The Silo Effect

by Gillian Tett

  But in the early years of the twenty-first century, or just as Goldstein was wondering what to do with his life, the Chicago police force underwent a bout of upheaval following significant incidents of brutality and corruption in the police force, which managed to stand out in a city whose history was littered with tales of corruption, violence, and other wrongdoing. In 2007, the superintendent of the police, Phil Cline, resigned. And in a bid to provide a fresh face, the city leaders decided to recruit the first outsider seen for fifty years into the role of police superintendent: Jody Weis.5

  Weis was appointed into his $310,000-a-year position on February 1, 2008. He looked like an archetypal policeman; indeed, he might have starred in a Hollywood cop movie, since he was broad-shouldered, with a chiseled face and wide-set eyes. But he had spent his twenty-two-year career with the FBI, not police, based in places such as eastern Pennsylvania. Many local police officers thought that made him unsuited to the job. “The Fraternal Order of Police, the powerful Chicago Police union, hated Weis before he got in the door,” Locke Bowman, the director of the Roderick MacArthur Justice Center, later observed. “The last thing the FOP has ever wanted is any outsider—it doesn’t matter who—meddling in police business.”6 But Weis preferred to think of this as an advantage. “Coming from the outside I didn’t really have any particular ties to the police department or really the city [and] I think that allowed me to take a very objective look at the department,” Weis observed. “I recognize that the Chicago Police Department is a very historic and proud organization, so I certainly wasn’t going to make it, you know, very FBI-centric. But I thought there were some best practices we could apply from the FBI.”

  In particular, Weis had some distinctly different ideas about how the police should be run. During most of his career at the FBI, Weis had worked in a world marked by extensive silos. Just as the different departments of a company such as Sony or UBS tended to hug data to themselves, so too the different sections of the police, FBI, or CIA tended to hoard information, out of a mixture of defensiveness, suspicion, or tribal rivalry. This created informational logjams, sometimes with disastrous results. One well-known example of this arose in the American intelligence services in the run-up to the 2001 attacks on the World Trade Center: several different parts of the American intelligence forces had received signals that al Qaeda was planning some attack, but did not take a coordinated approach to fighting the threat because different pieces of information were held in different corners of the bureaucracy. Nobody was able, or willing, to connect the different signals to get an overview.7 However, the problems with silos that were seen in the CIA before 2001 were repeated multiple times across other parts of the intelligence and security worlds, including the FBI. “When I was a young [FBI agent] one of my friends used to say: “You know what? If we wrote all of our intelligence paper on toilet paper, at least it would be of some value,” Weis observed. “We would have hundreds and hundreds of hours of taped conversations between a bad guy and one of our undercover agents, or a bad guy and one of our sources, and the tapes would sit in a box. Nobody would make use of it.”

  From time to time, police chiefs had tried to break down this fragmented bureaucratic culture. In New York, Police Commissioner Bill Bratton developed an innovative form of policing while he was running the NYPD in the 1990s, which tried to combine proactive community measures with crime fighting. This was called the “broken windows” approach to policing, since Bratton was convinced that policing could not be effective unless tough enforcement was carried out alongside initiatives to make communities stable and united; if the windows in the street were broken, this signaled that nobody had much sense of shared responsibility to the community, making it harder to fight crime. Bratton believed that a community needed to feel a sense of pride in itself to become safe. If policemen wanted to reduce crime, the argument went, they needed to care about fixing broken glass windows as much as simply making arrests. That required collaboration. It was essential to take a joined-up approach, Bratton argued.

  But Weis thought there was another way to combat crime: use better types of information flows. “When I was in the FBI, we realized that we had to change the way we do business after 9/11,” he explained. This meant that the different agencies which had traditionally operated as suspicious rivals needed to collaborate, to pool their data, contacts, and tips. During the time that he was running the FBI in Philadelphia, Weis had overseen a project to create an interactive live computer system that enabled the different agencies, such as the FBI, CIA, and police, to talk to each other about criminal and terrorist threats. “We took all our sources of information, plotted it on a map so we could show where we had coverage. And if there was no source coverage then we would try to target and develop sources, so that if something happened like a shooting we could go to our sources and say ‘You know, [somebody] just got shot, we need to find out who did it, so start working the streets,’ ” Weis recounted. “We would get all the outstanding warrants, all the sex offenders, various census data, and we could lay more and more data sets across the maps. So if there was a child abduction, and we knew it happened in this location, say, we could immediately see if there were a lot of registered sex offenders in that area.”

  To anyone outside the world of security operations, this data-sharing might have seemed like an blindingly obvious step. And in some parts of the security apparatus, other officials were trying to develop versions of this idea. Most famously, Bill Bratton, for example, in the 1990s had pioneered a policy known as compstat, which tried to take a much more rigorous approach to tracking crime statistics.

  But many of the long-serving police officers and FBI agents hated the idea of this holistic approach to managing security risks. They did not like the idea of changing their long-established practices or losing control of “their” data. “Even ten years after 9/11 some of my closest friends [in the FBI] were still saying that we shouldn’t do all this terrorism and intelligence stuff,” Weis admitted.” But he was convinced the security forces in America had to change how they operated: the world around them was becoming so interconnected and fluid, nobody could afford to stick in just one professional box. The FBI and CIA needed to break down their silos. So did the police, particularly in forces that were very large, like Chicago’s.

  IN THE MIDDLE OF a sweltering August in 2006, at the age of thirty-one, almost three long years after submitting his application, Goldstein reported for training at the Chicago Police Academy, a squat gray building on the city’s West Side. He was filled with idealistic dreams of making the world a better place. The reality of being a police recruit, though, was a shock. On the first day, he was put into a pack of other new recruits and marched into a cafeteria. “They are barking orders at you,” he recalled. “You are broken into what they call ‘home rooms’ and you’re marching single file, with this quasi-military organization and they are talking about uniforms. Stuff I know nothing about. You have all this gear you are supposed to get, shorts, T-shirts, shoes that have no logos! No logos!” Goldstein had never given much thought to whether sneakers had logos or not. In the trendy tech start-up world where he had worked there were no official rules about what anybody wore. However, sneakers were widespread and somehow it seemed that shoes always had logos. In the CPD auditorium, though, Goldstein realized he was standing in a world with different cultural rhythms; his idea about what was normal no longer worked. Suddenly he was forced to notice assumptions he had never pondered before.

  It was a brutal immersion. After marching in file, Goldstein’s unit was ordered to conduct a time-honored maneuver known as a “lean and rest,” a procedure where recruits were frozen still in a push-up, motionless, for long periods of time, while they were inspected. Goldstein could not see what the point of the ritual was; to him, the exercise seemed utterly useless. But the police considered the lean and rest a crucial part of the training; it was just there, part of the bonding experience. “That first week of August was exce
ptionally hot and there wasn’t any AC. So you are in your lean and rest, sweat dripping from your head and creating a pool under you. But you know if you break your position you’re going to get yelled at. And, then, just when you think you’re done, then you have calisthenics, which is jumping jacks, push-ups, and it just goes on and on. Then they take you on a run, and you are doing this exercise on this uphill driveway thing where you are lifting one of your classmates and carrying them up.”

  That night, Goldstein hobbled home, dazed and aching. He threw himself into an iced bath and gobbled down 800 mg of ibuprofen. “I just thought to myself: ‘So I quit my job for this? What have I done?’ ” But, the next day he returned, clutching ibuprofen, and went through the routine again. Then again, and again. He was older than almost all of the other recruits, and had far more academic qualifications. The course work seemed childishly easy. But it was the physical training that he struggled with. So he pushed himself to do endless runs and spent hours frozen, motionless, in a lean and rest, trying not to ask, “Why?” Then he was given a gun. He set about shooting at a target, with the same analytical intensity he had applied to all other areas of his life. “I actually found I was good at that,” he later recalled, with surprise. “I shot fourth out of the hundred people in my class.”

  The next phase of his training was even more brutal. Goldstein graduated from his class as Valedictorian, an honor which gave him the right to choose what part of Chicago he wished to work in. The range of options was very wide. Many parts of Chicago are safe and quiet, particularly around the wealthy suburbs. But Goldstein reckoned he needed to be near the action to learn the job. So he shunned the suburbs and asked to join a patrol in the 11th District, an area tucked into the West Side. This was one of the most violent, gang-ridden districts of Chicago. “I didn’t grow up in an urban environment, I grew up in the suburbs. So my first day on the West Side was—oh my God. There is drug dealing, there’s violence, there’s shooting, there’s everything.” He tried to bond with the other officers. But they were suspicious of a recruit who was obviously wealthier and better educated than most cadets. Eventually, during one lunch hour, Rod Gardner, assigned as Goldstein’s training officer, furtively pulled him aside in the gym in the basement of the police headquarters in the 11th District.

  “You know everyone thinks you’re a Fed [plant], right?” Gardner said.

  “Huh?” Goldstein replied. Chicago had a long history of corruption and plants, but it had never occurred to him that he might become entangled in that.

  “Yeah, everybody thinks you’re undercover FBI,” Gardner told him. “You learn exceptionally fast, you never ask questions about paperwork, you’re quiet, you’re older, and you’re professional. Everybody thinks you’re undercover FBI.”

  Goldstein was startled. However, he took it in stride. And as the months passed, and Goldstein continued to patrol his beat with the other police officers, he could sense that the experience was slowly changing him. Some days, he sometimes wondered if his whole adventure was as pointless as the lean and rest. On others, he realized that he was learning about a new world. A few years earlier, he had assumed that violence and poverty were something that happened to other people. To him, being normal was living in a calm, safe environment where children went to school and entrepreneurs made money by designing brilliant apps or websites. Normal was a place where everyone wore athletic shoes with logos. But viewed from a squadron police car in Chicago’s West Side, Goldstein could see that his former life was not the rule, but the exception; most people die not live like that. He had gained a bigger vision. His assumptions were changing too.

  Almost exactly three years after he had first turned up for training, in the summer of 2009, an incident happened that highlighted this sense of change. By then, he had spent innumerable hours with his police officers on patrol, moving around the streets of the 11th District. He no longer felt that they considered him a spy; he was adapting in that world. But he still wondered whether he would ever be a “proper” policeman; did he have the right instincts to react properly in a fight? Then one day, on a July day, Goldstein was in his family car with his pregnant wife and one-year-old son, on the way to get an ice cream when he saw a gang member pull out a gun and start shooting at the car in front of him. Shots were fired, and Goldstein realized that the gunman had just killed someone. Three years earlier, Goldstein would have fled the scene to protect his family, and then have called the police. But instead he slammed on his brakes, grabbed his gun, leaped out of the car, and ran toward the gunman, chasing him down the street and into an alley, where he managed to disarm him and conduct an arrest. “People say that time slows down during these things. Not for me. Everything just happened exceptionally fast. I went into an alley with someone who was armed and had just killed someone and I didn’t get shot,” he recalled. “Everything was training, because certainly if I had to think about it, what is the right choice in a situation like that? My pregnant wife is in the car, and the dude is shooting, so what do you do? It was instinct.” After the incident, he tracked down the family of the murder victim and discovered that he was a nineteen-year-old called Jeff Maldonado Jr., an aspiring black rap musician. The killer, Marcelina Sauseda, appeared have fired his gun as revenge for an earlier gangland fight. However, Maldonado’s family insisted that their son was not involved in gangs. He was attending community college and was simply caught in the crossfire while hitching a lift in an acquaintance’s van.8

  Goldstein subsequently received multiple awards for bravery.9 But he knew he had been lucky to have survived. He would never look at crime statistics in quite the same way again. Suddenly murder felt very personal.

  BY THE MIDDLE OF 2009, Jody Weis, the new head of the Chicago police, was feeling frustrated. He had arrived in Chicago determined not just to clean up the image of the Chicago police—but also, above all, to cut the city’s sky-high murder rate. But he faced a fearsome battle. In 2008, the first year of Weis’s command, the murder rate went up. “More people have been murdered in Chicago this year than in New York—even though New York’s population is three times greater,” the New York Daily News observed in shock in late 2008 after relatives of Jennifer Hudson, a well-known Oscar-winning actress and singer, was murdered on the city’s South Side.10 Weis blamed this soaring violence on the demise of the “special operations section” or SOS, which had been disbanded when it was mired in the scandals that preceded his arrival in Chicago. The SOS was designed to provide rapid response mobile teams of police experts that could rush to any place that violence erupted, and it had been lauded for its effectiveness in tackling crime. Unfortunately, a number of its then members had engaged in criminal activity themselves. Now, as 2009 got under way, he quietly reinstated the unit back onto the streets, under a new name, Mobile Strike Force.11 Weis thought that this policy was the only effective way to quell the worst gang fights. “If there is a gang shooting, you know there’s then going to be retaliation—and with Special Forces we had hundreds of officers that we could saturate an area and stop retaliation and keep everything calm.” The murder rate then fell slightly in 2009. By most measures, though, the death toll remained shockingly high. “When Chicago is compared to its big-city brethren, Chicago’s per capita murder rate is double that of Los Angeles and more than double that of New York City,” an editorial in the Chicago Sun-Times declared.12 “Violent crime is clustered most intensely on Chicago’s West and South Sides, terrorizing law-abiding citizens.” Indeed, the killings were so bad that from time to time local politicians would ask whether Chicago needed to deploy the National Guard onto the streets.13 Meanwhile Weis himself repeatedly faced questions about whether the city was spinning out of control. He insisted it was not. But with every month that passed, fresh stories broke about gangland killings, and the political impact of these was magnified as a blizzard of gory pictures and videos was disseminated across the Internet. “[This murder rate] is unacceptable in an American City, and it is totally unac
ceptable in a city as great as Chicago,” Weis declared, deploring parallels with a war zone. “We are not ‘Ch-iraq.’ ”14

  As the political pressure mounted, Weis cast around for new ideas. And in the summer of 2009, Weis’s chief of staff—former U.S. marine Michael Masters—tossed out a novel thought. Masters had briefly met Brett Goldstein in the mayor’s office when the recruits graduated from the Police Academy, and Goldstein had revealed to Masters that he was hoping to connect his dream of using his tech background to change how law enforcement used data. So maybe, Masters suggested, it would pay to ask Goldstein to develop some computer models to work out what was causing the murder rate to rise. Weis summoned Goldstein to hear his ideas. “We sat down, we chatted,” Weis recalled. “Brett is a little unusual personality, kind of a dry sense of humor, a little awkward, like your typical nerd.” Goldstein explained to Weis how he had once used advanced computer modeling techniques and sophisticated mathematics to help the OpenTable Internet site to connect customers to restaurants. That project had required a lot of tracking, modeling, and analysis and it did not seem obviously connected to the task of combating gangland murders. But maybe, Goldstein suggested, he could transpose the computer analytics he learned at OpenTable and the University of California to the police. If algorithms could work out which restaurants were hot, and match these to customers, maybe they could also track crime patterns.

  The Chicago police had never done anything quite like that before. At the start of the twenty-first century, they had tried to create more transparency about crime patterns by copying some elements of New York’s compstat platform (the statistical tracking system).15 But this did not have a big impact on the force. The police were used to chasing criminals with guns, not computers. However, Weis had seen what computer analytics could do at the FBI in Philadelphia. So he ordered Goldstein to move from the 11th District into a small, windowless office on the third floor of police headquarters. Goldstein installed a couple of old computers, collected all the crime data that he could, and started crunching the numbers with the help of some fellow technology experts at Carnegie Mellon University. As he had once done with restaurant orders, he was looking for patterns and connections in the data, in the hope these might predict future trends. Was there a rhythm to gangland shootings? A particular place and time when murders tended to occur? Then he plotted the death reports onto a big computer screen, alongside other data on violence, and then looked for other factors that might be correlated to these tensions. There was a folk belief inside the police that crime increased during a full moon or when the summers were hot, or when the winds were stiffly blowing in the city. So Goldstein compared data on past murders against the lunar cycle and temperature records, and then against the wind. This analysis showed little correlation between the moon and the murder rate. Nor was there much evidence that murders increased in heat or wind per se. But when the temperature moved by more than 15 degrees—say from 65 to 80—in a short period there was a big jump in crime. Conversely, when it became extremely hot, say above 90 degrees, crime fell.