Take two events this week, reported side by side in this paper. One was Kevin Barron's Private Member's Bill to ban tobacco advertising, which gets its second reading in the House of Commons tomorrow. The Government's position is that such advertising should not be banned, but there should be a new voluntary agreement to govern its use.
While there is some dispute as to whether such a ban would be effective in cutting smoking, there is no dispute at all about the life-threatening effects of smoking. Yet smoking among the young is not falling and may even be rising. The risk is simply not accepted by a large minority.
The other event was the attempt by Greenpeace in the High Court to halt the opening of the Thorp nuclear reprocessing plant. Greenpeace is joined by Lancashire County Council and backed by, among others, Sir James Goldsmith. This is not the place to get into the debate about Thorp's economic merits, but it is worth pointing out that in as far as we can accept the view of the experts, the dangers are very small.
The disparity between the perceptions of lay people and experts in the United States was described in a now famous article by Paul Slovic in Science magazine in April 1987. College students and the League of Women Voters ranked nuclear power as the most risky activity or technology; but experts put this way down their risk list, at number 20, far below X-rays at number seven and conventional electric power at nine. By contrast, the experts put motor cars, smoking and drinking in the first three places - all three were a little way down the 'worry list' of the non-experts.
The different way in which ordinary people and experts perceive risk might seem no more than an amusing oddity. But it has an enormous impact on public policy, which, particularly on environmental concerns, is skewed towards cutting risks which articulate middle-class voters think are important. When William Reilly became head of the US Environmental Protection Agency five years ago, he was advised by Senator Daniel Patrick Moynihan: 'Above all, don't allow your agency to be transported by middle-class enthusiasms.'
One frequently hears politicians saying something to the effect that any risk - even one-in-a-million - is unacceptable. Last month, Mr Reilly, now a visiting professor at Stanford University, highlighted the chasm between real risks and perceived ones by citing some things which attract a one-in-a-million chance of causing death (the calculations come from Richard Wilson at Harvard University).
They include: eating 40 tablespoonfuls of peanut butter, drinking Miami water for a year, living for two months with a smoker, living for two months in Denver (because of the extra radiation resulting from its height above sea-level), drinking half a litre of wine, smoking 1.4 cigarettes, spending one hour in a coal mine, flying 1,000 miles by jet, driving 300 miles in a car, riding for 10 minutes on a bicycle, or paddling six minutes in a canoe.
His point was not to alarm or depress; rather it was to show that one-in-a-million is a very remote risk indeed. The chance of being murdered in Washington was 900 in a million each year.
But public policy on both sides of the Atlantic is driven by an odd perception of risk. Take, for example, the removal of asbestos in schools. Bill Reilly points out that the dangers of a child being killed in a football accident are 100 to 2,000 times greater than the risk of death from exposure to asbestos - it is probably much safer to leave asbestos in place rather than disturb it.
Or take the flak which the Government here has received over not meeting fully the new European Union standards on drinking water. Given the enormous cost of eliminating minute quantities of impurities, it might well be more effective to spend the money on reducing other, quite different health risks: perhaps encouraging people to eat more healthy diets, certainly trying to cut smoking.
Or take still another concern: pesticide residues in food. There is widespread anxiety about the use of pesticides, yet 99.99 per cent of the pesticides in our diet are the natural ones the plants have developed to fight off insects or rodents which might eat them. A study by Bruce Ames, professor at the Environmental Health Sciences Center at Berkeley, California, lists 49 pesticides found in cabbage, of which several are known carcinogens.
If you feel you could manage without cabbage, what about coffee? Roasted coffee apparently contains 826 volatile chemicals, of which only 21 have been tested and 16 are rodent carcinogens. Clearly rats should go easy on the cappuccino.
We worry about the effect of diesel fumes in cities, and rightly so because they are very unpleasant. But apparently the amount of potentially cancer-causing mutagenic nitropyrenes that people ingest from diesel exhaust is far less than they get by eating grilled chicken. (And don't have potatoes with the chicken either: these are full of solanine and chaconine which apparently block nerve transmission.)
One can go on having fun at the expense of the worriers but of course there is an enormously important issue here: to what extent are we prepared to cut our living standards in order to reduce risks? Every intervention by government involves a trade- off of some sort. It may be in terms of a restriction on human liberties: banning smoking in public places is a restriction on people's freedom which has to be balanced against the health and safety case for so doing.
Or it may be a financial trade-off. Environmental protection is costing the US about 2 per cent of its GNP, and will reach 3 per cent by the end of the century. This has to be balanced against the benefits that it brings.
These are ultimately political choices. But it is difficult even to have a proper debate about the choices if people are not encouraged to understand the nature and scale of the risks and the costs of reducing these.
A final thought from Bill Reilly: a married male smoker will on average shorten his life by 2,250 days, while an unmarried one will shorten it by 3,500 days. 'One more argument,' he asks, 'for conventional family values?'Reuse content