Kill Chain: The Rise of the High-Tech Assassins by Andrew Cockburn

the German ball-bearing works at Schweinfurt in October 1943, with the attackers suffering huge losses and little damage to the plants, only one in ten of Norden-aimed bombs fell within 500 feet of the target. (The device was still being used to drop sensors for the electronic barrier in 1967.) While the Norden sight was an attempt to position a bomber in the correct spot to drop a bomb, postwar efforts were concentrated on ways of guiding a bomb after it had left the bomber. In December 1968, John Foster told an interviewer that although bombing Vietnam had produced “meager results … we’ve recently developed a series of weapons that permit us to get incredible accuracies, as compared with normal aircraft delivery systems. Instead of having accuracies of hundreds of feet, we now talk in terms of ten feet.”
    At the time, this had been another idle boast. Repeated efforts to hit “critical” targets in North Vietnam were still missing by hundreds of yards. One such target was a bridge over a river about one hundred miles south of Hanoi near a town called Thanh Hoa that was supposedly crucial to the enemy supply effort. The air force and navy bombed it obsessively with guided and unguided bombs between 1965 and 1972 to zero effect—apart from the loss of dozens of pilots. Finally, in May 1972, the bridge was cut with two laser-guided bombs. Though hailed as a momentous event then and since, it turned out that the Vietnamese had stopped using the bridge years before, while traffic flowed unmolested across an undetected river ford five miles upstream. Meanwhile the bridge itself was put to use as the center of what Pentagon wags termed “a flourishing anti-aircraft school.”
    The notion that this triumph of precision might have been irrelevant found little favor where it counted. Under the tutelage of Perry and Defense Secretary Harold Brown (a former nuclear weapons lab director who had also had Perry’s job directing defense research and development), billions of dollars poured into variants of precision guidance, some focused on directing the missile via a little TV camera in its nose or by tracking hot shapes with a heat-seeking infrared camera. Others followed the reflection of an infrared laser beam shone at the target by a pilot or a soldier on the ground. Once Ronald Reagan replaced Carter in 1981, defense spending, already inflated, went into a steeper climb, with the costs of all the revolutionary new weapons systems predictably following suit.
    Among these were various subsystems of Assault Breaker that took on independent but nonetheless prosperous lives after the program was officially ended. The heart of the original system had been the component that Perry hoped would make it possible to see “all high value targets on the battlefield at any time.” This radar was “side looking,” meaning that the antenna stretched along the plane’s fuselage and thus looked sideways, which, because of its size (bigger is better for radar antennae), promised to deliver sharper images. By filtering the data’s echoes to display only objects in motion, the system was billed capable of revealing Soviet tank armies moving up in the rear. Unfortunately, it proved all too efficient at detecting any moving object, not merely tanks, but also automobiles and even trees blowing in the wind. Though the problem proved intractable, the program lived on, to the recurring benefit of the Northrop Corporation, under a variety of code names that ultimately settled on JSTARS for Joint Surveillance Target Attack System.
    Soon after his appointment by Carter, Perry began assiduously promoting an even more ambitious concept, pouring huge amounts of money into a technology called radar cross-section reduction. This was first invented by the Germans to make their World War II submarine snorkels harder to detect with special shaping to reflect radar waves away from the sender and special materials to absorb radar. Perry renamed the