In 1840, the East India Company’s civil surgeon at Dacca, James Taylor, watched the Buriganga swell with the monsoon and reached for a comparison that has cast a shadow over the city ever since. The river in flood, he wrote in Sketch of the Topography and Statistics of Dacca, gave the city “the appearance, like that of Venice in the west, of a city rising from the surface of the water.” This was not idle flattery. The Mughal capital that Subadar Islam Khan founded in 1610 was built as an amphibious place. He ordered the digging of the Dholai Khal, which served as a moat, drain, and trade highway, stitching the Buriganga to the Balu and the Lakhya. Water was not the setting for Dhaka. It was its infrastructure.

However, read Taylor a few pages on, and a second, more consequential colonial instinct surfaces. Describing the decaying town of Panam, he dwells on its “stagnant creeks and ponds,” its rank vegetation, and the “sickly emaciated appearance of its inhabitants” — proof, to his eye, that it was among the district’s most unhealthy places. In the very volume that likened Dhaka to Venice lies the idea that would eventually unmake the city: the nineteenth-century conviction that standing water bred disease and that improving a city meant, above all, draining it of water. This is the part usually left out of the mourning for Dhaka’s lost khals (canals). Their disappearance is a tale of negligence, greed, and weak enforcement. However, beneath the encroachment and lack of maintenance lies something older and more deliberate: an imported idea that water is a nuisance rather than an asset. That idea has outlasted the empire that brought it, and it has proved far more destructive than any single land grabber.

The exact number of canals Dhaka has lost is uncertain, and that uncertainty is revealing. WASA's records list 54 canals until the mid-1980s; a widely cited figure suggests that the newly independent country inherited 57, of which perhaps 26 still function; other estimates range from 43 to well over 100. No agency keeps a definitive historical register, which is itself a measure of how casually the waterways were allowed to disappear. The reframing was gradual.

As the railways displaced river-borne commerce from the 1880s, the historian Iftekhar Iqbal has shown, the colonial city turned its back on the Buriganga and reoriented itself “towards land”. The waterfront, stripped of trade, lost its status. Water that no longer carried goods was recast as a sanitary liability—a breeding ground to be filled. Drainage schemes were drawn up not to preserve the khals, but to abolish them.

The clearest proof of this mindset is a single, remarkable document. In 1917, the Scottish biologist and town planner Patrick Geddes, among the most respected planners of his generation, was asked to advise on the future of Dacca. In his report, he records, with unmistakable dismay, an earlier government engineering scheme “for the improvement and sanitation of Dacca” that proposed, in plain terms, filling up the city’s khals—twelve and a half miles of them. The costs were tabulated like a merchant’s ledger: roughly seven and a half lakh rupees to fill the canals, against some fifty acres of building land to be gained. Water in one column and land in the other. It is difficult to imagine a purer statement of colonial calculus.

Geddes was horrified, and his answer was a path that Dhaka never took. He invited any European or travelling Indian who had seen the canals of Holland and Belgium to look afresh at “this splendid old Dolai Khal” and to weigh, against the cost of filling it, “the vast loss and waste” of leaving it to rot. He urged that the khals be dredged to revive their boat traffic, that their silt be spread to enrich the city’s gardens, and that their waters be stocked for pisciculture. Then he did what the sanitary engineers never contemplated: he turned their own argument inside out, suggesting that Dhaka’s relative freedom from malaria was itself bound up with its “ample water-supply.” In his reading, the khals were not the source of the disease. They were part of the defence.

Beneath the encroachment and lack of maintenance lies something older and more deliberate: an imported idea that water is a nuisance rather than an asset. That idea has outlasted the empire that brought it, and it has proved far more destructive than any single land grabber.

Geddes lost. Over the following decades, the Dholai Khal was buried until only a fragment remained open to the sky. That defeat changed the course of Dhaka's urban history. What followed was not inevitable. In 1917, Patrick Geddes, one of the world's leading planners, had already laid out a clear alternative: work with the water, not against it. It was a choice between two philosophies, and at every turn, the colonial one prevailed. Independence did not break the habit; it bureaucratised it instead.

When the British firm Minoprio, Spencely and Macfarlane drew up the first modern master plan for the Dacca Improvement Trust—the agency that became RAJUK in 1987—in the late 1950s, they defined the city’s central problem in terms that Taylor’s engineer would have applauded: a shortage of land above flood level, to be relieved by reclaiming more of it and by strict “economy in land use.” The scarce, valuable commodity was dry ground; water was merely low-lying land waiting to be filled. Geddes's question, what the water was for, had disappeared from the planners' vocabulary altogether.

Independence in 1947 did not change the thinking. It gave the state the machinery to put it into practice. Whereas a colonial engineer could only propose filling a dozen miles of khals, the modern state could actually do it, and far more, with hydraulic dredgers, imported sand, and fleets of earthmovers.

Comparing the cadastral survey maps of 1880–1940 with recent satellite imagery, a 2022 study by the River and Delta Research Centre counted roughly 175 canals, channels, and lakes in historic Dhaka. Of these, about 80 have vanished entirely and another 15 have shrunk drastically, while the city’s watercourses have contracted from around 326 km to 206 km. The figures for wetlands are worse. A 2012 study by the Centre for Environmental and Geographic Information Services found that Dhaka had lost roughly three-quarters of its perennial wetlands in four decades—some 15,000 of the 20,282 hectares that survived in 1967 are now built over. The rivers that once framed the city are being strangled in parallel, with recent surveys identifying hundreds of waste outfalls pouring into the Buriganga alone.

None of these losses is anonymous. In 1989, with funding from the World Bank, the Asian Development Bank, and the Japan International Cooperation Agency, a series of natural canals in the city’s heart—among them stretches of the Dholai Khal, Segunbagicha, and Panthapath—were encased in box culverts and paved over as roads, a decision urban experts have regretted ever since. The Dhaka Water Supply and Sewerage Authority (WASA) has been responsible for these channels for decades as they gradually silted up and shrank. The Bangladesh Water Development Board’s flood-control embankments, built after the catastrophic floods of 1988 and 1998, sealed the western part of the city off from its rivers and, by cutting the natural lines of drainage, turned ‘protected’ neighbourhoods into bowls that now flood from within after an hour of rain. RAJUK, which describes itself as the custodian of Dhaka's development, presided over the master plans that reclassified floodplains as buildable land, even as its own satellite townships rose on filled wetlands.

The consequences of this colonial mindset are written into Dhaka's ordinary streets. Before Panthapath became a major road, it was a canal that carried boats from Hatirjheel through Dhanmondi to the Buriganga, older residents recall. It was later sealed beneath a box culvert, leaving the water it once carried with nowhere to go. Multiply that small erasure across the city, and you have the modern condition: a metropolis that, planners note again and again, goes under after a mere 30 to 40 millimetres of rain, its drains overwhelmed by the still water that follows.

The Natural Water Bodies Protection Act of 2000 was meant to prevent the filling of canals, floodplains, and retention ponds. The Bangladesh Institute of Planners has documented that the metropolitan area lost more than half of its designated flood-flow zones in a single decade, much of it to housing, including schemes by government agencies themselves. Critics of the 2016–2035 Detailed Area Plan argue that, by permitting “conditional” construction in general flood-flow zones, it effectively converts encroachment into policy. The colonial ledger—water traded for land—has, in places, become the official plan.

Governments have continued to promise a fix. On the final day of 2020, WASA handed 26 canals and ten kilometres of box culverts to the two city corporations, promising that the capital's waterlogging would finally be solved. Five years and more than Tk 730 crore later, the same roads still disappear under water after heavy rain. Public money continues to be spent on the symptoms—clogged, narrowed and buried channels—while the underlying problem remains untouched: a planning philosophy, carried across a century and three states, that treats water as an obstacle to development rather than part of it.

Dhaka has already seen alternative approaches. The restoration of the Hatirjheel–Begunbari wetland into a functioning retention lake, despite the fact that later construction has compromised it, proved that a filled and fouled waterbody can be brought back to hold the city’s stormwater and cool its air. This is, in essence, precisely what Geddes proposed in 1917: cleanse and deepen rather than fill; treat the khals as sources of fertility, transport, and protection rather than as vacant real estate. A city serious about that idea would make its flood-flow zones inviolable public water rather than “conditionally” developable; it would trace every surviving canal from the old survey maps and defend its banks; and it would judge RAJUK, WASA, and the city corporations not by the projects they announce but by the retention capacity they actually restore.

The draining of Dhaka is often presented as a story of neglect, as though the city simply sleepwalked into being drowned. But it began with a deliberate idea: a colonial planning philosophy that taught a delta city to see its own water as a liability rather than an asset. That idea outlived the empire, was embraced by successive governments, and was ultimately carried to completion by modern machines. Recognising that history is not merely an academic exercise. It is the first step towards reversing it, because a mistake can only be regretted, but a decision can be undone.

Dhrubo Alam is Deputy Transport Planner at the Dhaka Transport Coordination Authority (DTCA), and Farah Mahboob is Deputy Manager at the Social Innovation Lab, BRAC. They can be reached at [email protected] and [email protected], respectively.

Send your articles for Slow Reads to [email protected]. Check out our submission guidelines for details.