Don't Be Gullible

Home Page

My Latest Articles

Buying the Book

Reviews by readers

Reviews by critics

More about the book

About me

Email me

This may be the most important chapter in the book. We are surrounded by tales of imminent food-borne disaster, of major risks from chemicals present in minute concentrations. Many of these claims are purportedly backed by solid statistical evidence. Only by learning how to reinterpret statistics in a commonsense way can we avoid daily hysteria.

There are several kinds of fallacies, mostly used to persuade us to part with our money or our votes. They include

    Confusing sensitive assays with inherent poisoning potential

    Confusing unitary (all-or-nothing) with distributed risk

    Data-mining, using sloppy statistical criteria     Ridiculous extrapolations from limited data

Risks that are spread around versus all-or-nothing risks

Most toxicology investigations don’t distinguish between risks that involve a certain amount of damage done to everyone, versus those that affect a few people but leave the rest unaltered. Here is an example. In the United States, government estimates are that about one egg in ten thousand is infected with salmonella. Imagine ten thousand eggnog-swilling Yanks. On average, one person will be exposed to salmonella, but the remaining 9,999 will not. There is a one-in-ten-thousand risk of disease from this source, but it is an all-or-nothing risk.

Next, imagine that some psychopath has put a lethal dose of poison into a vat of orange juice. These same ten thousand (imaginary) people drink a glass each of orange juice, thereby receiving one ten-thousandth of a lethal dose of poison. Is it conceivable that one person will die, and the rest will be unaffected? That’s nonsense. No mysterious mechanism is going to concentrate all the poison from the hundreds of litres of juice so that it all goes to one unfortunate victim.

Instead, everyone will ingest a tiny fraction of a lethal dose. At the very worst, each person will become one ten-thousandth less healthy. That’s an impossibly small figure, neither measurable nor testable. In fact, common sense and daily experience tells us that minute doses of any poison have no damaging health effect whatsoever. Most of the plant foods we eat have comparable low levels of ‘poisons’, yet vegetables make us healthier, not sicker.