Body count: how John Ashcroft's inflated terrorism statistics undermine the war on terrorism.

• Author: Gourevitch, Alexander

AT FIRST BLUSH, NEW JERSEY'S District Attorney's office seems like a model of federal law enforcement in the war against terrorism. In the year after 9/11, after all, they nabbed 62 individuals for acts of "international terrorism"--individuals who, arguably, would no longer be threatening American lives. But on closer inspection, there's less to this success story than meets the eye. Sixty of the 62 international terrorists, according to a March story in The Philadelphia Inquirer, turned out to be Middle Eastern students who had cheated on a test; specifically, they had paid others to take an English proficiency exam required for college or graduate school. Only one of the other two cases involved charges that might normally be understood as relating to an act of terrorism: Ahmad Omar Saeed Sheikh, who was indicted for his role in the kidnapping and murder of Wall Street Journal reporter Daniel Pearl in Pakistan.

With the exception of Pearl's kidnapping conspirator, in other words, none of the terrorists in question were actually terrorists. And these aren't isolated examples. Under post-9/11 rules promulgated by the Justice Department--which created a number of new terrorism-related categories by which to classify cases, but left it to district attorneys to determine which crimes fit the bill--federal prosecutors across the country are turning in creative anti-terrorism records to their superiors in Washington, who are under enormous pressure to produce results and have little incentive to double-check them. The result is an epidemic of phony reporting. According to a January report by the General Accounting Office, at least 46 percent of all terrorism-related convictions for FY 2002 were misclassified; of those cases listed as "international terrorism," at least 75 percent didn't fit the bill.

If it were just a matter of wasting a few million dollars, the shenanigans at Justice might be just another example of typical Washington game-playing. But it's far more consequential than that. In any bureaucracy, reliable statistics are an important feedback; for our anti-terrorism efforts, reliable statistics are a matter of life and death. Without them, it's impossible for policymakers and the public to guess how many terrorists are operating within the United States, how many of them we've caught, which of our anti-terrorism strategies are effective--and which are useless. But like Robert McNamara's generals, who inflated enemy body counts so politicians could claim the Vietnam War was going better than it actually was, federal prosecutors are giving us a false sense of security. "You have to have numbers if you are going to manage anything," says William Wechsler, a former director for transnational threats at the National Security Council who helped shut down al Qaeda's finances after 9/11. "If you're...