The Watchers (54 page)

The interrogation unit would be specially trained and stand ready to deploy anywhere in the world on a moment's notice. But that wasn't its most extraordinary characteristic. While the FBI director would lead the group, it would ultimately report to the National Security Council. Not since the days of Ronald Reagan had the NSC staff so directly taken on the management of a sensitive intelligence operation. Obama was going to keep terrorist interrogation in place, and he was going to keep it close to him.

Obama's order to close Guantánamo Bay also wasn't an emphatic rejection of Bush. It was politically audacious to order the prison shut down without a plan for dealing with the more than two hundred people still held there. But only a few months after he signed the order, Obama faced a revolt by members of his own party, who joined Republicans in Congress in strongly opposing any release of detainees into the United States. Lawmakers unanimously denied any funding for the prison closure until the White House came up with a plan for disposing of the prisoners.

The administration also found that some of the detainees' cases posed extraordinary legal challenges. They couldn't be let go, but they also couldn't be tried in U.S. courts, because some of the evidence against them was gathered through extreme techniques that the Obama administration said amounted to torture. The administration acknowledged that some detainees would have to be tried by military commissions, whose standards of due process differed from those of criminal courts. Bush officials had also opted to use military commissions rather than criminal courts, believing they'd offer a better chance to obtain convictions.

Obama had tried to distance himself from Bush in the first forty-eight hours of his presidency, but by the time his first hundred days were through, the president had become accustomed to a new critique, expressed most regretfully by the leftist base of his party: Real change was an illusion. The new president looked more like the old one every day.

On one cornerstone terrorism policy, however, Obama remained conspicuously silent. There were no executive orders issued on electronic surveillance. No speeches about shutting down the NSA programs or revisiting the policy of telecom immunity that he had come around to supporting. Despite continuing protests from the Netroots, the president showed no inclination to dismantle a set of tools that his predecessor had regarded as the “crown jewel” of America's counterterrorism arsenal.

Only weeks after Obama took office, he faced his first test on the NSA surveillance, which had run off the rails in a fashion that many critics of the “FISA fix” had predicted. During a periodic review of surveillance activities officials discovered that the agency had inadvertently collected the phone calls and e-mails of Americans. This was the very “incidental” collection that some had feared. In the course of monitoring supposedly foreign communications, the NSA had trouble distinguishing which phone numbers and e-mail addresses actually belonged to people in the United States. As a result the agency ended up directly targeting Americans without individual warrants—a basic violation of the new law.

This was not a onetime accident. The “overcollection,” as intelligence officials called it, was systemic, and a direct result of the difficulty of knowing for sure where a target was actually located on the globe. Members of Congress were only alerted to the overcollection after it had occurred. And that revealed another troubling aspect of the surveillance regime. The NSA hadn't been trying to hide what happened. Rather, officials apparently discovered the violation only after they checked their surveillance logs. The Watchers had been watching the wrong people and didn't know it.

During his campaign Obama had said that he voted for the dramatic changes to FISA because someone needed to “watch the watchers.” But it was unclear who was doing that job. The president had said that that duty fell to the FISA Court. But its members only learned of the infractions when they were told by the Justice Department, following a required review of the logs conducted once every six months. This was the new definition of oversight—a twice yearly checkup and a bill of infractions delivered after the fact.

The review began in the closing days of the Bush administration and carried over into Obama's first term. When officials became aware of the problem, the president issued no public remarks. He wrote no executive orders. Instead, his attorney general ensured that new, undisclosed safeguards were put in place. Then he appeared before the FISA Court, where he asked for and obtained a renewal of the program.

Things became clearer after that. Who was watching the Watchers? They were watching themselves.

 

 

 

 

John Poindexter is a man with few regrets. One might reasonably assume that, having been put through the political wringer twice, he has emerged a hardened cynic. And yet his grueling public ordeals have only made him a sunnier optimist. Perhaps that's because now, as he sails into the twilight of his life, Poindexter doesn't have to look very hard to see the evidence of his handiwork, and the beginnings of a legacy.

This isn't wishful thinking on his part. I have no doubt that his vision of a total information awareness society was not only prescient but accurately reflects where we have arrived as a nation. John Poindexter envisioned a world. Mike Hayden made it a reality. Mike McConnell enshrined it in law. And Barack Obama inherited it. In broad strokes, that's how we got where we are now.

When the news of the NSA's warrantless surveillance program first broke in December 2005, Poindexter was pleased to see that Hayden had been following his advice. He'd recognized that in order to find signals in the noise one had to collect information from far and wide. But Poindexter was deeply disappointed that the NSA had not pursued his ideas for privacy protection. Indeed, subsequent reporting revealed that the agency had rejected a new data-analysis system that had such protection built in because it was too expensive and too technically demanding. Apparently, adding a layer of protection technology increased the complexity of the software and made it less flexible. Poindexter knew it would take leaps of innovation to make a privacy appliance work, which is why he envisioned such work as part of a grand “Manhattan Project” for the information age. One can say Poindexter is wrong about some things, but technical engineering isn't one of them.

And there, I think, lies the national tragedy in the personal story of John Poindexter. If our leaders were truly committed to preventing acts of terrorism, and at the same time really wanted to protect individual privacy, then they should have kept Poindexter right where he was, as the lead visionary and chief proponent of a radical new way of thinking about how to secure people's lives and their rights. They should have closely scrutinized his work, kept it limited to research, and exploited his unparalleled technical expertise and his willingness to court controversy without regard for his political reputation. Our government should have used John Poindexter, if for no other reason than that he was willing to be used. Had that happened, I believe we might have avoided the scenario we're presented with today: a rising surveillance state and the near certainty that, in the wake of another major attack, the debate over how to properly balance the competing interests of security and liberty will simply become academic. There will be no time for deliberation, and our leaders will show little restraint. History tells us this is so.

We no longer live in a world where we can reasonably expect certain personal information to remain private. And we have
never
lived in a country where the government willingly restrains itself from collecting, analyzing, and disseminating information about our lives, our associations, and our thoughts. The intelligence surveillance laws and regulations of the late 1970s were short-lived. They were created in reaction to unquestionable abuses of power, but they began to prove their technical inadequacy in less than a decade. They have also been obsessed with regulating the collection of information rather than how it is used, which is the more important component of the intelligence system. Our elected officials could have rolled up their sleeves and dived into this complicated legal bramble, and we might have been better—and safer—for it. Instead, they have created laws that tilt the balance of power back to where it has usually been in the realm of national security—toward the executive branch, and to the president in particular. We have seen the excesses born of this arrangement. But we have never lived in a time when the government has had such remarkable technological ability to watch its own citizens.

Our government officials know that, and they have put us all on notice. One speech by a seasoned career intelligence official reaffirmed this point for me in a way that few public remarks ever have. In November 2007 Donald Kerr, then the deputy director of national intelligence, spoke about how the spy community was operating in the information age. Kerr had spent much of his career in the technical field and was one of the top CIA officials who met with Poindexter about the total information awareness program in 2003. I found his remarks revealing for two reasons: First, what he said, and second, the fact that so few people noticed.

“Nowadays, when so much correlated data is collected and available—and I'm just talking about profiles on MySpace, Facebook, YouTube here—the set of identifiable features has grown beyond where most of us can comprehend.” Traditionally, those “identifiable features” included the fragments of data that were unique to one person—a name, a date of birth, a Social Security number. Now we had to add profile pages, social networks, and personal blogs to the mix. “In our interconnected and wireless world,” Kerr said, “anonymity—or the appearance of anonymity—is quickly becoming a thing of the past.”

All the information needed to positively identify a person had been made public. “Protecting anonymity isn't a fight that can be won. Anyone that's typed in their name on Google understands that,” Kerr declared. That was an astounding admission. Not because Kerr was the first person to say it, but because he was the number two spy in the United States. The ability to positively identify an individual is one of the most important and powerful tools available to an intelligence agency that is in the business of tracking people. And now that tool was available to anyone with an Internet connection. A major shift had occurred. Kerr knew that. “I think people here, at least people close to my age, recognize that those two generations younger than we are have a very different idea of what is essential privacy, what they would wish to protect about their lives and affairs. And so, it's not for us to inflict one size fits all.”

Kerr's remarks still stand as the most frank public assessment by the government of where we are in our surveillance state. The meaning of privacy has changed and anonymity no longer exists. The nation should come to terms with that fact. As Kerr made clear, the intelligence community already has.

 

There's another candid assessment from deep inside the intelligence community that gives us an idea of how far the Watchers have come—and how far they have to go. In September 2006, the five-year anniversary of the 9/11 attacks, the Office of the Director of National Intelligence launched one more program aimed at Poindexter's ultimate goal—a fully integrated, self-teaching “system of systems” for detecting signals of terrorist planning. Quietly, officials spread the word about a plan to lash the old TIA programs into a new model. They called it Tangram, after an old Chinese puzzle that arranges seven geometric pieces into hundreds of distinct figures. The same set of shapes—five triangles, a square, and a parallelogram—can be made to look like a dog, a bird, a steamboat, even a woman pouring tea. It was a fitting metaphor for Poindexter's own narrative: breaking himself into pieces only to reemerge in new forms.

But Tangram was hardly a triumph. A document soliciting technical proposals for the system stated that intelligence agencies hadn't been able to move beyond “guilt by association” as the best indicator of whether someone was, or wasn't, a terrorist. That was an arresting fact. For all the money spent, the hours consumed, for all the programs run in secret and at extraordinary political and social cost, the government was not much better at detecting bad guys than they were on 9/11.

“To date, the predominant approaches have used a guilt-by-association model to derive suspicion scores,” the Tangram document stated. “In the cases where we have knowledge of a seed entity in an unknown group, we have been very successful at detecting the entire group. However, in the absence of a known seed entity, how do we score a person if nothing is known about their associates? In such an instance, guilt-by-association fails.”

The group running Tangram was the successor to the NSA outfit that inherited Poindexter's programs. This was where cutting-edge terrorism research was being conducted, and even these bright minds had come up short. Hunting terrorists in data was still a clumsy business, built on suspicion. The warnings of Erik Kleinsmith's Able Danger analyst echoed in the Tangram document: “Do you have any idea how many people on the planet would go to jail just because they
knew
somebody bad?”

We cannot truly protect the nation this way. The Watchers know that, which is why they've devoted so much energy to the considerable technical challenge that Poindexter first posed in 2002. But just because it's hard to identify terrorists with data doesn't mean that the government will stop trying. As Kerr's remarks made abundantly clear, the government has no shortage of access to information about people. Our current surveillance laws are engineered to allow the intelligence community to consume huge amounts of data, in the hopes that some of it will prove useful. Again we see that the Watchers have become very good at collecting dots but not at connecting them. When we come to terms with this we can begin to think about what function we want our intelligence agencies to serve in the twenty-first century. Now is the time to have this debate—not after the next assault on the United States. If we can't have an honest discussion and form sensible policy in a time of relative stability, with the last major attack more than eight years behind us, then we will end up making momentous decisions in a moment of panic, just as we did after September 11. In that case, we will all suffer the consequences.