Engineers of Dreams: Great Bridge Builders and the Spanning of America

by Henry Petroski

  The Canadian Commission, therefore, which spent the week from Oct. 14 to Oct. 19 in New York City, visited Mr. Cooper at his residence, conferred with him concerning the matter on several successive days, as his strength permitted, and finally formulated a series of questions covering the matters pertinent to the inquiry. Replies to these questions were dictated by Mr. Cooper at his leisure, and this testimony, after its reduction to writing, was fully reviewed and revised by the Commission and Mr. Cooper in further conference, until it represented as completely as the Commission could determine, the full testimony which Mr. Cooper was able to give.

  By the end of the month, the questions of the commission and Cooper’s responses were reprinted in full in Engineering News. The transcript showed that the commission was looking into the entire history of the bridge project, the involvement and interrelationship of Cooper, the Phoenix Bridge Company, and the Quebec Bridge Company, and the nature of the various organizations and their respective involvement in the design and oversight of the work. Among the leading questions posed to Cooper was whether the plans were approved to his satisfaction or whether he would have given them further study had he been able to do so. His reply was that of a man seeking sympathy:

  I should have been glad to have had the physical strength and the time allowed me to have given further study to many parts of this structure, but in my physical condition I have been compelled, and must accept the responsibility for the same, to rely to some extent upon others. I had and have implicit confidence in the honesty and ability of Mr. Szlapka, the designing engineer of the Phoenix Bridge Co., and when I was unable to give matters the careful study that it was my duty to give them, I accepted the work to some extent upon my faith in Mr. Szlapka’s ability and probity.

  Engineering News, in a prefatory editorial to its printing of Cooper’s testimony, cautioned engineers to “maintain a judicial attitude in considering the serious question how responsibility should be apportioned for faults in design, construction and erection.” Indeed, the trade publication reminded its readers, “every engineer will recognize the fairness of suspending judgment as he reads Mr. Cooper’s statement until the statements of the Phoenix Co. engineers are presented.”

  After taking Cooper’s testimony in New York, the commission traveled to Phoenixville and Philadelphia to collect further information and take testimony from the Phoenix Bridge Company and its officers and engineers, including Szlapka. Here the commission heard “vigorous language directed against Theodore Cooper,” which included charges that he had played down concerns over the incomplete bridge’s structural behavior when it was questioned by the company’s engineers, that he allowed more stress on the materials in this bridge than in any previous structure, that he ordered the main span increased to eighteen hundred feet, and that he refused to visit the Phoenixville plant where the first parts made for the bridge were being assembled.

  The commissioners delivered their report, to which was appended a “Report on Design of Quebec Bridge” by C. C. Schneider, within six months of the accident. Among the main findings of the inquiry were that the collapse was initiated by the inability of the lower chords near the main pier to withstand the high though not unexpected compression loads to which they were subjected. Szlapka had designed these chords, and Cooper had examined and approved them, and their failure “cannot be attributed directly to any cause other than errors in judgment on the part of these two engineers.” The report continued:

  These errors of judgment cannot be attributed either to lack of common professional knowledge, to neglect of duty, or to a desire to economize. The ability of the two engineers was tried in one of the most difficult professional problems of the day and proved to be insufficient for the task.… A grave error was made in assuming the dead load for the calculations at too low a value and not afterwards revising this assumption.… This erroneous assumption was made by Mr. Szlapka and accepted by Mr. Cooper, and tended to hasten the disaster.

  In short, what Szlapka had done was to let stand an educated guess as to the weight of steel that the finished bridge would contain. Such guesses, guided by experience and judgment, are the only way to begin to design a new structure, for without information on the weight of the structure the load that the members themselves must support cannot be fully known. When the loadings are assumed, the sizes of the various parts of the bridge can be calculated, and then their weight can be added up to check the original assumption. For an experienced engineer designing a conventional structure, a final calculation of weights only serves to confirm the educated guess, and so such a calculation may not even be made in any great detail. In the case of a bridge of new and unrealized proportions, however, there is little experience to provide guidance in guessing the weight accurately in the first place; a recalculation, or a series of iterated recalculations, is necessary to gain confidence in the design. (The situation is not unlike that of a veteran weight guesser at a carnival, who might be expected to predict quite well the weights of normal-sized fairgoers but not the weights of dwarfs or giants, who fall outside the range of even sideshow experiences.) According to the findings of the commission, “the failure to make the necessary re-computations can be attributed in part to the pressure of work in the designing offices and to the confidence of Mr. Szlapka in the correctness of his assumed dead load concentrations. Mr. Cooper shared this confidence.” Since Cooper was well known to have a “faculty of direct and unsparing criticism,” his confidence in Szlapka’s design work went unquestioned.

  Just as Cooper had confidence in Szlapka’s work, so did the resident engineer at the construction site have confidence in the work of them both. When a construction foreman expressed serious concern over the condition of the fatal member, the resident engineer thought the matter of little importance, telling the foreman, “Why, if you condemn that member, you condemn the whole bridge.” After the collapse, it was reported that the resident engineer “had confidence in that failing chord because it was to him unbelievable that any mistake could have been made in the design and fabrication of the huge structure over which able engineers had toiled for so many years.” As a result of the accident, however, Engineering News reported that mistakes of all kinds had become more believable.

  The underestimation of the true weight of the bridge had actually come to Cooper’s attention earlier in the design process, but only after considerable material had been fabricated and construction had begun. At this time, a recalculation of the stresses in the bridge led Cooper to consider that the error had meant that some stresses had been underestimated by 7–10 percent. All structures are designed with a certain margin of safety; he felt the error had reduced that margin to a small but acceptable limit, and so the work was allowed to proceed. In fact, some of the effects of the underestimated weights were, in the final analysis, of the order of 20 percent, and this was beyond the margin of error that the structure could tolerate.

  In its discussions of the various bridge-building organizations involved and their respective faults, the inquiry commission was clear regarding the sense of hubris and overconfidence that success can bring to an organization. In this regard, the Royal Commission anticipated in some ways by eight decades what the Presidential Commission would find in its investigation of the space shuttle Challenger’s accident. Although there do not seem to have been too few assistants in that more recent accident, there certainly seem to have been too many overconfident bosses, or at least too many bosses willing to make compromises for other than purely technical ends. According to the Royal Commission, reporting in 1908:

  Mr. Cooper states that he greatly desired to build this bridge as his final work, and he gave it careful attention. His professional standing was so high that his appointment left no further anxiety about the outcome in the minds of all most closely concerned. As the event proved, his connection with the work produced in general a false feeling of security. His approval of any plan was considered by every one to be final, and he has accepted a
bsolute responsibility for the two great engineering changes that were made during the progress of the work—the lengthening of the main span and the changes in the specification and the adopted unit stresses. In considering Mr. Cooper’s part in this undertaking, it should be remembered that he was an elderly man, rapidly approaching seventy, and of such infirm health that he was only rarely permitted to leave New York.

  Mr. Cooper assumed a position of great responsibility, and agreed to accept an inadequate salary for his services. No provision was made by the Quebec Bridge Company for a staff to assist him, nor is there any evidence to show that he asked for the appointment of such a staff. He endeavoured to maintain the necessary assistants out of his own salary, which was itself too small for his personal services, and he did a great deal of detail work which could have been satisfactorily done by a junior. The result of this was that he had no time to investigate the soundness of the data and theories which were being used in the designing, and consequently allowed fundamental errors to pass by him unchallenged. The detection and correction of these fundamental errors is a distinctive duty of the consulting engineer, and we are compelled to recognize that in undertaking to do his work without sufficient staff or sufficient remuneration both he and his employers are to blame, but it lay with himself to demand that these matters be remedied.

  The issues raised in the report were to reverberate throughout the engineering profession for many years to come, and in some form remain as issues today. Engineering work, especially relating to novel and untried projects, requires considerable time for thinking and rethinking about assumptions and tentative solutions, often among a broad range of colleagues and even in public forums. In cases like the Forth Bridge, the time and openness have been repaid in structures that stand as monuments. The very success of once bold endeavors like the Forth Bridge, however, can lead engineers like Cooper into a sense of security concerning ostensibly similar designs that may not be warranted. Incidents like the Quebec Bridge collapse provide rude awakenings, as Engineering News reported within weeks of the accident: “The Quebec Bridge collapse has been an object lesson to every structural designer; and we risk nothing in saying that in a thousand offices, stress computations are being checked over and details of design are being investigated and discussed with greater care and thoroughness than ever before.”

  The collapse of the Quebec Bridge, like that of the Tay, did not remove the need for a bridge at the location. Indeed, one could almost say there was renewed resolve to show that it could be done—and done right. The new design that was finally chosen was described at the time as “commonplace in appearance and costly to build.” This should not have been surprising, for matters of aesthetics and economy, so important when bridges are first planned, come to appear almost as luxuries in the wake of a tragedy of the magnitude of that which occurred on the St. Lawrence River in 1907. The bridge that was finally built at the site of the wreckage reinstated the look of straight bottom chords, which Szlapka had testified had been changed to curved ones “for the sake of artistic appearance.” Its outline did make the structure easier to analyze for load and stress distribution, but what was more significant about the new design was that it was a heavier and more substantial-looking bridge. If Cooper’s Quebec seemed to have the lightness of the Tay Bridge, the redesigned Quebec would appear to have the firmness of the Forth.

  5

  The collapse of the first Quebec Bridge in 1907 had a profound and immediate effect on the direction of bridge building worldwide. New York’s Queensboro Bridge, whose more lacy and graceful cantilever design with a maximum span of almost twelve hundred feet is often mistaken for that of a suspension bridge, was under construction when the Quebec collapse occurred. The Queensboro was completed in 1909 amid considerable protest and concern over its safety in particular, as well as over the safety of the entire genre. A second mishap in Quebec—which would occur in 1916, when the closing span of the redesigned structure fell to the bottom of the river while being hoisted into place, would reinforce reservations about the form. In spite of the resolve of the Canadian government to complete a successful cantilever design across the St. Lawrence River and thus vindicate the original decision, no other major cantilever bridge would be completed until the 1930s. To this day, none but the Forth comes within a hundred feet of the eighteen-hundred-foot span of the Quebec.

  The incidents at Quebec were naturally the subject of doubting editorials in newspapers and trade journals alike, for both the public and the profession took a keen interest in record-setting bridge building. Nevertheless, according to Engineering News-Record, “Twice the hopes of success have been dashed, but never in the heart of the true engineer was there doubt that the enterprise would be brought to a successful conclusion.” As with all failures, there were lessons learned in the collapse of the first bridge and the subsequent embarrassment during the final stages of the second, and it was the knowledge contained in these lessons that gave engineers the understanding to attack the problem of bridging the St. Lawrence with renewed confidence even in the wake of defeat, and enabled them in the end to “have vindicated the profession before a doubting world.”

  A variety of designs submitted for the rebuilding of the Quebec Bridge (photo credit 3.16)

  Years before the second Quebec accident, some members of the profession had their own concerns and prejudices about the whole process of choosing a bridge design. Among the most prominent and vocal of these was Gustav Lindenthal, who at the time of the Quebec collapse was a consulting engineer in New York City. Lindenthal had prepared the suspension-bridge proposal that the Phoenix Bridge Company had submitted along with its winning cantilever design in the original Quebec Bridge competition. He had also prepared a modified suspension-bridge design in response to the invitation by the board of engineers constituted to design a new bridge. This board had come up with specifications and with its own official concept, a cantilever structure with a straighter and bulkier outline than that of the collapsed structure. However, when bids were invited from construction companies, they were also given the option of submitting their own designs, though a company that did so had to assume “the entire responsibility not only for the materials and construction of the bridge, but also for the design, calculations, plans and specifications and for the sufficiency of the bridge for the loads” specified.

  A comparison of the cross sections of the lower-chord members of various late-nineteenth- and early-twentieth-century bridges (photo credit 3.17)

  Three members of the board of engineers for the redesigned Quebec Bridge standing inside one of the lower-chord members (left to right: Ralph Modjeski; C. N. Monsarrat, chairman and chief engineer; and C. C. Schneider) (photo credit 3.18)

  Lindenthal, in private practice as a consulting engineer, was personally and professionally outraged by such a condition, and he was greatly disappointed that the design chosen for the second Quebec Bridge was another cantilever, this one proposed by the St. Lawrence Bridge Company. The accepted design differed from the official one mainly in its central span of 640 feet, which was to be fabricated separately about three miles away, floated on barges to the bridge site, and hoisted into place between the 580-foot cantilever arms. After the choice had been made, Lindenthal wrote a lengthy tract entitled “Notes on the Quebec Bridge Competition,” printed in two successive issues of Engineering News, which was published every Thursday in New York City. Although it identified itself at the time as “a journal of civil, mechanical, mining and electrical engineering,” Engineering News concentrated on matters relating to large civil-engineering projects, especially those that had some connection to New York or to New York engineers like Lindenthal. The journal seems to have prided itself on publishing details of proposed, in-progress, and completed engineering projects, and the larger, more visible, and more expensive the projects it could report on, the better.

  The second Quebec Bridge accident, showing the central span buckling upon impact with the water (photo cr
edit 3.19)

  Lindenthal’s notes on the Quebec Bridge competition had been introduced by an editorial, which had called it “the most important international competition for the design of a great engineering structure which has ever been held.” The potential value of such a competition was, of course, that the various alternative designs emanating from the minds of creative engineers with a wide range of experience and vision provided an excellent opportunity for the comparison of the different proposed bridge types, and thereby the opportunity for understanding their relative strengths and weaknesses. A given engineer, especially within the limited time between the announcement of a design competition and the deadline for submission, may only be able to pursue in sufficient detail one or two designs. Decisions must necessarily be made at the outset as to whether to explore a cantilever or a suspension-bridge design or both, for example. If the engineer or the bridge company he is associated with is without work, considerable time may be available to be devoted to speculation on the competition, in the hopes of winning. If the engineer or his firm is already busy with other projects, a decision has to be made as to whether to hire additional engineers to work on those, so that a new design may be developed for the competition, or to try to find some spare time to think in new directions. In all cases, there is the question of how the time will be compensated for, and this was a point of major and emotional emphasis in Lindenthal’s notes on the Quebec competition.