The Lock
Dadaab does not move. More than 400,000 people live across four camps in Kenya’s northeastern drylands, some of whom have lived there for thirty-five years, and the settlement has never been permitted to become the city it plainly is. This immobility is not inertia; it is produced, and the reasons it is produced do not point in the same direction. This is the lock: not a single policy or wall, but the way a set of opposed interests jam together so that the settlement cannot move in any direction. Kenya has justified more than three decades of encampment on security grounds tied to al-Shabaab, and has periodically threatened to close the camp, most notably in 2016, as leverage. But the deeper arrangement is financial, which runs against dispersal: humanitarian aid flows to the camp, channeled through UNHCR and the agencies that deliver parallel services inside it. Yet a refugee who leaves for Nairobi drops out of the funded system altogether, surviving informally with no claim to rations or services. The donor system, for its part, has found the camp cheaper and more controllable than either integration or resettlement, and the international compacts of the last decade have let wealthy states fund containment at a distance while declining to resettle.
The host state and donors are not at odds over the camp; they share an interest in its continuation. Kenya’s periodic threats to close the camp are not a will to end it but a demand to be paid more for keeping it. Donors fund the camp for a reason of their own. The most efficient answer to Dadaab requires no design at all: Kenya grants the refugees citizenship, the camp is dissolved, and people go to Nairobi to build a life or leave Kenya altogether. It would fall to the Kenyan government to confer the status, and to the donor governments—the ones now paying to keep the camp—to underwrite the cost of dissolving it. Yet the donors fund the camp precisely because they fear where those people would otherwise go. Almost everyone I’ve spoken to in the camp wanted to leave Kenya, naming the same destination: the United States, which is exactly the movement the camp exists to prevent. The refugees survive by playing along: they register, comply, and keep their heads down, because visibility carries risk and refusal carries more.
In the camp, immobility turns inward: confined and stratified, violence is produced along its own hierarchies of ethnicity, religion, and gender. This violence is itself evidence of the lock, since a population free to leave would not be held in the conditions that generate it. Violence is worsening as the climate does: a recent multi-country study found that drought exposure raises the risk of non-partner sexual violence against adolescents by close to half, and the drylands around Dadaab are drying further.1 Every party with power over Dadaab has a reason to keep it exactly as it is, and their incompatible aims cancel into stasis. The result is a dirty game whose equilibrium is the camp’s perpetuation, and whose losers are the people inside it.
Against Upgrading
The standard response to a place like Dadaab is to make it better. UN-Habitat has developed upgrading plans for the complex. Upgrading is the humanitarian reflex: improve the drainage, widen the roads, formalize the market, add lighting, deliver services more efficiently. It accepts the settlement as given and works to raise the floor of its misery. This reflex should be refused. To upgrade Dadaab is to accept that Dadaab should exist and merely be managed more humanely. It is to treat a condition that should be intolerable as a problem of administration. The conditions in the camp are not an administrative shortfall; they are the result of a decision to hold people in suspension, and no amount of upgrading touches that decision. Improving the camp makes it more permanent while preserving the fiction that it is temporary. It is the most humane way of doing the wrong thing.
This is not an argument against improving lives. Rather, it is an argument about what improvement, framed as upgrading, conceals. Every road graded and every clinic added deepens the investment in the camp’s continuation and raises the cost of ever undoing it, while the legal fiction of temporariness remains untouched. As a result, the settlement becomes materially more permanent and politically more deniable at the same time. To refuse upgrading is not to abandon the residents. It is to refuse the trade in which their conditions are marginally eased in exchange for the indefinite postponement of the only questions that matter: whether they may stay, whether they may leave, whether they may become citizens of anywhere at all.
If the residents of Dadaab cannot leave because the lock holds them where they are, then the opposite of a better camp is a city so good that others would choose to come to it. Invert the condition completely. Today everyone in Dadaab wants to leave and cannot; What if we were to imagine instead a place people would want to move to? This is not a plan, but a provocation, and its purpose is diagnostic. Citizenship, the efficient answer, is a political act no design can perform. Yet a city desirable enough to attract residents cannot be dismissed as a holding pen, and every material objection such a city answers—the ground is too poor, the water too scarce, the climate too hostile, the site too remote—removes one more alibi behind which the lock hides. Strip the alibis and what remains, exposed, is the decision itself.
A Brief Written in the Field
The women of Dadaab survive by making themselves unseen. They cover to pass as members of the majority; they withdraw from public space after dark; they send their children to fetch water and firewood so their own bodies are not exposed; they contract their movement until the compound is the whole of their world.2 AbdouMaliq Simone established that in cities of the Global South, people become the infrastructure; that social relations substitute for the pipes and grids that were never built.3 Yaffa Truelove showed that it is women’s bodies, specifically, that perform this substitution, and that they pay for it in time, mobility, and safety.4 Dadaab compounds both insights into a single, brutal figure. The women are the infrastructure the camp runs on—water, fuel, food, care—and they must remain invisible in order to carry it, because to be seen in the camp’s public space is to be exposed to violence. Their invisibility is not incidental to how the camp functions. It is the condition on which the camp’s claim to be temporary depends, because a settlement that acknowledged the scale and permanence of the labor sustaining it would have to acknowledge itself as a city. What would the city look like that makes this labor visible? And what does this visibility expose in the process?
Over two visits in September 2024 and March 2025, I interviewed twenty-nine women across Dadaab’s four camps. Minorities were deliberately oversampled, because the camp’s official picture renders them least visible, and it is at the margins that the hierarchy inside the camp becomes most legible. These women’s testimonies are therefore not illustration; they are analytic. The camp has a majority—Somali, Muslim—and its official representation is built to that majority’s measure. The women who fall outside it—South Sudanese, Ethiopian, Congolese, Rwandan, Christian, the recently arrived, those living with HIV—experience a second camp inside the first, governed by a hierarchy the operational picture does not record. A Christian woman covers to pass as Somali not out of observance but to buy a measure of safety; a darker-skinned South Sudanese woman is marked and harassed on sight; a woman barred from selling to the majority is cut out of the economy that might have sustained her. Reading the camp from these positions makes the internal hierarchy legible, and legibility is the precondition for design that does not simply reproduce the existing order at higher quality.5 These women’s testimonies served as the basis for the development of four different scenarios developed by student teams. These are scenarios, not plans to be implemented. A scenario works by deliberate exclusion: it isolates one logic and brackets the others, so that a conventional master plan’s obligation to solve everything at once—housing, mobility, public space, energy, water—does not blunt it. Each takes Dadaab as a real and evolving urban condition and pushes its single chosen problematic as far as it will go. The contradictions between them are not failures to be reconciled away, but the very substance of the argument.
Provocation I: The Feminist City
The Feminist City takes the infrastructure of care that every city treats as peripheral—clinics, schools, kindergartens, kitchens, open space—and makes it the load-bearing spine of the plan. The reproductive labor that the camp renders invisible becomes the city's organizing structure, built and seen. The team relocated the settlement onto loam, rejecting the clay and sand the existing camp sits on as unfit for safe, walkable public space, and organized housing so that shared kitchens and a collective rooftop bind neighboring homes into units of mutual support. Moser named the systematic invisibility of women’s spatial needs in settlement design; the Feminist City is the inversion of that invisibility into the plan’s first principle.6 Where Massey and Kern show how urban space is gendered through the restriction of women’s movement, this scenario designs for the opposite—a city whose structure assumes women’s presence, mobility, and safety rather than their withdrawal.7
The plan’s detail follows from this principle. Housing is grouped into units of five blocks, each unit organized around shared provision rather than the isolated household. A large neighborhood kitchen serves two blocks, while a smaller one serves a single block of up to one hundred residents. A communal rooftop on the sixth level with collective washing and drying space links neighboring buildings and turns the roof into shared social ground. Green backyards hold room for play and rest. None of this is decorative. Each move takes a function that the camp privatizes and hides—cooking, washing, childcare, the labor of keeping a household alive—and renders it collective and visible, so that the work women do can be carried by the city’s structure rather than buried in the individual body. Where the camp forces women to withdraw, contract, and disappear, the Feminist City is built on the assumption that they will gather, move, and be seen. It does not solve the violence the testimonies record. It designs the physical opposite of the conditions that produce it: lit paths instead of dark ones, shared thresholds instead of unlockable doors, a public realm organized around care instead of one organized around risk.
Provocation II: The Climate-Resilient City
The Climate-Resilient City is less a design as it is an indictment. Its central finding, drawn from site analysis, is that all four existing camps sit mostly where heat, flood, and wind concentrate. Fanon described the colonial world as a compartmentalized world, a world divided in two, held apart by a line that space itself enforces.8 Dadaab updates this logic: the division is not between the settler’s town and the native’s town, but between the polity and its exiles. And the enforcer is not a barracks but the ground itself—hostile terrain chosen as the instrument of separation, so that the land does the work of punishment and the state need not. The scenario relocates the city to survivable ground and builds it as a learning from Shibam, the mud-brick desert city in Yemen that has stood for centuries on the passive-cooling logic the camp is told is impossible: dense fabric, narrow shaded streets, thermal mass, height worked against heat and wind. Precedents from Marrakech and Yazd extend the same knowledge. That a desert city of earth has endured for five hundred years asks the one question the camp cannot answer: why were these four sites chosen at all?
The design principles are specific and they are old. Narrow streets, kept to roughly nine meters, hold shade through the day. Small windows limit heat gain; white roofs reflect it. Building heights are set in proportion to street width so that the fabric shades itself. Wind towers serve the denser public buildings, drawing air through without machinery. The city is sited between the main roads, near water, and buffered by planting against the prevailing winds, with districts sized at around 20,000 residents each, so that daily needs sit within walking distance. The climate data that frames this scenario is not reassuring: the site already sees extended periods of heat above what the body can safely tolerate, and mid-century projections push the number of extreme-heat days sharply upward while making the short rains more violent. The scenario’s cooling is passive, drawn from the accumulated intelligence of ancient desert urbanism—orientation, mass, shade, and airflow worked into the built form itself rather than added as equipment. This is not the cheaper option forced by scarcity. Rather, it is the more durable one, proven across centuries in cities that never had mechanical cooling to fall back on. The knowledge of how to build a livable desert city already exists, but it was never brought to Dadaab, because Dadaab was never meant to last.
Provocation III: The Food-Producing City
The Food-Producing City confronts what is arguably the most pressing question of the century: how a city feeds itself in a hostile environment, on a planet where desertification is spreading and population is still climbing. Dadaab runs on food aid that is now being cut, and the ordinary response would be to shore up the rations or teach subsistence farming. This scenario refuses both. It proposes instead that Dadaab become a high-technology food laboratory: a city that produces its own food through science and, in doing so, produces knowledge the rest of a warming world will need. The provocation takes the aid-dependent refugee—the figure humanitarianism casts as a recipient—and recasts her as a food scientist, a producer of exactly the expertise that arid regions everywhere are about to require. The binding constraint of feeding a growing population is not calories but protein. Conventional animal protein is the most resource-hungry way to meet that need: to supply it for a population this size would demand thousands of square kilometers, many times the footprint of the entire complex. High-technology protein production, conversely, collapses the same requirement to a fraction. The choice of high-technology protein therefore follows directly from the numbers. Feeding people here by traditional means is not a failure of effort or will; it is a spatial impossibility. Naming it as such clears the ground for a different proposition entirely.
Humanitarian discourse fixes the refugee as a figure of need—someone to be fed, sheltered, counted, and kept. The Food Producing City refuses that fixity. It proposes that the same people, on the same ground, become the producers of knowledge that the rest of the world is about to need urgently, as the arable belt narrows and the number of mouths grows. The paternalism of the humanitarian system is not incidental but foundational, the assumption on which the whole apparatus runs. To place a camp’s residents at the front of advanced food science refuses that assumption at its root. Dadaab survives on food flown in by the World Food Programme, at high cost and never enough; the ration is now down to 20 percent of a full basket, no oil, only rice.9 A city that grows its own protein breaks the lock of dependency.
The same reversal shows in the money. A refugee camp attracts humanitarian funding: barely enough to sustain, never enough to transform. A food science city attracts investment, research partnership, climate finance: a different order of resource entirely, directed at a different order of future. The resources to make Dadaab a place people would choose to come to have always existed. They were withheld not because they were scarce but because the decision had been made about what this population was for.
Provocation IV: The Fossil-Free City
The Fossil-Free City imagines a settlement that produces no waste, generates its own energy, builds from earth, and can be taken apart without residue once it is no longer wanted. Its logic is circular: production, use, reuse, return to the ground. It builds from clay brick, clay pipe, trees and bamboo, agro-waste panels, and it runs on captured rainwater. Its provocation is chosen disappearance. It imagines a city designed to vanish cleanly, the deliberate inverse of the disappearance forced on women’s bodies. The scenario reaches its limit at energy. It can raise its own walls from clay, lay its own pipes, and harvest its own water from the sky, but figures show that biogas from animal waste would cover only a fraction of the power such a city needs. This gap marks the point where self-sufficiency stops and where a settlement built to owe nothing to the outside still cannot power itself without importing the technology it set out to refuse. The Fossil-Free City is most honest here, at the edge of what circularity can deliver.
The Lock Renamed
Taken together, these four scenarios share a single structure of refusal. Each identifies one material excuse for Dadaab’s impossibility—unsafe ground, hostile climate, food dependency, extractive infrastructure—and dismantles it, not to arrive at a buildable master plan but to remove that excuse from circulation. What no single scenario does, and what none was meant to do, is answer the question of status, and that silence is deliberate. The scenarios go as far as design can go and stop precisely at the boundary where the political begins, so that the boundary itself becomes visible. When every technical objection has been met and the settlement still cannot become a city, only one variable is left holding, and it is not a variable design can reach.
The contradictions between the four scenarios cannot be designed away. The obvious move is to combine the four: take the care spine, the survivable siting, the food laboratory, the circular material logic, and integrate them into one city. But the combination does not resolve anything; it produces a new contradiction. A settlement that is at once reversible and permanent, low-earth and high-technology, built to vanish and built to endure, is not a synthesis. It is a demonstration that the demand for a single, resolved, master-planned answer is itself part of the problem the scenarios diagnose. There is no master plan here because the master plan—the one right city, delivered whole—is the humanitarian fantasy wearing a designer’s clothes.
In March 2025 the Kenyan government launched the Shirika Plan, which states that refugee camp services will be transferred from UN agencies to the counties of Garissa and Turkana by the end of 2027, and by late 2025 Dadaab and Kakuma, another refugee camp, had been chartered as municipalities. While the reclassification is real, it does not change their status. Refugees remain confined to their hosting counties, without citizenship and without freedom of national movement, dependent on a plan whose funding fell away at the moment it was announced. The United States, which had provided some 70 percent of the World Food Programme’s budget for Kenya’s refugees, sharply cut its support, forcing deep reductions in food rations for a population already at the edge.10 A policy that demands self-reliance while withdrawing the aid that made survival possible, on grounds the receiving counties cannot afford to service, offered up through the language of blended finance and investable frontiers: this is the lock renamed, not loosened.
Shirika promises the city in name and withholds it in substance: a municipality with no design, no plan for permanence, no citizenship, arriving precisely as the money vanishes. The four provocations above do the reverse. Each pushes a single argument to its limit and refuses the role of solution, using the rigor of the design to expose what reclassification launders. Both are powerless over legal status, but one renames the cage and calls the renaming development, while the other names the lock and refuses the camp’s terms. The vocabulary is nearly identical—self-reliance, sustainability, urban future—and the function is opposite. Shirika’s self-reliance is aid withdrawal rebranded. The scenarios’ self-reliance is a provocation built to make the withheld city undeniable, and to make visible that its absence was always a choice.
Bothaina Eltigani et al., “Drought Exposure and the Risk of Sexual, Emotional, and Physical Violence against Adolescents in Zimbabwe, Mozambique, and Lesotho: An Observational Study,” The Lancet Planetary Health 10, no. 4 (2026), ➝.
Fabienne Hoelzel, “Invisible by Necessity: Tactical Disappearance and the Embodied Politics of Survival in Dadaab,” in Improvised Urbanism, ed. Gihan Karunaratne (London: Routledge, forthcoming).
AbdouMaliq Simone, “People as Infrastructure: Intersecting Fragments in Johannesburg,” Public Culture 16, no. 3 (2004): 407–29.
Yaffa Truelove, “Rethinking Water Insecurity, Inequality and Infrastructure through an Embodied Urban Political Ecology,” WIREs Water 6, no. 3 (2019), ➝.
Fabienne Hoelzel, “Voices from Dadaab: Women’s Narratives of Survival and Violence,” in Transitory Urbanism, ed. Gihan Karunaratne (London: Routledge, forthcoming).
Caroline O. N. Moser, Gender Planning and Development: Theory, Practice and Training (London: Routledge, 1993).
Doreen Massey, Space, Place, and Gender (Cambridge, UK: Polity Press, 1994); Leslie Kern, Feminist City: Claiming Space in a Man-Made World (London: Verso, 2020).
Frantz Fanon, The Wretched of the Earth, trans. Richard Philcox (New York: Grove Press, 2004), 1.
Fabienne Hoelzel, Women of Dadaab (Staatliche Akademie der Bildenden Künste Stuttgart, 2025).
Nita Bhalla, “Refugees in Kenya Risk ‘Slow Starvation’ as Donors Slash Aid,” Context, September 12, 2025, ➝.
