Fantasies of the end take many different approaches: funny, inventive, ghastly, far-fetched and scarily realistic. It is fertile territory for our imaginations. But if we look more closely at the way we tell our stories, we can see that how we portray the end can also tell us much about how we understand the world and the people around us, not just about how we think about our mortality. They can illustrate our dread of judgement, the importance we place on our societal connections, the darker side of our own human nature. From religious doomsdays and swarms of monsters to biological plague and technological doom, from the winding down of the universe to environmental catastrophe, in these pages we’ll explore not just our fear of death, but more importantly all the things we’re really afraid of in life.
But first I need to persuade you that the title of this introduction is correct: the world is going to end much sooner than you think. But first, like a stock character from a cartoon walking the streets wearing a sandwich board bearing the legend the end is nigh, I need to persuade you that the title of this introduction is correct: the world is going to end much sooner than you think. Humans have predicted the end of the world for thousands of years, yet it is always presented as imminent. If it appears contradictory to suggest that something can be imminent for millennia that’s because it is – imminent means that it’s happening very soon, and there’s nothing immediate about a timespan of tens of centuries. In the words of the Smiths, how soon is now?
So the question is this: is the end of the world millions of years away, or will it arrive momentarily?
According to the discipline of statistical probability, there is a scientific way of weighing up whether the world will end tomorrow or in trillions of years in the future, and the news is not good. I’m talking about the so-called ‘Doomsday argument’, which uses Bayesian probability to assess the odds of whether the world will end sooner or later.*
Probability, of course, is to do not with certainty but likelihood. Roll a die once and you have a one-in-six chance of guessing the number correctly; but roll a die a million times and the probability that each number will appear with one-sixth of throws increases. If you rolled the die a billion times and plotted how often each number appeared on a bar chart, you would have a chart in which all six bars were the same height. Probability is what happens when random instances accumulate to the point where randomness cancels itself out and the underlying pattern is revealed.
Bayes’ theorem is a branch of probability theory named after its inventor, the eighteenth-century English clergyman Thomas Bayes. Here’s how it works. If I asked you the likelihood that it was raining outside, you could go to the window and look; then you could tell me directly. However, let’s say that you’re too far from the window to see whether it’s raining, but you can see that lots of people outside have opened their umbrellas. Though this doesn’t absolutely confirm that it is raining, the alternatives are less likely than the most probable explanation.
That’s a trivial example, but there are plenty of nontrivial ways in which Bayesian reasoning is applied in the real world. For example, our likelihood of getting certain cancers increases under the influence of factors such as age, gender and lifestyle. Knowing this, doctors can use Bayes’ theorem to assess the probability of such cancers and so improve prevention and increase survival rates. In other words, feeding certain kinds of observation into Bayes’ theorem, in a medical context, can literally save lives.
What does this have to do with the end of the world? Well, a group of statistically minded philosophers recently used Bayes’ theorem to calculate the probability that the world is about to end. This was an exercise in probability theory rather than a specific prophecy. The idea was not to point the finger directly at environmental collapse, nuclear war or alien invasion but to establish the larger likelihoods of extinction. They did not feed data concerning the amount of carbon dioxide in the atmosphere or the number of nuclear weapons in the world into Bayes’ theorem; they fed into it only the fact that we are alive now. When they did that, the equation generated an intriguing result: the probability that human beings will become extinct in the relatively near future increased.
It is, of course, rational to be concerned that environmental stability is collapsing, or that nuclear weapons could, whether by accident or design, wreak terrible harm upon our world. But this hypothesis increases the likelihood of human extinction irrespective of other data. To be clear: the end of the world to which this analysis relates (because, of course, there are several different kinds of end of the world) is one in which there are no more humans alive, not in which the planet itself is necessarily destroyed. As humans, this really ought to concern us. Sensible people are right to have specific concerns – for instance, that the thawing of frozen methane in Siberia and the release of that greenhouse gas into the atmosphere makes the imminent end of a human-habitable world more likely. OK. The point of this analysis is: if we factor in Bayes’ theory, we should revise our estimate of that probability, whatever it is, upwards. It’s a strange, even a counter-intuitive, argument, but that doesn’t necessarily mean it’s wrong.
Consider two hypotheses: ‘Doom Soon’, the belief that human history will end in the
