The insurgency in Cabo Delgado rarely makes international headlines for long. It lacks the geopolitical legibility of conflicts that western audiences find easier to frame, no clear ideological villain, no simple narrative of good versus bad. What it has, in abundance, is displaced people. Over 600,000 of them, as of February 2026. Nearly half of them children. A humanitarian response requiring US$348 million that the world has not rushed to provide.
This is not a new crisis. It is a protracted one, which is a different and in some ways more damaging thing. Over 345,000 people have been pushed out of their homes across Cabo Delgado, Nampula, and Niassa, 60% of them still within Cabo Delgado, the province that can least afford to absorb them. At the worst of it, more than 100,000 people were crammed into over 100 temporary accommodation centres that were never built for permanence and never resourced for it either. North of 170,000 houses damaged or flattened. Water systems gone. Health facilities gutted. And then, as if the conflict hadn't done enough, early 2026 brought flooding, nearly 600,000 people hit by rains intensifying from late December 2025, another 12,000 homes destroyed, 126 schools knocked out, 13 more health facilities compromised in Cabo Delgado, a province already running on next to nothing. There is a particular cruelty to the way emergencies arrive in Mozambique. They don't wait. They compile.
Professor Bjørn Enge Bertelsen, a social anthropologist at the University of Bergen whose research spans Mozambique's colonial history, civil war and the current conflict, has little patience for the security framing that outsiders tend to reach for when they look at Cabo Delgado. The insurgency's roots, he argues, lie not in imported jihadist ideology but in the long-term political and economic marginalisation of a Muslim-majority province sitting outside the patronage networks of the FRELIMO (the ruling political party in Mozambique) state. Young, unemployed men who saw resources flowing south to Maputo's political elite, who watched a gas economy arrive promising development and deliver enclosure, these were the insurgency's original constituency. The grievance preceded the gun.
That gas economy is not incidental to the displacement crisis. It is structural to it. The Afungi LNG project displaced over 550 households and restricted coastal fishing beyond the armed attacks. TotalEnergies' 2021 decision to freeze the Mozambique project did more than stop construction, it halted critical community compensation and engagement efforts, leaving impacted populations in limbo. And the military response that followed, Rwandan troops, Southern African Development Community forces, has been oriented primarily around protecting the gas peninsula rather than securing civilian populations across the broader affected region. The extraction site grows more isolated, more enclaved, accessible only by sea or air. Where that leads, Professor Bertelsen argues, is not difficult to predict. Enclave development, external military protection of extraction sites, and chronic civilian neglect is a combination with a well-documented political endpoint: it manufactures the very antagonism it claims to be suppressing.
Against that backdrop, 1.7 million people in Mozambique require humanitarian assistance in 2026, within a response framework that needs US$348 million and is operating well below that threshold. The shortfall is not new, but it has been sharpened by the retreat of American engagement, cuts to USAID and reduced US participation in UN humanitarian coordination have pulled significant capacity from a response that was already stretched. Professor Bertelsen notes the geopolitical irony with characteristic directness: as the US steps back, China steps forward. Other actors fill the vacuum, but not equivalently and not immediately, and the communities dependent on that assistance live in the gap between what is withdrawn and what eventually replaces it. That gap has a human cost that transition narratives consistently underestimate.
What protracted displacement does to social fabric is something Professor Bertelsen has studied across Mozambique's longer history of violence. Drawing on his research into post-civil war reconstruction, he notes that displacement is a social rupture as much as a physical one, the relationship between people, land, and community that defines belonging cannot be restored simply by returning to a location. Mozambique after 1992 developed remarkable localised practices for processing collective trauma and reintegrating communities, including rituals of purification and the domestication of war through ceremony. The current conflict is disrupting the very structures those processes depend on. What is being lost is not just housing. It is the customary authority, the land tenure, the community networks that make return meaningful rather than merely a wish to a return to a specific geographical site.
Durable return, when conditions eventually allow it, will require reconstructing those foundations, not just reaching a security threshold. Seasonal patterns already reveal what civilian preference looks like when violence permits: the insurgency tends to ease during the rainy season, when movement is harder, and civilians use those windows to return to land, to farm, to maintain connection with places of origin. People want to go back. The question is whether anything worth returning to will still exist when they can.
In the meantime, 600,000 people wait. The gas flows, or tries to. The troops guard the peninsula. The floods come. And the civilians of Cabo Delgado, who were marginalised before the conflict began and have been displaced by it since, continue to absorb consequences they had no meaningful part in producing.
Further information and opportunities to engage with organisations working in this area are listed below:
https://unac.org.mz/
https://cap-anamur.org/en/projekte/mozambique/
https://www.caritas.org/where-we-work-country/mozambique/
Share your thoughts on this article
Get latest news delivered to your inbox