"I wish I could be half as sure of anything as some people are of everything," said the American humorist Gerald Barzan, which neatly summarises the motivation for this uneven but stimulating book. Dylan Evans thinks rightly that most of us are bad at thinking about, let alone managing, risk. The way he puts it is that different people display different degrees of "risk intelligence", defined as "the ability to estimate probabilities accurately". At the heart of risk intelligence is the ability to gauge the limits of one's own knowledge "to be cautious when you don't know much, and to be confident when, by contrast, you know a lot". And to have a rational strategy for obtaining relevant knowledge when you don't know much.
Most of us aren't good at this. Consider an everyday scenario: you're buying a big flat-screen TV. The salesperson asks if you are interested in purchasing a three-year extended warranty. Most of us hum and haw and wonder what commission the salesperson is getting, whereas in fact we should be considering the price of the TV, the cost of the warranty and the probability that the TV will fail within four years all data available to anyone prepared to do a spot of Googling. It's possible that the extended warranty is actually a good deal but most of those who buy it will do so mainly on the basis of a vague hunch that it makes sense.
And it's not just lay people who are bad at assessing probabilities. Medical students aren't too hot at it either. Evans cites a question he's posed to hundreds of them. Suppose you are tested for HIV and the test comes back positive. What is the chance that you have the condition? Most people think they must be pretty high. Only a few reply, correctly, that you need more information to work out a sensible answer. You need to know how good the test is, and how many people like you have the condition in question. And you need to know about Bayes's theorem, which calculates how much stronger (or weaker) beliefs should become when new information is taken into account. (Evans's answer to the question is 4%, by the way.)
Many years ago the psychologist Howard Gardner attacked the absurdity of measuring human intelligence on a single dimension. He identified eight different kinds of intelligence. Later, another psychologist, Daniel Goleman, added a ninth: emotional intelligence. Evans argues that we should now add a 10th: risk intelligence.
Low risk intelligence is an endemic problem in most societies. Signs of it are everywhere: in the "paranoid parenting" that prevents us from allowing our children to live normal lives; in the irrational fear of crime that blights the lives of Daily Mail readers; in the inability of our media to communicate accurately about uncertainty, which they interpret as "increased certainty that things are going to get worse"; in the worst-case scenarios that lead to the security theatre on display at airports; and in the general need for "closure" about troubling events that leads us to think any answer is preferable to none.
What's worse is that our society is irredeemably favourable to low risk intelligence. People who are confident and articulate tend to be admired while those who are more probabilistically informed are devalued. And this matters because our inability to handle risk leads to hysterical debates and makes rational policy-making on nuclear power, policing, immigration, terrorism or GM crops impossible. In relation to risk, stupidity seems to be humanity's default option.
