Maps in eastern Democratic Republic of Congo were redrawn quietly on Monday morning. Fighters from the Rwanda-backed M23 armed group withdrew from several positions north of Uvira, including the strategic town of Sange — a crossroads about 30 kilometres up the road from the Burundian border. Within 24 hours, M23’s own communications channels confirmed the pull-back, and the Congolese army cautiously confirmed it from the other side. The town that has changed hands more times than any of its residents can count is, for now, no longer under M23 control.
Whether this is the start of a genuine de-escalation in the eastern DRC, or a tactical repositioning ahead of a different offensive, depends on which map you are looking at. M23 has continued to hold Kamanyola, the strategic town 70 kilometres north of Uvira where the borders of the DRC, Rwanda and Burundi converge. That is not a town one gives up by accident. It is a chokepoint on trade, on troop movements, and on the contraband flows that finance the war. Holding Kamanyola while releasing Sange is the diplomatic equivalent of opening a hand while keeping the wrist locked.
The withdrawal has come under regional pressure. The Nairobi and Luanda processes, briefly revived after the Africa Forward Summit, have for the first time produced something more concrete than a joint statement: a phased pull-back schedule with verification by an East African Community military observer team. Kenya, Tanzania and South Africa have all reportedly contributed to a small, lightly armed observation unit that is now expected to deploy to the verification sites. None of that has been formally announced. The fact that M23 actually moved is the closest thing to a public confirmation that the schedule exists.
Against this hopeful note, the wider security picture remains grim. Rebel fighters in the northeastern Ituri province killed at least 69 people in a single weekend earlier this month, according to local civil-society reports. Around Minembwe town in the South Kivu highlands, pro-Congolese government forces clashed with M23-aligned militias even as the Sange withdrawal was being announced. On May 14, drones struck a market in Mushaki, with AFC/M23 issuing a statement calling the strike “a massacre of civilians.” Whether the drones were Congolese, Rwandan or third-party assets remains contested. The fact that drones are now part of the eastern DRC battlefield, alongside Sudan, is itself a story.
It is worth being honest about what a real settlement would require. The DRC government needs to demobilise the FDLR — the remnants of the 1994 Rwandan genocidaire armed group whose presence in eastern Congo is the official Rwandan justification for backing M23. Rwanda needs to verifiably stop arming and reinforcing M23. The 100-plus other armed groups in the Kivus need to be integrated into a credible disarmament programme. None of that has happened. None of it will happen on the back of a single town’s evacuation. But it could happen if Kinshasa, Kigali and the EAC can build a sequence in which each move is verified before the next is required.
For Kenya, the symbolism matters. The Nairobi process, launched under the Kenyatta presidency and inherited with mixed enthusiasm by the Ruto administration, is the only continental peace track on the DRC that has not collapsed. Kenya cannot afford for it to fail — not after Nairobi positioned itself, at the Africa Forward Summit, as Africa’s diplomatic capital. The cost of leadership is being available when the dossier gets harder, not just when there is a photo opportunity.
What to watch next: whether the East African Community verification team actually deploys to Sange in the next 14 days; whether M23 hands over Kamanyola, the real prize, on any timetable; whether Kinshasa moves credibly against FDLR positions in North Kivu; and whether Rwanda is willing to put its own troop redeployments under independent observation. Eastern DRC has seen ceasefires die in their first month before. The next thirty days will tell us whether this one has any chance of joining the small list of African settlements that actually held.
