On 16 July 1945, a device called "the Gadget" detonated in the New Mexico desert and turned the sand beneath it into a sheet of radioactive glass. The flash was visible from 250 kilometres away. J. Robert Oppenheimer quoted Hindu scripture. The world changed.
But what if it hadn't?
What if the Manhattan Project failed, or never started, or the physics simply didn't cooperate? What if nuclear fission remained a laboratory curiosity, useful for generating electricity but never weaponised? The 20th century plays out very differently. And not necessarily more peacefully.
Why it might not have happened
This isn't as far-fetched as it sounds. The Manhattan Project cost $2 billion in 1945 dollars (roughly $35 billion today) and employed over 125,000 people. It required the United States to build entire secret cities from scratch. Oak Ridge, Tennessee was consuming one-seventh of all the electrical power produced in the country just to enrich uranium. The whole thing was a staggering bet on unproven physics.
Several scientists argued it couldn't work. Others thought it could work but shouldn't. Leo Szilard, the man who first conceived of the nuclear chain reaction, spent the final months of the war circulating a petition to stop the bomb being used. If Szilard had been more persuasive five years earlier, or if Enrico Fermi's Chicago pile had failed to go critical in December 1942, the entire programme could have stalled.
So imagine it does. The war in the Pacific ends differently. And then everything after that ends differently too.
The Pacific War drags on
Without the atomic bombings of Hiroshima and Nagasaki, the most likely scenario is Operation Downfall: the planned Allied invasion of the Japanese home islands. The first phase, Olympic, was scheduled for November 1945, targeting the southern island of Kyushu with a landing force of 766,000 troops.
American military planners estimated between 250,000 and one million Allied casualties. Japanese casualties would have been far higher. The Japanese military had been training civilians, including teenagers, to fight with sharpened bamboo spears. They were not planning to surrender.
The invasion might have worked. It almost certainly would have, eventually. But "eventually" means months or years of grinding island warfare scaled up to a continental level. The firebombing of Japanese cities, already devastating, would have intensified. Soviet forces, which invaded Manchuria on 9 August 1945, would have pushed further into Japanese-held territory, potentially occupying Hokkaido.
Japan might have ended up partitioned, like Germany and Korea. A Communist north, an American-backed south. A DMZ somewhere around Sendai. Tokyo split into sectors.
That alone rewrites the next eighty years.
The Cold War without the bomb
Here's where it gets properly strange. The entire architecture of Cold War geopolitics rested on nuclear deterrence. Mutually Assured Destruction. The idea that any direct war between superpowers would end civilisation, and therefore nobody would start one.
Without nuclear weapons, that constraint vanishes.
The United States and the Soviet Union would still be rivals. The ideological conflict between capitalism and communism doesn't go away because the bomb doesn't exist. But without the existential risk of nuclear annihilation, the calculus for conventional war changes completely. Wars between major powers become thinkable again.
Korea might not stay a proxy war. Vietnam might escalate into something broader. The Berlin Crisis of 1961 might not end with a wall. It might end with tanks.
The optimistic reading is that without nuclear weapons, we'd avoid the constant low-level terror of potential extinction. No Cuban Missile Crisis. No children practising duck-and-cover drills under school desks. No Doomsday Clock.
The pessimistic reading is that nuclear deterrence actually prevented World War III, and without it, the second half of the 20th century becomes a bloodbath of conventional warfare between industrial superpowers. The great powers stopped fighting each other directly after 1945 for the first time in centuries. The bomb is the most obvious explanation for why.
Conventional deterrence fills the gap
Something would replace nuclear deterrence. Probably chemical and biological weapons, which were already well-developed by 1945. Both sides had stockpiles. Without the bomb monopolising the "existential threat" category, nerve agents and engineered pathogens might have become the backbone of strategic deterrence instead.
That thought is not comforting.
Chemical weapons are harder to contain, harder to detect, and much easier to manufacture than nuclear warheads. A world where sarin gas or weaponised anthrax serves the same deterrent role as a hydrogen bomb is a world where the barrier to catastrophic warfare is lower, not higher. The whole point of nuclear deterrence was that the consequences were so obviously total that nobody could miscalculate. Chemical deterrence doesn't have that same clarity.
Or maybe the arms race just goes conventional. Bigger armies, more tanks, faster jets. The military budgets of the Cold War era were already enormous. Without billions diverted to nuclear weapons programmes, those budgets might have produced even larger conventional forces. Europe becomes even more heavily militarised than it was.
The civilian side effects
Nuclear weapons research produced nuclear power. Without the bomb, nuclear energy still exists in theory, but the massive government funding that accelerated reactor development wouldn't be there. The first commercial nuclear power plant, Calder Hall in Cumbria, opened in 1956. It was directly derived from plutonium production reactors built for Britain's nuclear weapons programme.
Without the weapons programme, commercial nuclear power might arrive a decade or two later. Maybe three. That means more coal and oil in the interim, which means different patterns of industrialisation, different energy politics, and possibly faster climate change.
The space race also changes. ICBMs were developed to carry nuclear warheads across continents. The rockets that put astronauts in orbit were modified ballistic missiles. Without the military pressure to build intercontinental delivery systems, rocket development proceeds more slowly. Sputnik might not launch in 1957. The moon landing might not happen in 1969. Or at all.
Your smartphone exists partly because the Cold War nuclear arms race drove investment in miniaturised electronics, satellite communications, and computing. Remove the bomb, and you don't remove the eventual development of these technologies, but you change their timeline by decades.
The moral question
There's a version of this scenario that feels like a straightforward improvement. No Hiroshima. No Nagasaki. No 200,000 civilians killed by two bombs in three days. No decades of nuclear testing that poisoned Pacific islanders and downwind communities in Nevada. No Chernobyl, no Fukushima (if commercial nuclear power is delayed significantly). No existential dread hanging over every generation since 1945.
But there's also a version where the absence of the bomb leads to more total death, not less. Where conventional World War III kills tens of millions. Where the great powers never develop the mutual fear that kept them from direct conflict for seventy years.
The uncomfortable arithmetic is that nuclear weapons may have killed hundreds of thousands of people while preventing the deaths of tens of millions. That's not a comfortable moral position. It's not meant to be. It's the kind of calculation that the 20th century specialised in.
Oppenheimer himself said he had become "death, the destroyer of worlds." What he didn't say, and what no one could have known in July 1945, was that the thing he built might also have been the reason the world survived long enough to argue about it.