In a desperate attempt to explain why Cern's Large Hadron Collider has suffered a series of mishaps preventing it from commencing its search for the elusive Higgs Boson particle, respectable physicists have suggested (apparently in all seriousness) that nature abhors the Higgs so much that ripples from the future are travelling back in time to stop the Switzerland-based particle accelerator working.

Reports of the emergence of these theories have prompted renewed contemplation of the "granny paradox", which some think debunks the very idea of time travel. In this scenario, a time traveller goes into the past and inadvertently causes the death of his/her granny, before the traveller's parents are born. So the traveller never goes back in time, so granny doesn't die – and, well, so on. I have a much simpler explanation for the collider's plight. Its failure is related to the existence of other universes, the "parallel worlds" beloved of science-fiction writers.

This theory suggests there are many – perhaps infinitely many – universes, some more or less like our own, some very different. This is not an idea confined to science fiction; it is respectable scientific speculation. Such universes are thought to exist in their own sets of space and time dimensions, and include worlds where key turning points in history, such as the Battle of Hastings, turned out differently from the way things happened in our world. The physicist Hugh Everett proved half a century ago that this "many worlds" idea is completely compatible with everything we know about the way the world works, and is a natural feature of quantum physics.

In the classic "thought experiment" to demonstrate this, a moggy, known as Schrödinger's cat, is either killed or not killed by what the physicist Erwin Schrödinger called a "diabolical device" operating on quantum principles. After the "experiment" (I should stress that nobody has ever actually subjected a cat to this indignity), according to the quantum rules the universe divides so that there is one universe with a dead cat and one with a live cat. Extrapolating this to cover every event that has ever happened in the universe implies that there are many universes in which experiments equivalent to the one at Cern are being attempted. But there is a problem with such experiments. When the Large Hadron Collider was planned, some scientists speculated that it might destroy the universe we live in. This would happen if the empty space that surrounds us is in a state called the false vacuum.

The best analogy to the false vacuum is a large, placid lake of water, behind a dam, high in the mountains. Everything is calm and peaceful – but if the dam breaks, the lake disappears as water rushes to a lower level. Conceivably, if the universe is in a false vacuum state, a collider such as Cern's could punch a hole in the fabric of space, like a hole in the dam, allowing the entire universe to fall out of the false vacuum and settle at a lower level.

We would never know if this happened, because the entire universe as we know it would disappear in a split second. But perhaps this has happened – not once, but many times, in the universes next door. If the universe – a universe – can be destroyed by the successful activation of a particle accelerator such as the Large Hadron Collider, the only universes that survive will be the ones in which a series of freak accidents prevent the collider from working. And that is why we are still here to puzzle over the repeated failure of the LHC. Our cousins next door have not been so lucky.