Causality Defies Simple Explanations

News of the terrible bombing in Oklahoma City has traveled around the world and with it has journeyed this startling revelation: common farm chemicals were used to make the explosive.

Combine fuel oil, ammonium nitrate fertilizer and a few other components and you've made a bomb, or at least that's the message being broadcast. Every farm is a potential weapons plant. The ingredients for disaster lurk in every farmer's shed.

The scientific truth about fertilizer-grade ammonium nitrate is that making a bomb from the material requires a level of technical know-how that few people possess. The proportions of the ingredients must be combined in exact amounts under controlled conditions and triggered in some fashion to cause an explosion. The chances of a lunatic acquiring this knowledge, having the means by which to apply it, and being disturbed enough to bomb a public building are astronomical.

And just because one Michigan farmer may have been involved doesn't mean that all farmers are potential terrorists, nor that everyone from Michigan is crackers.

Perhaps the national news media should explain this, but it won't, at least not while the story is breaking and everyone is paying attention. The reasons things happen is usually hard to figure and news must be reported quickly. False assumptions about cause-and-effect are broadcast immediately and may take weeks or even years to correct.

John Allen Paulos, a mathematics professor who specializes in logic and probability, suggests that journalists enlarge their list of standard questions (Who, What, Where, When, Why and How) to include: "How many? How likely? What fraction? Does the correlation suggest a causal relationship or is it merely a coincidence."

In his book A Mathematician Reads the Newspaper, Paulos cites the example of cellular phone use being linked to brain cancer. Appearing on a TV talk show, a widower blamed his wife's death from brain cancer on her phone and that triggered a wave of lawsuits, media speculation and confusion.

The annual incidence rate for brain cancer among Americans is 6 per 100,000. The number of cellular phone owners is estimated at 10 million. Consequently, 600 cellular phone owners can be expected to develop brain tumors each year whether they use their phones or not.

"Since the evidence for an association between cancer and cellular phones consists of only a handful of people, far short of even one year's contribution of 600, we conclude that cellular phones effectively ward off brain tumors," Paulos explains. "Absurd to be sure, but no more so (in fact, less so) than the original hysteria."

In the same sense, it can be shown that farming prevents terrorism and that ammonium nitrate, by providing a vital nutrient to food crops, saves lives.