As physicist Niels Bohr said (and it wasn’t original then), ‘Prediction is very difficult, especially about the future.’ The difficulty of foreseeing what is to come is not always a problem for science. We can, for example, predict exactly when Halley’s Comet will return and sweep around the Sun on its 75-year orbit (the next time it will be sighted from the ground will be in 2061), but we are incapable of forecasting the weather more than a few days ahead.

Knowing how the weather is going to develop is incredibly important, but how can we know the future when dealing with something as intangible as this? In principle it seems it should be feasible. Ever since Newton gave us a mechanical picture of the universe, it has seemed possible that predictions of the future could become near-perfect. The eighteenth-century French academic Pierre-Simon Marquis de Laplace noted that because of the apparently clockwork-like nature of everything, if a hyper-intelligent being could have access to all the data in existence ‘nothing would be uncertain and future, as the past, would be present to its eyes’.

First attempts

Until recent times, even the vaguest possibility of Laplace’s vision has been inconceivable for anything other than a deity. Yet forecasting how the weather will change is a challenge that has been faced for as long as human beings have thought about nature and how it works. Before a scientific approach was possible, predicting the weather was largely left to the priest or the shaman. At first sight, what they did appears to be not so much a prediction as an attempt at weather control. The idea was to intercede with the gods who were thought to inflict the weather on the people of the Earth as a reward or punishment for their actions.

In practice, though, a savvy priest could indeed be capable of spotting signs that helped predict the weather. If he or she had a fair idea of what was coming, they could then claim to be asking the gods to bring bad weather as a punishment or good weather as a reward. If they got the prediction right, they would appear to wield enormous power, keeping the priesthood safe in their position of esoteric knowledge.

This supernatural view of the weather as something caused by gods, spirits or magic would last well into the medieval period, when witches were sometimes blamed for bringing on a patch of bad weather that damaged an opponent’s crops (something that sadly can still happen in parts of the world). As with many topics, one of the first to think scientifically about the weather was the Greek philosopher Aristotle, who lived between 384 and 322 BCE. We now know that Aristotle was often wrong about science – but in a sense that doesn’t matter.

The problem with the proto-science of Aristotle’s time is that it worked more like the law does today than a true science. Instead of relying on experiment and careful observation to decide the most likely scientific explanation for a phenomenon like the weather, different theories would be debated, and whichever idea came out best in the debate would be accepted as the truth, even if it bore no resemblance to reality. Infamously, Aristotle opined that women had fewer teeth than men. He never bothered to check and people took his word for it.

Aristotle’s scientific legacy was not so much in the accuracy of his theories, as his ability to set the agenda. He covered a vast range of subjects that science would come back to again and again. Some of the topics were abstract, such as the nature of infinity, but others were very pra