The Congolese army woke up on May 11 to find villages it had not controlled in months suddenly empty. M23 fighters had begun withdrawing from the Ruzizi Plain in South Kivu two days earlier, leaving Sange and a string of surrounding localities behind. By the end of the week, Wazalendo militia units and Forces armées de la République démocratique du Congo (FARDC) detachments were filing back into territory that, until a fortnight ago, belonged to a group Kinshasa had given up trying to dislodge.
It looked, at first glance, like a victory. It is not. The withdrawal is best read as a tactical repositioning by a movement that retains its strategic depth — and, far more importantly, the backing of the Rwandan Defence Force. Three pieces of evidence make the case.
First, the geography. M23 did not retreat from areas it was about to lose. The group pulled out of the Ruzizi Plain — a relatively flat, indefensible agricultural corridor — while reinforcing its grip on the highlands of South Kivu and on key positions in North Kivu’s Rutshuru territory. On May 13, the group attacked Wazalendo fighters near Nyamilima in Rutshuru, deep in territory it has held since 2024. A movement in genuine collapse does not launch a fresh offensive in another sector the same week it withdraws.
Second, the diplomacy. The pullback dovetails almost exactly with the implementation timeline of the Washington-brokered peace framework that the Tshisekedi and Kagame governments signed in stages through 2025. The UN repatriated 359 Congolese to Rwanda via the M23-controlled Goma corridor on May 10. That is not the choreography of a defeated insurgency; it is the choreography of a negotiated handover, in which M23’s territorial holdings are being consolidated rather than dismantled, and Rwanda’s leverage over eastern DRC is being formalised through international machinery rather than denied.
Third, the sanctions context. The United States in March sanctioned the Rwandan army and senior officials for supporting M23, and earlier this year sanctioned former DRC president Joseph Kabila over his alleged ties to the same group. Those sanctions imposed political costs but no operational ones. Rwandan-backed M23 forces did not reduce their footprint after the sanctions; they reorganised it. The Ruzizi withdrawal is the visible side of that reorganisation.
For Kinshasa, the moment is ambiguous. President Félix Tshisekedi can claim — and his ministers already are claiming — that government forces are reclaiming territory. The propaganda value matters domestically, particularly given the unresolved Ebola crisis: health authorities reported 148 suspected deaths and nearly 600 cases by mid-May. A government that controls more villages, even quietly handed back ones, is a government that can argue it is restoring sovereignty. But sovereignty in eastern DRC is not measured in vacated villages. It is measured in whether Kinshasa can station administrators, collect taxes, and protect civilians in those villages without inviting another round of M23 reprisals. That test has not yet been run.
For Rwanda, the calculus is also subtle. Kigali has secured what it most wanted from this phase of the conflict: a Western-mediated peace framework that treats the DRC-Rwanda relationship as a state-to-state problem rather than as the sponsorship-of-armed-groups problem the UN and the United States have long documented. The withdrawal of M23 from the Ruzizi Plain is therefore a useful gesture. It demonstrates compliance without ceding the strategic positions — the Goma corridor, the Rutshuru highlands, the mineral-rich pockets around Rubaya — that make M23 a permanent fact of life in eastern Congo.
The Wazalendo are the wild card. The umbrella term covers a heterogeneous collection of pro-Kinshasa militias, some of which have their own scores to settle with M23 and with Rwandan-aligned populations in the highlands. Once Wazalendo fighters move into territory that M23 has vacated, the historical pattern in eastern DRC is that ethnic-based reprisals follow. International monitors will be watching for the next few weeks for any pattern of revenge killings or property seizures targeting Kinyarwanda-speaking communities. Those incidents would give Kigali — and M23 — a politically usable pretext to re-enter the same territory the group has just left.
For Kenya and the East African Community, the moment matters because the East African Community Regional Force, withdrawn after acrimonious disputes with Kinshasa, has been replaced by a Southern African Development Community mission that is itself winding down. The question of who provides security in the spaces between M23’s tactical withdrawal and FARDC’s slow re-deployment is, for now, unanswered. Kenya has political and economic interests in eastern DRC — from Equity Bank’s footprint to KenGen’s regional contracts — that depend on those security gaps not metastasising into new flashpoints.
Three things to watch. First, whether the UN-brokered repatriations through Goma continue, and at what scale; the cadence will reveal how durable the peace framework actually is. Second, whether the Wazalendo-FARDC re-deployment into the Ruzizi Plain triggers any ethnic-targeted violence; if it does, expect Kigali to react fast. Third, whether the Tshisekedi government uses the political space to pivot to the Ebola response or burns it on premature victory speeches. The first option buys time. The second invites the next collapse.
M23 has not been defeated. It has been repositioned. The difference will matter more in the next six months than the headlines from this week suggest.
