washingtonpost.com  > Opinion > Columnists > Eugene Robinson

Sum of Our Fears

By Eugene Robinson
Friday, March 11, 2005; Page A23

Today we eschew both cotton-candy indulgence (the Michael Jackson trial) and eat-your-peas sobriety (whither Social Security?) to tuck into more starkly existential fare: If you're the worrying kind, what do you choose from the smorgasbord of dread laid out upon the table of modern life?

Osama bin Laden still out there scheming, somewhere at large? Terrorists with a radioactive "dirty bomb" or just a Ryder truck full of fertilizer and fuel oil? A laboratory accident that looses a real-life Andromeda strain? Iran (almost) with nukes? North Korea (it boasts) with nukes? Something as general as global warming, or as specific as the dangerous intersection on the way to your daughter's school?

_____What's Your Opinion?_____
Message Boards Share Your Views About Editorials and Opinion Pieces on Our Message Boards
About Message Boards

Now comes avian influenza, a virus that has jumped from chickens, ducks and geese to humans. Avian flu has infected fewer than 100 people we know of, all in faraway Southeast Asia, yet the head of the Centers for Disease Control and Prevention already sees "a very ominous situation for the globe," while a World Health Organization official warns of "the gravest possible danger."

So are we supposed to stop fretting about everything else and focus a laser beam of anxiety toward Vietnamese barnyards? To help make sense of it all, I rounded up a few experts in the burgeoning field of risk analysis.

Paradoxically, the exercise was a comfort. It turns out there's no "right" or "wrong" way to calculate the sum of our fears.

Elizabeth Anderson, who runs a risk-analysis firm and edits an academic journal devoted to the field, pointed out that in deciding what to worry about, "you're comparing risks with varying degrees of certainty, and some may be low probability but high consequence."

In other words, few people (except the parents of teenagers) worry themselves sick about car accidents, which are common but usually minor. An explosion at a nuclear plant, on the other hand, is vastly less likely -- but utterly devastating if it ever happens. Also, one man's risk can be another man's breakfast: Europeans see grave dangers in genetically modified "Frankenfood"; Americans generally don't care if their corn flakes were engineered in a lab.

Paul Slovic, a professor at the University of Oregon, did a striking experiment in which he asked experts and laymen to rank a list of 30 dangers. "Nuclear power," ranked most dangerous by regular folks, was just 20th on the experts' list. The experts, who knew the statistics, were far more concerned about such things as medical X-rays and surgical mishaps.

Further research has led Slovic to conclude that this subjectivity is inherent. We tend to overestimate risks that strike us viscerally with dread -- a dirty bomb that spews radiation, say -- and underestimate risks that don't. Slovic found that we even change our evaluation of a risk depending on the language used to describe it: Illogically, we perceive "20 out of 100" as a greater risk than "one out of five" or "a 20 percent chance."

The experts agreed that there's one thing that can be fixed: how authorities communicate with the public about the risks we face.

Health authorities can give us more than blurted pronouncements of certain doom. They can give us enough information to make an informed assessment. Is there still a good chance the flu will fizzle out? What is that chance, realistically? What's the plan if and when the first human case shows up in this country?

The Department of Homeland Security should lead a similar conversation about terrorism threats. Basically, all we get now is the green-blue-yellow-orange-red color chart, which tells us nothing. Occasionally there's an all-points bulletin for a few people whose only crime may be their swarthiness, but the alert always comes with the admonition to go about our normal lives. Michael Chertoff, new DHS secretary, here's a job for you: Talk to us constructively about what "normal" means these days.

Baruch Fischhoff, a Carnegie Mellon University professor who heads the Society for Risk Analysis, said, "Risk analysis is easier when you have a historical record, but it's a matter of theory when you don't." Trying to judge the risk of any kind of terrorist attack is hard, because by definition, it's an exercise in reading the terrorists' minds.

On the other hand, he noted, there's a long historical record of how influenza outbreaks grow into epidemics that explode into pandemics, and "it's not as if we have to read the minds of the microbes." That record includes the Spanish influenza of 1918-19, which killed between 20 million and 40 million people worldwide.

So if eminent epidemiologists are running around with their hair on fire over avian flu, I'm afraid we might want to pay attention.


© 2005 The Washington Post Company