You've nothing to fear but fear itself

Reckoning with Risk: learning to live with uncertainty

Gerd Gigerenzer <em>Allen Lane, The Penguin

Good front cover: man balancing on precarious pile of furniture atop New York skyscraper, circa 1930. Great subtitle: "learning to live with uncertainty", which is something we all need to do. For a moment, I thought this book might offer insight into our unhealthy obsession with potential risks to our health and welfare. But, in essence, it is nothing more than another psychobabbling self-help manual, designed for those of us with low self-esteem when it comes to statistics.

Gerd Gigerenzer is good on how the presentation of statistics can distort our perception of risk and increase unnecessary anxieties about everything from breast cancer to "wife-battering". He is particularly concerned about the tendency for the authorities to present risk in terms of percentages and probabilities rather than "natural frequencies" (simple counts of events). For example, he criticises the conventional wisdom that "one in ten women develops breast cancer". This figure is a distortion, he writes, because it "refers to a woman's cumulative chances of developing breast cancer by the age of 85. But most women die before then, and those who contract cancer at this advanced age will most likely die of some other cause."

Gigerenzer is concerned about how we are bombarded with warnings based on relative rather than absolute risks. He cites the case of Britain's Pill scare of the 1990s, when health officials announced that certain types of oral contraceptive were "associated with around a twofold increase in the risk of thromboembolism" (blockage of blood vessels by clots). In response to being told that their relative risk had doubled, many women stopped taking the Pill, leading to an increase in unwanted pregnancies and abortions. Yet, as Gigerenzer points out, if the same information had been presented in terms of absolute risk, the more reassuring news would have been that "the chance of thromboembolism increases from about 1 to 2 in 14,000 women".

Better education and clearer presentation of evidence are obviously worthy aims. But these insights are really just a variation on Disraeli's warning about there being three kinds of lies - "lies, damned lies and statistics". Such a timeless generalisation does little to explain why public concern about all kinds of risks has intensified in recent years. Problems with statistics cannot tell us why there now seems to be a major new panic every month, focusing on anything from mobile phones to cooked food. If the problem is simply statistical ignorance and misunderstanding, why are we so much more prone to believe the worst these days?

Gigerenzer suggests that we do not understand percentage analysis because, as a result of our long evolutionary history, "our minds are adapted to natural frequencies", in the same way that "rats are able to 'count' up to about 16". Similarly, he thinks that "John Q Public" (aka Mr Ordinary) mostly fears those things that "were dangerous in our evolutionary past - such as snakes, spiders, large cats, darkness . . ."

The risk-averse, cautious society in which we live is shaped by powerful cultural forces, not statistical fallacies. A free-floating predisposition to panic, moving from one issue to another, was discernible long before the terror of 11 September. That is why, even when there is no hard evidence of any significant risk, and all of the known facts and statistics are explained, many people simply refuse to accept that anything is safe today. See, for example, the never-ending, yet entirely irrational, scare about the link between the MMR triple vaccine and autism.

In the end, Gigerenzer's mind-tools cannot ease our anxieties. There is no one, simple technical or statistical solution to our contemporary culture of fear. He refers to the problem of "the illusion of certainty" in society, an illusion marketed by insurance companies and the medical industry. But if that were ever true, it isn't now. If anything, our age is dominated by an exaggerated sense of uncertainty and by an inflated fear of the unknown. When people are continually given the signal (by politicians, scientists and doctors, among others) that uncertainty is a problem, and that society does not know how to deal with it, it should be little wonder that we recoil from the idea of risk.

Mick Hume is editor of