The Absentee

On April 9, the March 23 Movement announced that its delegation was en route to Switzerland for a new round of talks with the Congolese government, scheduled for April 13. The venue was supposed to be Doha. Qatar has facilitated the direct negotiation channel between the DRC and M23 since the Doha Framework was signed in November 2025. But Doha experienced repeated missile strikes during the Iran war, and the diplomatic capacity that sustained the process was consumed by the larger conflict. DRC Foreign Minister Therese Kayikwamba Wagner told diplomats that the process is in a state of “paralysis.”

The talks move to Switzerland. The framework stays the same. So does the structural reason nothing has been resolved.

There are three parallel negotiation tracks for eastern Congo’s war. Each produces agreement by excluding the party who would prevent it.

The Washington Accords (March 17-18, 2026): DRC and Rwanda, mediated by the United States. The parties agreed to a scheduled disengagement of Rwandan forces from defined areas in DRC territory, time-bound DRC efforts to neutralize the FDLR, and a joint security coordination mechanism within thirty days. Absent from the room: M23 — the armed group that holds the territory, controls the mines, and administers a parallel government across North and South Kivu.

The Doha Framework (November 2025): DRC and M23, facilitated by Qatar, now moving to Switzerland. Eight implementation protocols outlined. Two signed — ceasefire monitoring and prisoner-of-war exchange. Discussion on the remaining six has not begun. Absent from the room: Rwanda, which the United States sanctioned on March 2 — the RDF as an entity and four named commanders — for “direct operational support” of M23, including training, combat alongside, and providing GPS jamming systems and drones.

The AU, SADC, and EAC residual tracks: overlapping regional frameworks, all parties nominally present, no enforcement mechanism.

Each track’s absentee is the party whose consent makes implementation possible. Washington produces an agreement between DRC and Rwanda about M23’s territory — but M23 is not party to it. Doha produces protocols between DRC and M23 about a ceasefire — but Rwanda, which controls M23’s military capacity, is not party to it. An agreement M23 signs without Rwanda’s military endorsement is unenforceable. An agreement Rwanda signs without M23’s political consent is unimplementable.

These exclusions are not accidental. DRC cannot seat M23 at a state-to-state table without legitimizing it as a sovereign counterpart. Rwanda cannot appear at the Doha table without acknowledging the control it formally denies. Each exclusion serves the political needs of the party that designed it. The result is three processes, each structurally incapable of producing what the other two require.

The CIC/NYU Congo Research Group published “The M23: A Fractious, Entrenched Rebellion” in April 2026, as the Switzerland talks were being announced. M23’s stated demands include sweeping governance reforms and the potential resignation of the Congolese government. Its political umbrella — the Congo River Alliance, led by former electoral commission head Corneille Nangaa — has publicly stated its aim to overthrow President Tshisekedi. The military wing wants to consolidate control over North and South Kivu. These objectives diverge. The negotiating partner does not agree on what it is negotiating for.

High desertion rates among recruits. Administrators with no governing experience running a parallel state. Internal rifts between M23 leadership and its Rwandan backers spanning decades. The rebellion is entrenched but not cohesive — a fractured occupation producing contradictory demands from different factions, held together by Rwanda’s military infrastructure, which does not appear at the table where M23 sits.

Rwanda’s condition for withdrawing its forces from eastern Congo: the FDLR must be neutralized first. The FDLR — Forces Democratiques de Liberation du Rwanda — is a Rwandan Hutu armed group whose origins trace to perpetrators of the 1994 genocide, operating in eastern Congo for three decades. Rwanda’s security concern is real. Its condition is also, functionally, a veto.

The DRC launched FDLR disarmament Phase 1 on March 30. General Ychaligonza deployed from Kisangani with three battalions. “The FDLR must disarm, whether willingly or by force.” A reception center was established to process ex-combatants. As of April 10: no confirmed surrenders.

The US assessment of FDLR neutralization progress is due April 15. If Phase 1 has produced nothing by then, Rwanda’s condition remains unmet. Kagame, earlier this month: “No amount of sanctions or insults will change Rwanda’s defensive measures.”

The sequence is circular. Rwanda conditions withdrawal on neutralization. The DRC cannot effectively operate against the FDLR in territory controlled by M23 — which is there because Rwanda has not withdrawn. The condition produces the conditions for its own permanence.

What is being fought over, and who is dying in the interim.

M23 has controlled the Rubaya coltan mines since May 2024, taxing production at over $800,000 per month. Rubaya accounts for roughly 15 percent of the world’s tantalum supply.

On January 28, a mine collapse at Rubaya killed more than 400 people — artisanal miners, children, traders from the surrounding villages. On March 4, a second collapse killed more than 200, including 70 children.

In February, the DRC offered the Rubaya mine to the United States in a proposed minerals partnership, seeking $50-150 million in investment to restart operations. The state is offering an asset it does not control, to attract a power whose leverage it hopes will help recover it. The mine produces revenue for the group that holds it, deaths at a rate no present authority prevents, and a diplomatic instrument that assumes the state has something to trade.

The international architecture that might enforce any agreement is contracting at the moment it would need to expand.

South Africa announced in January that it would withdraw its SANDF contingent from MONUSCO — ending 27 years of South African peacekeeping in the DRC. Four hundred fifty troops remain on a bridging deployment through July 31. Full withdrawal by end of 2026.

The EU confirmed it will end funding for Rwanda’s deployment to Mozambique’s Cabo Delgado province in May — approximately 4,000 RDF troops protecting ExxonMobil and TotalEnergies LNG terminals from an Islamic State-linked insurgency. Rwanda has stated it will withdraw if funding is not secured. The same force the United States sanctioned for supporting M23 in Congo is the security guarantee for European energy infrastructure in Mozambique. The EU cannot politically fund a US-sanctioned force. It cannot afford the security vacuum if the force leaves.

On April 2, M23 withdrew from Katondi — its most forward position in Lubero district. FARDC entered Kipese. Headlines noted progress. But M23 had simultaneously redeployed elsewhere in Lubero and suspended withdrawals after FARDC attempted to retake other localities. Each side executes a claimed withdrawal while accusing the other of violation. The pattern repeats before every round of talks.

Two of eight Doha protocols have been signed. The two that were signed — ceasefire monitoring and prisoner exchange — are the two that do not require the absent party’s active consent to function. The remaining six do. Discussion on them has not begun.

The talks resume Monday in a new country. The venue moved. The absentee didn’t.

Sources