At the beginning of 2016, I stood on an improvised road in Mocuba, Zambézia, watching families who had once waded through waist-high floodwaters carrying salvaged doors, chickens, and the few remaining bags of food and clothing. On that day, they were returning to houses that had already been condemned, to neighborhoods officially declared uninhabitable.
International disaster responders called it irrational behavior. It did not last long. The houses were destroyed again, deepening a rupture of trust between the state and its citizens. Although experts condemned these acts as risky and irrational, the returnees called it something else. They called it survival.
What I witnessed that day—and in the months that followed in Angoche, Maganja da Costa, and the cyclone-struck coastal cities of Pemba and Beira—forced me to reconsider everything I thought I understood about climate displacement. We speak of migration as a choice. Even more, when we speak about migration, we remain trapped in the assumption that migration means external movement, neglecting internal displacement. We treat relocation as a solution and measure success by counting the number of houses built and families resettled.
But the people I spoke with did not appear to be making choices. It felt like a trap—like navigating a maze of constraints. And at the center of every constraint was land.
The title paradox
Mozambique’s Land Law is, on paper, progressive. It recognizes customary occupation as a valid form of land rights through the DUAT—Direito de Uso e Aproveitamento da Terra (Right to Use and Benefit from Land). Communities that have cultivated the same soil for generations can obtain formal recognition without a cadastral survey. Women, in theory, have equal rights.
In practice, something far more troubling is unfolding.
My research in Mocuba examined the relationship between DUAT allocation and flood risk. The result was statistically clear: the state is significantly more likely to grant formal land titles to residents who have lived longer in a given area, regardless of whether that area is officially designated as a flood zone. In neighborhoods like Mademo, families hold legal tenure over land that the government’s own territorial planning laws declare lethally dangerous.
Is this a policy implementation failure—or a political contradiction made visible?
From a holistic perspective, we have created a perverse incentive structure in which the strongest legal claim to land is rooted in the most dangerous locations. Then we express perplexity when families refuse to abandon those claims for resettlement sites without tenure security, without infrastructure, and without pathways to livelihood.
We must recognize that the problem is not that people are irrational. The problem is that our framework treats land as a logistical obstacle rather than the foundational asset of resilience.
Legality within Environmental Illegality in Angoche
Angoche, on Mozambique’s northern coast, should be a cautionary tale taught in every climate adaptation course. Its mangroves—critical storm buffers—are legally protected. Its high-tide zones are formally prohibited for construction. Yet the prohibition signs do not exist. Licenses are not required. Enforcement does not arrive.
My fieldwork documented that 91.2 percent of constructions in these prohibited zones were built without licenses. Yet they were erected with the approval of local community leaders and the tacit consent of municipal authorities. When public infrastructure itself is built in flood zones and subsequently inundated, the message to ordinary citizens is unmistakable: the law is merely a suggestion.
This goes beyond informality. The state exposes itself deliberately.
Mozambique is not exceptional in this regard. From Bangladesh’s riverine lands to the hillsides of Port-au-Prince, from the neighborhoods of Caracas to the outskirts of Jakarta, governments simultaneously criminalize informal settlements while refusing to provide accessible, serviced, and safe land that would render informality unnecessary. We build a global architecture of exposure and then blame the exposed for their vulnerability.
The Mirage of Resettlement
After Cyclone Idai, approximately 300 families were relocated from high-risk zones to a site called Nomiua in Maganja da Costa. The land had been ceded by a private owner. By any objective measure, it was inadequate—too small to accommodate the designated population, too remote from markets and health facilities, too unfamiliar to the agricultural and fishing livelihoods on which these families depended.
Within months, the resettlement collapsed. Families returned to the risk zones they had never wanted to leave. The legal and logical justification for return appeared at the opportune moment.
This story, with minor variations, repeats itself across disaster-prone regions I have studied. We treat resettlement as a territorial transaction—a land acquisition and transfer of bodies—rather than the creation of a new and sustainable foundation for life. We measure success by the number of households relocated, yet fail to count how many return. We finance emergency response while starving long-term planning.
The literature on climate displacement has long warned that poorly designed resettlements create land conflicts, overburden host communities, strain local administration, and sow the seeds of future tensions. These outcomes are not unpredictable. They are foreseeable consequences that we consistently fail to mitigate.
Women at the Center of the Dilemma
Across all study sites, one pattern emerged with painful consistency. Women—who constitute the majority of Mozambique’s agricultural workforce and bear primary responsibility for water, food, and household fuel—hold the most insecure land rights. Customary systems that recognize male inheritance, joint titling programs that are never implemented, and climate forums from which women are systematically excluded.
This is a first-order resilience failure, not merely a matter of gender equity.
Women hold critical knowledge of local ecosystems, water sources, and adaptive agricultural practices. Their exclusion from planning processes ensures that this knowledge never translates into policy. Programs that could strengthen their tenure—joint DUAT titling, legal aid, community land delimitation—remain underfunded and ignored.
We are designing climate adaptation with one hand tied behind our backs.
What I Learned
After four years of research across Mozambique’s disaster corridors, I arrived at an uncomfortable conclusion.
We do not lack the technical knowledge to manage climate mobility. We know which areas are high risk. We know how to design participatory resettlement. We know that secure tenure enables investment and recovery. We know that women’s inclusion strengthens outcomes. The evidence base, though incomplete, is sufficient for action.
What we lack is the political will to treat land rights as a first-order climate adaptation strategy.
We continue to spend vast sums on reactive relocation and emergency shelter while starving the preventive investments that would make much of this displacement avoidable. We continue to treat land as a secondary concern—a problem to be worked around rather than the foundation upon which resilience is built.
This is a failure of justice.
There is an alternative, and it is not utopian. It requires:
First, aligning land rights with safety. No government should issue formal tenure in zones its own laws designate as uninhabitable. Where such rights already exist, they should be exchangeable—a legal pathway to convert dangerous tenure into safe, well-located land elsewhere, with a resilience premium that recognizes the sacrifice of place.
Second, making zoning legally enforceable. This requires not only amending laws but empowering inspectorates, protecting inspectors, and establishing personal accountability for officials who authorize illegal construction in prohibited zones. Paper prohibitions are not prohibitions at all.
Third, planning land needs before disaster strikes. A national climate land bank—a portfolio of safe, well-located, publicly owned sites held in readiness for future resettlement or ecological restoration—is not extravagant spending. It is the minimum precondition for dignified relocation.
Fourth, investing in hybrid defenses that integrate ecological restoration with livelihood support. The mangroves that once protected Angoche’s coast were destroyed not out of malice but poverty—families had no alternative but to harvest wood for charcoal. Reforestation funds that provide direct income to vulnerable households are simultaneously climate adaptation, social protection, and ecological restoration.
Fifth, guaranteeing transparency. The corruption that diverted humanitarian aid after Cyclone Idai did more than misallocate resources. It catastrophically eroded the public trust upon which any large-scale adaptation program depends. Independent oversight is not bureaucratic formality. It is the social contract made visible.
When an elderly woman in Pemba asked me, “If I leave my land, who will remember that it is mine?”, she was not expressing ignorance of the risks she faced. She was expressing a truth our policy frameworks have yet to accommodate.
Land is not merely property. It is memory inscribed in soil. It is collateral for credit, the site of livelihood, the place of identity. Asking people to abandon it without securing their rights elsewhere is not relocation. It is dispossession.
Environmental justice in the era of climate mobility will not be achieved solely through emissions targets, nor through adaptation finance dispersed through existing channels. It will be achieved when a woman cultivating salt-tolerant rice in a Mozambican estuary holds a title the state will defend, a voice planners will hear, and a destination safer than the place she now inhabits.
That day has not yet arrived. But it is the only destination worth pursuing.