Perhaps as early as this month, President George W. Bush is expected to declare that a handful of prototype missiles in California and Alaska are ready to protect the U.S. from long-range missile attacks. The Pentagon calls the system a "test bed," one that still needs more sophisticated radar, interceptors and space-based lasers to realize Ronald Reagan's dream of a "Star Wars" antimissile program. The Defense Department, however, maintains that it can defeat North Korea's small arsenal of long-range missiles--a claim that may be hard to swallow given the limited number of tests done so far.

The ground-based midcourse defense system, as it is called, will start off with no more than 10 "hit-to-kill" interceptors designed to collide directly with incoming missiles in space. To date, the program has intercepted target missiles in five of eight heavily scripted tests.

But critics say those trials prove little. The Union of Concerned Scientists, in a report released earlier this year, concluded that the initial system “will be ineffective against a real attack” and also slammed the administration for “irresponsible exaggerations” about its abilities. In June opponents in Congress tried unsuccessfully to postpone deployment on grounds that the system should be tested further. During a Senate debate, Senator Barbara Boxer of California likened the plan to the Wizard of Oz, who “was scary, but when you pull back the curtain, it was just some little guy,” she said.

The Pentagon's Missile Defense Agency (MDA) has long held that live tests, which are costly, difficult to plan, and limited by range and safety concerns, are not the only means of proving the system's efficacy. According to the Pentagon, sophisticated modeling, simulations and exercises can offset the paucity of real intercepts. In April, Air Force Lt. Gen. Ronald Kadish, then head of the MDA, told Congress: “We use models and simulations, and not flight tests, as the primary verification tools.”[...]