Urban mortality

Why did cities kill more people than they birthed for most of recorded history?

London grew for centuries. So did Rome, Paris, Edo. It is natural to assume that a growing city is one where more people are born than die — that is, after all, what growth means for a country, or a species.

For most of recorded urban history it was not true of cities. John Graunt, working through London's Bills of Mortality in the 1660s, noticed that burials in the city persistently outran christenings. The city was growing while consuming its own population. So what was arriving to make up the difference — and why should putting people close together kill them at all?

Take the second question first, because the answer to the first follows from it.

What actually changes when you concentrate a hundred thousand people into a square mile? Consider what each person produces daily besides work: waste, and the by-products of cooking and heating. In a village these are absorbed by land — spread thin, they break down. Density does not create the waste; it removes the space that was quietly disposing of it. Volume rises with population while the absorbing area does not, so the load per acre climbs faster than the headcount.

Now ask where a city's water comes from. Typically from wells, or from the same river that receives the waste, downstream of somebody. Here is the mechanism that does most of the killing: waterborne disease. Cholera, typhoid, and the dysenteries spread by the route from human waste back to drinking water, and a dense settlement without sanitation closes that loop tightly. Contamination is not incidental to density — it is what density does to a water supply.

Density does something else too. An infection that dies out in a village because it runs out of susceptible people can circulate indefinitely in a city, which is large enough to keep supplying new hosts — and above all a steady supply of children with no immunity. So the killer diseases of childhood become endemic rather than episodic. Add a constant stream of arrivals from the countryside, each carrying no exposure to the local pathogens, and the city is continuously restocked with the vulnerable.

Put the pieces together and you get what demographers call the urban graveyard effect: death rates in pre-modern cities that ran above their birth rates, in some periods substantially, with infant and child mortality doing most of the work. Historical estimates for early modern London have burials exceeding baptisms across long stretches of the seventeenth and eighteenth centuries.

Which forces the first question. If the city was in natural decrease, how did it grow? Only one answer is available: in-migration. Cities were sustained by a continuous inflow from the countryside — overwhelmingly young adults, drawn by work and wages, arriving in numbers large enough to cover the deficit and add to it. A pre-modern city was less a population than a process, and if the countryside had stopped sending people, it would have shrunk.

Now the part that is genuinely counterintuitive. When did the cities stop being graveyards? In western Europe, across the nineteenth century — and the turn came from civil engineering, not from medicine. Sewers, drainage, filtration of water supplies, and the separation of drinking water from waste were doing the work well before germ theory was settled, and long before there was any effective treatment for cholera or typhoid. Snow's 1854 investigation of the Broad Street pump identified a route of transmission without identifying an organism, and the great London sewer system built under Bazalgette after the Great Stink of 1858 was justified partly by a theory of bad air that was wrong about the cause and right about the fix. Antibiotics arrive in the 1940s, decades after the urban penalty had already largely closed.

Three refinements hold this together.

First, density is not the villain by itself; density without sanitation is. This matters because the same crowding that spread cholera also made the sewers affordable — piped water and drainage have enormous fixed costs and cheap marginal ones, so they only make economic sense where people are packed together. Density created the problem and then paid for the solution.

Second, the migration flow and the mortality were connected, not merely simultaneous. The inflow of non-immune young adults was part of what kept endemic disease burning, so the very process that sustained the city's numbers also fed the thing that consumed them.

Third, an honest caveat about the cause of the decline. The claim that sanitation rather than medicine turned it is well supported but not uncontested. Thomas McKeown argued in the 1970s that rising living standards and nutrition mattered more than either public health or medicine; Simon Szreter's influential response in 1988 marshalled the case that municipal public-health investment did the heavy lifting, and that view is now the more widely held. The weighting between nutrition, sanitation, and falling exposure is still argued — but on one point there is no real dispute: curative medicine was not what did it, because it did not yet exist.

• Why rapidly growing cities today reproduce or avoid the same pattern, depending on infrastructure rather than income alone.
• How the demographic transition interacted with urbanisation, and which came first where.

Every term the collection defines is gathered in the glossary.