In October 2007, dozens of Twin Cities residents gathered to discuss a new bridge between Minneapolis and St. Paul. Their feedback would help the designers of the bridge that would replace the one that had collapsed earlier that year; on August 1, the central 1,000-foot (305-meter) span of the eight-lane I-35W Bridge experienced a catastrophic failure and fell 108 feet into the Mississippi River. Thirteen people died and scores of others were injured. Somber but optimistic, residents were ready to support the rebuilding. Those at one table, however, were gray-faced and silent. They were Minnesota Department of Transportation engineers, later exonerated, who had been entrusted with the fallen bridge’s maintenance. Beyond the immediate tragic costs, other, less visible losses are suffered when a bridge fails.

Still, there was a silver lining: the I-35W collapse opened an overdue conversation about decaying infrastructure in the United States. Built in the heyday of the Eisenhower era, when the number of roads and bridges increased exponentially under the Federal-Aid Highway Act of 1956, many bridges are reaching the end of their structural lives; typically, a span has a life expectancy of fifty years. In 2013 the American Society of Civil Engineers reported that one in nine of the 607,380 bridges in the United States are structurally deficient, with an average age of forty-two years.

The bridge seemed to be among the things that last forever; it was unthinkable that it should break.

This was not the first investigation. In 1889 engineer Theodore Cooper (18391919) documented the loads exerted by trains. He called for the careful calculation of stresses, material testing, and inspection, but added that a bridge’s stability relied more on an engineer’s instincts than “merely upon a theory of stresses.” His words would come back to haunt him when he played a role in a notorious bridge collapse eighteen years later.

By the end of the nineteenth century, with the expansion of the railroad, bridge collapses were noted with increasing regularity and hysteria. Well-qualified scholars still debate how many bridges failedsome say one in four; others, one in four thousand. Many collapses were attributable to causes other than structural inadequacies, including train derailment, fire, and flooding.


Ithiel Town’s slogan, “Build it by the mile and cut it off by the yard,” describes the attitude and methodology of the many companies competing to build more and longer railroad bridges. Though this fostered the development of an economical and practical style of metal truss bridge, it was also rife with corruptionskimping on materials and on the review of the bridge’s technical design were not uncommon. Bridges designed by highly esteemed engineers also collapsed, however, indicating the pervasive role of human error in failure.

Robert Stephenson’s Dee Bridge (1846) across the Dee River in Englandthe longest metal truss built to that dateprovides the first instance of a metal bridge collapse that might have been avoided if information known at the time had been made available to engineers. Success with earlier truss bridges led to a progressive increase in their span length without considering the additional structural needs that the extra length imposed. The bridge girders buckled under these unconsidered stresses and fell on May 24, 1847, claiming five lives.

After Hurricane Katrina ripped through the Mississippi Gulf Coast region in 2005, all that was left of the U.S. Highway 90 Bridge between Biloxi and Ocean Springs, Mississippi, were its support columns.

The Twin Towers’ collapse on September 11, 2001, destroyed several bridges in lower Manhattan, including the North Bridge, a 95-foot (29-meter) span. Its twin, seen here, was farther south and survived.

On December 28, 1879, Thomas Bouch’s Tay Bridge over Scotland’s Firth of Tay went down in a gale with the loss of seventy-five souls. When completed in 1878, the rail bridge had been hailed as an “engineering triumph” and earned a knighthood for Bouch. Later found to be fatally complacent, Bouch had relied on outdated data that neglected the dynamic forces of violent gusts.

In the 1907 collapse of the Quebec Bridge over the St. Lawrence River in Canada, an ill-fated combination of technical error, corporate parsimony, miscommunication, and professional hubris caused the deaths of seventy-five workmen. The Quebec Bridge Company had hired Theodore Cooper to consult on the Phoenix Bridge Company’s cantilever design, which Cooper had selected as the “best and cheapest” of those submitted; as is often the case, Quebec wanted to spend as little as possible. Cooper increased the length of the bridge’s main span from 1,600 to 1,800 feet (488 to 549 meters) without recalculation of stresses. When asked about the soundness of his work, Cooper replied, “There is nobody competent to criticize us,” according to the November 28, 1907, edition of Engineering News. By August 1907, when the south arm of the bridge had been cantilevered out about 600 feet (183 meters), the discovery that the truss members along the bottom and top of the bridge, or chords, were bent was followed by three weeks of debate about when they were bent and not why. Cooper, who never visited the site when the superstructure was under construction, sent a telegram on August 27 ordering the work stopped until the bent chord issue was resolved. The telegram was ignored, either undelivered or unread, and work continued until August 29, when the bridge collapsed. Ironically and terribly, its replacement would also collapse in 1916, taking thirteen lives. The Quebec Bridge disaster would influence the development of suspension spans for long distances. Yet the suspension bridge over Tacoma Narrows collapsed in 1940 because the designer failed to consider aerodynamic forces (see here). In 1970 two bridges of a new type, the steel box girder, went down, one in Wales and the other in Australia, due to a combination of design and construction issues. In 1977 Paul Sibly and Alastair Walker noted a thirty-year cycle of bridge failures and their shared characteristics. Each of the bridges involved was a new type of bridge (trussed girder, truss, cantilever, suspension, and box girder) that was being developed in response to the collapse of an earlier type of bridge. Engineer and author Henry Petroski extends Sibly and Walker’s findings by thirty years and theorizes that the next collapse might involve a cable-stayed design, the newest type of bridge. While human error will always be a variable, improvements in other areas more reliable materials, better construction methods, expanded testing, and the recognition that failures, having the most to teach about successful design, should be widely documented and sharedhave removed some uncertainty. Bridges remain vulnerable. Extreme weather, wrought by climate change, has exacted a catastrophic toll, as has global unrest. Much remains to be done. Every four years, America’s civil engineers assess the nation’s infrastructure and issue an old-fashioned report card. In 2017 America’s bridges earned a C+. “Railway mania” during the first half of the nineteenth century saw the number of bridges in Britain double from 30,000 to 60,000. The Britannia Bridge’s gigantic iron girders were 30 feet (9.1 meters) high and nearly 15 feet (5 meters) wide.

Maybe You Like Them Too

Leave a Reply

72 − = 67