At first the terrifying prospect seemed plausible. Detonate an atom bomb in water, scientists had already realised, and you can trigger an even bigger explosion — a technique would later become the hydrogen bomb. But why, Edward Teller wondered, would water be needed? What would happen if you used air instead?
The concept was simple. When an atom bomb explodes in a tank of purified heavy water, the result is an explosive chain reaction. The energy of the bomb forces hydrogen atoms in the water to collide; each collision releasing a powerful pulse of energy. Each pulse adds a bit more energy to the explosion, until a nuclear inferno erupts.
That was fine, the scientists thought, and perfectly controllable, as long as your aim was to blow up a city. But Teller took things a step further. If heavy water could detonate a hydrogen bomb; why not sea water? The sea is full of hydrogen too, and thus susceptible to the same chain reaction. Unlike the hydrogen bomb, however, the sea is vast in size, and the potential chain reaction uncontrollable.
Could, then, detonating an atom bomb in the sea trigger a chain reaction that would swallow the planet in a nuclear inferno? That idea was bad enough, but things could get worse. What if you detonated the bomb in air, as the scientists at Los Alamos planned to do in six months time?
Air doesn’t contain much hydrogen, but it does contain nitrogen, an atom that can undergo the same type of explosive chain reaction. When they set off the first atom bomb, Teller wondered, might they also ignite a nitrogen chain reaction throughout the atmosphere? Would the bomb destroy humanity and turn Earth into a miniature star?
The prospect was terrifying enough to make its way up the chain of command, attracting the attention of senior figures in the Manhattan Project. Scientists were ordered to study the risk. A wonderfully titled paper, “Ignition of the Atmosphere With Nuclear Bombs” was written, deeming the chances irrelevant.
The idea of burning the atmosphere, the authors concluded, was impossible. The nitrogen in the air was too spread out, the heat of the bomb too low. Six months later the Trinity test went ahead, as planned. The explosion in air was controlled. The atmosphere did not catch fire, and the Earth was not incinerated.
Fears that scientists might accidentally destroy the world have lasted beyond the nuclear bomb. Today few worry that we might accidentally set the atmosphere on fire. Instead they fear black holes produced by powerful super-colliders or new forms of strange matter that consume the planet.
Though many scientists consider these ideas ridiculous, are they really? What level of risk is acceptable when you might accidentally destroy the world? And though physicists may reassure us their experiments are safe, isn’t the whole point of them to discover something unknown, something potentially dangerous?
Part of the fear comes from the feeling that scientists are doing something unnatural. Atom bombs and particle colliders seem far from our everyday lives — and indeed they are. These are experiments probing extreme physics — but not unnatural physics.
An exploding nuclear bomb, even the most powerful ever made, has a fraction of the energy of the Sun. Particle colliders reach high energies, true, but even more energetic particles hit our planet every day from deep space. Nothing bad ever happens when they do — no black holes or strange matter is created by them.
This argument is the strongest weapon in the physicist’s toolbox. There is nothing we can do on Earth, at least for now, that is not already happening somewhere close by. The only difference is that it is happening in a controlled environment, one where we can watch closely and track what is happening.
Even if some unknown physics takes place and creates some bizarre particles, we can confidently say that they will not destroy the planet. Of course, this approach cannot be used for every situation. The example of the Sun — a self sustaining nuclear bomb — is hardly reassuring when trying to argue the same won’t happen on Earth.
For these situations scientists need a different approach. The risk of catastrophe must be judged based on known physics and on what we can say about unknown physics.
In the case of the atom bomb, the physicists already had reasonably good ideas about nitrogen-based nuclear reactions. From these ideas they could calculate what might happen after the bomb exploded. The equations indicated that nothing troubling would come to pass. Any nuclear reaction in the air that did happen would quickly fade away. No chain reaction could be sustained.
But what about the unknowns? Physicists cannot be so arrogant as to claim they know every possible outcome. There was a chance, slim though it may have been, that some unexpected nuclear reaction could occur. In this situation, the best the scientists could do was come up with a probability of something terrible happening.
This was done based on what we already know about physics, and what we can see in the world around us. For instance, we know that in four billion years of history, the Earth’s atmosphere has never once exploded in a nuclear inferno. This might seem obvious, but it tells us something important about the stability of the atmosphere.
We also know the properties of nitrogen atoms, and other atmospheric gases. We can draw upon centuries of past experimentation, to set limits on what is possible. Ultimately, in the case of the atom bomb, the scientists concluded that there was less than one chance in three million that the atmosphere would be incinerated. That, facing the threat of Nazi domination, was good enough for the managers of the Manhattan project.
Should we accept a similar level of risk for modern physics experiments? One could argue that the benefits of the atom bomb made a higher risk than normal acceptable. Particle physics experiments, though they may bring breakthrough scientific advances, are unlikely to bring the same kind of benefits.
One extreme stance is to say there should be no risk at all. After all, destroying the planet would be catastrophic. Not only would it wipe out every known living thing in the Universe, it would erase our cultural history and prevent the lives of thousands of unborn generations.
But this stance is akin to saying we should never do anything. We cannot eliminate all risks. We all go about our lives knowing there is a small chance of death each and every day. We know, too, that the world may end even without our intervention. An asteroid could strike the Earth, or a nearby supernova could suddenly erupt, blasting the planet with radiation.
These things are very unlikely, but they are not impossible. One stance of scientists has been to say experiments should have a lower risk than this. The odds of a civilization killing asteroid — roughly one in a hundred million — are sometimes set as the threshold.
At this rate science experiments would not do much to alter the background level of risk we are forced to accept in everyday life. Arguments can be made to go lower — shouldn’t we try to minimise the risk as much as possible? — or to accept higher — the odds are already very low — but in the end the risk is one for society to judge.
These kinds of discussion are useful to have. They force scientists to talk about what might go wrong and to justify why we should take the risk at all. Governments and scientific organisations need to properly assess their experiments, both in terms of the benefits and the risks.
Doing so transparently doesn’t weaken science — it strengthens it, and builds confidence in what is being done. The risk is not zero — it never can be — but neither are the experiments without benefit. In the long run, breakthroughs in fundamental science can improve life for everyone — as the wonders of the computer age have proven.
Ultimately, perhaps all these calculations miss the real risks. The threat of the nuclear bomb turned out not to be the risk of incinerating the atmosphere, but rather the chance we would use them to destroy the world. The odds of that seem much higher than one in three million.
So much reporting around health, science and space exploration is unrealistic, hyperbolic and misleading. These are complicated topics, and there are often no easy or straight forward answers. Instead what is needed is analysis, discussion and an exploration of the possible ways forward.
Follow me and subscribe to my free Substack newsletter to get the latest. Articles are published regularly, and by signing up you will never miss an update.