I’ve been away for four weeks, so I’ve only just heard about the horrible events in Wales, where a five-year-old girl, April Jones, is missing and presumed dead; a man has been charged with her murder.
But it is horrible, and for parents, no doubt, utterly dismaying. How can we allow our children out on the streets on their own, when monsters like these are about? Cristina Odone reports this morning that her nine-year-old daughter is too scared to walk the 200 yards to the bakery on her own now: “What if someone takes me?”, the little girl asked.
Now, Cristina is unsure what to do. “I worry about how to proceed,” she says. “We want to shield the little ones from paranoia, and are mindful that mollycoddling them will turn our offspring into gormless cowards.” She knows that abductions are “extremely rare”. But against the “frightening soundtrack” of paranoia, parents are less willing to let their children out on their own.
I’d like to suggest something, to Cristina and to all other parents who are – understandably – worried about this sort of thing. In fact I’d like to suggest it to everyone who has ever worried about shark attacks, about terrorist attacks, about train crashes and plane crashes and Ebola and all the other high-profile, dramatic risks that occupy disproportionate amount of column inches and airtime. It is to learn about a faintly obscure finding of psychological research carried out in the 1970s, which labours under the unglamorous title of “the availability heuristic”. I honestly believe that teaching parents about it would make the risks of modern society much easier for them to bear.
Two psychologists, Amos Tversky and Daniel Kahneman, found that when we are asked a difficult question, our brain frequently substitutes the answer to an easier one. The availability heuristic is a rule of thumb that our minds use, which, when asked the question “How common is Event A?” – a difficult question, which relies on a lot of knowledge – slips in the answer to an easier question: “How easily can I think of an example of it?”
It’s actually quite a useful rule of thumb, most of the time. How frequently does it rain in London? I don’t know the days-per-year statistic, but I can easily think of lots of rainy days this summer, so I’m happy to answer “Quite a lot”. How often do I get pay rises? I can’t think of any examples, so I’m happy to answer “not very often”.
But for the assessment of risk in a modern society, it’s dreadful. It probably worked quite well in prehistoric times: how easily can you think of a someone getting eaten by a sabre-tooth cat? Quite easily, because even though it only happened the once, to Urg in Cave 4B, it was pretty dramatic. And because people were living in tribes of perhaps 150 members, dramatic, unpleasant things that happened would tend to be things that were likely enough to be worth avoiding.
Today, it’s very different. Today, if a terrible thing happens to one person in a society of 60 million – or even in a global society of seven billion – it can be disseminated rapidly to everyone in that society. Dramatic events make better pictures; rare events are more newsworthy. So our newspapers and our television are full of dramatic, rare events. And dramatic events are more memorable. So when you ask yourself the question “how likely is it that my child will be the victim of an abduction?”, a question that requires significant statistical information, you are likely to substitute the answer to the question “how easily can I think of other children being abducted?”, and the answer, of course, is “very easily”. So we naturally think that it’s not safe to let our children out, because of paedophiles.
But that’s completely false. According to the NSPCC, an average of 56 children are killed “at the hands of another person” in England and Wales each year. In two-thirds of those, the chief suspect is a parent. It’s hard to tease out numbers, but if the 2007-2008 Ofsted statistics are comparable, at least another 10 or so will be killed by other young people. That leaves about eight child killings a year in an unspecified “other” category, which will include various things, including abduction-murders. Judith Levine, the author of Harmful to Minors, puts the risk of a child being murdered by a non-family member in the US at between 1 in 364,000 and less than 1 in 1,000,000. “A child’s risk of dying in a car accident is twenty-five to seventy-five times greater,” she says. Paedophile killings are even rarer, in single digits annually even in the US: a likelihood of one in millions.
None of this is to take away from the utter horror of what has happened in Wales. I can’t, for even a second, imagine what it must be like for the family. But it’s important to remember that these horrors are rare. Dramatic, rare events are always going to be more newsworthy than duller, more common ones; they’re also going to be more memorable. It’s silly to want our news services to give a rolling list of statistically likely causes of death (“Today, diabetes moves up a place to third, while liver cirrhosis is steady at 62”); no one would buy them or watch them. But for parents, and others, who are concerned about the risks that we face in modern society, it would be helpful to know why it is they’re disproportionately frightened of terror and murder, instead of the real killers, like disease and accidents. The availability heuristic is the treacherous tool we use to determine risk, and it’s just not very good. Not realising that condemns parents, and people in general, to live in unnecessary fear. I seriously think it should be taught in schools.