Collective Self-Defense and the Internationalization of Armed Conflicts in Eastern DRC

Nearly a decade after the Democratic Republic of the Congo (DRC) signed a ceasefire agreement with the March 23 (M23) rebel movement, violence has once again erupted in the North-Kivu province. On December 15, 2023, the Southern African Development Community (SADC) deployed a Mission to the DRC (SAMIDRC), claiming in part it was acting in collective self-defense. Later, on February 17, 2024, the United States condemned the “Rwanda-backed” M23 armed group and called on it to withdraw all Rwandan Defence Force personnel from the DRC.

SADC’s entry into the conflict and the accusations against Rwanda, regarding both its support for the M23 and its direct involvement in the conflict, raise a number of questions under the jus ad bellum and jus in bello. From the jus ad bellum perspective, SADC’s reliance on collective self-defense is particularly interesting to note. It may suggest that SADC interprets the right of self-defense as applying against non-State actors (the M23). Alternatively, SADC may consider that the actions of the M23 are attributable to Rwanda (although it does not seem to have publicly accused Rwanda of supporting the movement).

The latter would in turn provoke questions concerning the jus in bello. The situation in the DRC is considered a non-international armed conflict (NIAC). That Rwanda may be directly supporting and exercising overall control over the M23, and that its troops may be present in the DRC, raises questions whether the situation has reached the threshold of an international armed conflict (IAC).

The SADC Deployment to the DRC and Collective Self-Defense 

SADC, composed of sixteen Member States and forming one of the African Union’s (AU) eight regional economic communities, is the leading organization involved in matters of peace and security in the Southern African region. Responding to the increase in hostilities, on July 11, 2023, SADC approved the deployment of SAMIDRC (see also here and here). It deployed in December 2023, involving troops from Malawi, South Africa, and Tanzania.

Although the SADC Extraordinary Summit makes no mention of the legal basis for the deployment, a press statement in January 2024 noted the deployment was “in accordance with the principle of collective self-defense and collective action” under the SADC Mutual Defence Pact of 2003. The SAMIDRC mandate itself also invokes the provisions of the Mutual Defence Pact relating to collective self-defense.

Article 6(1) of the Mutual Defence Pact states, “An armed attack against a State Party shall be considered a threat to regional peace and security and such an attack shall be met with immediate collective action.”

The SADC Summit authorizes each State party to participate in collective action “in any manner it deems appropriate” (art. 6(2), (3)). The fact that SADC has invoked its Mutual Defence Pact suggests it is indeed acting in collective self-defense. However, whether SADC considers an armed attack has taken place in the DRC is open to question. Similarly, whether it interprets the actions of the M23, a non-State actor, as constituting an armed attack sufficient to invoke collective self-defense remains unclear.

Such an interpretation would seem to be contrary to the recently adopted Common African Position on the Application of International Law to the Use of Information and Communication Technologies in Cyberspace. Albeit concerning international law in cyberspace, the AU affirms that “the right of self-defense is triggered solely if an armed attack is attributable to a State according to the applicable rules of customary international law of State responsibility” (para. 43).

Whether the actions of the M23 are attributable to Rwanda is a separate question. Recent statements by some States suggest Rwanda is not only supporting the M23 but also has its own troops present in the DRC. In addition to condemning Rwanda’s support for the M23, the United States called for the immediate withdrawal of all Rwandan forces from the DRC and the removal of its “surface-to-air missile systems” (see also here and here). A day later, France called on Rwanda to “cease all support to the M23 and to withdraw from Congolese territory.”

SADC leaders have previously sought to invoke collective self-defense when they deployed the SADC Mission to Mozambique in 2021 to counter an Islamic State affiliate in the Cabo Delgado province. Ultimately however, the deployment was authorized under the SADC Protocol on Politics, Defence and Security Co-operation and with the consent of the Mozambican government. SADC’s position on whether the actions of a non-State actor are capable of amounting to an armed attack to which self-defense could be invoked therefore remains unclear.

Collective Self-Defense Against Rwanda?

SADC’s invocation of collective self-defense in the DRC could be explained by Rwanda’s involvement in the conflict. SADC’s previous communiques have condemned the M23 and other terrorist groups operating in the DRC “with support from foreign aggressors.” Although SADC has not explicitly named Rwanda as an aggressor or accused it of supporting the M23, there is clear animosity between the two.

On February 12, in a letter to the UN Security Council, Rwanda objected to proposals that MONUSCO, the UN peacekeeping mission in the DRC, should provide logistical and operational support to SAMIDRC, claiming it would fuel further violence. Days later, following what it called a dramatic military build-up by the DRC, Rwanda indicated it had taken “measures to ensure complete air defense of Rwandan territory, and to degrade offensive air capabilities, following the introduction of advanced Chinese CH-4 attack drones by DRC in 2023 . . . .” On February 18, the DRC accused Rwanda of orchestrating a drone attack against Goma International Airport.

Considering Rwanda’s involvement in the conflict, it may well be that SADC considers that it is responding to an armed attack by a State. This assessment could be based either on the direct actions of Rwanda’s armed forces or Rwanda’s “substantial involvement” in the activities of the M23 (para. 195).

Be that as it may, a further query that arises regarding SADC’s invocation of collective self-defense concerns the requirements laid down in the ICJ’s Paramilitary Activities (Nicaragua) case, that a State which has been attacked must declare itself the victim of an armed attack and make a request for assistance (paras. 195, 199). There is little evidence to suggest either of these requirements has been met.

The DRC has not publicly declared itself the victim of an armed attack and there is no record of any request for assistance made to SADC (at least not within SADC official decisions and statements). While State practice suggests a declaration per se may no longer be required, Professor James Green recently concluded that State practice indicates “sufficient opinio juris” that a request for assistance is mandatory for the exercise of collective self-defense (p. 164-165).

The only publicly available information regarding a request by the DRC pertains to a statement on X by (troop-contributing) Tanzania’s Minister of Foreign Affairs and East African Cooperation, stating that the SADC deployment was “a result of the August 2023 SADC Summit decision to accept a request by the DRC government, in conformity with SADC Mutual Defence Pact, to deploy a military mission to assist it [to] address security challenges on its Eastern part.” While no explicit request by the DRC is yet on record, it is certainly clear that the SADC deployment has the full consent of the DRC government.

In addition, it is also worth noting that neither SADC nor the DRC has informed the UN Security Council that they are acting in self-defense (individual or collective), a requirement expressed in Article 51 of the UN Charter. Moreover, Article 6(4) of the SADC Mutual Defence Pact obliges it to report “any such armed attack, and measures taken in response” to both the UN Security Council and the AU Peace and Security Council. As has been well noted, while the reporting requirement is procedural in nature and does not affect the invocation of the right of self-defense, the absence of reporting may be indicative when assessing the legality of the exercise of the right (Ruys, p.70-72; Green, p. 131-132).

Given the above, it remains unclear whether the deployment of SAMIDRC is indeed based on collective self-defense or simply on the consent of the DRC government. Rwanda’s alleged involvement in the DRC and its support to the M23 may lend credibility to SADC’s invocation of self-defense. Meanwhile, both SADC and Rwanda’s involvement in the DRC necessitate a reconsideration of the classification of the conflict.

Conflict Classification: Why it Matters 

It is well established that the M23 is a party to a NIAC with the armed forces of the DRC. This NIAC exists in a tangled web of parallel NIACs ongoing in the DRC’s Kivu region. The M23 (which laid down arms in 2013) has seemingly taken up arms once again on the territory of the DRC, perhaps from as early as 2021.

There are numerous suggestions that the M23 has close ties with Rwanda and that in turn Rwanda has directly supported the M23 (see here, here and here). The nature of this relationship is particularly relevant for the purposes of conflict classification as Rwanda’s control over the M23 may internationalize the pre-existing NIAC.

The classification of the conflict involving the M23 has practical and legal implications for all the parties to the conflict, including those States that are supporting the DRC. First, there is growing support for an argument that the gap between the law of IAC and the law of NIAC is narrowing (Stewart, p. 323; Dwyer & McCormack, p. 50-52; Sassòli, p. 210-213). However, these are separate bodies of law and the distinction between the laws that apply to each type of conflict reflects a reality, regardless of the practice of some States in applying the corpus of the law of IAC to NIACs (Bartels, p. 595; Milanovic & Hadzi-Vidanovic, p. 256-57).

There are many legal justifications for having two distinct sets of rules. These include: the hesitation of States to recognize non-State armed groups as legitimate; the fact that prisoner of war status applies only during an IAC; that combatant privilege is not extended to members of organised armed groups participating in NIACs; and that the deprivation of liberty during IAC is regulated robustly whereas only basic rules apply in the treatment of detention during NIACs (Bartels, p. 595).

Consequently, conflict classification from an operational perspective is important. Practically, this means that if the M23 fighters belong to Rwanda (for the purposes of GC III, art. 4(a)) or if Rwanda exercises overall control over the group, they are entitled to prisoner of war status. In addition, if the conflict is internationalized through Rwanda’s exercise of overall control, the full laws of IAC apply. If, however, the M23 does not “belong to” or is not subject to the “overall control” of the Rwandan Armed Forces, its members are not entitled to prisoner of war status and the laws of NIAC apply.

How do NIACs become Internationalized?

There are conflicting reports on the nature (or even existence) of a relationship between Rwanda and the M23. In addition to the United States’ condemnation of Rwanda for supporting for M23, the UN has suggested that Rwandan Armed Forces are fighting alongside the group in the DRC. Rwanda continues to deny these allegations. However, to classify the conflict and determine whether the law of IAC or the law of NIAC applies, it is necessary to establish the precise relationship between Rwanda and the group.

If Rwandan Armed Forces are fighting alongside the M23 in the DRC, then two parallel conflicts exist: an IAC between the DRC and Rwanda; and a NIAC between the DRC and the M23. If, however, the links between the M23 and Rwanda are sufficient to show that Rwanda exercises overall control over the group, the NIAC may be reclassified as an IAC.

Thus, a foreign State or States can internationalize an armed conflict by offering outside support to an organised armed group party to a conflict. Professor Kubo Mačák defines internationalization as “the process of transformation of the legal nature of a prima facie NIAC, which renders the law of IAC applicable to such a conflict” (p. 27). Djemila Carron similarly describes “internationalization” as “the shift from a NIAC to a single IAC” (p. 1032).

Overall Control

The level of intervention required to change the classification of a conflict is that of overall control (Tadić para. 137). This test is widely considered the standard for the purposes of conflict classification by scholars (Cassese; Carron). According to the test,

The control required by international law may be deemed to exist when a State (or, in the context of an armed conflict, the Party to the conflict) has a role in organising, coordinating or planning the military actions of the military group, in addition to financing, training and equipping or providing operational support to that group . . . .

Discussions at the UN suggest weapons have been directed to the M23, including surface-to-air missile systems (see here and here). The most recent discussions at the UN Security Council indicate the M23 is in possession of sophisticated weaponry, serving as evidence of external support to the group. It is also pertinent that the UN Group of Experts on the DRC has previously noted that Rwanda is providing the M23 with weapons, ammunition, and uniforms.

To satisfy the overall control test, evidence of more than “financing, training, equipping or providing operational support” is required (Tadić, para. 137). To qualify as overall control, it is necessary to show that Rwanda has a role in the organisation, coordination, or planning of the military actions of the M23 (Tadić, para. 137). In the context of the DRC, the UN, European Union, United States, and several other States (see here, here, and here), have requested Rwanda to withdraw its own forces from the territory (see also here and here). It must be stressed however that mere presence of the Rwandan Armed Forces in the proximity of the M23 is insufficient to show overall control; a sufficient organisational relationship must also be shown (Prosecutor v Prlić et al, para. 284).

The final report of the UN Group of Experts on the DRC provides insights concerning the extent to which the Rwandan Armed Forces coordinate their military activities with the M23. The report notes that,

The Group of Experts obtained further evidence of direct interventions by the Rwanda Defence Force (RDF) on Democratic Republic of the Congo territory, either to reinforce M23 combatants or to conduct military operations against the Forces démocratiques de libération du Rwanda (FDLR) and local armed groups.

While Rwanda is highly critical of the report, previous reports by the Group of Experts similarly found evidence of co-ordination between the M23 and Rwanda. A mid-term report from 2022 highlighted Rwanda’s provision of reinforcements for specific operations concerning territorial gains “in particular when these were aimed at seizing strategic towns and areas” (para. 49). This organised movement of Rwandan forces and M23 fighters indicates a certain level of cooperation.

At present, it remains unclear who takes battlefield decisions. However, this is not necessarily a determining factor for purposes of overall control. In his monograph, Professor Mačák comments,

. . . the requirement of a joint use of force is not equivalent with the demand that the insurgent group be subordinated to the intervening state, or even subjected to one line of command. It suffices that the two entities partici­pate in the organization, co- ordination, or planning of their military activities. In such a case, neither of the parties can be said to be using force autonomously. Their activities have become sufficiently intertwined to justify the conclusion that ‘a single armed conflict is being fought, albeit with multiple actors participating in it’ (p. 103).

Professor Mačák finds support for his reasoning in the Prlić appeal judgment (para. 284) which held that to satisfy the overall control test it is not a legal obligation to establish that the outside State “maintained the ultimate decision-making authority and command over each and every military operation” (Mačák, p. 103). Therefore, the overall control test does not require that Rwanda makes all or even the majority of battlefield decisions. Evidence of organization, co-operation and/or planning—which is substantiated in the final report of the Group of Experts on the DRC—is sufficient to suggest overall control.

The available public information therefore suggests that Rwanda does exercise overall control over the M23, meaning that the preexisting NIAC is transformed into a single IAC. It should however be stressed that the Rwandan Government has repeatedly denied its troops are present in the DRC and claims that it is not supporting the M23 (see here and here). The M23 has also refuted any cooperation with the Rwandan government.

Concluding Remarks

Should Rwanda’s relationship with M23 meet the threshold of overall control, its involvement in the conflict has changed the nature of the pre-existing NIAC between the M23 and the DRC. The conflict is therefore transformed into a single IAC between the DRC and Rwanda.

Taking this hypothetical further, States contributing forces in support of the DRC would be engaged in a conflict with Rwanda. This would mean that the SADC deployment in the DRC (composed of troops from Malawi, South Africa and Tanzania) are engaged in an IAC with Rwanda. This could in turn indicate that a major escalation of hostilities in the region is underway.

On this basis, the situation in the DRC should be approached with caution. Considering SADC’s recent involvement in the conflict in the DRC, its reliance on collective self-defense, combined with the potential that Rwanda exercises overall control over the M23, the conflict may in fact be international in nature. Any reclassification of the conflict in the DRC therefore requires restraint and an objective assessment of the evidence (in particular of Rwanda’s involvement therein). Open hostilities between the various actors currently involved may have major repercussions for the entire sub-region, one already plagued with untold violence and suffering.

***

 

 

 

Photo credit: MONUSCO Photos