Читать книгу Global Warming and Other Bollocks - Stanley Feldman - Страница 11

Illnesses caused by exposure to noxious substances, such as arsenic, have long been recognised. Those that take many years of exposure before the illness becomes apparent are much more difficult to identify.

Richard Doll, the 20th-century British epidemiologist, demonstrated a relationship between cigarette smoking and the risk of developing cancer of the lung. He suggested that smoking caused cancer of the lung after he had surveyed people’s smoking habits and found very little effect in those smoking fewer than five cigarettes a day, whereas above this level the frequency with which they developed cancer increased with the number of cigarettes they smoked. This dose-related phenomenon added considerable credibility to his findings. When he further demonstrated that people – doctors, it so happened – who voluntarily stopped smoking reduced their risk of cancer dramatically, he had sufficient evidence to convince all but the most sceptical of a causal rather than chance association between the two. Without the supporting evidence the relationship between smoking and cancer of the lung would have remained a tentative hypothesis.

Failure to seek evidential confirmation can produce half-baked theories, often presented in the media as ‘new scientific evidence’, that can be very misleading.

Some years ago there was an epidemiological study that showed that the residents of the island of Okinawa, in the Pacific, lived longer on average than those on neighbouring islands. It was claimed that this was due to their high consumption of yams. The researchers clearly demonstrated that they did eat more yams and they did live longer. To have turned this hypothesis into proof they would have needed to show that the lives of the residents of Okinawa were shortened if they gave up eating yams and that by eating yams those living on nearby islands increased their life expectancy. They did not do this and the hypothesis remains unproven. It is highly probable that the conclusions are wrong and that the observations are better explained by the high proportion of the inhabitants of Okinawa who are of Japanese descent. Japanese have, in general, a longer expectation of life than almost all the other people living in the Pacific.

Just as important as producing evidence to support a particular theory is the ability to demonstrate that any alternative explanation is wrong. It was this concept that led the distinguished 20th-century philosopher Karl Popper to propose that science advances by disproof. He pointed out that, if you made the hypothesis that ‘all swans are white’, it did not matter how many white swans you found, or how much evidence you gathered to support your theory, as soon as a black swan was found, the theory becomes untenable. However, it is always possible that the black swan was a white one that had been painted black or an odd genetic freak.

In order to demonstrate the validity of the disproof, it would be necessary to show that it was a swan and that it was naturally black and bred other black swans. It is just as important to support any disproof with evidence as it is for proof. Popper pointed out that disproof is much more potent as an evidential weapon than proof. One element of disproof will dispel ideas supported by volumes of positive evidence. A tablecloth with a dirty stain in one corner is ‘dirty’ regardless of the fact that 99.9% of it is clean!