In modern warfare, non-international armed conflicts, or NIACs, are increasingly characterized by the presence of a myriad of armed non-State actors, often fragmented, that operate independently or in fluid, fragile coalitions. The Democratic Republic of the Congo (DRC), the Sahel region, and Central African Republic (CAR) are just a few examples of situations where State forces engage in hostilities against a significant number of armed groups, sometimes organized horizontally or in sub-groups, that may fight against each other as well. These circumstances pose crucial challenges for the purposes of classification.
It is well known that a non-international armed conflict exists ‘whenever there is … protracted armed violence between governmental authorities and organized armed groups or between such groups within a State.’ Specifically, the assessment is based on two criteria: the organization of the parties and the intensity of violence. Below this threshold, a situation of armed violence should be considered as ‘internal disturbances and tensions, such as riots, isolated and sporadic acts of violence and other acts of a similar nature,’ namely instances when international humanitarian law (IHL) does not apply.
According to the majoritarian view, whether an armed conflict exists should be determined looking at the bilateral situation between the parties. The so-called ‘fragmented approach’ has the merit to look at the factual relationship between each party to the conflict. Nevertheless, it may seem unfit to face situations when a considerable number of non-State actors are fighting against the government as well as against each other. This is particularly true when the armed non-State actors (ANSAs) are opposing the same enemy and/or form coalitions.
For instance, let’s imagine a country in which several ANSAs are fighting against governmental forces, some as part of a coalition and others operating independently. How can we determine whether these groups are party to a NIAC? After having identified the non-State actors that meet the organization requirement, the problem lies in analyzing whether the threshold of the intensity of violence is met. According to the fragmented approach, the intensity of violence should be assessed separately for each armed group, but this may not reflect the reality on the ground, and it might be practically unfeasible.
IHL becomes applicable as soon as there is intense violence between an organized ANSA and governmental forces, or between organized ANSAs. In practice, this means that the government has to apply IHL or law enforcement rules against different organized ANSAs fighting against it, depending on whether the violence between the single group and State forces meets the threshold of violence. But what if none of the organized groups active in the country engage in sufficiently intense violence to be party to a NIAC? This might lead to the conclusion that no NIAC exists in the country, even when the global intensity of violence is extremely high.
In order to address these and related challenges and to avoid ‘a legal and operational loophole,’ it might be preferable to aggregate the intensity of violence for classification purposes. Three approaches can be envisaged. First, in a recent post on this blog, Nikolic, de Saint Maurice, and Ferraro suggest aggregating the intensity of organized non-State actors provided they are members of the same coalition. Second, Kleffner has recently affirmed that, ‘where acts of violence by several organized armed groups occur on a geographical and temporal continuum,’ the intensity should be assessed cumulatively. After analysing these two approaches, I offer a third, whereby the intensity of violence should be aggregated when ANSAs are fighting in the same area and at the same time, against a common enemy, even if they are not party to the same coalition.
Coalitions of non-State actors
The fragmented approach presents crucial challenges when a myriad of non-State actors are militarily active in a country: assessing the intensity of violence for each and every one of them could be nearly impossible. Accordingly, it has been suggested that it could be more realistic and legally sound to aggregate the intensity of non-State actors that are members of a coalition. The criteria for the classification of a new NIAC would be threefold. First, each group would have to be organized according to IHL standards. Second, the ANSAs should be part of the same coalition and present ‘a collective approach to fighting.’ Third, the intensity of violence between the coalition and State forces should meet the threshold required by IHL.
Aggregating the intensity of violence of ANSAs only when they are party to the same coalition raises a number of questions. First, international law does not provide a definition of coalition of armed groups. To be sure, a number of indicia can be considered to this end. For instance, Nikolic, de Saint Maurice, and Ferraro suggest indicators such as the:
[E]stablishment of centralized joint command, allocation of areas of responsibilities, sharing of operational tasks (detention, procurement, equipment, transport of troops and other logistics), declarations / agreements describing the tasks assigned to coalition members, existence of common Standard Operating Procedures (SOPs) and/or Rules of Engagement (RoEs), exchanges of tactical/strategic information, existence of an umbrella platform dealing with political issues and communication in the name of the members of the “coalition”, facilitating military operations of one of the actors in the areas under control of another actor, coordinating simultaneous attacks against the common enemy, conducting joint operations, etc.
While these elements could indeed guide the assessment as to whether a coalition exists, determining in practice whether an armed group is party to a coalition or not might be challenging. This is particularly true if we consider that coalitions are often fluid and unstable.
Another crucial conundrum concerns the case when a group leaves the coalition. To address this issue, it is worth recalling the criteria to determine the end of a non-international armed conflict. As specified in the ICRC Commentary to Article 3 common to the Geneva Conventions, a NIAC ends when ‘one of the Parties ceases to exist’ or when there is ‘a lasting cessation of armed confrontations without real risk of resumption.’ On the other hand, ‘a temporary lull in the armed confrontations must not be taken as automatically ending the non-international armed conflict.’
Let’s imagine that a situation is classified as a NIAC by aggregating the violence carried out by members of a coalition, and that not long after that a member leaves the coalition and starts fighting against former co-members of the alliance. As long as it was part of the coalition, the ANSA was considered party to a NIAC against governmental forces. Should the armed group still be party to a NIAC after leaving the coalition?
In order to answer, we should distinguish two different scenarios. If the intensity of violence between the single ANSA and State forces is sufficient for IHL purposes, then we can conclude that there would have been a NIAC between the rebel group and the State regardless of the coalition. Accordingly, to determine whether leaving the coalition has an impact on the classification of the relationship between the ANSA and the State, the aforementioned rules regulating the end of armed conflicts should be applied. But what if the intensity of violence between the ANSA and State forces has never been sufficiently intense per se, and the opposition group was considered party to a NIAC only by aggregating the violence of the members of the coalition? If the violence between the ANSA and governmental forces remains low after the group leaves the coalition, it seems problematic to claim that this is simply ‘a temporary lull in the armed confrontations.’ Should we then conclude that the NIAC between the ANSA and the State ended?
Temporal and geographical continuum
In a recent article, Kleffner posited that, in complex situations like Syria and DRC, ‘there are good grounds for applying a cumulative approach to the determination of whether the requisite level of intensity has been met.’ This would be the case when several organized armed groups engage in armed hostilities ‘on a geographical and temporal continuum.’ In other words, as long as ANSAs fight in the same area at the same time, the intensity of violence should be assessed cumulatively. Similar to the approach analysed before, here the objective is to suggest classification criteria that reflect the reality on the ground. Nevertheless, it seems to excessively relax the intensity requirement: admitting that IHL would be applicable to any armed hostilities taking place between any ANSA and/or State forces operating in the same area during the same period of time would risk to excessively broaden the scope of application of international humanitarian law.
Armed groups fighting against a common enemy
While these alternative approaches present a number of challenges, applying the fragmented approach to complex realities is simply unfeasible, and may lead to the legal paradox of concluding that there are no NIACs in a country even when the cumulative intensity of violence is extremely high. The drawbacks of this are twofold. First, the government would be forced to react to high intensity armed hostilities with law enforcement measures, which are clearly unfit to face these difficult realities. Second, applying IHL might increase the protection of the victims of war. While human rights law is more protective than international humanitarian law, as long as it is still controversial whether non-State actors are bearer of human rights obligations, IHL enhances the protection of the civilian population.
It is therefore suggested that the intensity of violence between organized armed groups and the government or between such groups should be considered in an aggregated fashion when three cumulative conditions are met: namely when a number of organized ANSAs are fighting (1) in the same geographical region; (2) during the same period of time; and (3) against a common enemy. These criteria will be analyzed in turn.
First, the requirement of a geographical and temporal continuum is necessary to distinguish ‘acts of violence that are reasonably grouped together because they occur in an identifiable location and within sufficient temporal proximity of one another from a loose series of acts of violence by different organized armed groups that occur over a wide geographical area.’ In practice, the geographical and temporal boundaries should be determined on a case-by-case basis. Generally speaking, hostilities should not be ‘separable in temporal and geographical terms.’
We may wonder what happens if two groups are based in two different areas and decide to coordinate and act together as members of the same coalition. Does the fact that the violence is taking place in different areas prevent the aggregation of intensity? In other words, is it possible to replace the geographical continuum criterion with the membership to the same coalition? In my opinion, intensity should not be aggregated in this case, even if the two groups are members of the same coalition. According to the fragmented approach, in order to determine the existence of a NIAC the level of intensity should be assessed for each bilateral situation between the actors engaging in armed confrontations. The suggestion to aggregate violence – regardless of the criteria and conditions to do so – serves to answer a specific challenge. Notably, in complex situations such as Syria, CAR, or the DRC, the application of the fragmented approach might be unfeasible because it would be nearly impossible to ‘disentangle the level of violence by each of the armed groups involved.’ Nevertheless, if two groups are based in two different regions, it would not be hard to assess the level of violence separately, even if they are members of the same coalitions. Accordingly, it is submitted that a geographical continuum is necessary to aggregate intensity.
The third criterion, namely fighting against the same enemy, serves the objective of limiting the application of the cumulative intensity approach, without raising the aforementioned issues regarding the coalition approach. In practice, this would mean that when a number of organized ANSAs are fighting against State forces – some of them as part of a coalition and others independently – the intensity of violence would not be assessed looking only at the coalition, but also at other organized groups that are fighting against the government, even if they have not joined the coalition. Similarly, let’s imagine that a group is party to a coalition, but following internal fighting it splinters into two groups: one of them remains in the coalition, while the other leaves it and starts fighting independently. In this circumstance, provided that the two splinter groups meet the organization requirement, it would seem unreasonable to continue applying IHL to the first group, which is still party to the coalition, but not to the other simply because it is fighting independently, albeit against the same enemy.
Considering aggregated violence for the purposes of classification raises a crucial question with regard to the number of armed conflicts taking place in one country. It is submitted that, when a number of organized ANSAs are fighting against a common enemy, in the same geographical area, at the same time, this instance should be regarded as one single NIAC.
A second question regards whether recurring to aggregated violence for classification purposes would have an impact on the criteria to determine the end of an armed conflict. As aforementioned, according to the ICRC Commentary to Article 3 of the first Geneva Convention, a NIAC ends when one of the parties ceases to exist and in case of ‘a lasting cessation of armed confrontations without real risk of resumption.’ The conclusion of a peace agreement or ceasefire per se does not guarantee that hostilities will not continue. Indeed, it is not uncommon that armed confrontations resume after the formal adoption of these instruments. When a NIAC is classified on the basis of aggregated violence, should its end be assessed cumulatively? It is suggested that this should not be the case and that the end of the conflict should be determined using the fragmented approach. Indeed, if one of the ANSAs cease to exist or in case of lasting cessation of hostilities between one non-State actor and governmental forces, it seems reasonable to conclude that the ANSA is not party to the NIAC anymore.
Classifying armed conflicts presents increasingly practical challenges due to the current operational environments in which NIACs take place. The proliferation of armed groups, the creation of coalitions often based on fragile and volatile alliances, and the increasing number of splinter groups make it nearly impossible to apply the traditional fragmented approach in practice. In order to address these pressing difficulties, this post has suggested to aggregate the intensity when a number of organized ANSAs fight in the same geographical region, during the same period of time, and against a common enemy.
It might be argued that this would lead to an over-classification, whereby situations that should not be covered by IHL are classified as armed conflicts. This is not the case. The aim of the suggested approach is not to loosen the intensity criterion in order to allow the application of IHL to situations that should be regulated by human rights law. On the contrary, the objective is to avoid the risk of under-classifying situations where the global intensity of violence calls for the application of IHL, but where the classification of NIACs would be nearly impossible with the strict application of the fragmented approach.
*** Editor’s note: This blog has been written in the author’s personal capacity and may not be interpreted as reflecting the views of the ICRC or the Geneva Academy.
- Jelena Nikolic, Thomas de Saint Maurice & Tristan Ferraro, Aggregated intensity: classifying coalitions of non-State armed groups, October 7, 2020
- Anki Sjöberg, Where are the carrots? Positive discipline for armed groups, March 19, 2020
- Jelena Pejic, Irénée Herbet & Tilman Rodenhäuser, ICRC engagement with non-State armed groups: why and how, March 4, 2021
 ICTY, Prosecutor v. Tadić, Decision on the Defence Motion for Interlocutory Appeal on Jurisdiction, IT-94-1, 2 October 1995, §70.
 ICTY, Prosecutor v. Tadić, Trial Judgment, IT-94-1, 14 July 1997, §562.
 Article 1(2) of the Protocol Additional to the Geneva Conventions of 12 August 1949, and Relating to the Protection of Victims of Non-International Armed Conflicts (Additional Protocol II).
 T. Ferraro, ‘The ICRC’s legal position on the notion of armed conflict involving foreign intervention and on determining the IHL applicable to this type of conflict’ (2015) 97(900) International Review of the Red Cross 1227, at 1229-1230.
 J. Nikolic, T. de Saint Maurice, T. Ferraro, ‘Aggregated intensity: classifying coalitions of Non-State Armed Groups’, Humanitarian Law and Policy, 7 October 2020.
 J. K. Kleffner, ‘The Legal Fog of an Illusion: Three Reflections on “Organization” and “Intensity” as Criteria for the Temporal Scope of the Law of Non-International Armed Conflict’ (2019) 95 International Law Studies 162, at 177.
 J. Nikolic, T. de Saint Maurice, T. Ferraro, ‘Aggregated intensity’ cit.; see also ‘International humanitarian law and the challenges of contemporary armed conflicts. Recommitting to protection in armed conflict on the 70th anniversary of the Geneva Conventions’, Geneva, October 2019, submitted to the 33rd International Conference of the Red Cross and Red Crescent, Geneva, Switzerland, 33IC/19/9.7, at 40-41.
 J. Nikolic, T. de Saint Maurice, T. Ferraro, ‘Aggregated intensity’ cit.
 J.-M. Henckaerts and others, ‘Article 3: Conflicts not of an International Character’ in K. Dörmann, L. Lijnzaad, M. Sassòli, and P. Spoerri (eds.), Commentary on the First Geneva Convention: Convention (I) for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field (Cambridge University Press, 2016), at §488.
 J. K. Kleffner, ‘The Legal Fog of an Illusion’ cit., at 175.
 Ibid., at 177.
 J. K. Kleffner, ‘The Legal Fog of an Illusion’ cit., at 177.
 J. Nikolic, T. de Saint Maurice, T. Ferraro, ‘Aggregated intensity’ cit.
 J.-M. Henckaerts and others, ‘Article 3’ cit., at §488.