In September 1882, four hundred lamps lit up on Pearl Street in Lower Manhattan, and Thomas Edison proved that electricity could be a business. Ten years later, his system lost—but what if it hadn’t?
The Pearl Street station supplied an area of Manhattan roughly one square mile in size. This was not a limitation of imagination, but of physics: at 110 volts, direct current lost so much energy to the resistance of the wires that running a line any farther became pointless—copper turned into a street heater. The formula for losses is brutally simple: the power dissipated in a conductor is proportional to the square of the current. To transmit the same power at low voltage, you need a large current, which means thick, heavy, expensive wire. The three-wire system patented by John Hopkinson in 1882 raised the voltage to 220 volts between the outer wires and saved nearly two-thirds of the copper—but the radius still remained within a few miles, and that was the ceiling of DC technology at the time. Every city block aspiring to electric lighting needed its own small power station—with a boiler, steam engine, generator, and coal store.
Edison, though, saw this as a business model. A local station serves an affluent neighborhood, collects payment for light, pays for itself. Manhattan, London, Milan—he opened stations in city centers, where customers with money lived close together. The system worked, made money, and seemed sufficient to him. Why haul electricity twenty miles if you can put a generator next to the consumer? The question sounded perfectly reasonable right up to the moment someone wanted to light more than Wall Street.
A transformer is a disarmingly simple device: two coils of wire on a shared iron core. Alternating current in one coil induces alternating current in the other, and the ratio of turns determines how many times the voltage rises or falls. The key word is “alternating.” Direct current creates no changing magnetic field, and a transformer is dead to it. There is nothing to refine here: it is Faraday’s law.
George Westinghouse founded Westinghouse Electric in 1886, and his bet was precisely on the transformer. The idea was to generate electricity at a large station, raise the voltage to thousands of volts for transmission over thin wires across dozens of miles, then lower it again to a safe level at the consumer end. Losses fell dramatically, because at high voltage the same watt is transmitted with a small current. Copper savings multiplied. One large station by a waterfall or a coal mine could replace dozens of small urban boiler houses.
Westinghouse licensed Nikola Tesla’s patents for the multiphase AC motor—and got what the AC system had lacked to be complete: a motor capable of turning machine tools, pumps, and trams. I am still struck by the elegance of Tesla’s solution: a rotating magnetic field created by phase-shifted currents makes the rotor chase the field like a donkey chasing a carrot—no brushes, no commutator, almost no wear. Edison’s DC motors required carbon brushes that wore down and sparked. Across hundreds of factories, the difference in maintenance became a difference in profits.
When the technical argument began tipping in favor of alternating current, the struggle moved into territory we would now call a smear campaign. In the summer of 1888, engineer Harold P. Brown—with financial and technical backing from Edison Electric—began a series of public demonstrations in which he killed animals with alternating current. Dogs, calves, horses died before audiences, and every death was meant to hammer one thought into the public mind: AC was the “current of death.”
Brown actively lobbied for the use of a Westinghouse alternating-current generator in New York State’s first electric chair. The logic was cynical and inventive: if the state kills criminals specifically with alternating current, then alternating current is an instrument of execution, not a household convenience. Westinghouse, understanding the reputational blow, tried to block the sale of his generators for that purpose. The first execution by electric chair—that of William Kemmler, in 1890—was monstrous: it took two shocks, the procedure lasted minutes, and witnesses described the smell of burning flesh. Newspapers called it “a disgrace to civilization,” but blamed the failure not on alternating current, but on the execution itself.
Separately—there is a story here that needs untangling. The execution of the elephant Topsy at Luna Park on Coney Island took place in 1903, a decade after the “war of the currents” had ended. The decision to kill the animal was made by the park’s owners, not Edison; Topsy had a terrible reputation—in 1902, she killed a spectator, and rumor also credited her with the deaths of two circus workers. Edison Manufacturing sent a cameraman and filmed a short movie—an awful document that survives on YouTube. But to attribute the organization of this execution to Edison as an act of propaganda against AC is to confuse chronology and motives. By 1903, the “war of the currents” had long been over, Edison himself no longer controlled the company that bore his name, and General Electric was already producing alternating-current equipment in earnest.
The “war” ended not in public demonstrations and not in court, but in a bankers’ conference room. In 1892, Edison General Electric merged with Thomson-Houston—a company that, unlike Edison, was already working actively with alternating current. The new corporation took the name General Electric, and the word “Edison” vanished from the title. This was more than a rebrand: Edison lost control of the electrical business he himself had created. He turned to ore mining, cement production, improving the phonograph—but the electrical empire drifted away.
General Electric immediately began manufacturing equipment for AC systems. Niagara Falls, whose power was transmitted to Buffalo by alternating-current lines from 1896, became the showcase of a new era: a large hydroelectric station far from the city, high-voltage transmission, transformer step-down at the point of use. A model impossible with Edison’s direct current. Cities across America and Europe began building centralized power systems, and by the beginning of the 20th century, the question had been settled—at least for the next hundred years.
Suppose Westinghouse went bankrupt earlier, Tesla’s patents found no buyer, and regulators, frightened by Brown’s campaign, banned high-voltage alternating current in cities. What then?
The first and most visible consequence would be the architecture of cities. A power station every one to three miles means a city studded with small generating facilities. Coal yards, smokestacks, access tracks for fuel—in every block. Manhattan’s density could still accommodate this, but the suburbs that sprawled for dozens of miles in the 20th century would have remained without electricity, or required their own stations with their own staff. Rural electrification—the REA (Rural Electrification Administration) program that ran lines to American farms in the 1930s—would simply have been impossible in the DC paradigm: transmitting direct current a hundred miles over wires at low voltage is like carrying water in a sieve.
The second is batteries. In a world where DC won, they would have become a strategic resource much earlier. Lead-acid batteries already existed in the 1880s; Edison himself spent years developing the nickel-iron battery. If a centralized AC grid had never appeared, demand for energy storage to smooth load peaks and back up small stations might have pushed electrochemistry half a century ahead. Perhaps lithium-ion technology, which in our reality became commercial only in 1991 (Sony), would have appeared earlier—though that is speculation, because lithium-ion batteries required materials and knowledge that are difficult to accelerate through market pressure alone.
The third is power electronics. To change direct-current voltage, you need electronic converters: choppers, inverters, converters. In our reality, mercury-arc rectifiers appeared in the 1900s, thyristors in the 1950s, IGBT transistors in the 1980s. In a world of DC victory, the pressure to develop such devices would have been colossal, but the pace at which semiconductors appear is set by solid-state physics. You cannot invent the silicon transistor faster just because it is desperately needed—you need quantum mechanics, pure silicon, a culture of semiconductor laboratories. Roughly speaking, the world might have remained stuck in electrical twilight for those same fifty to seventy years, until electronics matured enough for the task.
The fourth is household appliances. Alternating-current motors—induction motors, based on Tesla’s patents—are cheap and nearly indestructible. They powered the fans, washing machines, refrigerators, and vacuum cleaners of the 20th century. Brushed DC motors require maintenance, spark, and wear out. Household appliances in a DC world would have been more expensive, more temperamental, and probably less widespread. The home refrigerator, which in our reality became a mass-market product by the 1930s, might have been delayed until the postwar period.
The fifth, and most paradoxical, is industry. Factories in the 19th century ran on steam engines with belt drives from a single shaft. The electric motor freed the machine tool from its attachment to the shaft: every machine got its own motor, and workshops were rebuilt around that fact. But in a DC world with a short electrical supply radius, a factory would have had to stand beside its own station. Industrial suburbs stretching along railroads—a characteristic landscape of the 20th century—would have formed differently: more tightly, closer to coal yards and waterways.
The most unexpected detail in this story is that direct current is coming back. HVDC (High Voltage Direct Current) lines transmit energy across thousands of kilometers with lower losses than AC—but they require precisely the semiconductor converters Edison did not have. Solar panels generate direct current. Batteries store direct current. All the electronics inside a laptop, phone, or server run on direct current—the power supply in the wall converts AC from the outlet into DC for the processor. Data centers are seriously discussing a shift to internal DC distribution to eliminate double conversion and save a few percent of energy, which at the scale of Google or Amazon means megawatts.
I like that history has come full circle, but landed at a different point. Edison was right that direct current is convenient for the end user. He was wrong that the world could be electrified without long-distance transmission. Westinghouse and Tesla were right that the transformer solves the problem of scale. They could not have foreseen that a hundred years later, silicon switches would do for DC what the transformer did for AC—and direct current would return, but now at voltages of hundreds of thousands of volts, not Edison’s one hundred and ten.
The “war of the currents” is often told as a parable: evil Edison killed dogs, good Tesla invented the future, justice prevailed. This version is convenient, but inaccurate. The decision was made not by moral arguments, but by accounting.
Westinghouse’s system won because one large station with a turbine is more efficient than twenty small ones with steam engines. Because copper wire can be thinner at high voltage. Because a transformer—a passive device with no moving parts—works for decades without maintenance. Because Tesla’s induction motor is cheaper and more reliable than a commutator DC motor. Because Niagara Falls lies twenty-seven miles from Buffalo, and those miles had to be bridged somehow. Each of these factors alone could have been overcome; together, they formed a system with no DC alternative at the time.
Edison, for all his genius as an inventor and entrepreneur, bet on the component—the light bulb, the generator, the meter—and built a local business around it. Westinghouse bet on infrastructure—and built a network that could grow. The 1892 merger that created General Electric sealed that outcome: the company bearing Edison’s name switched to his rival’s technology within a few years of its founding. New York’s last DC customer—an old building on East 40th Street in Midtown—was disconnected by Con Edison only on November 14, 2007. One hundred and twenty-five years after that September evening on Pearl Street: a stubborn, anachronistic line serving a handful of buildings whose owners had never rewired them.