The City That Drinks Its Dead
Accra has been flooding since before independence. The water is not the problem. We are.
There is a photograph that should be required viewing for every Ghanaian elected to public office. It was taken in April 1960, published on the front page of the Daily Graphic under the headline ‘When the Rains Came to Accra,’ and it shows exactly what you think it shows: streets submerged, residents wading, traffic at Kwame Nkrumah Circle locked for hours while people bailed floodwater from their homes with buckets. The photograph is sixty-six years old. It could have been taken last Sunday.
This is not a metaphor. It is a measurement. Sixty-six years of the same image, the same locations, the same stunned expressions, the same official promises and the same silence that follows the promises, lasting precisely until the next rainy season when the water rises again and we are invited, once more, to act surprised. Accra’s floods are among the most documented, most predicted, most thoroughly analysed urban disasters on the African continent. They are also, with a consistency that would be remarkable if it were not so deadly, among the least effectively addressed.
What we are dealing with is not a natural disaster. To call it one would be lazy, and worse, dishonest. Accra sits on a coastal plain that is, by the standards of West African geography, well positioned for drainage. It is not below sea level, like Amsterdam. It does not sit astride a great river prone to catastrophic inundation, like Vienna on the Danube. It receives no meteorological surprises, we know the rains come in May, in June, every year, as reliably as the calendar. What Accra suffers from is not nature’s caprice. It is the slow, cumulative violence of a city built against itself, by people who knew better, overseen by institutions that chose not to act, in a democracy that has held seven elections and produced five presidents and has not, in thirty years of constitutional governance, managed to drain its own capital.
The floods are not an act of God. They are a ledger. And the entries have been accumulating since before independence.
What the Water Remembers
Cities have long memories. Water has longer ones. The Odaw River, which drains much of the Accra basin into the sea at Korle Lagoon, has been following the same course for longer than the city has existed. The wetlands that once lined its banks, the swampy, unglamorous, economically unproductive-looking land that colonial planners and postcolonial developers alike regarded as waste, were not waste. They were the city’s immune system. They absorbed excess water. They slowed runoff. They held, in their saturated soil, the overflow that now holds the city hostage every rainy season. We filled them with concrete. We built on them. We issued permits for structures on floodplains because the assemblies needed the revenue and the developers needed the land and the voters needed the housing and nobody, at any point in this chain of mutually convenient decisions, was required to answer for what would happen when the rains came.
This is the first and most fundamental truth about Accra’s floods: they are not caused by rain. Rain is merely the auditor. The floods are caused by every building permit issued on a waterway, every drain blocked with solid waste, every wetland filled for a car park or a church or a residential estate. The rain arrives, performs its audit, and presents the bill. We pay it in property. We pay it in livelihoods. We pay it, in the worst years, in lives.
The heaviest recorded rainfall in the city’s history fell in June 1959, a volume of 7.56 inches that, a meteorological spokesman reported at the time, nearly brought normal life in the city to a standstill. That was before Accra became a metropolis of four million. Before the peri-urban sprawl consumed the green buffer zones. Before the Odaw basin was treated as a municipal dumping ground. The city of 1959 flooded under extraordinary rainfall. The city of 2026 floods under ordinary rainfall. That is the measure of what six decades of accumulated negligence has cost us.
Two Cities and Their Water
Consider Amsterdam. The Dutch capital sits, on average, two metres below sea level. This is not a metaphor either, it is a physical fact that has governed Dutch urban policy, Dutch engineering ambition, and Dutch national psychology for the better part of five centuries. The Netherlands has approximately 3,500 kilometres of primary flood defences, a network of pumping stations, canals, dikes, and water storage basins so sophisticated that the rest of the world sends engineers to study it. The Dutch do not merely manage water. They have incorporated the management of water into their national identity, their planning law, their constitutional framework. The word they use, waterstaat, has no precise English equivalent, because no English-speaking nation has needed to make water governance a foundational concept of statehood.
The Dutch built this system not because they are inherently more disciplined than other peoples, but because the alternative was extinction. When a nation lives below the sea, it learns, with great urgency, to take water seriously. That urgency produced institutions. Those institutions produced engineers. Those engineers produced infrastructure. And that infrastructure has held, imperfectly, always imperfectly, and with continuous reinvestment, for centuries.
Now consider Vienna. The Austrian capital sits on the Danube, one of Europe’s great flood rivers, and was catastrophically inundated throughout the nineteenth century. The great Danube flood of 1830 submerged much of the city. The flood of 1862 killed dozens and destroyed hundreds of buildings. Rather than accept periodic inundation as the cost of living on the river, Vienna embarked, from 1870 onward, on one of the most ambitious flood-mitigation projects in European history: the construction of a regulated Danube channel and, a century later, the Neue Donau, the New Danube, a parallel relief channel that doubles as a recreational waterway and public park. The Donauinsel, the long island created between the two channels, is today one of Vienna’s most beloved public spaces, hosting festivals, cycling paths, and beaches. Vienna converted its flood liability into a civic asset. It did not do this cheaply. It did not do this quickly. It did so because its institutions possessed the memory of disaster and the will to prevent its recurrence.
Amsterdam sits below the sea and stays dry. Vienna sits on a flood river and has turned its floodplain into a park. Accra sits above sea level on a coastal plain and drowns every June. The difference is not geography. It is governance.
Accra asks for no such heroism from its geography. It does not need to hold back the sea or tame a continental river. It needs to stop filling its wetlands, clear its drains, enforce its planning laws, and remember, from one rainy season to the next, what happened the last time it did not. These are not feats of engineering genius. They are feats of institutional memory and political will. And they are precisely what Accra has lacked, consistently, across six decades and five presidential administrations.
The Architecture of Forgetting
There is a particular political ritual that follows every major flood in Accra, and it is worth describing with precision because its very familiarity is part of the problem. The rain falls. The city floods. Officials tour the affected areas, usually accompanied by cameras. Statements are issued: the drainage will be cleared, the illegal structures will be demolished, the waterways will be restored, the budget will be found. There is a task force. There is sometimes an aerial survey. There is, occasionally, a specific sum of money, a sixty million cedis here, a World Bank facility there, announced with the gravity of genuine intent. And then the water recedes, the cameras leave, the task force is disbanded or forgotten, the structures remain standing, and the city waits, with the patient fatalism of long experience, for next year’s rains to come and begin the ritual again.
This is not incompetence. Incompetence would be easier to fix. This is something more structural and more troubling: a political system that has learned to treat flooding not as a governance failure to be solved, but as a recurring spectacle to be managed. The flood is an event. The promise is the response to the event. The forgetting is what happens between events. And the cycle continues because no one, in the architecture of Ghanaian democratic accountability, is required to answer for the gap between the promise and the delivery.
One Member of Parliament put it with unusual candour: regulation in Ghana is failing, he said, because revenue has been placed at the heart of it. Building permits have become income streams for district assemblies rather than instruments of planning control. A structure on a waterway generates permit fees. The same structure, when the rains come, generates floodwater. The assembly collected the fee in January. The family lost their home in June. No one has been held responsible for the connection between those two events. No one is ever held responsible. And so the permits keep being issued, the structures keep being built, and the city keeps drowning, one rainy season at a time.
Who Actually Drowns
The floods of Accra are not democratic. They visit the whole city, in recent years the flood map has expanded so relentlessly that no neighbourhood can claim immunity, not Legon, not the Airport residential area, not Dzorwulu, but they do not kill equally. The dead, in Accra’s floods, are drawn almost exclusively from the same population: the people who could not afford to live anywhere else, who built or rented in the only places available to them, which happened to be floodplains, riverbanks, and drainage corridors, because those were the lands no one else wanted until they were needed, at which point they were allocated not to the poor but to developers.
This is the human geography of urban flooding the world over, and Accra is not unique in it. What Accra is unique in is the particular cruelty with which the state has addressed it. The standard political response to flooding in the waterways has been the threat of demolition, the removal of structures built on drainage corridors to restore the flow of water. This is, as an engineering proposition, entirely correct. Structures built on waterways block drainage. They must be removed.
But here is what is almost never said in the same breath: the people living in those structures built them, or rented them, or inherited them, because Accra has no coherent policy of affordable housing for its working poor, no land titling system that distributes serviced plots to low-income residents, and no planning framework that has ever seriously grappled with the question of where a city of four million people, growing by hundreds of thousands every decade, is supposed to put the people who cannot afford the places where building is safe. The illegal structures in the waterways are not evidence of reckless indiscipline on the part of their inhabitants. They are evidence of a city that has grown without governance, that has invited millions of people to come and build its economy and then left them to find their own housing in the only spaces remaining.
To demolish without resettling is not flood management. It is the displacement of the poor in the language of engineering. And it has been the preferred instrument of Ghanaian flood policy, announced after every disaster, executed selectively and incompletely, and producing neither genuine flood relief nor genuine justice for the communities it displaces.
The floods do not discriminate. But the dying does. It happens, with terrible regularity, in the places where the people who built this city were left to house themselves as best they could.
June 3, 2015
There are moments in the life of a city that should function as absolute thresholds, events so catastrophic that they permanently alter the relationship between the state and its obligations to the people it governs. For Accra, June 3, 2015 should have been such a moment.
The sequence is almost unbearable to recount. The rains fell for hours. The streets flooded, as they always flood. People sought shelter where they could; in shops, under awnings, at a filling station near Kwame Nkrumah Circle where dozens crowded beneath the forecourt canopy to wait out the storm. The floodwater swept stored fuel from the station’s tanks. The fuel met a fire from a nearby building. The explosion that followed killed over two hundred and fifty people. The morgue at 37 Military Hospital reached capacity. Corpses were retrieved from floodwater. The President visited the site and called the death toll catastrophic and pledged that steps would be taken to ensure such a disaster would never occur again.
The steps were not taken. Or rather: some steps were taken, some money was allocated, some task forces were constituted, and the city flooded again the following year, and the year after that, and the year after that. The dead of June 3 were mourned for three days of national mourning and then, with the gentle efficiency of institutional forgetting, consigned to the past, referenced in subsequent flood seasons as evidence of severity but never as evidence of accountability.
Two hundred and fifty people died because the city flooded and there was nowhere safe to shelter and the infrastructure of their city had been neglected for so long that even the act of waiting out a rainstorm had become, for some of them, a death sentence. This is not tragedy in the classical sense, the working out of fate against which human will is powerless. This is something colder and more culpable: preventable death produced by governable conditions that were not governed. The distinction matters enormously. Tragedy invites mourning. Culpability demands accountability. We have had ten years of the former. We have had almost none of the latter.
The Grief We Owe the City
There is a way of writing about urban failure in African cities that is, in the end, a form of condescension that treats the broken infrastructure, the neglected drains, the flooded streets as evidence of an inherent incapacity, a civilisational backwardness that separates African cities from their European counterparts. This essay has no interest in that argument, because that argument is false, and because it conveniently locates the problem in culture rather than in the specific, traceable, accountable decisions of specific, traceable, accountable people and institutions.
Amsterdam did not become a city of functional water management because the Dutch are inherently disciplined. It became one because centuries of existential threat produced institutions with long memories and genuine consequences for failure. Vienna did not turn the Danube’s floodplain into a public park because Austrians are inherently visionary. It did so because engineers and politicians, over generations, made decisions that accumulated into infrastructure. The difference between those cities and Accra is not cultural destiny. It is institutional memory, the capacity of a city to remember what it has learned, to hold its decision-makers responsible for the gap between promise and delivery, and to insist, through the instruments of democratic governance, that what was promised last year be delivered before next year’s rains arrive.
Ghana has the instruments. It has a free press that has documented every flood, every promise, every forgotten task force. It has a parliament that can summon ministers and demand answers. It has courts that can enforce planning law. It has civil society organisations that have produced detailed, evidence-based analyses of what needs to be done and how much it would cost. What it lacks; what it has lacked, across administrations of both parties, through years of relative prosperity and years of economic difficulty is the political will to convert knowledge into action and promise into delivery.
That lack is not mysterious. It is the product of an electoral system in which the costs of flooding are borne by the poor and the benefits of inaction accrue to the developers, the permit issuers, the contractors who are paid to desilt drains that will be choked again before the next rainy season. The incentive structure of Ghanaian urban governance is, with respect to flooding, almost perfectly inverted: those with the power to fix the problem have the least to lose from its continuation, and those with the most to lose have the least power to compel action.
Ghana has the knowledge. It has the documentation. It has, in the World Bank’s assessment, even some of the money. What it has not found, across sixty-six years and five presidencies, is the will.
A Letter to the Next Rainy Season
The rains will come again. They will come in May, as they always come, because the climate of this coastline has not changed in the ways that matter to this argument. What will come with them depends entirely on what happens between now and then, in the assemblies, in the courts, in the offices of the planning authorities, on the banks of the Odaw.
There are things that could be done. The waterways could be restored, not by demolishing the homes of the poor without alternative, but by a serious, funded, politically sustained programme of resettlement that treats the inhabitants of flood-prone land as citizens owed a solution rather than squatters owed an eviction notice. The drainage network could be maintained, not through the annual theatrical exercise of pre-rainy-season desilting that is forgotten by July, but through institutionalised, budgeted, year-round maintenance with clear lines of accountability. The wetlands that remain could be protected; legally, physically, permanently, from the development pressure that has destroyed so much of the city’s natural flood buffer. The planning system could be reformed so that the permit fee is no longer the end of the state’s interest in a building’s relationship to its surroundings.
None of this is technically difficult. None of it requires expertise that Ghana does not possess or resources that Ghana cannot access. All of it requires something that has proven, across six decades of the same headlines, to be the scarcest resource in Ghanaian public life: the willingness to govern continuously rather than respond episodically, to treat the city as a living system that requires sustained attention rather than a periodic crisis that requires a press statement.
The 1960 photograph is still there, in the Daily Graphic archives, waiting to be reprinted alongside the photograph from last Sunday. The streets are the same. The water is the same. The buckets are the same. The only thing that has changed is the number of people who have died in the interval between the two images, and the number of promises that have been made and forgotten in the space between those deaths.
Accra is not a city that cannot be saved. It is a city that has not yet decided to save itself. That decision, unglamorous, expensive, politically thankless, requiring the sustained attention of institutions across multiple electoral cycles, is the only thing standing between the city and the water. It is also, it must be said, long overdue.
The rains are coming. They always come. The question, the only question that has ever mattered, the question that 1960 and 2015 and last Sunday are all asking in the same voice, is whether this time, finally, we will have done something worth remembering before they arrive.