The Second Sinking Of The ‘Maine’


That August the Army evaluated two methods for raising the ship. The first required attaching cables to the hull, which would then be lifted. The second involved building a cofferdam around the wreck, pumping out the water, and then refloating the ship or, if that was not possible, breaking her up. As far as anyone could guess, both plans would cost about the same. But the use of cables would jar the fragile, twisted remains and thus might destroy clues to the cause of the destruction. Engineers were not convinced that the cofferdam would work either, but if it did, they would be able “to examine the wreck thoroughly just as it lies and before there has been any change whatever in the relative locations of any of [the ship’s] parts.” Thus the cofferdam was the only approach likely “to accomplish the wishes expressed in Congress.” So the Army determined to part the waters of Havana Harbor.

The cofferdam was “without precedent in engineering history,” said the Army’s chief of engineers.

As described by the chief of the Army Engineers, the cofferdam was “in area, height and the pressure to be resisted without precedent in engineering history.” It would be an elliptical ring of twenty cylinders, each fifty feet in diameter, formed by interlocking sheet-steel piles driven to a depth of seventy-five feet. The cylinders would be connected on their outer perimeter by short arcs of similar sheet piles, then filled with stiff mud and clay. As the filling settled, it would press upon the cylinder walls to create a watertight barrier around the wreckage.

The building and draining of cofferdams were common practice among engineers working on inland waterways, but the Maine project was something else. This cofferdam would have to withstand the pressure of thirty-seven feet of seawater on a harbor bed layered with some twenty feet of soft mud and sand. Moreover, it had to hold up against waves, tidal currents, and gale-force winds. The Army had no illusions about how difficult the project would be, and troubles beset it from the start.

The first sheet pilings were slated for delivery in early November, but bad weather held them up in New York. Two hurricanes in Havana damaged much of the equipment promised by the Cuban government, causing further delays and adding to the cost. In its progress report of January 11, 1911, the Army requested an additional $350,000 to supplement its original $300,000 appropriation. The wide publicity may have put the Army on the spot, but it also encouraged the government to be generous. By the time the work was finished, the cost exceeded $750,000.

The Army continued its pile driving through the spring of 1911. Electrical power for light fed by underwater cable from Havana turned the project into an island of activity that went on twenty-four hours a day. By June all twenty cylinders and arches had been constructed and filled, and the slow, painstaking process of unwatering began.

Uncertain how the cofferdam would react under external pressure, the engineers devised a system of line measurements and gauges to detect the slightest movement of the cylinders. As the water level inside the cofferdam decreased to minus fifteen feet, the walls began to bulge inward. Immediately engineers increased the water level until the filling inside the cylinders could settle and become more stable. Thus attuned to the shifting personality of their handiwork, engineers continued pumping through June and into July, when some leakage became apparent. Although the amount of water seeping into the cofferdam was minimal, eight thousand yards of stone were placed against the inside sectors of the cylinders. By midsummer most of the water had been drained out, and the cofferdam stood like a fortress around the Maine. At the end of July the ship lay exposed, pungent with marine growth drying under the tropical sun.

The Navy stuck to its original conclusion that the ship had been destroyed by an external explosion.

The Navy’s man at the site, Capt. William B. Ferguson, found the ship’s superstructure covered with two to three inches of mud and reported that three feet of sludge clogged the cabins, engine rooms, and boiler compartments. He noted that most of the visible wreckage was a “tangled mass” and that the damage was far more extensive than the 1898 investigation had indicated.