On February 9, 2011, Janet Napolitano, the secretary of the Department of Homeland Security, delivered some terrible news to Congress. The terrorism threat against the country, she announced, had not diminished despite the enormous counterterrorism effort and decade-long war—in fact, it had only gotten worse. In some ways, “the threat facing us is at its most heightened state” since the attacks a decade ago, she said, slowly reading every word of her prepared testimony.
Of particular concern was the inclination of some potential terrorists living in the United States to brew plots and carry them out by themselves, without the help of larger, more easily detectable networks. Traditional counterterrorism methods will not be enough to find the lone wolf, a terrorist operating alone, without a network, Napolitano stated. “State and local police will more often be in the best position to notice the signs of a planned attack.”
For all the drama in her words, there was something amiss about the moment. Napolitano read her statement in a monotone, as if reciting an obscure budget document. No one in the House committee room looked fazed. Neither did members of the audience, who sat expressionless, even bored. Reporters on deadline didn’t rush her for details afterward. In fact, they looked a little bored, too. Napolitano’s testimony ended up being a one-day story, and, in most newspapers, didn’t even make the front pages.
If the people in charge of Top Secret America were really convinced the situation was as dire as Napolitano claimed, they weren’t sharing enough information with an impatient public and a press by now numb to the drone of perpetual yellow alerts for their warnings to ring true. The few known attempted bombings seemed to be minor incidents, as there had never been a danger to public safety. Weeks earlier, a construction worker had been arrested during an FBI sting operation in Baltimore; he had attempted to set off a bomb outside a military recruiting center. There was little public or press interest in that, either. When a Somali-born student was arrested in Portland, Oregon, during a similar FBI operation—this time the bomb was supposed to explode at a downtown Christmas tree lighting ceremony—members of Congress barely said a thing.
But it was hard not to notice all of the portable traffic signs cropping up along the eastern seaboard. The signs displayed alarming messages: “Report Suspicious Activity.” “Terror Tips?”
“What’s that all about?” friends and colleagues would ask. “Is there something going on?”
Yes, something was going on, and it wasn’t as easily identifiable as a group of military officers behind closed doors at NorthCom trying to map the nooks and crannies of the United States. It was more obscure and decentralized, harder to touch and feel.
There had been a disturbing trend in the last five years, in settings around the country: Pennsylvania, Virginia, Tennessee, Ohio, South Dakota, and California. No single event was particularly troubling in isolation—but together they began to form a picture. An undercover police officer in Maryland had been sent to spy on nuns protesting the wars. Environmentalists showed up on terrorism bulletins in Pennsylvania. Public park cleanups and gatherings of animal rights activists were monitored by authorities in Virginia. The Ohio, Kentucky, and Indiana regional fusion center declared that terrorist sleeper cells thrived in diverse neighborhoods where there was “an easygoing attitude toward different cultures,” as if multiculturalism itself was a threat to the republic.
The increasing frequency of such events coincided with investment by police departments around the country in the kind of technology that overseas had helped elite military units find a single terrorist in a haystack of nobodies. Police departments also bought equipment to identify large numbers of people without their knowledge—the same kind that U.S. soldiers and spies used in foreign wars, hot and covert. These tools were not hard to find or purchase: the same corporations that developed and sold so many of these useful gadgets and software programs to American military antiterrorism squads were now setting their sights on places like Phoenix, Memphis, and Sioux City.
The transition from public to private commerce was easy for the corporate side of Top Secret America. With slick advertising and marketing, a variety of companies, large and small, presented themselves as the nation’s protectors, bragging about how their products had helped the U.S. Army identify insurgents, track clandestine activity, and disrupt all things criminal and nefarious. Their campaigns tapped into the patriotism and culture of vigilance inherent in law enforcement agencies, and also the feeling that they could become connected to the center of power, where the action was, by using these technologies.
L-1 Identity Solutions, for example, sells American police departments the same type of handheld, wireless fingerprint scanners used by U.S. troops to register entire Iraqi villages during the insurgency. Other companies sell local law enforcement authorities devices to detect the location of mobile phones, a technology used by troops and intelligence agencies in Iraq and elsewhere.
Thermal infrared cameras made by the FLIR Corporation, like the night vision devices it makes for the military, were deployed by police in several American cities. Those cameras could see through metal, alerting police to someone hiding, say, in the trunk or on the floor of a car. (They could even tell, by the heat signature underneath its chassis, whether the car had just been turned off.) In Arizona, the Maricopa County sheriff’s office purchased the sort of facial recognition equipment prevalent in war zones, using it to record some nine thousand biometric digital mug shots a month, many of them of illegal immigrants. And, just as soldiers in the field did when trying to keep towns free of insurgents, many American police departments purchased equipment allowing them to record images of license plate numbers belonging to every car going through toll booths and tunnels.
Such surveillance was especially intense around larger cities, especially those that had felt the direct impact of the 9/11 attacks. Soon, said authorities in the Washington area, everyone who drives into the nation’s capital will have his car tracked and recorded, a high-tech, invisible version of the so-called ring of steel that the British government imposed on London during the Irish Republican Army killings there in the early 1990s and broadened after 9/11. Lower Manhattan was the first U.S. location to set up a web of similar surveillance, a system also used by U.S. troops in Afghan and Iraqi communities.
That many of these advances were initially developed for use in foreign wars was a double bonus for some of the companies that had made them. Often the technology development itself had been partly subsidized by the government. Elite Special Operations units tasked with killing and capturing terrorists drove technological advances in rapid analysis, allowing operators to fuse biometric identification, captured computer records, and cell phone numbers to map the hierarchy of al-Qaeda and insurgent networks, making thousands of connections within minutes so commandos could launch surprise raids within hours. Here at home, the Department of Homeland Security and its state affiliates were increasingly enamored with the idea of using similar technology to collect photos, video images, and other personal information about U.S. residents in the hopes of teasing out terrorists.
As governor of Arizona (2003–9), Napolitano built one of the strongest local intelligence organizations outside of New York City. Now the public face of the administration’s aggressive efforts to capture and coordinate domestic information throughout the country, she wanted every state and local law enforcement agency to feed data to the FBI and to DHS. “This represents a shift for our country,” she told New York City first responders on the eve of the ninth anniversary of 9/11. “In a sense, this harkens back to when we drew on the tradition of civil defense and preparedness that predated today’s concerns.”
Her statement was startling because of the clear reference to the cold war, which also became the dark days of McCarthyism, when citizens were encouraged and pressured to turn in people they suspected of being Communist sympathizers. Back then, an obsessed and paranoid FBI had drawn up a black list, vacuuming up not only many Americans whose association with communism was tangential but also the names of countless individuals who had little or no sympathy for the doctrine or its practice. Without enough evidence to prosecute many of these suspects on espionage or sedition charges, the FBI instead ruined their careers and reputations. FBI director J. Edgar Hoover’s covert COINTELPRO (Counter Intelligence Program) sent undercover agents to disrupt and discredit political figures and groups it deemed subversive. These included civil rights leaders such as Martin Luther King Jr., organizations like the National Association for the Advancement of Colored People, and Vietnam-era antiwar protesters. As far as Hoover was concerned, the burden to prove innocence rested on the accused, not the accuser, a complete reversal of the system enshrined in the Constitution.
At a congressional hearing in March 2010, California Representative Jane Harman, a liberal advocate of stronger domestic intelligence, reminded colleagues about such past domestic abuses: “Let’s not fool ourselves. If homeland security intelligence is done the wrong way, then what we will have is… the thought police and we will be the worst for it.” The solution, she added, was clarity and openness, and neither existed as of yet. “We need clear definitions about what we are doing. We need transparency and a process to hold people accountable. We need to shut down what doesn’t work and we know can’t work. The rule of law must always apply.”
Transparency remains embryonic, but the federal-state-corporate partnership has produced a vast domestic intelligence apparatus that collects, stores, and analyzes information about tens of thousands of U.S. citizens and residents, many of whom have not been accused of any wrongdoing. It involves a web of 3,984 federal, state, and local organizations, each with its own counterterrorism responsibilities and jurisdictions, according to Arkin’s calculations. At least 934 of these organizations have been created since the 2001 attacks or reorganized since then; or they became involved in counterterrorism for the first time after 9/11.
Just as in other parts of Top Secret America, the effectiveness of these programs, as well as their cost, is difficult to determine. Since most of the money for these programs came from state budgets, information about spending on state security should have been accessible. But it wasn’t, or at least the records weren’t easy to find. Public accountability is limited to a handful of statistics. There is little disclosure about how things really work—and few people, if any, truly understand how the entire system is woven together. Local reporters were constantly frustrated by state offices that simply refused to explain how the state’s new intelligence center and data collection worked. Most of the time, none of it was classified top secret or even secret until it reached the FBI. But much of it was classified “law enforcement sensitive,” which meant it could be withheld from the public.
When Harman spoke of the need for transparency, she was also addressing government agencies themselves. For even the institutions at the core of Top Secret America are often in the dark. The Department of Homeland Security, for example, does not know how much it spends each year on state fusion centers, which bring together and analyze information from various agencies within a state. The DHS has given $31 billion in grants since 2003 to state and local governments for all kinds of projects, including fusion centers, a department spokeswoman said, but she said the federal agency doesn’t actually track all the programs the money is used for. Nor does it bother to track which programs are effective. At least four other federal departments also contribute to local efforts, but the bulk of the spending every year comes from state and local budgets that are too disparately recorded to figure out an overall total.
In addition to the new Department of Homeland Security and all the other organizations created in the wake of the 9/11 attacks, Napolitano, the FBI, the National Security Agency, and even the CIA, in a more limited way, had inherited a structure pioneered by the Bush administration that allowed for closer coordination not only against foreign terrorist networks but also against American citizens who appeared to authorities to be acting suspiciously.
This new normal, enshrined in the 2001 Patriot Act, was at the core of an all-out effort to prevent terrorism before it happened. To achieve this, the Patriot Act had taken a giant step by abandoning a measure enshrined in American law since 1978 to prevent any more COINTELPRO abuses: it dismantled the separation between criminal cases, which require a high standard of evidence of a crime to initiate, and intelligence investigations, whose goal is to obtain more information, not pursue criminal cases, and which therefore can be launched with much less solid information. The new Patriot Act unleashed the FBI once again, allowing it to spy more, use more informants, listen in on more conversations, infiltrate more groups, collect more email and voicemail, access and store more financial and personal records, and cross-reference more data than it had before. The justification needed for these investigations had to begin with more than a hunch, but could be based on considerably less than the kind of evidence that was required to justify such tactics before 9/11.
This meant the bureau was now cleared to gather information not for criminal prosecution in a court of law but to further its own understanding of how suspected groups and networks inside the United States operated. Formal preliminary investigations could be opened with less actual proof of wrongdoing than in the past. And while on paper racial profiling was banned, in practice it happened all the time. A man who looked Middle Eastern couldn’t be stopped for simply walking down the street, but he could be stopped for walking in front of a federal building several times looking curious.
With advances in technology and the right approvals, the government could also now capture a person’s digital exhaust, the revealing data a human being gives off in the course of daily life—when buying groceries or gas or beauty supplies, surfing the Internet, using a cell phone or ATM, flying from country to country or driving from state to state. This data could be married with biometric and law enforcement records, such as fingerprints and previous arrests, and stored on law enforcement servers, allowing officials to build and share lengthy profiles of both suspected terrorists and ordinary citizens who someone believed were acting suspiciously.
By allowing information about individuals not subject to a criminal investigation to be collected without their knowledge, the Bush administration had also weakened the safeguards written into the Watergate-era Privacy Act of 1974. The act guaranteed that, with the exception of ongoing criminal investigations, individuals could know what the government had collected on them, and it ensured that the information would not be inappropriately shared. Now all sorts of data and observation about private citizens circulated freely among the FBI, state and local police and fusion centers, and the Department of Homeland Security. The default assumption became to err on the side of the nation’s safety. Most citizens were not allowed to find out if their names were among the circulating files. This prohibition was to prevent them from being tipped off and modifying their behavior.
Napolitano was the first to publicly advocate a more aggressive use of these Bush-era revisions. Arguing that the wars in Iraq and Afghanistan hadn’t stopped the pipeline of terrorists into the United States, she in essence declared the Bush administration’s rationale for going to war in Iraq a failure. “The old view that ‘if we fight the terrorists abroad, we won’t have to fight them here’ is just that—the old view,” she told police and firefighters in a speech. The new problem, she explained, was “home-grown terrorists,” even though most of them were not homegrown at all but young immigrants from Somalia, Afghanistan, and Pakistan with easy access to inspirational jihadist websites and how-to manuals.
Napolitano and many in the Obama administration believed that the next iteration of terrorism to hit the United States would be attacks by disaffected immigrants. These so-called lone wolves were difficult to detect and stop. Their only hope was to turn ordinary citizens and hometown beat cops into informants for the FBI. The slogan she chose for her campaign was “See Something, Say Something.” The implication of Napolitano’s strategy was to extend the terrorism battlefield beyond the nation’s capital and its largest cities and into the American heartland, and to turn local law enforcement agencies into surrogates for federal counterterrorism teams. Ideally, citizens themselves would become counterterror informants, paying close attention to any anomalies they noticed.
From the beginning, determining an effective and proportionate response to the threat of terrorism was challenging for local and state authorities. In Tennessee in 2001, state officials canceled flights and banned parking within three hundred feet of airport terminals after airplanes were turned into fuel-filled missiles on 9/11. National Guardsmen with rifles appeared. So did canine patrols. Tennessee lawmakers gathered at the governor’s mansion for secret briefings on possible targets. They were advised to switch their own official license plates to ordinary ones so that they could keep a low profile. The Tennessee National Guard and military installations went on alert. Within three weeks, the governor appointed a retired brigadier general to head the state’s antiterrorism campaign and a new Homeland Security Council to coordinate the actions of a dozen agencies charged with protecting the state. Tennessee governor Don Sundquist promised the new effort would cost “very little” money. Emblematic of the economizing, the new Office of Homeland Security squeezed its tiny staff into the Veterans Affairs office downtown.
But frugality quickly gave way to fervor. Just eighteen months later, state politicians had coughed up millions for a new bureaucracy, and the federal funding floodgates opened. Today Tennessee’s Homeland Security Department alone has a staff of twenty-eight and an annual budget in the millions, and occupies two floors of the looming Tennessee Towers State Office building, as well as three regional offices. Forty-five terrorism liaison officers have been appointed to keep small towns and rural communities up-to-date on threats identified by Washington and elsewhere. The FBI opened three new Joint Terrorism Task Forces in Tennessee (in Nashville, Knoxville, and Memphis). All members of the JTTFs needed security clearances, some at the top secret level; such clearances were highly coveted. To transfer the classified information from FBI headquarters on Pennsylvania Avenue to Tennessee, workmen laid encrypted cable lines and built SCIFs in Memphis. In Tennessee alone, over a thousand law enforcement officers participate in counterterrorism and homeland security training, learning to stop possible terrorists, communicate with each other quickly, and respond to catastrophic attacks. Despite the expensive buildup, however, the reality in Tennessee and most other cities and towns across the country is that there just isn’t enough terrorism-related work to keep everyone busy. A sample day in Utah saw one of five intelligence analysts in the state’s fusion center writing a report about the rise in teenage overdoses of an over-the-counter drug. Another was making sure the visiting president of Senegal had a safe trip. Another had just helped a small town track down two people who claimed to be selling magazine subscriptions but who were pocketing the money themselves, a far cry from a national security problem.
In nearby Colorado, at the state’s Information Analysis Center, a few investigators were following terrorism leads, but others were looking into illegal Craigslist postings and online World of Warcraft gamers. In fact, the vast majority of fusion centers across the country have transformed themselves into analytical hubs for all crimes, from school vandalism to petty drug dealing. They use federal grants, handed out in the name of homeland security, to combat everyday offenses.
This overcapacity arose because after 9/11, local law enforcement groups did what every agency and private company did in Top Secret America: they followed the money. The DHS helped the Memphis Police Department, for example, purchase ninety surveillance cameras, including thirteen that monitor bridges and a causeway. It helped buy fancy video monitors that hang on the walls of a new Real Time Crime Center, as well as radios, robotic surveillance equipment, a mobile command center, and three bomb-sniffing dogs. All came in the name of river port security and protection to critical infrastructure such as bridges, dams, highways, and power stations.
Since there hasn’t been a solid terrorism case in Memphis yet, the equipment’s greatest value has been to help drive down local crime. Where the mobile surveillance cameras are set up, criminals scatter, said Lieutenant Mark Rewalt, as he spent a Saturday night scanning the city from an altitude of 1,000 feet. Flying in a police helicopter, Rewalt pointed out some of the numerous DHS-funded cameras below. The devices constellated the entire city; they were found in mall parking lots, in housing projects, at popular street hangouts. “Cameras are what’s happening now,” he marveled.
Looking at the Tennessee fusion center’s website, you’d never guess that there hasn’t been much terrorism around these parts. Click on the incident map, and the state appears to be under furious attack: red icons of explosions dot Tennessee, along with blinking exclamation marks and flashing skulls. The map is labeled “Terrorism Events and Other Suspicious Activity.” But roll over the icons and the explanations that pop up have nothing whatsoever to do with terrorist plots: “Johnson City police are investigating three ‘bottle bombs’ found at homes over the past three days,” reads one description. “The explosives were made from plastic bottles with something inside that reacted chemically and caused the bottles to burst.” Another told a similar story: “The Scott County Courthouse is currently under evacuation after a bomb threat was called in Friday morning. Update: Authorities completed their sweep… and have called off the evacuation.” Nine years after 9/11, this map is part of the alternative geography that is Top Secret America.
Among the millions of people now assigned to help stop terrorism is Memphis Police director Larry Godwin, and he has his own version of what that means in a city where there have been more than ninety murders in 2010. “We have our own terrorists, and they are taking lives every day,” Godwin said. “No, we don’t have suicide bombers—not yet. But you need to remain vigilant and realize how vulnerable you can be if you let up.”
Memphis wasn’t about to let up, not while dollars continued to flow from DHS and crime remained high. In addition to the surveillance cameras that monitor residents near high-crime housing projects, problematic street corners, and bridges and other critical infrastructure, the federal agency helped the city pay for license plate readers and defrayed some of the cost of setting up a crime analysis center for the city. All together, since 2003, it has given Memphis $11 million in homeland security grants.
“We have got things now we didn’t have before,” acknowledged Godwin, who has produced record numbers of arrests using all this new analysis and technology. “Some of them we can talk about. Some of them we can’t.”
What he is more willing to discuss is how everything an officer does out on the street—all warrants issued, arrests made, subjects apprehended—is automatically transferred to the Memphis Real Time Crime Center, a command center with three walls of streaming surveillance video and analysis capabilities that rival those of an army command center. The tentacles of Godwin’s operation reach deeply into the city’s streets and neighborhoods. I got a whiff of just how deeply while riding around a Memphis public housing complex with the police department’s ingenious database geek, a former cop named John Harvey who traded in his badge to become a tech guru.
Harvey, who retains arrest powers and a patrol vehicle, mostly to test and demonstrate his new gadgets, drove slowly past the cars in the parking lot in front of the rundown two-story town houses. Inside his vehicle, between the two front seats, Harvey had mounted the civilian equivalent of the Special Operations equipment used by the military—a computer that can ingest a person’s name and spew back a package of data on the individual collected from various government and commercial databases. With these new tools and money from state and federal grants, Harvey and local police officers like him are building ever more sophisticated localized intelligence systems. When officers were wasting time knocking on the wrong doors to serve warrants, Harvey persuaded the local utility company to give him a daily update of the names and addresses of customers. When he wanted more information about phones captured at crime scenes, he programmed a method of storing all emergency 911 calls (which often include names and addresses) that the police could later associate with a found phone or other data and documents. He created another program to upload new crime reports every five minutes and to mine them for phone numbers of victims, suspects, witnesses, bystanders, and anyone else listed.
Then he persuaded the department to buy seventy military-grade infrared cameras to mount on the hoods of patrol cars. Now, instead of having to decide which license plate numbers to type into a computer console in the patrol car, an officer can simply drive around while the automatic license plate reader moves robotically from left to right, snapping digital images of one license plate after another and then automatically running them against the computer databases. That allowed Harvey to drive through a housing project, or park at the side of a busy street, and just wait for one of the sounds he’d programmed into the computer to signal a hit on a license number: a “boing” for lesser offenses, such as driving on a suspended license or traffic violations; a gunshot or siren for more serious alerts, including convicted gangster, sex offender, felon, murderer.
When he got a hit, the tap of a key on the keyboard brought up the name of the car’s owner. Another tap brought up the owner’s criminal history. If there was one, he’d hit another key, and this was when it got weird: the names of all the other people in his database who shared the same address—family, friends—would pop up, along with past offenses, aliases, Social Security numbers. Sitting in the car with him, aiming the camera at a certain car or a certain apartment, I felt like a Peeping Tom, as though I were peering through the window of a house and could see the people inside.
Harvey explained that the depth of information was valuable for officer safety. “You want to know who you’re dealing with.” That made sense. The data could also be used to flush out a hunch, the kind police officers are trained to feel in their gut. That man stuffing something in his jeans—is he living with his mother, or is he a convicted drug dealer? Let’s watch what house he goes into and see if that address is in the database, and see if any convicted criminals live there, and if any of them are drug dealers.
The system can also track observations that may have no criminal connotation alone but which, when correlated, could be suggestive. An officer might wonder who owns the red truck parked under the bridge; by checking the system, he can see if it has ever been seen at another bridge before, and, if so, how many times. The more data captured, the more connections made.
As Harvey approached a parking lot, a young woman standing on the sidewalk noticed him. She walked around to the back of her car to block his view of her license plate. He stopped the car five feet from her and waited. She waited, too. He inched the car up a bit. She repositioned herself. He inched the car back. She moved again. “I could wait her out,” he said, as she stood with her back to him, now only three feet from his bumper. But he moved on instead.
On another night, a police patrol car rolled slowly through the parking lot of a large discount store. The camera on the hood snapped digital images of one license plate after another and analyzed each almost instantly. The officers inside the car waited for a hit.
Suddenly, a red light flashed on the car’s computer screen, along with the word warrant.
“Got a live one! Let’s do it,” the officer called out.
When an officer got a hit, he could pull over the driver, and, instead of having to wait twenty minutes for someone back at the office to manually check records, he could use a handheld computer to instantly call up eight databases. He might find a mug shot or a driver’s license photo, a Social Security number, the status of the driver’s license, traffic violations, past charges, aliases, any outstanding warrants, and even pawn shop sales—basically everything but fines on overdue library books. Such data came from throughout the state of Tennessee and was available to every law enforcement officer who had access to the databases.
The rationale for having such data available during routine traffic stops, as it applies to terrorism, is not to miss the next Mohammed Atta. Shortly before 9/11, three of the hijackers were separately stopped for minor traffic offenses. Atta, the leader of the operation who piloted one of the planes into the twin towers, was stopped and fined in Florida for driving without a valid driver’s license. He failed to pay the fine, and a warrant was issued for his arrest. When he was stopped again several weeks later, this time for speeding, he was let go because the officer was not aware of the outstanding warrant. Had a nationwide or even statewide integrated computerized system been in place, he might have been detained, and history might have turned out differently.
Linking databases is a major goal for people like Napolitano and Harvey. When the technology works correctly, people doing bad things get caught. But when it doesn’t work right, such reliance on technology can cause a different kind of problem, like the one encountered on the night I took a second ride with Harvey and two police officers.
I had joined Harvey, the two police officers, and Washington Post photographer Michael Williamson as the officers patrolled the streets with automatic license plate readers mounted on two patrol cars. They were looking for drivers with outstanding warrants and other infractions, and hoping for something more exciting than that. Alas, nothing dramatic was coming up. One of the most interesting hits, after several hours on the job, was the plate on a black car they saw in the discount store parking lot. Its registered male owner, said to be aged thirty-five, was wanted on three drug charges. The officers waited at a distance for him to come out of the store and get back into the vehicle, where they figured they would arrest him. After ten minutes, however, they were disappointed to see three teenage girls hop in the car instead.
Two other times the license plate readers beeped and the person driving the car was not the owner who had unpaid tickets but an elderly parent instead. Slightly confused and scared by the posse of police cars demonstrating a sudden interest in them, the drivers cooperated with every instruction given. (In both cases, they were poor African Americans with rundown vehicles in a town whose racial divide is more starkly apparent than in other more integrated cities.)
“It’s a target-rich environment,” Harvey joked as the patrol cars headed off. Even if the pullovers were false alarms, at least they provided a chance to enter more data.
The computer monitors began boinging again. The radio crackled.
“I’ve got five cocaine hits on this black Toyota,” one of the officers announced as they moved through another parking lot. “Let’s see who gets in the car.”
All the information from Harvey’s night on patrol, down to the last license plate number, was fed into the Real Time Crime Center back at headquarters. It was plotted on a map, along with information about the other cars stopped, warrants written, and arrests made that night and every night, to produce a visual rendering. This information would help analysts predict trends so the department could figure out what neighborhoods to swarm next with officers and surveillance cameras. These police sweeps, called Blue Crush by the Memphis Police Department, sometimes netted thousands of arrests, far too many for the local jail system to handle.
“They throw them out as fast as we put them in,” Harvey grunted. But that was still not the end of it, either, because the fingerprints from the crime records would also go to the FBI’s data campus in Clarksburg, West Virginia. There are ninety-six million sets of fingerprints in Clarksburg, including those of all military personnel and all U.S. prisoners, and including those of citizens of Saudi Arabia and Yemen, Iraq, and Afghanistan. It’s a volume that government officials view not as daunting but as an opportunity. In 2010, for the first time, the FBI, the DHS, and the Defense Department were able to search each other’s fingerprint databases, said Myra Gray, head of the Defense Department’s Biometrics Identity Management Agency, speaking to an industry group. “Hopefully in the not too distant future,” she said, “our relationship with these federal agencies—along with state and local agencies—will be completely symbiotic.”
At the same time that the biometric and fingerprinting staff are building their database in West Virginia, and the police department in Memphis is building its database, and the state of Tennessee and other states across the country are building their databases, in the nation’s capital, the FBI is building an even bigger, more powerful repository of information on American citizens and legal residents with an Orwellian name: Guardian.
The Guardian database is controlled by people who work in a top-secret vault on the fourth floor of the J. Edgar Hoover FBI Building on Pennsylvania Avenue, near the Capitol. Guardian stores the profiles of tens of thousands of Americans and legal residents who are not accused of any crime. Most are not even suspected of one. What they have done is appear, to a town sheriff, a traffic cop, or even a neighbor, to be acting suspiciously. The federal government defines a suspicious activity in a fairly loose way, as “observed behavior reasonably indicative of pre-operational planning related to terrorism or other criminal activity.” In fact, the very effectiveness of this database depends on collecting the identities of people who are not now known criminals or terrorists, and on being able to quickly compile in-depth profiles of them on the theory that someday in the future a nugget of information will come in that will clarify whether the person is or is not a threat. In this way, it is a giant dragnet with which the FBI hopes to snare some gold. Like any dragnet, it is bound to sweep up at least some of the innocent.
If the new Nationwide Suspicious Activity Reporting Initiative, or SAR, works as intended, the Guardian database may someday hold files forwarded by all police departments across the country in America’s continuing search for terrorists within its borders. That, certainly, is the hope.
In some places, citizens eagerly joined the fight. In Kentucky, the Office of Homeland Security released a free mobile phone application, created by the publicly held NIC corporation, that allows users to send suspicious activity reports immediately to police authorities with a description of the fishy person—represented on the app by a figure of a person fleeing—as well as a cell phone photograph, a map of the suspicious person’s location, the subject’s vehicle license plate number, the time of day, and a description of “the incident.”
FBI officials say anyone with access to Guardian has been trained in privacy rules and the penalties for breaking them. But time and again enthusiastic local police have used the suspicion of terrorism to collect intelligence on perfectly legal protest groups, which is exactly what got the FBI in so much trouble more than three decades ago. Without appropriate training and without clear privacy guidelines, bad things happen all the time, as they did in Pennsylvania after the state’s director of homeland security, a former Army Special Forces officer with years of experience overseas but none in U.S. law enforcement, contracted with a former New York City police officer to write intelligence bulletins.
The former New York police officer, Michael Perelman, had cofounded a nonprofit security organization called the Institute for Terrorism Research and Response. Three times a week, beginning in October 2009, ITRR sent its intelligence reports to 1,800 law enforcement and homeland security offices and to state employees’ email accounts. The group was supposed to monitor real threats to Pennsylvania’s critical infrastructure, resources, and special events. Instead, the bulletins reported on lawful meetings and protests of groups as varied as the Pennsylvania Tea Party Patriots Coalition, the Libertarian Movement, antiwar protesters, animal rights groups, and environmental activists dressed up as Santa Claus and handing out coal-filled stockings.
After the Philadelphia Inquirer discovered the existence of a Homeland Security Department sole source contract with Perelman worth $102,000 and some of his intelligence reports on lawful groups, the governor ended the contract and apologized, the legislature held hearings, and Major George Bivens, the head of the Pennsylvania Bureau of Criminal Investigations, revealed that he had complained about the reports but was unsuccessful in stopping them. “I would liken it to reading the National Enquirer,” he wrote in one email he gave to legislators. “Every so often they have it right but most of the time it is unsubstantiated gossip.”
State intelligence analysts and FBI investigators say they use Suspicious Activity Reports to determine, for example, whether a person is buying fertilizer to make a bomb or to plant tomatoes; whether she is plotting to poison a city’s drinking water or studying for a metallurgy test; whether, as happened on a Sunday morning in late September, the man snapping a picture of a ferry in the Newport Beach harbor in Southern California simply liked the way it looked or was plotting to blow it up.
Photographing the ferry had turned up in Suspicious Activity Report N03821, a local law enforcement officer noting that he had observed “a suspicious subject… taking photographs of the Orange County Sheriff Department Fire Boat and the Balboa Ferry with a cellular phone camera.” The confidential report, marked “For Official Use Only,” noted that the subject next made a phone call, walked to his car, and returned five minutes later to take more pictures. He was then met by another person, both of whom stood and “observed the boat traffic in the harbor.” Next, another adult with two small children joined them, and then they all boarded the ferry and crossed the channel.
All of this information was forwarded to the Los Angeles fusion center for further investigation after the local officer ran information about the vehicle and its owner through several crime databases and found nothing. Authorities would not say what happened from there, but there are several paths a Suspicious Activity Report can take. At the fusion center, an officer would decide either to dismiss the suspicious activity as harmless or to forward the report to the nearest FBI terrorism unit for further investigation. At that unit, the information would immediately be entered into the Guardian database, at which point one of three things could happen. The FBI could collect more information, find no connection to terrorism, and mark the file closed but leave it in the database. It could find a possible connection and turn it into a full-fledged case. Or, as most often happens, it could make no specific determination, which would mean that Suspicious Activity Report N03821 would sit in limbo for as long as five years, during which time many other pieces of information about the man photographing a boat on a Sunday morning could be added to his file: employment, financial, and residential histories; multiple phone numbers; audio files; video from the dashboard-mounted camera in the police cruiser at the harbor where he took pictures; and anything else in government or commercial databases “that adds value,” as the FBI agent in charge of the database described it. The FBI is even working on a way to attach biometric data, such as iris scans and facial images, to files. Meanwhile, the bureau will also soon have software that allows local agencies to map all suspicious incidents in their jurisdiction.
Traditional law enforcement channels are not the only ones taking advantage of Guardian. The Defense Department recently transferred one hundred reports of suspicious behavior into the system. Over time it expects to add thousands more as it connects eight thousand military law enforcement personnel to an FBI portal that will allow them to send and review Suspicious Activity Reports about people suspected of casing U.S. bases or targeting American military personnel.
As of December 2010, there were 161,948 suspicious activity files in the classified Guardian database, according to the FBI. These were mainly leads from FBI headquarters and state field offices. Back in 2008, the FBI also set up an unclassified section of the Guardian database so that state and local agencies could send in suspicious incident reports and review those submitted by their counterparts in other states. Some 890 state and local agencies have sent in 7,197 reports so that far; the FBI has turned 103 of those into full investigations. From those investigations have come five arrests, the FBI said. There have been no convictions yet, but FBI agents point out it can take years for an arrest to come to trial. An additional 365 reports, explained the FBI’s database manager, have added information to ongoing cases.
While the list of SAR success stories supplied by the FBI’s public information office filled a page, only a few were significant. Last year, the Colorado fusion center helped the FBI’s Denver office analyze information obtained through an FBI search warrant on Najibullah Zazi, an Afghan-born U.S. resident who was planning to bomb the New York subway system. The FBI was already furiously at work on the case and had used the Colorado fusion center’s databases to help it quickly understand the information it had obtained. Zazi was arrested before he could carry out his alleged plot (though not before another police department, New York City’s, got involved and unintentionally tipped him off to the investigation).
And in 2007, according to the FBI public affairs office, a Florida fusion center provided the vehicle ownership history that helped an ongoing FBI investigation identify and arrest an Egyptian student who later pleaded guilty to providing material support to terrorism, in this case transporting explosives. Some FBI agents said they would have eventually turned up the same information in both cases themselves, but were grateful for the help of their local counterparts.
“Ninety-nine percent doesn’t pan out or lead to anything,” said Richard L. Lambert Jr., the special agent in charge of the FBI’s Knoxville field office. “But we’re happy to wade through these things.”
In practice, most SAR reports—and the names of the people included in them—don’t go anywhere; they remain in the uncertain middle and just sit in the database, which feeds into the debate over the privacy implications of retaining so much information on U.S. citizens and residents who have not been charged with anything.
Most of the FBI agents who have doubts about the system won’t publicly say so, given that their views are contrary to official policy. But if you asked the question another way, whether more terrorism cases come about as a result of this digital dragnet or from more focused, old-fashioned agent work, the agents responded like Richard A. McFeely, special agent in charge of the FBI’s Baltimore division. Talking about the Baltimore suspect in the attempt to bomb the military recruitment center, I asked him whether the new technology had helped crack the case.
“This was good, old-fashioned police work by a lot of different police agencies coming together.”
“Okay, so not so heavy on the technology?” I asked.
“That’s correct.”
Still, McFeely defended the large database. “We need it because you never know,” he said. “And it’s that one question mark that is out there.”
We need it because you never know is the answer to so many questions about the size, expense, and effectiveness of Top Secret America. But is that really an answer? “You never know” was the same as saying that all the spending, all the effort, even all the waste was worth it because, well, it might stop one attack. Nowhere else in American life has this kind of logic been an acceptable answer, except perhaps during the cold war, when a first strike by the Soviet Union could have resulted in mutual obliteration.
In every other arena, more rational cost-benefit calculations prevail. The government isn’t deploying a million people and spending hundreds of billions of dollars to stop illegal drug sales and use, even though many more people die each year from drug-related violence than die in terrorist attacks, and authorities know for certain that at least as many will die from drug-related violence the following year. Most people drive cars, even though 24,474 of them died in auto accidents in 2009. Parents don’t keep their children locked at home all day because they might be killed—even though 1,096 children were murdered in 2007 alone.
But if someone is taking pictures of a bridge in some city and a citizen reports it, it will probably end up in the FBI’s database, said Lambert. If there’s no other information connecting any of that to even a whiff of something suspicious, “that name will lie dormant there” until the same person “at a later time takes a picture of another bridge across the country or starts taking pictures of the gates at Langley [CIA’s headquarters].” Explained Lambert, “Unless we have the ability to go back and look at that [information], we can’t do this type of what we call predictive analysis.” If the American public is worried about the privacy implications, “my message back, I guess, is that you really can’t have it both ways…. We are very careful… but if we want to get to the point where we want to connect the dots, the dots have to be there. And if we’re being told that the dots have to be erased every time we have contact with a dot and there’s no derogatory information there,” the FBI will never be able to forecast an attack.
Other democracies—Britain and Israel, to name two—are well acquainted with the sort of domestic security measures that have become increasingly commonplace in Top Secret America. But for the United States, the sum of these activities represents a new level of governmental scrutiny of its population. Nonprofit organizations like Secrecy News, the Electronic Privacy Information Center, and the American Civil Liberties Union (ACLU) have worked around-the-clock to counter the tidal wave of new attempts to collect and mine data on Americans. Congress has held briefings, but they were never well attended. People seem not to notice the incremental changes taking place across the country, the eroding of privacy and the tabulation of personal information in government hands.
Charles Allen, who left the CIA to accept the job of leading DHS’s new intelligence arm, has his own misgivings about the dragnet nature of the SAR system, and he knows other senior intelligence officials who are also skeptical. “It’s more likely that other kinds of more focused efforts by local police will gain you the information that you need about extremist activities,” he concluded.
Besides that, “You make mistakes,” Allen said. “You put information in that’s fallacious, because in the world of terrorism, counterterrorism, false reporting, exaggerated reporting seems to be a norm.” As a result, he added, “I found that we were reacting, overreacting to virtually everything.” Overreaction doesn’t just end up wasting time and money; it also undermines morale, vigor, and credibility. American intelligence has long been haunted by its struggles to make wise connections within the enormous amount of information it has on hand. In the end, technology is only as good as the people who use it.
As with so much in the new world of Top Secret America, everything about this nationwide, technology-assisted, database-driven, distended dragnet of names relies on local police department employees to make the right decisions. The ones I met were enthusiastic, for sure. They wanted to dive into the toughest cases. In heartfelt testimony, police association representatives told Congress that they needed funds to develop ties with foreign intelligence services. They wanted security clearances and access to more and more classified information. Many departments longed to be like the New York City Police Department, which routinely sent investigators overseas when attacks occurred to look for links to Big Apple residents and terrorist networks.
Even in the smallest local police agencies, excitement over the post-9/11 national security mission was palpable. Idaho State Police sergeant Russell Wheatley, who manages the state’s intelligence fusion center, did not have any international problems to investigate. The closest he got was arresting violent white supremacist and survivalist groups. Timothy McVeigh and Terry Nichols were never far from his mind. But he was ready to join the bigger fight against global terrorism anytime. Sitting in his squad car one day, his enthusiasm bubbled over. “It kinda gives me a chill to think that something a state trooper does will someday evolve into something that has to do with national security.”
Most of Wheatley’s colleagues had little training in terrorism analysis. They weren’t FBI agents. Instead, they were often people like Lacy Craig, who was a police dispatcher before she became an intelligence analyst at Idaho’s fusion center. Or they are like the detectives in Minnesota, Michigan, and Arkansas who can talk at length about the lineage of gangs or the signs of a crystal meth addict, but don’t know the difference between a Shi’a and a Sunni Muslim. Yet these days, they are terrorism analysts, too.
“The CIA used to train analysts forever before they graduated to be a real analyst,” said Allen, the former top CIA and DHS official. “Today we take former law enforcement officers and we call them ‘intelligence officers,’ and that’s not right, because they have not received any training on intelligence analysis.”
State fusion center officials say their analysts are getting better with time. “There was a time when law enforcement didn’t know much about drugs. This is no different,” said Steven W. Hewitt, codirector of the Tennessee fusion center, considered one of the best in the country. “Are we experts at the level of [the National Counterterrorism Center]? No. Are we developing an expertise? Absolutely.”
Becoming expert is only partly helped by the quantity and quality of information the fusion centers receive from Washington. DHS’s daily reports were meant to inform agencies about possible terror threats. But to some officials they seemed like a never-ending stream of random details—vague, alarmist, and often useless.
We reviewed nearly a thousand DHS reports dating back to 2003 and labeled “For Official Use Only” that confirmed that view. Typical is one from May 24, 2010, titled “Infrastructure Protection Note: Evolving Threats to the Homeland.” It tells officials to operate “under the premise that other operatives are in the country and could advance plotting with little or no warning.” Its list of vulnerable facilities seems to include just about everything: “Commercial Facilities, Government Facilities, Banking and Financial and Transportation…”
As Harvey, the two police officers, the photographer, and I traveled the streets of Memphis looking for problems, during one stop I was reminded just how all-powerful the local police can be as a seething tension quickly emerged between the officers and a young woman who had done nothing wrong at all.
It happened after the officers spotted a car several blocks away that fit the description of one they had been searching for since the week before. They drove over to the vehicle as it slowly made its way down a residential street of small, modest homes. As they drew closer, they turned on their flashing lights and the motorist, now at nearly a crawl, came to a stop. But instead of waiting in her car for an officer to approach as instructed over the megaphone, the driver opened the door and jumped out. “What’s the matter?” she asked them.
Get back in your car, she was ordered.
Instead of immediately complying, the woman stiffened. “Why? What’s this about? Why are you stopping me?”
“Ma’am, just get back in your car.”
“Why? What did I do? This is my neighborhood.”
“Ma’am, please just do as I say.”
She had angled her car as if it were about to go down the driveway of the house where she had come to a stop, which happened to be her elderly father’s home. He was now walking briskly toward her.
“Honey, just do as the man asks,” he pleaded with his daughter, who was having none of it.
“What did I do? Why are you pulling me over?” she demanded, as one of the officers focused intently on his handheld computer while the other, hand on his gun, chest slightly thrust forward, told her again: “Get back in your car.”
She looked at me and the photographer, trying to figure out what was going on and getting more incensed as the situation sank in.
“What did I do?” she demanded.
“Obstructing a lane of traffic,” the officer finally said.
“What! This is my street. This is my house. Why are you doing this? I work for the city of Memphis—”
“Honey, just let the officers do their job,” her father called out, clearly worried about the escalating tone of the argument, as the unequal balance of power became more and more obvious and ominous. They could make her life miserable right now. I had a powerful urge to tell the woman what they should have told her right away, that they had been looking for a car just like hers but had obviously made a mistake.
“Please give me your driver’s license,” the officer demanded. She handed it over.
“Is it valid?” he asked.
The young woman exploded. “Of course it’s valid! I work for the city of Memphis. I work hard. You gonna write me a ticket? For what? Being in my own street. Why are you doing this? Why do you think it’s okay to pull over a black person like this?”
“Whoa, whoa, whoa, you callin’ us racist?” the officer demanded.
“Why are you doing this?” she tried again. “Now I’m going to have to spend a whole day off work fighting this. Why don’t you go find some real crime? There’s plenty of it in this city. Why are you doing this when nothing happened? This isn’t right. Why do you think you can just come and do this?!”
“Just calm down,” the officer ordered.
“Honey…” her father tried again.
“Man…,” she said under her breath, trying her best to maintain calm.
He wrote out a ticket and handed it to her. It would cost her a half-day’s wages or at least a day off to fight it.
“Okay, here,” he said, handing her the ticket. “You can pay this downtown.”
“One hundred twenty-five dollars!” she yelled out. “Ahhh! For what? Being in my own neighborhood?”
“Good night, officers,” her father called out. “Thank you.”
The officers were already walking back to their cars. The police car doors closed. As they pulled away, a small crowd of family members and neighbors gathered around the woman.
“Can you believe it? Did you hear that?” one of the officers said as the police car rolled down the quiet street. “She called me racist. I can’t believe it…. Well, we just made her day.”