In the year 2023, the Oxford-based Philosopher Toby Ord published a book entitled The Precipice about the possibility of extinction for humans. Ord estimated the likelihood of an “existential catastrophe” for our species over the next 100 years at one out of six.
It’s a significant number and a worrying one. The story made headlines in the early days, and has continued to be influential it was recently mentioned through Australian political figure Andrew Leigh in a speech in Melbourne.
It’s not difficult to agree with the notion that we are facing worrying future scenarios in the next decades, ranging from the threat of climate change, nuclear weapons, bio-engineered pathogens (all important issues to me) as well as insidious AI and huge asteroids (which I’d consider less so).
What is this number? What is its source? And what does it mean?
Coin flips and forecasts for weather
To answer these questions, we must answer another first: What is probability?
The most common approach to probability is known as frequentism. It gets the name of its roots in the game of dice and cards. Based on this perspective it is clear that there’s an one in six chance of a fair die coming up with a triple (for instance) by looking at how often threes occur on many rolls.
Take the more complicated scenario of forecasts for weather. What is the meaning when a meteorologist tells us there’s a one in six (or 17 percent) possibility of raining tomorrow?
Read more: The science of weather forecasting: what it takes and why it’s so hard to get right
It’s hard to believe the weatherperson means us to imagine a large collection of “tomorrows”, of which some proportion will experience precipitation. Instead, we need to look at a large number of such predictions and see what happened after them.
If the forecaster is competent at what they do We should be able to see when they stated “one in six chance of rain tomorrow” It actually rained the next day, once in six.
The traditional probability rely on observations and procedures. In order to calculate it, we have to have a list of repeated events to base on our estimates.
We can take lessons learned from Moon?
What is this saying in terms of the likelihood of the human race dying? This kind of event is likely to be a singular one. That is, after it occurred, there is no chance for repeats.
We could also come across some parallel events we could take lessons from. In fact, in the book of Ord, the author discusses a range of possible extinction incidents, a portion that could be looked at in the light of a historical perspective.
Recognizing craters on the Moon could provide information about the possibility of asteroid impact on Earth. NASA
For instance, we could determine the likelihood of an asteroid that is extinction-sized hitting Earth by studying the number of meteors from space have hit the Moon in its past. A French scientist, Jean-Marc Salotti performed this research in 2022 and calculated the likelihood of a extinction-level event within the next century to be about one out of 300 million.
Of course, this estimate is not without uncertainty, but it is nevertheless supported by something that is close to an accurate frequency calculation. Ord is, in contrast, estimates the chance of extinction caused by an asteroid to be one in a million, but he also acknowledges that there is a significant amount of uncertainty.
A system of ranking results
Another way of thinking about probability that is called Bayesianism in honor of Bayesianism, which is named after the English statistician Thomas Bayes. It is focused less on the actual what happens in the world and more on the things we expect, understand and think about them.
In a very basic way, it is possible to say that Bayesians view probabilities as an element of a ranking system. According to this perspective, the particular number assigned to a probability cannot be taken as a whole. However, it should be compared with other probabilities to determine the outcomes that are more or more likely.