American Scientist

To the Editors:

Henry Petroski’s Engineering column “Tappan Zee Bridge” in the May–June issue states that the Tappan Zee Bridge has a lifetime of only 50 years and that it is going to be replaced. But as Petroski mentions in the article, there are other bridges such as the Brooklyn Bridge, the Manhattan Bridge, and the Williamsburg Bridge in New York City that are more than 100 years old. Nobody talks about these structures having reached the end of their useful life. In Europe there are bridges built 2,000 years ago by the Romans that still function well, even under modern traffic. Why do modern U. S. bridges have such short life spans?

Philipp Kornreich
Electrical Engineering &
Computer Science Department
Syracuse University
Syracuse, NY

Dr. Petroski responds:

The difference in lifetime among bridges across different eras and designs may be explained at least in part by four interrelated factors: materials, money, maintenance, and mathematics. Stone is generally more durable than steel, which needs to be protected against corrosion. Stone construction, however, is much more time-consuming and hence expensive, and so is not considered practical for modern bridge projects. The long 14-year construction time for the Brooklyn Bridge was due in part to the building of its masonry towers. Still, that bridge might have collapsed long ago had engineers not discovered that inferior steel wire, provided by an unscrupulous supplier, had been incorporated into its main suspension cables. It was only because a greater quantity of high-quality wire was added at the time of construction than was originally planned, and because the cables were properly protected against corrosion, that they have lasted as long as they have. The lifetime of the Brooklyn Bridge has been extended further by restricting its use to automobiles.

After over a half-century of use, parts of the Manhattan and Williamsburg bridges became badly corroded, a condition created by some design decisions that proved to be unwise, and hastened by maintenance being deferred during the 1970s because of poor fiscal conditions in New York City at the time. It was only the extensive rehabilitation of the bridges performed after the fiscal crisis was over that obviated the need for replacement bridges.

Ancient bridges were designed and built using rules learned by trial and error, which incorporated a great deal of conservatism. Early 20th-century suspension bridges such as the Manhattan were designed with the aid of mathematical theories that gave engineers increased confidence that they were calculating stress distributions more accurately than previously possible. This advance, coupled later with computer modeling, resulted in lighter structures with lower factors of safety, not only for suspension bridges but also for truss bridges such as the Tappan Zee. The lower cost of lighter-weight bridges is understandably attractive to government agencies paying for them, but these structures necessarily come with a limited lifetime, determined at the outset on the basis of how many cars and trucks, and of what weight, are projected to be using a bridge year after year. The structural deterioration of lightweight bridges was hastened when political rather than technical considerations prevailed in allowing heavier and heavier truck traffic on the roads that the bridges carried.