On this day in 1945, Enola Gay dropped her little boy. Seconds later, the city of Hiroshima was obliterated. Sixty eight years on, the nuclear bombings that marked the end of the war in the Pacific remain the only time such weapons have been used in combat. Did we reach peak destructive force with nuclear arms?

Make no mistake, our bombs got bigger. The largest ever tested, the Tsar Bomb, was 1,400 times more powerful than the combined yield of the bombs that fell on Japan. But it was more effective strategically to develop smaller bombs, faster bombs, more stealthy bombs.

Since then, no weapon system has come close to the sheer devastation that nuclear warheads can unleash. Conventional explosives peaked with bunker busters, especially heavy and hardened missiles created to penetrate and destroy fortified military buildings. (These have their roots in WWII, where they were given the more evocative name of earthquake bombs because the explosion was designed to collapse the ground underneath bunkers without disturbing the surface.)

More powerful weapons have been imagined, if not built. German rocket scientist Hermann Oberth drew up calculations for a giant space-bound parabolic mirror that would focus the sun’s rays onto the Earth’s surface, producing a concentrated beam that could incinerate entire cities. More pragmatic is to simply drop inert lumps of metal onto your enemy from an orbital platform. Kinetic bombardment — so-called “rods from God” — was conceptualised in the 1960s by Boeing employee Jerry Pournelle before he decided he had a better future in writing science fiction. With a destructive yield on par with a small nuclear bomb, and no messy fallout, we’d probably have them if the Outer Space Treaty didn’t forbid placing superweapons in orbit.

The answer to why we stopped building superweapons probably lies in the fact that nuclear bombs were so destructive we had to invent an entirely new military strategy in which to contain them: Mutually Assured Destruction. It’s a frightening concept (especially when the criteria fail, which I’ll get to in a minute), but it is nevertheless an arguably stable one. To date, no nuclear-armed nation has gone to war with another. It’s perhaps one of the few times the Prisoner’s Dilemma actually works, because you must work in absolutes. As Wikipedia succinctly puts it: “neither side, once armed, has any rational incentive either to initiate a conflict or to disarm”. And once you have a nuclear bomb, there’s really no need to develop another superweapon. We don’t talk much about MAD as a global political structure these days, even though we live on peace brokered by it. It’s become sort of invisible by its ubiquity.

However, there are exceptions to the rule. The balance of MAD can be destabilised by a weapon that prevents a retaliatory strike. If the US could launch a secret attack, or shoot down incoming Soviet missiles, it gains a first strike advantage that encourages the Soviet forces to launch first. For this reason, Reagan’s “Star Wars” missile defence shield was sharply criticised as just such a destabilising technology. It upset the essential balance of MAD by giving one player an advantage over another.

Another constraint of MAD is that it considers rational state actors who fear for the survival of their own citizens. There are no such restrictions placed on fanatical leaders or radical individuals. While we’re fighting against vague concepts such as “drugs” and “terror”, our bullets and bombs are really aimed at small pockets of individuals who can wreak relatively large amounts of damage. Asymmetric warfare, pitting states against small bands of insurgents, cells, lone gunmen, is the dominant paradigm of the day, and the basis on which we build huge surveillance systems. Attach the gunsights of PRISM to the bomb bays of drones and you could argue the whole thing is a superweapon of sorts, something that sees all, knows all, and will blow you into pieces before you have a chance to download those ricin-making instructions.

Here’s a final esoteric thought. Another constraint of MAD is that it can be dissolved by survivability. With a bunker deep enough, and food supply large enough, destruction is no longer assured. But to look at it another way, we might one day reach a level of post-humanism that our physical bodies are no longer necessary to live. As intelligences distributed and backed up across multiple platforms, be them biological, mechanical or something else, we need not fear annihilation from a point source weapon. We might then, finally, be freed from the company of Fat Men and Little Boys.