In the summer of 1902, the U.S. Congress voted on the route by which an artificial river would join two oceans. Panama won—but by such a thin margin and through such a string of accidents that counterfactual history here stops being a parlor game and becomes an engineering experiment.
By the beginning of the twentieth century, the idea of an interoceanic canal through Central America already had half a century of turbulent history behind it. The Spanish had dreamed of it under Charles V, but it was the French who tackled it seriously—and got burned. After the Spanish-American War of 1898, when the United States began to see itself as a two-ocean power with a fleet that needed to move quickly between the Atlantic and Pacific, the question became practical. During the war with Spain, the battleship Oregon made the voyage around South America—some 14,000 nautical miles and nearly 70 days. That journey became a living argument: the country needed a canal, and it needed one soon.
Several routes were considered—through Tehuantepec in Mexico, through Darién, through Panama, and through Nicaragua. By 1901–1902, two serious candidates remained. The Panamanian isthmus—narrow, about 80 kilometers from ocean to ocean, with an existing railroad and an enormous amount of earth already moved by the French. The Nicaraguan route—longer, roughly 280 kilometers, but with much of the way covered by a natural water system: the San Juan River, flowing out of Lake Nicaragua, and the lake itself—the largest in Central America, with an area of around 8,200 square kilometers. From the lake’s western shore to the Pacific, all that remained was a relatively short cut through a low watershed. At first glance, nature had already done half the work.
The Walker Commission—the Isthmian Canal Commission, at work since 1899—recommended Nicaragua twice. The first time, because the French New Panama Canal Company wanted $109 million for its assets, which the commission considered excessive. When the French lowered the price to $40 million, the commission changed its recommendation in January 1902 in Panama’s favor. The difference between the two decisions was the price tag. The engineering merits of the routes had not changed by a single meter.
On June 28, 1902, President Theodore Roosevelt signed the law known as the Spooner Act. Its wording was designed as a conditional arrangement: the president was authorized to acquire the French assets in Panama for no more than 40 million dollars and negotiate with Colombia for a strip of land for the canal. If that could not be done within a “reasonable time,” the government was required to switch to the Nicaraguan route.
The law was, in essence, a double-barreled gun. One barrel pointed at Panama, the other at Nicaragua. And the Colombian Senate pulled the trigger on the second barrel when, on August 12, 1903, it unanimously refused to ratify the Hay–Herrán Treaty. Colombia wanted more money and was unwilling to surrender sovereignty over the canal zone on Washington’s terms. Formally, at that point, the Spooner Act should have sent the Americans to Nicaragua.
But something else happened. On November 3, 1903, Panama declared independence from Colombia. The American warship Nashville stood in Colón harbor. Roosevelt recognized the new republic with astonishing speed—within two days. And by November 18, the Hay–Bunau-Varilla Treaty had already been signed, giving the United States control over a ten-mile canal zone “as if it were sovereign.” Philippe Bunau-Varilla, who signed the treaty on Panama’s behalf, was a French engineer, a shareholder in the New Panama Canal Company, and a man who had spent years lobbying for the Panamanian route in Washington. No Panamanian citizen had seen the text of the treaty before it was signed.
That is the turning point. If Roosevelt had not supported the Panamanian revolt—and he had both legal and moral grounds not to—the Spooner Act would have sent engineers to the shores of Lake Nicaragua. And then we find ourselves in a world where an entirely different canal has to be dug.
The Nicaraguan route looked attractive on a map and frightening in cross-section. The San Juan River runs from Lake Nicaragua to the Caribbean Sea for roughly 200 kilometers, but it is shallow, winding, and subject to seasonal changes in water level. Lake Nicaragua lies about 32 meters above sea level—significantly higher than the artificial Gatun Lake eventually built in Panama at an elevation of 26 meters (Gatun had to be created from scratch; here, the lake already existed, but its level would have needed to be controlled). The cut from the lake’s western shore to the Pacific was roughly 20 kilometers across a watershed whose highest point was around 45 meters.
The project would have required a system of locks at both ends—on the Caribbean side, to raise ships to lake level, and on the Pacific side, to lower them again. The San Juan River would have needed massive deepening and straightening, and in its lower reaches, probably replacement of the natural channel with a canal. Flood runoff was a separate headache: dozens of rivers flow into Lake Nicaragua, and during the rainy season the water level rises sharply. Managing that runoff would have required dams and spillways comparable in scale to what was later built on Panama’s Chagres River.
And then—volcanoes. Lake Nicaragua lies in a zone of active volcanism. Momotombo, Concepción on Ometepe Island in the middle of the lake, the chain of volcanoes along the Pacific coast—none of this was scenery but geological reality. Earthquakes struck the region regularly. What has always fascinated me is how this argument—entirely legitimate from an engineering standpoint—was turned into a theatrical gesture.
The story of the postage stamp is usually told as an anecdote that decided the fate of two countries. In reality, it was the final chord of a long and well-organized campaign.
Philippe Bunau-Varilla and attorney William Nelson Cromwell were the two principal lobbyists for the Panamanian route. Cromwell represented the interests of the New Panama Canal Company and the Panama Railroad, while Bunau-Varilla acted as both shareholder and engineering evangelist. They worked with senators, published pamphlets, arranged speaking engagements. Senator Mark Hanna of Ohio, a close ally of the late President McKinley and an influential figure in the Republican Party, became their chief voice in the Senate.
Against them stood Senator John Tyler Morgan of Alabama—a committed supporter of the Nicaraguan route, who saw a canal through Nicaragua as a way to revive Southern trade through Gulf of Mexico ports. Morgan had promoted a Nicaraguan canal for decades, and the House of Representatives had already passed the Hepburn bill in its favor.
In May 1902, while the Senate was debating the Spooner Act, disaster struck the Caribbean island of Martinique: on May 8, Mount Pelée destroyed the city of Saint-Pierre, killing around 30,000 people. The event electrified the debate. Bunau-Varilla seized the moment. He sent every senator a Nicaraguan postage stamp depicting the Momotombo volcano with smoke pouring from its summit. The stamp’s caption said it was the official symbol of the country through which the proposed canal would be built. The argument was simple: Nicaragua advertised its own volcanoes.
Effective? Absolutely. Decisive? No. The Senate voted 42 to 34 for the Panamanian route. An eight-vote margin is substantial. The stamp was one piece of a mosaic that included the reduced price of the French assets, Cromwell’s lobbying, the authority of the Walker Commission, the eruption of Mount Pelée, and Hanna’s political skill. To attribute the decision to one stamp is to confuse an elegant detail with the mechanism.
To understand why Panama ultimately worked, you have to understand why it failed first. Ferdinand de Lesseps, the triumphant builder of the Suez Canal, began constructing a sea-level canal in Panama in 1881—no locks, a straight trench from ocean to ocean. Suez had been cut through flat desert. Panama was tropical forest, a mountain range, the Chagres River with its monstrous floods, geology made up of soft clays and hard basalts all jumbled together, and two kinds of deadly disease whose nature was still not understood in the 1880s.
Over eight years, the French removed an enormous volume of earth—by various estimates, from 30 to 60 million cubic meters. But excavating the Culebra Cut through the Continental Divide at sea level required literally shaving down mountains—and every rainy season, landslides filled the trench back in. Workers died by the thousands from yellow fever and malaria: the total death toll during the French period is estimated at 20,000 and more. De Lesseps’s company went bankrupt in 1889, the scandal shook the Third Republic, and the word “Panama” became synonymous with fraud in French.
The New Panama Canal Company, established in 1894, operated at a minimal level—maintaining equipment, carrying out minor work—but its main purpose was to preserve the concession and sell the assets. It was these assets—equipment, the railroad, maps, completed excavations—that the United States bought in 1904 for those same 40 million dollars.
The Americans received from the French not only machinery and trenches but—perhaps more importantly—negative knowledge. They knew that a sea-level canal through Panama was a task at the edge of the possible. But here is the curious part: that knowledge did not automatically lead to a choice in favor of a lock canal. In 1905–1906, after American work had already begun, a fierce dispute raged on the engineering board. A minority led by chief engineer John Frank Stevens, with Roosevelt’s support, insisted on a lock canal. The majority of international experts on the board—including several European engineers—recommended a sea-level canal, considering it the more durable solution.
Congress approved the lock design in 1906. Its key element was the decision to dam the Chagres River and create the artificial Gatun Lake. The lake raised ships 26 meters above sea level, and much of the journey across the isthmus took place over water—with no digging required. The volume of excavation in the Culebra Cut was reduced radically: instead of cutting the watershed down to sea level, it only had to be lowered to lake level. Water ceased to be the enemy and became an ally—it filled the valley and made the canal itself.
The sanitary history of the Panama Canal is one of those chapters that sounds like a triumph of reason but, on closer inspection, turns out to be a triumph of bureaucratic stubbornness. William Crawford Gorgas, an army doctor, arrived in the Canal Zone in 1904 with experience fighting yellow fever in Havana after the Spanish-American War. He knew the work of Carlos Finlay and Walter Reed, who had proved that yellow fever was carried by the Aedes aegypti mosquito. Malaria was carried by Anopheles.
The first year was nightmarish. The Canal Zone leadership—the first Isthmian Commission—did not believe in the mosquito theory and considered Gorgas an alarmist. They denied him funds for screens, kerosene to treat standing water, drainage work. Yellow fever flared up again in 1905, and workers fled. Only after Stevens became chief engineer and personally demanded Washington’s full support for Gorgas did the situation change. Gorgas received men, money, and authority. His team drained puddles, poured kerosene and oil into every patch of standing water, installed window screens, and provided piped water so people would not keep water in open containers. By 1906, yellow fever in the Canal Zone had effectively been eliminated. Malaria was reduced dramatically, though it could not be completely defeated before construction ended.
Could this system have been transferred to Nicaragua? Gorgas’s methods were universal—mosquitoes breed in standing water everywhere in the tropics. But the scale of the problem in Nicaragua would have been different. Lake Nicaragua is a gigantic freshwater expanse with marshy shores. The San Juan River, with its floodplain, tributaries, and swampy lowlands, is ideal Anopheles habitat. Draining the surroundings of the construction site in Panama was difficult, but the territory was compact. In Nicaragua, the construction zone would have stretched across hundreds of kilometers. Controlling mosquito breeding grounds along the entire route was a challenge of another order. Not impossible, but demanding substantially greater resources and time. I think Gorgas would have managed it—but more slowly, more expensively, and with greater losses in the early phase.
There is another factor: political control. In Panama, the Americans governed the ten-mile zone as their own territory—with their own police, courts, and sanitary inspectors. In Nicaragua, the terms of the concession would probably have been different. Nicaragua was a sovereign state, if one under strong American influence—the U.S. Marines landed there repeatedly. Imposing a strict sanitary regime over a long stretch of another country’s territory is diplomacy, not engineering.
A lock canal in Nicaragua would paradoxically have looked both simpler and more complicated than Panama’s. Simpler—because the lake already existed; there was no need to create it by flooding a valley. More complicated—because a natural lake does not obey an engineer the way an artificial reservoir does.
Gatun Lake in Panama is a human creation. Its level is controlled by a dam and spillways. Engineers decide how much water to release and can keep the level within the required range. Lake Nicaragua is fed by dozens of rivers and streams, its drainage basin is vast, and its only natural outlet is the San Juan River. In the rainy season, the lake rises; in the dry season, it falls. The range of fluctuation is around 1–2 meters, but for a lock canal every meter is critical: it determines channel depth and the operation of the lock chambers.
To stabilize the level, regulating works would have been needed on the San Juan—essentially a dam or series of dams turning the river from a natural outlet into a controlled spillway. But the San Juan is a border river between Nicaragua and Costa Rica for part of its length, and any construction on it would have affected the interests of a third country. The Clayton–Bulwer Treaty of 1850 between the United States and Britain, though replaced by the Hay–Pauncefote Treaty of 1901, set a precedent for international restrictions on such projects. Costa Rica repeatedly challenged Nicaragua’s rights over the San Juan.
Dredging work on the lake and river would have been colossal. Lake Nicaragua is shallow on average—around 12–15 meters. To allow ocean-going vessels with drafts of 10 meters or more to pass, the channel would have had to be deepened along the entire lake portion of the route. The lakebed material—silt, volcanic sand, clay—would have had to go somewhere. In Panama, the spoil excavated from the Culebra Cut was used to build Gatun Dam and embankments. In Nicaragua, the logistics of moving earth along an extended water route would have been more complicated.
The Pacific cut through the watershed was the most straightforward element of the project, closest to classic earthworks. But even here, the region’s volcanic geology would have produced its own surprises: tuffs, lava flows, unstable soils. The Culebra Cut in Panama, with its landslides, was a warning that tropical geology does not forgive optimism.
The Panama Canal opened on August 15, 1914—ten years after American work began and thirty-three years after the French began. The final cost of American construction came to around 375 million dollars (plus 40 million for the French assets). If the U.S. Army Corps of Engineers had begun work in Nicaragua in 1903–1904, when would the canal have opened?
A direct extrapolation is impossible—there are too many variables. But a few benchmarks can be set. A Nicaraguan canal would not have had the benefit of the French groundwork: no completed excavations, no railroad, no equipment on site. Everything from scratch. The preparatory stage—building ports, roads, housing, hospitals, warehouses—would have taken at least two or three years, as it did in Panama. The main earthworks along a route three to four times longer than Panama’s would have taken more time, even allowing for the fact that part of the route was already open water. Dredging the lake, canalizing the San Juan, building locks at both ends, the Pacific cut—roughly speaking, this meant 12–18 years of construction with American organization and resources. The canal’s opening would have shifted to 1918–1922—meaning it would not have been ready for the First World War.
Cost is an even more speculative question. In 1901, the Walker Commission estimated a Nicaraguan canal at roughly 190 million dollars, and a Panamanian one at 144 million—not counting the purchase of French assets. Both estimates proved two and a half times too low for Panama. Apply the same multiplier to Nicaragua, and the total comes to around 450–500 million—and that is without the additional complications that would inevitably have emerged along the way.
If the canal is built in Nicaragua, Panama remains a province of Colombia. There is no separation in 1903, no Canal Zone, no republic whose fate is tied for a century to a ten-mile strip of foreign territory in the middle of its own country. The Panama Railroad continues operating, Panama City remains a regional trading center, but without the canal it does not become one of the hubs of global logistics. Colón does not become the canal’s Atlantic gateway. All of these are unlived lives.
Nicaragua, by contrast, finds itself in the position Panama held in reality: a canal country, a state whose economy and politics are shaped by the transit flow between oceans. The American military presence, already substantial—the Marines were in Nicaragua intermittently from 1909 to 1933—would have become permanent and institutionalized. A canal zone in Nicaragua—long, running through the heart of the country—would have created an even deeper intrusion into sovereignty than the ten-mile strip in Panama. It is hard to imagine Augusto Sandino’s movement developing in the same way in the 1920s–1930s if the stakes of the American presence had been many times higher.
Military logistics: a canal through Nicaragua would lie farther north than Panama’s, closer to Gulf of Mexico ports and the U.S. East Coast. For the American fleet, that is an advantage—a shorter route from Norfolk or New Orleans. But the canal would be longer, transit through it would take more time, and capacity would probably be lower because of the limitations of the lake channel and river section. During the Second World War, when thousands of warships and transports passed through the Panama Canal, that difference in capacity might have proved critical.
Global trade: shipping routes would have reorganized around a different port of call. Instead of Panama and Colón, the key points would have been Greytown (San Juan del Norte) at the Caribbean end and Brito or San Juan del Sur on the Pacific. The geography of world trade has inertia: ports that grow around a canal create infrastructure that attracts still more cargo. Nicaraguan ports in this alternate world would have become what Panama and Colón became in ours.
What gets me most about this story is how tightly intertwined are the things we are used to placing on separate shelves. The geology of the Culebra Cut and the financial scandal of the Third Republic. The life cycle of the Aedes aegypti mosquito and a Senate vote. The price demanded for its shares by a bankrupt French company and the independence of an entire country. A postage stamp from a small state and the strategic map of a two-ocean power.
The Panama Canal was not inevitable. It became possible through a chain of decisions, each of which could have gone another way. If the French had not lowered the price—Nicaragua. If the Colombian Senate had ratified the treaty—Panama, but without secession and probably on different terms. If Roosevelt had not supported the revolt—Nicaragua. If the engineering board in 1906 had insisted on a sea-level canal—construction would have dragged on for decades and perhaps would never have been completed at all before new excavation technologies were invented.
The canal that opened in 1914 was the result of the right decisions being made in the right order, often for the wrong reasons and by people whose motives were far removed from engineering perfection. Bunau-Varilla wanted to save his shares. Cromwell wanted his fee. Roosevelt wanted a canal in his lifetime and was not too concerned about the means. Stevens wanted to build it properly and left when he grew tired of fighting the bureaucracy; he was replaced by army engineer George Washington Goethals, who finished it. Gorgas wanted to prove the mosquito theory and save lives. Out of these competing wills came a mechanism that still works today—expanded in 2016 with a third set of locks, but fundamentally the same as the one that opened two weeks before the First World War began.
The Nicaraguan canal was never built. Projects surfaced again and again—in the 1910s, 1930s, 1960s, and quite recently, in 2013, when Nicaragua granted the Hong Kong company HKND Group a concession to build a canal costing 50 billion dollars. The project stalled after several years of protests, environmental objections, and financial trouble for the concessionaire. Two oceans are still joined by one canal—the one that passed through the narrower isthmus, following in the tracks of bankrupt Frenchmen, through mountains that slid back into the trench, in a country that came into being so that this canal could be built.