Deciphering the M23 Conundrum in the Democratic Republic of Congo: Armed Non-State Actor Control, Proxy Occupation and the Tale of Re-incarnated Evolutions – Part II

About the author(s):

Joshua Joseph Niyo

Dr. Joshua Joseph Niyo has recently completed a Swiss National Science Foundation Visiting Research Fellowship at the University of California Los Angeles (UCLA) School of Law. He was also formerly a Teaching Assistant at the Geneva Academy of International Humanitarian Law and Human Rights and Assistant Lecturer at Uganda Christian University School of Law. He recently completed his PhD in International Law (magna cum laude) at the Graduate Institute of International and Development Studies in Geneva, Switzerland, with a research project focused on the normative, conceptual and contemporary considerations in defining the control of territory by armed non-state actors in armed conflict.


Proxy engagement manifests in different situations, but especially in the conduct of hostilities and in the control of territory. Although both occurrences may not be effected contemporaneously by the state in concert with its proxy, it is nearly always the case that any significant advances in both situations, will necessitate the proxy to operate in partnership with its benefactor. Certainly, evidence of support in either situation, provides highly probable indication of joint conduct in the other. This is also a legal question of recurring debate among international law commentators, that is, the general or specific nature of support given to the proxy. The legal examination in this Part addresses this question, teasing out the more appropriate framework of proxy engagement applicable to the M23’s operations. 

Moreover, while scrutinising the legal implications of the structure, operations, and control of territory by the M23 in eastern DRC, it is important to recall the socio-political context (examined in Part I), as well as the impact of foreign states in the evolution and establishment of the insurgency sustained by the ANSA. Is the M23 a linear continuation of the ANSAs that preceded it? Are all these ANSAs – in the end – one and the same in substance? The case for an affirmative answer is very strong, certainly. 

However, even if the answer to these questions could be, “No”; the fact that the M23 carries on the regional territorial focus (eastern DRC), aspirations, motivations, benefactors, and perhaps some members of the previous ANSAs, speaks to its current formation, and whether the armed operations and control of territory carried out in the name of the M23, have been, and could be, linked to foreign states. Consequently, the analysis in this Part commences with an examination of the level of organisation and current formation of the M23, including the suggested links with foreign states. This provides a basis for the legal examination carried out subsequently, on the critical matter of the indirect occupation of territory through control by the ANSA.   

M23 Structure, Organisation and its Territorial Control and the Question of Proxy Engagement

In 2017, almost half a decade after the peace declarations in Nairobi involving the M23, some former M23 fighters under Sultani Makenga fled Uganda to recommence their insurgency, establishing a base at Mount Mikeno, at the border between Rwanda, Uganda, and the DRC. (See International Crisis Group, Easing the Turmoil in the Eastern DR Congo and Great Lakes, pp. 9-10) Subsequently, this small contingent of the M23, launched an unsuccessful armed attack against the DRC state forces in 2021. However, at the beginning of 2022, as the M23 began to regroup – with many more former fighters fleeing Uganda and returning to the DRC – the ANSA launched a major offensive in February 2022, citing the non-fulfillment of the conditions of the 2013 December peace deal by the DRC. Furthermore, the purported need to protect persons of Tutsi heritage in eastern DRC from attacks by the Hutu-led FDLR armed actor, is mentioned as another motivating factor for the upsurge.

In June 2022, the M23 took control of the town of Bunagana, which has a population of about two million people. Although this is not as significant as its seizure of territory in DRC previously, a spokesperson in the DRC state forces called this both a violation of DRC territorial integrity, and effectively an occupation of Bunagana by the Rwandan state forces. Indeed, DRC forces have reported the presence of individuals wearing uniforms of the Rwanda Defence Force (RDF) in M23 camps located in the DRC – an allegation reportedly confirmed by aerial footage and photographic evidence. (See Final Report of the United Nations Group of Experts on the Democratic Republic of the Congo, S/2022/479, para. 67) Certainly, the United Nations Group of Experts have, in a confidential report, confirmed that they have “solid evidence of the presence of, and military operations conducted by, RDF (Rwandan Defence Force) members […] between November 2021 and July 2022”.

Essentially, the M23 displays the features of a centralized fighting hierarchical system (See ICRC’s The Roots of Restraint in War, 38-39). This is complemented by other non-fighting branches of the entity that focus on the political and public communication aspects of the M23. Previously (2012-2013), the M23 had established an administrative authority in the territory under its control, a political cabinet (a president, military leader, executive secretary, and heads of administrative and other departments), a ministerial department on justice, a security department – including an intelligence service, as well as a police force with five divisions. (See Report of The United Nations Joint Human Rights Office on Human Rights Violations Committed by the M23 in North Kivu Province April 2012 – November 2013, paras. 8-9) 

However, with specific regard to the current outfit of the M23, Sultani Makenga continues to provide overall command to the fighting forces, with Yusuf Mboneza leading ground operations. Bertrand Bisimwa is the president of the M23 and is, therefore, its official political leader. Currently, the ANSA is reported to have a fighting force in the mid-hundreds. Furthermore, the M23 maintains its main base on Mount Sabyinyo in the DRC, at the confluence of the borders of the DRC, Rwanda and Uganda, with other camps on Mount Visoke in the DRC. M23 camps are used for training new recruits. The ANSA is reportedly well-equipped with AK-type assault rifles, as well as PKM machine guns and light machine guns. It is also stated to possess 12.7 mm heavy machine guns, and rocket-propelled grenades, with 60 mm mortars and night-vision binoculars. (See Final Report of the United Nations Group of Experts on the Democratic Republic of the Congo, S/2022/479, paras. 64, 65, 66 and 68)

Accordingly, the M23’s high level of organisation and territorial dominance, alongside the apparent support of foreign states, particularly in the present case – Rwanda, prompts the need – at the very least – of a legal assessment as to whether they both operate, and control territory as a proxy. Certainly, their historical evolution does make such an assessment inevitable, in view of the interconnectedness between the M23 and its previous outfits, with the different states in the region.

The Legal Question of Indirect Occupation through Control by an ANSA (Occupation by Proxy): Is M23 Operating as a de facto Agent for Rwanda and other Foreign Sates?

Traditionally, the law of occupation applies where a state establishes its military presence and control over territory in another state. However, contemporary trends suggest that, instead of using de jure members of the armed forces and other similar governmental authorities, a state may exercise such control over another state’s territory through de facto agents. In such situations where de facto agents either contribute to such territorial control, or conduct such control on their own, it is critical in both situations that the de facto agents act on behalf of, or under the control of the foreign (occupying) power, for the controlling state to be considered effectively as an occupying power. (Sassòli, 1399, para. 22) Three paradigms are put forward as illustrating the existence of this notion known as occupation by proxy, or in this case, indirect occupation through control by the armed non-state entity M23. 

First, is the “agency argument”, which is drawn from a purposive construction of Article 29 of 1949 Geneva Convention IV (GC. IV). With reference to both own and occupied territory, the provision establishes that a (state) party to a conflict in whose hands protected persons may be, is responsible for the treatment rendered to them by its “agents”. Consequently, it is argued that although Article 29 GC. IV is not explicit on occupation by ANSAs, it can be interpreted to incorporate circumstances where a state occupies territory of another, through a proxy – including an ANSA – as an “agent”. (Gal, 65-66) Agency under Article 29 GC. IV, which disregards formal criteria of subordination, is argued to be a reasonable legal determination of the situation of occupation, especially regarding conduct by ANSAs under the control of – or specifically, “in the service of” – the foreign state. (Prosecutor v Tadi?, 1997, 298-299; 1958 ICRC GC. IV Commentary, 211-212) This notion of agency, incidentally, tracks rather closely with the concept of “complete dependence” espoused by the ICJ, where entities (like ANSAs) – used ultimately as mere instruments of the state – could, for purposes of international responsibility, be equated with state organs, even where the status does not flow from internal law. (Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v Serbia and Montenegro), Judgment, 2007, para. 392)

Therefore, where ANSAs operate informally “in the service of” the foreign state – or, for that matter, ultimately as mere instruments of the foreign state (as “state organs”), it is critical that they independently demonstrate a capacity to conduct effective control of the territory for the occupation rules to apply to it and the foreign state. (Gal, 66) In this regard, it would be essential to establish whether the M23 operate and control territory in the DRC “in the service of” Rwanda – for the most part – or Uganda. What is clear from reports and the analysis carried out – including in Part I of this commentary, is that the M23’s purposes of the insurgency align with those of Rwanda. However, this does not necessarily mean that the M23 is an “agent” or “state organ” of Rwanda. There would need to be more to establish – even informally – that some sort of proximate subordination exists on the part of the M23 towards the authority of Rwanda. Whereas Rwandan forces might have been sighted with the M23 in recruitment camps, working alongside, or assisting the ANSA, this does not necessarily reflect the agency relationship envisaged within this paradigm. The socio-political history of the ANSA, as well, does not reveal this close affiliation of agency and (or) subordination, even though both Rwanda and Uganda have seemingly supported, sustained and financed the M23 and its predecessors. In a sense, more direct information would be needed to make an affirmative determination of such a high standard of affiliation and subordination.

The second paradigm is termed here as the “effective control argument”, which is derived from the test used to determine attribution of responsibility and (or) the classification of some armed conflicts involving ANSAs. It is contended that, for an ANSA to be determined as functioning on behalf of a state, it is critical that it be operating under the definite instructions of the foreign state, with the state having “effective control” over the ANSA and its actions. The ICJ has been the major exponent of the “effective control” test, essentially maintaining that ascription of conduct by non-state entities (like ANSAs) to a state, requires that the support given by the state to the entity – even if extensive – be sufficient to demonstrate that the state specifically directed every impugned action of the entity. (Military and Paramilitary Activities in and against Nicaragua (Nicaragua v United States of America), Merits, Judgment, 1986, para. 115; Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v Serbia and Montenegro), Judgment, 2007, paras. 399-400, 406) 

M-23 launch attack on MONUSCO in Kiwanja” by MONUSCO is licensed under CC BY-SA 2.0.

In this regard, for a foreign state to be deemed as an Occupying Power, it must exercise “effective control” over the ANSA, which is its proxy. (Sassòli, para. 8.205) Consequently, regarding the M23, the “effective control” test maybe applicable, depending on the conduct under scrutiny. With the proximity of Rwanda to the ANSA, its reported presence in M23 camps, and its ostensible conduct of military operations with the M23, it is not inconceivable that some military activities have been directed by the state. But this is tenuous at best. Moreover, although the socio-political history of the group(s) indicates a high level of support of both Rwanda and Uganda, the M23 maintains a measure of autonomy in its operations, which is reflected in: 1) its local socio-political (DRC) specific motivations for war (unrelated to the Rwandan question), and attempts at integration into the DRC state armed forces; 2) situations where major defeats were suffered by the ANSA(s) with the apparent (tactical and physical) absence of their benefactors at those critical stages of the armed conflict(s); and, 3) the fact that at many instances the states (Rwanda especially) provide military support in fighting alongside the ANSA(s) but not necessarily in directing the latter’s fighting forces.

Third, are what Gal refers to as “Tadi?-type IACs” (international armed conflicts), which flow from the “overall control argument”. (Gal, 60) Accordingly, the International Criminal Tribunal for the Former Yugoslavia (Appeals Chamber) has emphasised that conduct of ANSAs are attributable to the foreign state – where the state exercises “overall control”, whether or not each operation or conduct was specifically imposed, requested or directed by the state. (Prosecutor v Tadi?, 1999, paras. 120, 122) Importantly, within this line of argumentation, the internationalisation of the armed conflict, due to the level of control by the state on the ANSA, suggests that the ANSA can occupy territory on behalf of the state, and have the law of occupation apply to its activities in the occupied territory, even if the foreign state neither necessarily directs the way the ANSA occupies the territory, nor has control over every operation of the ANSA. In effect, the ANSA could be considered as carrying out the occupation. (Gal, 66-67) Although this construction may not be the intended outcome, most expert commentators still endorse the idea that the first part of the occupation by proxy test is whether the foreign state exercises “overall control” over the surrogate armed forces (ANSAs), and the second facet is whether the surrogate (ANSA) in turn exercises effective control on the territory in question. (Ferraro, 158; ICRC Experts Meeting Report, 23) 

It is within this particular paradigm of proxy engagement and occupation of territory that the case can be made that Rwanda (and Uganda) have previously occupied parts of eastern DRC (Goma, especially) through their overall oversight, and provision of material support both to the development, and involvement of the ANSA(s) in the eastern part of the DRC. Indeed, this could apply, as well, to the current control of territory in Bunagana, especially in view of the claims made by both the DRC and the United Nations Group of Experts regarding tangible evidence of the overall military support given directly by Rwanda. Crucially, this overall level of control of the M23 and its previous manifestations is exemplified in the fact that – for the most part – the peace agreements signed to end the various versions of the conflict(s), mainly focused on the regional states as the major protagonists. The ANSA(s) only seemed secondary in priority, at times signing later versions of the agreements. In a sense, there has been an understanding that discussions surrounding the end of violence by M23 and its various pre-forms in eastern DRC requires deals to be drawn at the inter-state level, since they hold sway over the activities of the armed actors. This remains the case with current ongoing inter-state efforts, to end the recent M23 resurgence of armed violence.    


Proxy engagement is a complicated matter. It is especially so where an ANSA represents socio-political motivations of both a wider inter-regional, and local nature. The tension between proxy interests, and autonomous ambitions could make the assessment of responsibility for unlawful conduct particularly tenuous. The M23 does represent a confluence of regional interests, but also, internally, it is illustrative of incarnational evolutions that perpetuate the same insurgency with different faces. In the midst of this conundrum, it – at the very least – is clear that a measure of responsibility lies with regional states, like Rwanda, for the conduct of the M23, especially in areas where it controls territory. With overall control over the activities of the M23, both presently, and overtime, Rwanda and – in the broader sense – Uganda, must maintain cognisance of the effects of the activities of the M23 as potentially acts, which could be attributed to them as Occupying Powers.  As the M23 seeks to expand its territorial influence in eastern DRC – potentially to proportions witnessed in 2012-2013 – state(s) that provide material, tactical as well as in-person military support on the battlefield, will need to pay particular attention to the conduct of the ANSA towards populations in the areas under their de factoauthority.

(Visited 514 times, 1 visits today)

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: