Kill Chain: The Rise of the High-Tech Assassins (21 page)

But this was an illusion. As Colonel Milani, the Special Forces officer tasked with assessing the events on Takur Ghar reported (much to the irritation of the SEALs), nothing of the kind had happened. Roberts’ body was discovered very close to where he had landed in his fall. Postmortem examination of his wounds indicated he had been shot almost immediately. On the other hand, the body of Sergeant Chapman, who had been left for dead in the hurried flight following the second landing on the mountain, was discovered in a bunker several yards from where he had been seen to fall and from where he may have been firing on the enemy. The wounds that actually killed him apparently came from the U.S. bombs that destroyed the bunker. Given that Chapman, wounded in the initial attack, had been abandoned on the battlefield, this was not an appealing conclusion. As Milani ultimately concluded in his unsparing after-action report, “Roberts’ colleagues desperately wanted to see him alive and taking it to the enemy. They not only saw what they wanted to see, they saw what they needed to see.” They were, he wrote, “necessarily enmeshed in a network of preconceptions.”

Those officers saw what they needed to see even though they had the leisure of several weeks to review the video in minute detail. But more and more, drone pictures were guiding life-and-death decisions made in a matter of hours or minutes, with consequences that would endure for years.

8

KILL THEM! PREVAIL!

George W. Bush had arrived at the White House with a pledge, as outlined in his 1999 speech at the Citadel, to “begin creating the military of the next century” as well as to boost overall defense spending. The Afghan operation had put precision-guided bombing on display, but the revolution in military affairs held the promise of further wonders. “Millennium Challenge 2002,” the largest and most elaborate war game ever held, was accordingly designed to put the revolutionary “military of the next century” on full display. Three years in the planning, budgeted at $250 million, involving 13,500 participants waging mock war in 9 training sites across the United States as well as 17 “virtual” locations in the powerful computers of the Joint Forces Command, the exercise, to be held in the summer of 2002, enjoyed the personal attention of Defense Secretary Rumsfeld himself. As Rumsfeld declared during a visit to Joint Forces Command headquarters in Norfolk, Virginia, where the major players in the game would be based, the game would show “the progress that we have made this far in transforming to produce the combat capability necessary to meet deep threats and the challenges of the 21st Century.”

Escorting Rumsfeld round the premises, the commanding general, William Kernan, took care to keep his distinguished guest away from a tall, bald-headed man with a military bearing but clad in civilian clothes; he was the enemy. Paul Van Riper, the three-star marine general, now retired, who had poured such scorn on David Deptula’s theories in the years following Desert Storm, had been called back to command the red team in the Millennium war game. In such exercises, the enemy is always red; the U.S. side is always blue. But Van Riper was a twofold enemy; not only was he playing the role of an opponent, he also was making no secret of his contempt for the concepts underpinning the blue team’s plan for the game.

“None of it was scientifically supportable,” he told me later, after delivering a droll recitation of the full range of acronyms pumped out by the command. “They claimed to be able to understand the relationship between all nodes or links, so for example if something happened to an enemy’s economy, they could precisely calculate the effect on his military performance. They talked about crony targeting [the destruction of the property of Slobodan Milo
š
evi
ć
’s friends’ property during the Kosovo conflict in hopes of affecting his behavior] a lot.” In short, the blue plan encapsulated the core belief system of U.S. military doctrine. High-value targeting, as Van Riper was well aware, was inherent in the official doctrine’s assertions of the capabilities of effects-based operations. But given his low regard for the entire concept of “total situational awareness,” he was not unduly worried.

In the scenario designed by the exercise planners, Van Riper was playing the role of a rogue military commander somewhere in the Persian Gulf who was willfully confronting the United States. Though there were thousands of troops as well as planes and ships taking part in the game across the country, much of the action would be “virtual,” occurring in computers and displayed on monitors. It was to be the ultimate video game. Needless to say, each of the services foresaw a useful role for their expanding fleets of drones as well as for other novel systems.

Among the digital tools available to the blue team was an enormous database labeled Operational Net Assessment (ONA), which they believed contained everything they needed to know about their opponent and how he would behave. But they did not even know what he looked like. The blue commander, a three-star army general, worked in full uniform, surrounded by his extensive staff. As the game was getting under way, Van Riper, dressed in casual civilian clothes, took a stroll, unrecognized, through the blue team headquarters area to take the measure of his opponent. With his own staff, he was informal, though he forbade the use of acronyms. “We’ll all speak English here,” he told them.

In the first hours of the war, the blue team knocked out Van Riper’s fiber-optic communications, confidently expecting that he would now be forced to use radio links, which could be easily intercepted. He refused to cooperate, however, turning instead to motorcycle couriers and coded messages in the calls to prayer from the mosques in preparing his own attack. He was no longer performing an assigned part in a scripted play. Van Riper had become a real, bloody-minded, Middle Eastern enemy who had no intention of playing by the rules and was determined to win.

Just a month earlier, the Bush administration had unveiled a new national security policy of preemptive attacks, justified as “our inherent right of self-defense.” So, when a blue team carrier task force loaded with troops steamed into the Gulf (at least in the computer simulation) and took up station off the coast of his territory, Van Riper assumed that they were going to follow the new policy and attack him without warning. “I decided to preempt the preempter,” he told me. Oddly enough, the blue general sensed this, saying: “I have a feeling that Red is going to strike,” but his staff was quick to assure him that their ONA made it clear that this could not happen.

Van Riper was well aware of the U.S. Navy’s Aegis antimissile capabilities and of how many missiles it would take to overwhelm them. “Usually Red hoards its missiles, letting them out in dribs and drabs,” he told me in retracing the battle. “That’s foolish, I did a salvo launch, used up pretty much all my inventory at once.” The defenses were overwhelmed. Sixteen virtual American ships sank to the bottom of the Gulf, along with twenty thousand virtual servicemen. Only a few days in, the war was over, and the twenty-first-century U.S. military had been beaten hands down. Van Riper, who had been an attentive student of the theories of John Boyd, the fighter pilot and theoretician of conflict, won by adapting quickly and imaginatively to changing circumstances (such as his use of motorcycle messengers and calls from the minarets of mosques when his phone links were destroyed). In contrast, his opponent’s rigid “effects-based” approach had locked him into a preset vision of how the battle would play out.

For General Kernan, the Joint Forces commander, there could be only one solution to this crisis. Van Riper was informed that the sunken ships had magically refloated themselves, the dead had come back to life, and the war was on again. But this time there would be no surprises. He was not allowed to shoot down vulnerable blue team V-22 troop transports. The red team was ordered to switch on their radars so that they could be more easily detected and destroyed. The umpires announced that all of the red team’s missile strikes had been intercepted. The game was now unashamedly rigged to ensure a U.S. victory as well as validation of the new theories. Van Riper resigned in disgust as red leader but stayed on to monitor the predictable rout of his forces under these new conditions. Afterward he wrote a scathing report, documenting how the exercise had been rigged and by whom, but no outsider could read it because it was promptly classified.

Undaunted, in the very next real war, the March 2003 invasion of Iraq, the U.S. military deployed many of the same concepts recently deployed in the Millennium Challenge. Initially, things appeared to go well. The invasion was planned strictly in accordance with the concept of rapid dominance, defined as “near total or absolute knowledge and understanding of self, adversary, and environment; rapidity and timeliness in application; operational brilliance in execution; and [near] total control and signature management of the entire operational environment.” This happy state being achieved, the enemy must inevitably be reduced to a state of “shock and awe.”

Decapitation fit neatly into this approach, indicating that the planners misunderstood how useful it would be to keep Saddam alive and in command of enemy forces. According to a postwar Pentagon assessment, “The largest contributing factor to the complete defeat of Iraq’s military forces was the continued interference by Saddam (Hussein),” posthumously affirming the argument advanced by British intelligence officer Lieutenant Colonel Thornley against killing Hitler in 1944 on the grounds that the Nazi leader’s blunders were of inestimable help to the Allied cause. Nonetheless, despite Saddam’s previous demonstrations of military incompetence during the 1991 war and before, the U.S. pursued a decapitation strategy to an almost obsessive degree. Attempts to kill the Iraqi leader in 1991 had been lightly cloaked in euphemism, but this time there was little such pretense.

In the first minutes of the war a fusillade of bombs and no less than forty cruise missiles rained down on a collection of farm buildings on the outskirts of Baghdad, prompted by a CIA report that Saddam and his sons were lurking there in an underground bunker. Early reports were optimistic. “We were sure we’d got him,” one of the targeting team told me. “Cheney came out and said he was dead. It took us three days before we were sure he had survived.” In fact, neither Saddam nor either of his sons had been at the farm. Nor did it have a bunker.

The effort was coordinated by a High-Value Target Cell in the Pentagon, an elaboration of the system that had tracked Milo
š
evi
ć
in 1999. The office coordinated other such cells at CIA, NSA, and Centcom, the military command overseeing the Iraq and Afghan wars. “Between 1999 and 2002 it was growing into a new science,” a Pentagon analyst formerly assigned to the operation told me. “If you’re doing HVT, on Saddam Hussein, for example, you have to know where he is at all times, who are his security retinue, where they are. You look for patterns, but our predictive ability was low. We got very, very good on where he had been; sometimes we knew where he was. But predicting where he would be, that was hard, and we needed that because at that time the kill chain, the time between getting the intelligence and the bomb or missile impacting, was too long, a minimum of forty minutes, and often more. The shortest kill chain we managed in the 2003 war was forty-five minutes,” the analyst recalled. “That was the strike on the al-Saath restaurant in Baghdad. We thought that Saddam was there. He wasn’t, but we did kill a bunch of civilians.”

“I did not know who was there. I really didn’t care,” Colonel Fred Swan, weapons officer on the B-1 bomber that hit the restaurant, later told reporters. “We’ve got to get the bombs on target. We’ve got to make a lot of things happen to make that happen. So you just fall totally into execute mode and kill the target.”

Drones, with their ability to wait and watch for a target to appear and then launch a missile can, at least in theory, shrink the kill chain almost to zero. “But that means you’re taking the decision on the fly, with no time to really assess potential collateral damage, like who else is in the house or whatever,” the former inmate of the High-Value Target Cell pointed out to me.

The former specialist raised another occupational hazard of this particular strategy: the difficulty of assessing success. “After that first strike, it took three days before we knew for certain he wasn’t dead. Even when he appeared on TV, sitting at a desk, reading a speech, with glasses on, a lot of people in the office were saying ‘It’s not him, it’s a double,’ or even suggesting it was prerecorded earlier in case he got hit. We even had analysts going over the video to see if the angle of light in the window was right for that time of year.”

Precision strikes were targeted over the ensuing weeks of the invasion on the purported lairs of various Iraqi commanders, though without success. According to the former Defense Intelligence Agency analyst Marc Garlasco, the United States selected fifty specific “high-value individuals” to be targeted and killed during the invasion. All survived. Not so lucky were the “couple of hundred civilians, at least,” according to Garlasco, who were killed in the strikes. Many of them may have died thanks to what appeared to be an ingenious innovation in targeting technology. The war coincided with the introduction in the Middle East of a new model of satellite phone for civilians, the Thuraya. Unlike its unwieldy predecessors, this device, about the size of an old-fashioned telephone handset, could be easily carried and used, and it was believed that many of the fugitive Iraqi leadership carried them. Like all such devices, the Thuraya inevitably transmitted information regarding its location, thus providing a convenient mark for GPS-guided bombs. In effect, the target would be guiding the bomb that killed him. But there was a flaw: the Thuraya’s GPS system was not so precise in fixing its position and was accurate only within a 100-meter (109-yard) distance, which meant that the bomb could land anywhere within an area of 37,500 square yards. This was lucky for the target but not so lucky for innocent passersby who happened to be in that area. In other words, as a Human Rights Watch report subsequently observed, innovative technology had turned “a precision weapon into a potentially indiscriminate weapon.”