Kill Chain: The Rise of the High-Tech Assassins Read online

  However, in 1966, the group was asked to turn its attention to a real and ongoing war. The principal topic of discussion at Dana Hall that June was “technical possibilities relating to our military operations in Vietnam.” The war was going badly; although the Johnson administration had been pouring troops into South Vietnam, the North Vietnamese had apparently matched this escalation, sending ever more troops and supplies down the jungle routes through Laos known as the Ho Chi Minh Trail. “Rolling Thunder,” a sustained bombing campaign against North Vietnam’s infrastructure, appeared to have had little effect on these supplies while drawing worldwide condemnation. It was a problem that had plagued the air force throughout its history. Deep in the service’s DNA was the traumatic memory of its early life as the “army air corps,” a mere branch of the army no different in status and with a budget lower than that of the artillery. In those days the air corps had nurtured dreams and schemes of revolt and independence that were based on a dogma that strategic bombing of an enemy heartland can win a war without any need for armies or navies. In-house theoreticians took as their gospel the writings of an Italian artillery officer, Giulio Douhet, who argued in the aftermath of World War I that since modern technology favored the defense on land and at sea, airpower alone could bring an enemy to his knees.

  On the eve of World War II, these crusaders had convinced themselves that the destruction of no more than 154 key targets identified as critical to the German economy, such as power-generation plants, would bring the enemy war machine to its knees after six months. As it turned out, this ambitious assumption, based on the most assiduous analysis of the enemy economy, turned out to be wholly wrong. The 154 targets, even when they could be located and hit, either turned out not to be so vital to the enemy war effort as supposed, or the Germans adapted by replacing them or using substitutes. The attackers suffered heavy losses, and the enemy had to be defeated the old-fashioned way, by massive armies slogging across Europe. Nevertheless, the bomber generals’ campaign for independence ultimately bore fruit in 1947, without their having to shed any of their core beliefs in the utility of strategic precision bombing, which they accordingly proceeded to apply in Vietnam once unleashed by President Johnson.

  Early in 1966 air force planners believed they had identified the “critical node” underpinning the North Vietnamese war economy: storage tanks in Hanoi and Haiphong allegedly holding most of the country’s supply of oil. Furious raids on the tank farms produced gratifyingly fierce fires and towering columns of smoke, but it soon emerged that the Vietnamese stored most of their oil in hidden sites elsewhere, and that these were not the magic targets after all. As had been and would be the recurring pattern in such affairs, the targeting committees set about expanding the target list, always in hopes of finding the crucial node whose elimination would prove mortal to the enemy. Meanwhile however, Defense Secretary Robert McNamara, a determined technocrat, was eager for an alternative approach and thus sought counsel from the Jasons.

  After a series of highly classified intelligence briefings on the war, especially the results of the bombing campaign, the group moved west from Massachusetts to the University of California at Santa Barbara. Delving into the reams of classified reports and briefings at their disposal, they concluded that the ultimate critical node in the enemy war effort was the route from North Vietnam by which men and supplies reached the communist insurgents fighting in South Vietnam, the legendary Ho Chi Minh Trail that led though the jungles and rugged terrain of southeastern Laos. They were confident that if the trail could be blocked, thereby preventing any further communist reinforcements and resupply, the war would slowly but inevitably come to an end. Over the next two months they crafted the blueprint of a fence to straddle the trail across thousands of square miles. This would not be a physical barrier but something never seen before in the world: an invisible electronic network that would detect, identify, and destroy any enemy seeking to cross it.

  The first stage consisted of arrays of sensors distributed across the jungle. Some were acoustic, microphones listening for the sound of trucks or footfalls or voices. Others were seismic, ready to detect movements of these same trucks or people. Yet more “sniffed” the air for telltale traces of ammonia, denoting urine and therefore people. Another variety was on the alert for the ignition spark of an engine. If and when a sensor picked up an indicator of enemy presence, it would transmit the signal to planes circling constantly overhead, which would in turn relay the message to distant computers programmed to sift through the necessarily ambiguous information—a column of troops or a herd of elephants?—and match up the apparent source of the signals with the map of enemy trails implanted in the computer memory. To process the data Garwin, the IBM scientist, recommended the IBM-360 computer to analyze the signals, he explained, and “try to characterize the sounds so you wouldn’t be bombing birds or peasants but convoys, trucks, or whatever.” Once birds and peasants had been eliminated, promised Garwin, the computer would order “response, immediate response” from attack aircraft. (Even though they were dealing with the movements of humans and trucks, the scheme echoed their solutions for detecting and intercepting incoming Soviet missiles.)

  This was no casual back-of-the-envelope scheme. Ensconced in Santa Barbara, the scientists, with occasional breaks for surfing or walks on the beach, delved deep into finer details, such as the necessary camouflage for sensors impaled on spikes in the ground (disguised to look like weeds native to the area) and munitions suitable for use against enemy formations once they were located. Their preferred choices were SADEYE/BLU-26B cluster bombs, which blew open after release to disburse 600 yellow shrapnel bomblets over a radius of 800 feet. For sensors, they recommended acoustic devices adapted from the traditional microphone-equipped buoys used by the navy when searching for submarines. Since these might fail to pick up the sound of sandal-clad Vietnamese or their trucks, the area would also be seeded with 300 million tiny firecrackers the size of aspirins. When detonated by a rolling tire, or a stealthy footfall, they would make a sharp bang and so trigger the sensors. The cost of the entire effort was estimated at about $800 million a year (roughly $10 billion in 2015 dollars), which they deemed a bargain because the war would consequently “taper off.” In any case this initiative was positively humane, in the scientists’ view, compared to the wholesale bombing of North Vietnam then underway.

  In September, leading Jasons returned east to brief Defense Secretary Robert McNamara on the carefully thought-out scheme. The secretary, who was very fond of neat technical solutions to human problems, was highly enthusiastic and ordered the air force to get to work immediately. Though initially irked at having an operation dreamed up by civilian eggheads foisted on them, the service chiefs soon reconciled themselves to the limitless funds available, and by mid-1967 the system was largely in place.

  F-4 fighter-bombers and other aircraft strewed hundreds and then thousands of sensors across the jungle. Fleets of assorted aircraft were deployed to circle day and night and relay radio signals from the sensors back to Nakhon Phanom, a military base on the west bank of the Mekong River in northeast Thailand that was so secret it officially did not exist. The base hosted a whole variety of unacknowledged “black” activities, but at its heart, behind additional layers of razor wire and guard posts, sat an enormous air-conditioned building, the largest in Southeast Asia, that was home to Task Force Alpha, the “brain” of the automated battlefield. Behind air locks pressurized to keep the omnipresent red dust of northeast Thailand away from the delicate machines, technicians monitored incoming sensor signals as they were fed to two IBM 360/65 mainframe computers, the very fastest and most powerful in existence at that time. Teams of programmers on contract from IBM labored to rewrite software that would make sense of the data. Not coincidentally, the layout of the darkened, aseptic “war room” resembled that of the command centers of the air force ballistic missile early warning system back in the U.S. waiting for signs of a Soviet nuclear attack.
<
br />   In the view of the military command, the highly classified project represented the first step into a world in which human beings, with all their messy, unpredictable traits, would be eliminated, except as targets. By the time Task Force Alpha began operating Igloo White, the secret code name for the overall electronic barrier (the military likes to preserve the illusion of security with a proliferation of code names), this approach was already failing against the Vietnamese “people’s war,” but there was little inclination for a change of heart.

  General William Westmoreland, the army chief of staff and former commander in Vietnam, expressed the vision most concisely in October, 1969: “On the battlefield of the future,” he declared in a lunchtime speech to the Association of the U.S. Army, a powerful pressure group, “enemy forces will be located, tracked and targeted almost instantaneously through the use of data links, computer assisted intelligence evaluation, and automated fire control. With first round kill probabilities approaching certainty, and with surveillance devices that can continually track the enemy, the need for large forces to fix the opposition will be less important.”

  As it so happened, other components of the air force were fighting a very different kind of war. Marshall Harrison, a former high school teacher, spent 1969 piloting a slow-flying air force plane with excellent visibility called an OV-10 Bronco over South Vietnam as a forward air controller tasked with tracking and fixing the enemy for bombing by jet fighter-bombers. Equipped with no surveillance device more sophisticated than his own eyes, he learned to look for signs that no sensor would ever catch: “fresh tracks along a trail, smoke coming from areas where there should be no smoke, too many farmers toiling in the paddy fields … small vegetable patches where they shouldn’t be.” No computer would calculate, as he learned to do, that footprints on a muddy trail early in the morning if it had rained only during the night probably belonged to the enemy, since civilians were wary of moving at night and being killed if caught breaking curfew.

  Needless to say, Harrison did not encounter senior officers at the barebones strips where he was usually based. The futuristic complex at Nakhon Phanom, on the other hand, could move privileged visitors to awe. “Just as it is almost impossible to be an agnostic in the Cathedral of Notre Dame,” reported Leonard Sullivan, a high-ranking Pentagon official who visited in 1968, “so it is difficult to keep from being swept up in the beauty and majesty of the Task Force Alpha temple.”

  Sullivan’s boss, Dr. John S. Foster, was even more unbridled in his enthusiasm, telling a congressional committee in 1969 that “this system has been so effective … that there has been no case where the enemy has successfully come through the sensor field … It is a very, very successful system.” Foster’s support was potent. His title, director of defense research and engineering, a post he occupied from 1965 to 1973, belied the immense power of his office, since the research projects he authorized and paid for could turn into multibillion-dollar production programs. The suave, smooth-talking physicist-bureaucrat was an ardent proponent of high-technology weapons programs, the more esoteric the better, dispensing billions of dollars for weapons development without excessive concern for cost or practical results.

  Among the aspects of the electronic fence that most excited Foster, an avid model plane hobbyist, was the plan to deploy unmanned planes—drones—not only to relay sensor signals back to Thailand but also ultimately to attack targets. Remotely piloted aircraft had been a topic of military interest ever since World War I, when a prototype radio-controlled biplane designed to attack enemy trenches had been tested and discarded for lack of accuracy and reliability, not to mention frequent crashes. Further radio-control experiments in the interwar years led to the actual coining of the term drone by a pair of naval scientists in 1936, according to an official history, “after analyzing various names of insects and birds.” In World War II the U.S. Navy had brought about the death of the heir to the Kennedy fortune by enlisting him in Operation Aphrodite, a scheme to fly remote-controlled B-24 bombers packed with explosives into German submarine pens. Human pilots were required to handle takeoff and to switch on the radio controls. When Joseph Kennedy Jr. flipped the switch prior to bailing out, the plane promptly blew up. None of Aphrodite’s other eleven attempts were successful. By the 1960s drones had carved out a useful niche mission as semirealistic targets for aerial gunnery training. Come the Vietnam War, they were adapted for reconnaissance, though without much success, being easy targets for enemy gunners. Foster, however, cherished the notion that they could soon begin replacing manned attack planes in various roles, and so money poured into a variety of drone programs under development by corporations such as Boeing, Vought, and Teledyne Ryan.

  None of them worked very well. Most were canceled after a suitable interval, including those assigned to the electronic fence. Yet the desire for aircraft that could be controlled from some remote location was already deeply ingrained at high levels, perhaps because their missions could be more easily kept secret. In 1969, for example, President Nixon and Henry Kissinger decided to begin bombing Cambodia without any notice to Congress or the public. The raids were therefore conducted in deepest secrecy. The flight paths of the B-52 bombers assigned to Operation Menu were under the direct control of a ground station that also precisely controlled the release of their huge bomb loads. Most of the crews thought they were bombing Vietnam, as usual, with only pilots and navigators briefed on the true target. As we shall see, this attribute of remotely piloted aircraft would become even more attractive to higher authority in the following century.

  Drones were not the only elements of the electronic fence that failed to perform as advertised. The plan that appeared so complete and elegant on the beach at Santa Barbara or in a congressional hearing did not prosper in contact with the real world, a world that shifted and adapted even as the mighty computers struggled to make it fit the maps and patterns programmed into their memories. The network of trails, totaling 12,000 miles, radically changed in each rainy season. The ingenious notion of having the enemy announce his presence by stepping on the pill-sized firecrackers had not taken into account the moist Laotian climate, which almost immediately rendered the explosives as dangerous as a wet match.

  Further complicating matters was the inaccuracy of the high-tech LORAN (long-range navigation) radio-navigation system used to drop the sensors, so no one knew for sure where they, and consequently any targets they might detect, actually were. “If we got within four or five miles of the aim point we were doing pretty good,” remembers Rex Rivolo, an F-4 pilot who dropped “hundreds” of sensors into the darkness. LORAN, which was also used to guide bombing runs, once led Rivolo inadvertently to bomb the huge American base at Da Nang.

  Most important of all, Task Force Alpha was not confronting a fixed target that would faithfully behave as predicted but a living enemy skilled in camouflage and deception that could watch, think, and adapt. Even before the arrival of Task Force Alpha, large sections of the trails system led nowhere, decoys that were purposely put in place to confuse American reconnaissance planes. Fake bridges were erected to draw attack, while real ones, resting on inflated inner tubes, lay invisibly submerged during the day. Back in Santa Barbara, the Jasons had entertained the possibility that the enemy might eventually adopt countermeasures of some kind against the sensors, though they were confident that this would take “some period of time.”

  It took the Vietnamese a week.

  General Dong Si Nguyen, a veteran revolutionary and transportation genius who commanded the entire Ho Chi Minh Trail for most of the war, later reminisced how “the devices were dropped over an area as large as 100 kilometers covering our transportation network. We spent seven days trying to arrive at a solution. We brought vehicles to the area and ran them back and forth throughout the day (to make listeners believe the area was active) … While we distracted the Americans in this manner, the actual convoys were then able to safely move by means of a different route.”

 
; Specialized teams were meanwhile set up in every section of the trail to hunt for sensors. “They were hard to find,” one hunter said later. “Sometimes they were in an area that was not really important to us, so we deliberately triggered them.” Otherwise, the Vietnamese ran herds of cattle down the trail to simulate troop columns to fool infrared cameras or bottles of human and animal urine to confuse the sniffer-sensors.

  The electronic barrier cost almost $2 billion to set up and roughly $1 billion a year to operate. The funds for the secret operation were so artfully hidden in the defense budget that for years Congress had almost no knowledge or oversight of the operation for which it was voting huge sums of money. There was therefore little official incentive to undercount the amount of damage being inflicted on the enemy. To calculate enemy losses without sending men to look for themselves, deemed an impossibly dangerous task, the air force simply multiplied the number of bombs dropped by the number of people who could in theory be killed by varying types of bomb, such as the SADEYE. The tally produced by this arithmetic—20,723 for Igloo White’s first season—conveyed an air of precision that had little basis in reality. “This process,” an official U.S. Air Force historian tartly noted some years later, was “based on so many assumptions that the end product represented an exercise in metaphysics rather than mathematics.”