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

by Henry Petroski

  John Roebling’s Suspension Bridge and the cantilever bridge over the Niagara Gorge, with Whirlpool Rapids in the foreground, in an etching from one of the many late-nineteenth-century tour guides of the area (photo credit 3.5)

  The invitation to the opening of the bridge below Niagara Falls was signed by Chanute’s protégé, Schneider, as chief engineer of the project. The engraving on the invitation shows Schneider’s bridge in the foreground and the famous suspension bridge in the background—symbolic and prophetic of the relative positions the two bridge types were to hold, in the minds of some at least, for the next three decades or so. Also indicative of the climate in which bridge building was taking place in the closing decades of the nineteenth century, the 1883 invitation associated no personal name with the great structure, which was called simply, directly, and technically the “Canti-lever Bridge below Niagara Falls.”

  The hyphenation of the word “canti-lever” attests to how newly coined it was, at least with reference to bridge building; it required explanation, especially when applied to Fowler and Baker’s bridge, under construction across the Firth of Forth, less than fifty miles south of where the Tay Bridge had collapsed. Though Eads had in fact begun his explanation of the principle of his arch bridge with a discussion of a canted lever, that fifteen-year-old report to the Illinois and St. Louis Bridge Company had been generally forgotten. In fact, the cantilevered method of construction used in erecting the Eads Bridge, once the most striking visible feature of the project, was now seldom referred to.

  With the growing publicity in both Britain and America surrounding the supposedly new type of bridge, the hyphen was quickly dropped—but not the curiosity about the form. A reader of Engineering News, who wrote to the editor in late 1887 from an engineer’s camp near Danielsville, Georgia, asked, “Whence comes the term ‘Cantilever’ as applied to bridges; or in other words what is a Cantilever bridge?” The editor responded:

  This is a question quite frequently asked and we might as well answer once for all. The term, as applied to a bridge, is of comparatively recent origin, but the principle is as old as the Hindoos and the art of building itself. It has been applied to wooden bridges for centuries, and it is only its later scientific solution by modern builders of steel and iron bridges that has brought it forward again prominently. Its advantage over other forms of truss construction is, that by a proper method of anchoring or balancing and the arrangement of its tension and compression members, it can be erected over space without supporting false-work. The Niagara and the Forth bridges are the latest examples of its application to a site where the conditions made false works impossible, or very expensive.

  The idea of a cantilever played a central role in a remarkable lecture that Baker delivered at the Royal Institution in 1887, in which he also gave some indication of the public scrutiny under which the new bridge was being built. Eight years after the Tay Bridge disaster at Dundee, he felt it necessary to preface his remarks with a declaration that the Forth and Tay bridges were quite different structures, albeit still confused in the public’s mind. He related an exchange he fully expected to have with every second Britisher: “ ‘How are you getting on with the Tay bridge?’ I suggest ‘Forth Bridge,’ and the correction is generally accepted as a mere refinement of accuracy on my part.” Not surprisingly, Americans were no more sure about Scottish geography. A report in Scientific American in early 1888 on the progress of construction on the cantilever bridge at Poughkeepsie, New York, misidentified the setting of the Forth cantilever as “between England and Scotland.”

  It was not just the location of the bridge that Baker was at pains to explain; he had also to convey a sense of its great size, each span being almost four times as long as the longest tubes in Stephenson’s famous Britannia Bridge. Baker appealed to common points of reference to enable his London audience to appreciate the size of the Forth’s spans:

  To get an idea of their magnitude stand in Piccadilly and look towards Buckingham Palace, and then consider that we have to span the entire distance across the Green Park, with a complicated steel structure weighing 15,000 tons, and to erect the same without the possibility of any intermediate pier or support. Consider also that our rail level will be as high above the sea as the top of the dome of the Albert Hall is above street level, and that the structure of our bridge will soar 200 ft. yet above that level, or as high as the top of St. Paul’s. The bridge would be a startling object indeed in a London landscape.

  Benjamin Baker, circa 1890 (photo credit 3.6)

  The image of the bridge amid familiar projections into the London skyline, as well as transported landmarks such as the Great Pyramid at Giza, St. Peter’s in Rome, the Cathedral at Chartres, and a host of others, was to be made real in a mural in the South Kensington Museum. But Baker wished to convey to his audience not only the monumental size of his bridge but also the principles by which it stood. He asserted that it had “excited so much general interest” in part because it was “of a previously little known type.” He would “not say novel, for there is nothing new under the sun,” but made no references to Gerber’s bridge, the American examples, or the cantilever method of construction of the Eads Bridge.

  Baker did acknowledge that a visitor being shown about the construction site had suggested a Chinese precedent, to which the engineer replied, “Certainly.” He went on to elaborate:

  The Forth Bridge drawn to scale before familiar structures and landmarks (photo credit 3.7)

  Indeed, I have evidence that even savages when bridging in primitive style a stream of more than ordinary width, have been driven to the adoption of the cantilever and central girder system, as we were driven to it at the Forth. They would find the two cantilevers in the projecting branches of a couple of trees on opposite sides of the river, and they would lash by grass ropes a central piece to the ends of their cantilevers and so form a bridge. This is no imagination, as I have actual sketches of such bridges taken by exploring parties of engineers on the Canadian Pacific and other railways, and in an old book in the British Museum, I found an engraving of a most interesting bridge in Thibet upwards of 100 ft. in span, built between two and three centuries ago and in every respect identical in principle with the Forth Bridge. When I published my first article on the proposed Forth Bridge some four years ago, I protested against its being stigmatised as a new and untried type of construction, and claimed that it probably had a longer and more respectable ancestry even than the arch.

  Baker seemed willing to acknowledge the ancient roots of his bridge, but nowhere in his lecture did he emphasize the precedents of his near contemporaries, Gerber, Shaler Smith, or Schneider. Indeed, after establishing the time of his first publication on the topic, Baker made remarks that cause the modern reader to wonder whether he read nothing but old books or simply did not wish to acknowledge the competition. Charles Schneider’s fifty-page article on a cantilever bridge at Niagara Falls, followed by even more pages of discussion, had appeared in the Transactions of the American Society of Civil Engineers two years before Baker’s lecture. However, at the time, British engineers generally seemed little interested in recognizing or acknowledging American precedents. In a passage remarkably reminiscent of Eads’s feud with Roebling in the pages of Engineering, Baker continued:

  An Asian cantilever bridge with a central girder span, which Baker found identical in principle to the Forth Bridge (photo credit 3.8)

  The best evidence of approval is imitation, and I am pleased to be able to tell you that since the first publication of the design for the Forth Bridge, practically every big bridge throughout the world has been built on the principle of that design, and many others are in progress.

  Interestingly, Baker avoided the use of the term “cantilever” in this passage, perhaps to make his assertions literally correct. Earlier in his lecture, however, he had introduced the new term with some elaboration, and reservation as to whether his bridge was a true cantilever:

  One of the first que
stions asked by the generality of visitors at the Forth is—why do you call it a cantilever? I admit that it is not a satisfactory name and that it only expresses half the truth, but it is not easy to find a short and satisfactory name for the type. A cantilever is simply another name for a bracket, but a reference to the diagram will show that the 1700 ft. openings of the Forth are spanned by a compound structure consisting of two brackets or cantilevers and one central girder.

  The anthropomorphic model that Baker used in his lectures on the Forth Bridge (photo credit 3.9)

  Baker then described how, in preparing for his lecture, he “had to consider how best to make a general audience appreciate the true nature and direction of the stresses on the Forth Bridge, and after consultation with some engineers on the spot, a living model was arranged.” The elusive Baker did not make clear whether he, the engineers, or all together came up with the idea, but the striking anthropomorphic model was very effective and often reproduced, both literally and visually, at the time:

  Two men sitting on chairs extended their arms and supported the same by grasping sticks butting against the chairs. This represented the two double cantilevers. The central girder was represented by a short stick slung from one arm of each man and the anchorages by ropes extending from the other arms to a couple of piles of brick. When stresses are brought on this system by a load on the central girder, the men’s arms and the anchorage ropes come into tension and the sticks and chair legs into compression. In the Forth Bridge you have to imagine the chairs placed a third of a mile apart and the men’s heads to be 300 ft. above the ground. Their arms are represented by huge steel lattice members, and the sticks or props by steel tubes 12 ft. in diameter and 1¼ in. thick.

  Nor did Baker mention who represented the “load on the central girder” in the human model. In fact, it was Kaichi Watanabe, a young Japanese engineer, apparently among the first from his country to study in Britain, who was sitting, hands to his sides, grasping what looks like a very narrow swing-like seat representing the central span of the bridge. He was a student of Fowler and Baker and “was invited to participate in the human model of the cantilever to remind audiences of the debt the designers owed to the Far East where the cantilever principle was invented.” In fact, New York’s Engineering News actually reproduced the “ingenious illustration of the cantilever bridge principle” a full six weeks before it appeared in London’s Engineering and called the model “a Japanese idea, as may be suspected from the central figure,” Watanabe. The American journal also reported that the model was “received with loud and general applause” during Baker’s lecture.

  The report in Engineering News was no anomaly, for in America there was considerable interest in the design and construction of the Forth Bridge and large and unique engineering projects generally, even if their American precedents were not acknowledged. Engineers and information could and did travel freely and relatively quickly by ship across the ocean, as Eads and Roebling knew, and the transatlantic cable had been operational for two decades. In the case of the human model of the Forth Bridge, Engineering News acknowledged its indebtedness to Thomas C. Clarke, a former director and future president of the American Society of Civil Engineers, for the use of the original photograph from which the illustration could be reproduced only weeks after Baker’s speech. This makes Baker’s silence about the existence of not insignificant contemporary cantilever bridges in America all the more inexcusable, and it reinforces the judgment that he at least, if not the entire British engineering profession, was unwilling to acknowledge that there was much to be learned from contemporaneous American experience. If anything, Baker offered a bit of gratuitous ridicule of the eighteen-hundred-foot cantilever bridge that the American Thomas Pope had proposed in 1810 to join New York and Brooklyn. He called his bridge a “flying lever pendant” and, reminiscent of a more famous Pope, described it in heroic couplets. Baker quoted from these in his lecture, no doubt because they gave so apt a description of his own project:

  Two of many possible ways proposed to bridge the Firth of Forth, with the accepted design below (photo credit 3.10)

  Each semi-arc is built from off the top,

  Without the aid of scaffold, pier, or prop;

  By skids and cranes each part is lowered down,

  And on the timber’s end grain rests so sound.

  Sure all the bridges that were ever built,

  Reposed their weight on centre, pier, or stilt;

  Not so the bridge the author has to boast.

  His plan is sure to save such needless cost.…

  Baker admitted approvingly that, should he have thought of describing his own bridge in verse, he would have “appropriated bodily Mr. Pope’s lyrical version.” However, Baker chose “sober prose” to describe the Forth Bridge, which after all was soon to be a completed reality rather than the unrealized dream of an American poetaster.

  Construction of the Forth Bridge began, as did that on the Eads and Brooklyn spans, with the sinking of caissons, and Baker again avoided mentioning the singular achievements of those recently completed American bridges except to note that almost one out of every five men who worked on the foundations of the St. Louis Bridge was attacked by some form of caisson disease, with sixteen deaths, whereas his bridge had “no deaths directly attributable to the air pressure.” There were accidents, of course, especially among workers assembling the superstructure. Baker closed his speech by speaking of the risk that the “zealous and plucky workmen” performed high above the firth. Speaking on behalf of the engineers, he said, “we never ask a workman to do a thing which we are not prepared to do ourselves, but of course men will, on their own initiative, occasionally do rash things.” He concluded, rather insensitively:

  Happily there is no lack of pluck among British workers; if one man falls another steps into his place. Difficulties and accidents necessarily occur, but like a disciplined regiment in action we close up the ranks, push on, and step by step we intend to carry on the work to a victorious conclusion.

  The Forth Bridge under construction (photo credit 3.11)

  Construction of the Forth Bridge was under the personal direction of the contractor William Arrol, whose company built the second Tay, but here he was in partnership with Joseph Phillips, who had considerable experience building large bridges, and with Sir Thomas Tancred and Travers H. Falkiner. Work was concluded in 1890, and early that year the definitive and comprehensive account of the project was written by Wilhelm Westhofen, the engineer who had supervised the building of the central section. Among the “chief desiderata,” according to Westhofen, was the need for the “maximum attainable amount of rigidity,” not only vertically, under railroad trains, but also horizontally, against wind pressure, so that the completed bridge “may by its freedom from vibration gain the confidence of the public, and enjoy the reputation of being not only the biggest and strongest, but also the stiffest bridge in the world.” This was necessary, of course, because the memory of the decade-old Tay disaster was still fresh. It might not have been necessary, however, for Westhofen to remind his readers that Bouch’s “original suspension-bridge design complied with none” of the desiderata, which also included assurances of fully tested materials, facility of erection (during which time the incomplete bridge was expected to be as safe against the wind as the completed structure was to be), and maximum economy consistent with safety.

  The Eads Bridge is generally acknowledged as the first to use steel, and its success may have led to the Board of Trade’s 1877 lifting of its ban on the use of steel in British bridges. The Forth was thus the first major bridge to be made fully of steel, which was manufactured using the Siemens-Martin open-hearth process. The bulk of the material was supplied by the Steel Company of Scotland, with some also coming from Wales. The Clyde Rivet Company in Glasgow supplied the forty-two hundred tons of rivets required. Although steel was 50 percent stronger than wrought iron, Baker assured his audience at the Royal Institution that the material used
in the bridge was “not in any sense of the word brittle, as steel is often popularly supposed to be, but it is tough and ductile as copper.” He went on to make the point in more familiar terms: “You can fold ½ in. plates like newspapers, and tie rivet bars like twine into knots. The steel shavings planed off form such long, true and flexible spirals, that they are largely used for ladies’ bracelets when fitted with clasps and electro plated.”

  The completed Forth Bridge, showing Inchgarvie under the central pier and the South Queensferry landing (photo credit 3.12)

  For all the flexibility of the steel shavings, the bridge certainly was stiff—even during construction, when its great cantilever arms grew out symmetrically from each pier. It looked not unlike the Eads Bridge during its construction, but whereas the towers and cables that held up the Eads could be removed once the arches were complete, there would be no such extraneous falsework to be removed once the Forth Bridge was completed. Only the riveting cages and cranes would have to be taken off the finished bridge. Baker had predicted that each 1,710-foot span would deflect no more than four inches under the heaviest train, and measurements on the finished bridge showed an actual deflection of only three and a half inches.

  The bridge was built “straddle legged” not only to achieve a great stiffness against the wind but also to look as if it did. The columns over the piers are as much as 120 feet apart at the base but only thirty-three feet apart at the top, giving the bridge the appearance of being able to withstand the severest blow. The bridge has been referred to as having a “Holbein straddle,” after the stance that characterized male portraits by the German artist Hans Holbein. Fowler was apparently quite aware of this, reportedly having once remarked to the Scottish-born mechanical engineer James Nasmyth, at a London exhibition of Holbein’s work, that the Tay might not have fallen had its piers had such a straddle.