Skip to main content

Why Cities Don’t Die

Winter 1990 | Volume 5 |  Issue 3

More cities were destroyed during World War II than in any other conflict in history. Yet the cities didn’t die. The modern technological city, held together by electricity, telephones, water lines, and highway and rail networks, was still a recent phenomenon. No one knew how strong or vulnerable a machine it was. The consensus was that it was too reliant on an easily shattered infrastructure to survive a well-planned military attack. The consensus turned out to be wrong.

As hundreds of thousands of conventional bombs fell on Europe’s metropolitan centers, the devastation appeared at first to fulfill prewar expectations about urban vulnerability and the superiority of air power. It had been expected that ports and shipbuilding centers would be hard hit, and indeed, some of the worst-hit cities—Plymouth, London, Hull, and Glasgow in Great Britain, Brest and Lorient in France, and Hamburg in Germany—were critical either to the Allied effort to maintain Atlantic sea links or to the German U-boat campaign to sever them. But the war at sea was never decided by the struggle of port cities. And the cities themselves suffered remarkably little social and economic disruption considering the physical ruination they endured. What did happen provides valuable lessons about how twentieth-century cities function. Clearly the nature of how metropolises work and how they can be made not to work had been greatly misunderstood.

The surprising lessons of precision bombing in World War II and Vietnam

Two types of bombing strategy were developed during the years before World War II: area bombing and precision bombing. Area bombing, favored by the British, assumed that the indiscriminate destruction of any part of a city or town would induce panic and perhaps riots and that civilians subjected to such bombing would pressure their governments to end the war.

Precision bombing, favored by the Americans, sought the destruction of electricity-generating plants, factories producing critical supplies, and key linkages in transportation systems. It was preferred not only because it promised a way of rationing aircraft and crews to achieve maximum damage of an enemy’s war economy but also, and perhaps more subtly, because it claimed to limit civilian casualties. Yet its targets were most often found in large industrial cities. In practice, if not in theory, cities would be destroyed. They would be attacked as machines of production, specifically of war production.

Edgar S. Gorrell, an American colonel in World War I, developed many of the features of precision bombing theory. In a memorandum dated November 28, 1917, to Gen. Benjamin Foulois, chief of the Army Air Service, Gorrell recommended the destruction of factories producing critical parts for war equipment. He assumed that the activities of a given factory could not be continued once the factory itself had been destroyed, that factory production in war was likely to proceed at full capacity, and that the various parts of an urban and national economy were functionally interdependent, so that crippling one would hurt others.

In the succeeding years British and American strategists anticipated the need to identify those potential targets whose destruction could best disrupt war production. To that end the British established the secret Industrial Intelligence Centre around 1930; this evolved into the Ministry of Economic Warfare. American neutrality impeded planning on the other side of the Atlantic. Maj. Donald Wilson, an instructor at the Air Corps Tactical School, proposed the establishment of an “air intelligence section” in 1934, but the step was delayed until 1940.

In the meantime American strategists were thrown back on the study of the American economy as a model. On April 5, 1939, Maj. Muir S. Fairchild of the Air Corps Tactical School gave a lecture on the weak points in a national economic structure. Taking the northeastern United States as an example, he observed that no new electricity-generating plants had been built in recent years. Because the electricity grid was poorly connected by transmission lines to other regions, the destruction of electric power plants in the Northeast, he surmised, would have national repercussions and might force the nation out of a war.

A day later Fairchild gave a lecture specifically on the vulnerability of the New York industrial area. He pointed out that although the electric-power systems of other nations might not be as easily destroyed as America’s, all industrial nations had great cities. He observed that only a few plants—three near Newark, New Jersey, and a dozen in New York—produced most of the region’s power. Their destruction would, he estimated, take between nine and eighteen months to remedy.

Without power New York City’s offices and rapid-transit systems, its cranes at waterside, cold-storage warehouses for food, pumps for water supplies, and 270,000 electric motors in manufacturing plants would be unable to function. Although Fairchild and other theorists of precision bombing considered their mark to be physical targets, such as factories, plants, and transportation facilities, they also took seriously the impact of such destruction on life in great cities and the consequences of urban collapse on an industrial economy.

When blackouts and electricity shortages struck American cities in the 1930s, air strategists had a ready-made situation with which to analyze the possible effects of air raids against electricity supply systems. A blast underground in the vicinity of Fifth Avenue and Thirtyninth Street in New York City caused a power failure on October 24, 1935. There was no panic as office buildings and Lord & Taylor, the department store, were evacuated. Another power failure, on January 15, 1936, was much more severe. The second-largest power plant in the world, at Locust Avenue and 132d Street in the Bronx, had to be shut down at 4:16 P.M. following a short circuit in the generator cables leading from the turbines. Power remained off in the Bronx and in Manhattan north of Fifty-ninth Street for most of that evening. Sixty thousand people were trapped in subway trains. Crowds surged onto the platforms of two other underground lines that continued to operate with their own power supply, creating problems in crowd control. Police from unaffected areas were called to blacked-out areas to prevent looting. Tens of thousands of people found their way out of darkened stores, office buildings, and theaters without ever panicking.

The blackout called attention to the extent to which New York was dependent on its own resources for power, but the social disruption it caused did not seem severe. Air-war strategists could only argue that an attack that did not merely interrupt power but actually destroyed several power plants would have far more serious results.

Social critics at the time were keenly aware of the growing importance of machinery-driven utilities in the life of great cities. Infrastructure systems for mass transport, electric power, telephones, gas, water, and sewage had been built up mainly between 1880 and 1920; their importance was greatest in the largest metropolitan centers. It was widely agreed that the benefits gained in comfort, increased investment, faster communications, and greater mobility incurred hidden costs. Some social commentators worried that cities might be vulnerable to infrastructure collapse even in peacetime. Using the same terms as air strategists, civilians wrote about vital points, bottlenecks, and breakdowns, with frequent comparisons between urban infrastructure and the human central nervous system.

E. B. White, amazed at what New Yorkers endured with patience, observed in Here Is New York : “By rights New York should have destroyed itself long ago, from panic or fire or rioting or failure of some vital supply line in its circulatory system or from some deep labyrinthine short circuit.” By 1949, when White’s book appeared, such sentiments were an established convention of urban commentary. Stuart Chase wrote in Harper’s Magazine in 1929 of two kinds of urban collapse: a social collapse born of the frustration and exhaustion of everyday life in the city and a technological collapse that would follow the disruption of key systems of supply and control. The fact that infrastructure systems had not failed, argued Chase, is no proof that they are not susceptible to failure. And he emphasized how air raids could exploit the same kind of collapse that might come in peace. Others echoed this theme.

Considering the ruination they endured, cities suffered only modest social and economic disruption.

Niles Carpenter, a social scientist, wrote in The Sociology of City Life in 1931 that infrastructure systems spelled “decreased security, for the longer the lines of communication the greater is their liability to interruption, and the more intricate they are the more easily they can be thrown into disorganization.” He envisioned urban life crumbling after air raids, with people unable to repair or operate infrastructure systems and unable to cope with daily life without them.

The British writer L. E. O. Charlton foresaw similar chaos in his 1938 book The Air Defence of Britain . “If it had been done deliberately,” he wrote, “we could not as a nation have produced a social pattern, and a set economy, more favorable for aggression from the air. Our millions are bottle-fed, and all their needs cared for, by a system of distribution and supply so intricate, and so haphazardly evolved, that once dislocated beyond the power of immediate repair they would be as helpless as newborn babes to fend for themselves.”


Such criticisms shared certain assumptions about the nature of modern urban life in a technological age. By implication, the role of great modern cities in the life of a nation was taken to be so significant that cities of lesser size and importance would be unable to take over their functions. In that sense these forecasts of the ravages of war paid homage to the power of cities. But they also indicted modern urbanization.

By projecting an image of urban dwellers as lacking self-confidence and resourcefulness, by dismissing the ability of people to form vibrant communities in cities, by emphasizing the social and economic differences that might divide urban people into factions, such forecasts portrayed the effect of technology on urban life in entirely negative terms. No wonder these prophecies made problems of civil defense appear nearly insurmountable.

The test came, of course, when war came. The British studied the effects of German air raids on their cities to better understand the theory of air war. German raids in mid-March and early May 1941 on Clydebank, a shipbuilding suburb of Glasgow, were typical. Approximately sixty tons of explosive fell per square mile, a high ratio, but fewer people were killed or injured per ton of high explosive than pre-war forecasts had predicted, even though damage to housing was considerably higher. All the essential utilities were knocked out, and transport links were severed. The complete loss of 10 percent of the city’s housing stock and damage to as much as half the rest seriously affected a city already suffering from residential overcrowding. Although most residents left Clydebank to seek shelter during the period of restoration, workers continued to travel back every day to work. Most factories lost between three and four weeks of production. About half the loss of production was due to damage to ships at shipyards, and about a third to damage to engineering works (principally firms that made templates for ships). General disorganization accounted for the rest of the loss.

Clearly the raids had not caused panic. The worst social problems that arose involved incompetent municipal personnel and poorly led relief teams, not the behavior of the masses. Furthermore, people learned to cope with daily life in a city without vital services through the simple expedient of relocating to unaffected areas. From German air raids on British cities in 1940 and 1941, the British learned that cities would have to be hit repeatedly if production levels were to remain depressed. And they also learned how such attacks could actually strengthen the nation by mobilizing its resolve against the enemy.

Pre-war concern had focused on very large cities, on the assumption that life in such cities could be disrupted more easily and effectively than in smaller communities. But experts at the British Ministry of Home Security, in a report entitled “The Total Effects of Air Raids” in May 1942, concluded that the exact opposite was true: Large cities can adjust better to the punishment of air raids because they have better transportation, larger housing stocks, and larger and more mobile labor markets. The report did not, however, question the fundamental assumption that urban economies and societies are perilously dependent on a fragile technological infrastructure of supplies and services.

The Allies’ first major air attacks against cities on the Continent were intended to weaken the German U-boat campaign. As Allied shipping losses mounted in 1942 and 1943, the decision was made to attack German submarine bases and production yards. At first Allied bombers could reach only the German bases in western French ports. From the fall of 1942 until the spring of 1943, Lorient, Brest, and St.-Nazaire, among other sites, were hit repeatedly by British and American bombers. Between . January 14 and February 16, 1943, forty-four hundred out of five thousand buildings in Lorient were destroyed; not until the last months of the war would air raids against German and Japanese cities achieve comparable damage.

If cities were vitally dependent on a high level of utilities, and if small cities were more vulnerable than large ones, then the destruction of St.-Nazaire and Lorient should have seriously hampered German naval operations. But the impact on the U-boat campaign was negligible. First, the Germans had constructed self-contained U-boat bases of reinforced concrete, which the precision air raids did not destroy, and had designed the bases to be much less dependent on the adjacent city than shipyards usually are. And second, because most people fled from these cities before and during air raids, civilian casualties were very low; even though all vital urban services were lost during the raids, the residents, many of them from the surrounding regions, could adapt and rebuild.

Would a massive Allied attack against a great German city, an attack on a scale far exceeding any German air assault against a British city, be a better test of pre-war bombing theory? The British created a fire storm in Hamburg, one of Germany’s two principal centers for U-boat production, in late July 1943. After two days and nights 56 percent of the city’s housing was destroyed; nine hundred thousand people were homeless. The number of German civilian casualties in this single raid was nearly as great as the number of British casualties from all the German air raids of the war.

Yet within days mail service had been resumed, transport links had been restored, and the electricity supply actually exceeded demand. The British air raids against Hamburg resembled area bombing insofar as they sought the destruction of the city’s whole center, but the attack was akin to precision bombing insofar as it was grounded in assumptions about the interdependence of urban social, economic, and technological life. The Allies tried to reduce German production by both direct and indirect means—directly by destroying systems on which industrial activity depended and indirectly by destroying workers’ housing. The interrelationships between technology and urban life proved to be far more flexible than the Allies had thought.

Only direct hits on merchant and naval ships, not the destruction of port cities, were truly detrimental to an enemy’s sea power. Yet air attacks on port cities in France, Britain, and Germany destroyed few ships. British intelligence reported that only two completed* U-boats were sunk, and six hulls under construction damaged, in the Hamburg raids. Work at shipyards was interrupted for a month. A November 1943 British intelligence report revealed that raids against Kiel and other smaller U-boat assembly sites as well as Hamburg had reduced total U-boat production by seventeen ships out of a possible two hundred.

The devastation of Hamburg had not destroyed the city’s industrial economy. Si- multaneous air raids against several cities might have seriously crippled the German war economy, but by the time the Allies were able to launch air raids on such a scale, in late 1944 or early 1945, the war was already being won at sea and on the ground. The final air raids of the war were intended purely to weaken German morale, an objective the theorists of precision bombing had previously scorned.

After the war most air strategists rationalized the limited effect air raids had had on the German economy by citing imperfect target selection and technical problems with Allied air forces. The famed United States Strategic Bombing Survey of 1945 did this while generally praising the effectiveness of the air war. But there were scholars who questioned the assumptions about cities on which strategic-bombing theory had been based. Clearly, urban residents were not as disoriented or emotionally upset during air raids as had been expected. They coped because they were resourceful and adaptive in stressful situations.

Although in peacetime city people relied on technologically sophisticated networks, this dependence had not weakened the relation between individual and community. In addition, infrastructure facilities had consistently been repaired or replaced more easily than expected. Pre-war strategists had overlooked the extent to which substitutes for elements of infrastructure could be found or an existing resource base stretched.

The strategists’ fundamental error all along had been to see urban infrastructures principally as technological systems. That was how the public at large saw cities, too, but it gave technology too much credit, and responsibility, for making cities work—and gave people too little. Immense levels of physical destruction simply did not lead to proportional or greater levels of social and economic disorganization.

For all their technological complexity, cities remain, above all, masses of human energy and resource.

There has been one sustained example since 1945 of precision bombing of urban targets: the American air war on Hanoi in the late 1960s and early 1970s. That experience proved how difficult forecasting the impact of urban air raids still remained.

Despite the low level of industrialization in North Vietnam as a whole, Hanoi and its surroundings supported modern transport and manufacturing, the technological base for the war against the South. In the decade between Vietnamese independence and the commencement of American air raids in 1965, Hanoi grew faster than any other city in North Vietnam, its industrial sector gaining at the expense of small merchants and its outlying area redeveloped to provide food to replace imports. The American air raids between 1965 and 1968 (Operation Rolling Thunder) were a failure, not only because antiaircraft defenses were unexpectedly effective but also because political considerations about American opinion, more than military objectives, governed the selection of targets.


During the Linebacker offensives of 1972 and 1973, the Americans targeted industrial facilities for their military value. American precision bombing in 1972 was technically far more sophisticated than in 1965, let alone in 1943; the air raids were a success operationally, destroying antiaircraft defenses, power plants, and transport facilities and incurring barely 2,000 civilian deaths, a very low rate considering the 15,287 tons of explosives that fell. (The largest number of civilians to die in a city under siege since 1945 has been 30,000 in Beirut, from artillery shelling.)

The low number of civilian casualties did not result from improved American bombing techniques alone; some 750,000 people, about 75 percent of the population, had been evacuated from Hanoi. The evacuation not only saved lives but enabled the North Vietnamese to overcome the impact of the bombing.

Assuming that the combination of blockade and bombing would put enormous pressure on the North Vietnamese economy, the United States had underestimated the ease with which the North Vietnamese could find alternative routes for imported supplies once their ports had been mined. Supplemental rail routes that were built during the fall of 1972 enabled sixteen thousand tons of goods per day, far more than Hanoi needed, to be carried to that city. Because evacuation had reduced Hanoi’s population so drastically, the North Vietnamese, no longer needing to devote a large percentage of their transport to civilian needs, were actually able to move more war matériel through Hanoi in the fall of 1972 than in the late 1960s. What importance these militarily ineffectual air raids had in hastening the final diplomatic resolution of the war is still an open question.

Since World War II the study of tornadoes, fires, floods, hurricanes, earthquakes, typhoons, pollution disasters, epidemics, plane crashes, blackouts, and other emergencies has led to a far better understanding of how cities recover from disaster. If there has been a central lesson, it is that the processes at work in cities during and after disasters are the same as those that account for concentrated social and economic development in less stressful times. Just last year, San Francisco suffered less and recovered faster from a major earthquake than anyone at first expected. The ability of cities to recover from disasters, once thought to be very limited, now appears to be broadly based on a variety of mutually reinforcing conditions and factors.

For all their technological infrastructure and complexity, cities remain, above all, great concentrations of human energy and resourcefulness. Indeed, Eric Jones, an economic historian, has argued that the rise of the West since the Middle Ages can be explained in part by the ease with which Western societies have recovered from disaster, as compared with African and Asian societies. Yet the myth of terrible urban vulnerability endures.

It is worth noting that contemporary prognostication about the effects of limited nuclear war resemble earlier predictions for air raids: both assume that the destruction of cities would come early in a conflict, that a relatively small number of warheads or bombs would spread destruction over a large area, and that such a level of destruction would cripple an enemy’s economic and social systems. Obviously, nuclear war is a new kind of war, of a new order of magnitude. But given that fact, is it possible that we have simply updated our old assumptions about the importance of technological infrastructures in cities to keep pace with the nuclear age?

We hope you enjoyed this essay.

Please support America's only magazine of the history of engineering and innovation, and the volunteers that sustain it with a donation to Invention & Technology.


Stay informed - subscribe to our newsletter.
The subscriber's email address.