on the issue, see ICRC position paper: Jelena Nikolic, Irénée Herbet and Tilman Rodenhäuser, “ICRC Engagement with Non-State Armed Groups: Why and How”, Humanitarian Law & Policy, 01, available at: https://blogs.icrc.org/law-and-policy/01/03/04/icrc-engagement-non-stat… (all internet references were accessed in October 01).
• 3The ICRC considers that it becomes a “neutral intermediary” when it acts as a third party between two or several parties in dispute and with their agreement in order to facilitate the resolution of the dispute or the implementation of a settlement agreement.
• 4Not necessarily all of these armed groups are a party to a non-international armed conflict for the purposes of IHL.
• 5All the data reported in this article are drawn from our internal reports.
• 6See, for example, ICRC, The Roots of Restraint in War, Geneva, 2018, available at: https://www.icrc.org/en/publication/roots-restraint-war.
• 7ICRC, “ICRC: Study Shows More the Conflicts, Greater the Danger for People”, News Release, Geneva, 2018, available at: https://www.icrc.org/en/document/icrc-more-conflicts-more-sides-conflic….
• 8Support relationships are a major feature of contemporary warfare; more information can be found at: IRC, Allies, Partners and Proxies: Managing Support Relationships in Armed Conflict to Reduce the Human Cost of War, Geneva, 2021, available at: https://www.icrc.org/en/publication/449-allies-partners-and-proxies-man….
• 9For example, Assaf Moghadam, Nexus of Global Jihad: Understanding Cooperation Among Terrorist Actors, Columbia University Press, New York, 2017.
• 10Jelena Nikolic, Thomas de Saint Maurice and Tristan Ferraro, “Aggregated Intensity: Classifying Coalitions of Non-State Armed Groups”, Humanitarian Law & Policy, 2020, available at: https://blogs.icrc.org/law-and-policy/2020//07/aggregated-intensity-cla….
• 11The changes induced by the arrival of the Islamic Emirate of Afghanistan in Kabul are not taken into consideration by this article whose data and writing predate 15 August 2021. However, the order of magnitude conveyed by the overall figures would not be significantly changed.
• 12This preliminary estimate is aggregated from the ICRC’s delegations worldwide.
• 13For a full legal discussion on this issue, see Tilman Rodenhäuser, “The Legal Protection of Persons Living Under the Control of Non-State Armed Groups” in this edition of the International Review of the Red Cross.
• 14The ICRC previously published research on this theme, above note 6.
• 15Above note 11.
• 16Armed groups usually evolve in larger social movements that can be based on religious, nationalist or other ideological foundations. Intensive competition within the same social movement can exacerbate armed groups’ contending positions on COVID-19.