BEBETO MATTHEWS/ AP PHOTOMining social media has helped prosecutors deal with gang violence and get offenders and guns off the streets.

On December 5, 2011, an East Harlem gang member nicknamed “Ice” posted a Facebook message to a fellow gangster: Don’t be “stingy wit dat bitch.” Ice wanted “dat bitch 2 give at least 2 more nigga hiv . . . ASAP like dis month.” Ice was not a pimp. The bitch in question was a gun, to be used in a gang war that was tearing apart three East Harlem housing projects and their surrounding areas. From October 2009 to April 2013, the turf, covering less than half a square mile, saw 66 nonfatal shootings and seven homicides. Residents learned to look for cover if two groups of young males started congregating on opposite sides of a street.

Then the shootings all but stopped. From April 2013 to April 2014, only three took place in that same area—an 85 percent drop from the annual average over the previous four years. The sudden reduction in violence was the outcome of a gang conspiracy indictment, filed by the Manhattan district attorney in April 2013, that induced an unprecedented 62 gang members, most still in their teens, to plead guilty and accept long prison sentences. That indictment, in turn, was the product of the Manhattan D.A.’s effort to bring prosecution into the intelligence-driven crime-fighting era. The East Harlem gang takedown is part of the moment when prosecution—at last—starts to get smart.

In 1994, New York police commissioner William Bratton, Mayor Rudolph Giuliani’s first police chief, revolutionized policing by declaring that the NYPD would prevent crime, not just respond to it. This new proactive mission required the department to analyze crime data minutely and to refine tactics constantly, as new intelligence came to light. The police began holding themselves accountable for public safety, measuring success not by arrests but by crime declines. Bratton’s revolution, known as Compstat (short for Computer Statistics), quickly spread to other police departments across the country.

Prosecutors’ offices, however, remained in a purely reactive mode. As lawyers, prosecutors should be the analysts and thinkers of the criminal-justice system, says Elizabeth Glazer, a former special counsel to the New York State attorney general and now New York mayor Bill de Blasio’s criminal-justice coordinator. “Instead, they’ve turned themselves into assembly-line workers, cabined in the case, missing the opportunity to think as creatively about fighting crime as the police do.” Viewing their mission as doing justice in the individual cases that the police bring to them, district attorneys rarely consider the effect that an indictment might have on broader lawlessness, or how a defendant fits into the criminal landscape. Vital information about offender networks gleaned in the course of preparing a case for trial usually remains on a prosecutor’s legal pad, rarely conveyed back to the police or shared with other prosecutors. With few exceptions, such as then–U.S. attorney Giuliani’s strategic attack on mob influence in New York in the 1980s and 1990s, prosecutors have gauged success by convictions, not by crime declines.

San Francisco’s district attorney can attest to the relative backwardness of prosecution. George Gascon ran the Los Angeles Police Department’s Compstat program in the 2000s before eventually becoming San Francisco’s D.A. “I knew when I was coming in to prosecution that there would not be the level of information that the police have,” he says. “But I didn’t know how little information there was available to me; it was shocking. The other side of the law enforcement house is nowhere near where policing is today; it is still stuck in the culture of one case at a time, one set of facts, without a lot of ability to link the dots.”

Manhattan district attorney Cyrus Vance aimed to upend that reactive mind-set when he took office in 2010. The D.A.’s mission was henceforth to lower violent crime, he declared, embracing the Compstat ideal. Vance then created a new entity called the Crime Strategies Unit, the sole purpose of which is to gather and deploy intelligence on Manhattan’s crime patterns and serious offenders. The unit’s five senior assistant district attorneys, assigned to one of five geographical areas in Manhattan, no longer prosecute cases but instead learn everything they can about who is driving crime in their jurisdiction. Based on daily communication with local police commanders and precinct field-intelligence officers, the Crime Strategies Unit has compiled a database of Manhattan’s most significant criminal players and other persons of interest (such as elusive or uncooperative witnesses). If any of those individuals—now numbering about 9,000—gets arrested anywhere in the city, one of the five senior CSU attorneys is immediately alerted, day or night. The attorney will then contact the local D.A. assigned to the case—whether in Manhattan or another borough—and make sure that the defendant does not slip through the cracks.

The arrest-alert system embodies the recognition that a defendant’s official history of arrests and convictions may fail to convey his position in the criminal food chain. A 16-year-old gang member from Harlem may be responsible for numerous shootings in his neighborhood, as attested to by his Facebook page, but never have been arrested for any of them because his victims and witnesses refused to cooperate with the police. Or he may himself have been shot by a rival and refused to assist with the investigation. He may be the prime suspect in several robberies, for which the police have yet to develop probable cause to make an arrest. If he winds up nabbed for shoplifting in SoHo, the misdemeanor prosecutor will have, at most, a few minutes to decide whether to pursue the charge. Seeing simply a petty vandal, the charging attorney would likely let him walk free. But armed with the intelligence that the CSU has gathered on the gangsters, the attorney can try to bump up the charges and can argue to the judge for high or no bail pretrial. Meantime, if detectives from the precinct where the gang member is suspected of committing his most serious crimes have asked to be included in the CSU’s arrest-alert system, they can pay him a visit in jail and see if he will now cooperate with them.

After Hurricane Sandy, the Crime Strategies Unit received an arrest alert for looting and burglary on Staten Island. The suspect: a robbery recidivist from Manhattan’s Lower East Side. The unit gave the Staten Island prosecutor the suspect’s criminal history and a screenshot of his Facebook page, which declared his intent to go to Staten Island and loot after the storm. A case that might otherwise have been dismissed was instead treated seriously and prosecuted. In 2012, police arrested a leading gang member in East Harlem for running toward people in a brawl brandishing a metal lock tied into a bandanna. The defendant had been shot in the past and had also likely witnessed a homicide, without cooperating with police after either crime. The attempted assault would ordinarily have gone nowhere, had the CSU not closely tracked the assailant. Instead, the prosecutor indicted him for criminal possession of a weapon in the third degree—a felony charge. After the charges were read, the defendant absconded while on bail. Arrested a second time on the open warrant, he was eventually sentenced to two to four years in prison on the weapon charge and two to four years on the bail-jump charge—outcomes that would have been unthinkable but for the information that the CSU had developed.

The arrest-alert system is also based on the theory that a small number of individuals commit a disproportionate amount of crime. “It’s not so much that there are crime hot spots as hot people,” says Lauren Baraldi, a prosecutor in the Philadelphia district attorney’s office, which is seeking to replicate Manhattan’s Crime Strategies Unit. Incapacitate those “crime drivers,” in Manhattan D.A. parlance, and you will produce an outsize effect on public safety, the thinking goes.

Vance’s mounting gang conspiracy cases have put that crime-driver theory to the test. Many prosecutors shy away from conspiracy cases because of their complexity and the risk of getting hit with dozens of defense motions filed by each of the defendants. Nevertheless, a famed Manhattan assistant district attorney, Walter Arsenault, used conspiracy law to take out Jamaican and Dominican drug gangs in the 1990s. Vance has expanded the form from drug cases to conspiracy to possess guns and commit murder. Prosecuting gang members for actual murder or attempted murder can be exceedingly hard, given their victims’ usual refusal to cooperate with law enforcement. Conspiracy law lets prosecutors focus on the agreement among gang members to obtain firearms and use them against their rivals. Such cases are quintessentially proactive, since the painstaking investigations that lead up to a grand-jury indictment precede the defendants’ arrests, rather than follow from them. Having a large number of defendants joined in one proceeding allows a prosecutor to tell a panoramic story about a gang and its life on the street.

And a central component of building up that story today is the use of social media. Law enforcement officials like to say that social media is their best confidential informant. Once, the FBI would take photos of mobster John Gotti talking to associates outside his favorite Queens social clubs; these days, after obtaining a subpoena from a grand jury, an agency would watch them communicating on Facebook. The NYPD got an early clue about the value of social media after a gang homicide in Brooklyn’s 73rd precinct in Brownsville. Officers arrived at the home of a gang member who had just been shot. As the gangbanger lay dying before them, they noticed the Tweets from his fellow gang members, bent on revenge, popping up on his open cell phone.

The Crime Strategies Unit has been as quick as the NYPD to spot the crime-fighting usefulness of social media. Crime analysts in the unit constantly track the Internet footprints of suspects in the arrest-alert database. The indictments in the 2013 East Harlem gang conspiracy case, for instance, consist almost exclusively of the Facebook postings of the defendants, as well as recordings of their phone calls whenever they were confined at Rikers. Those communications limned the progress of an escalating turf war that began in October 2009, with the fatal broad-daylight shooting of 19-year-old John Williams at a basketball game. Williams was a resident of the Johnson Houses, one of East Harlem’s many sprawling public-housing complexes, and a well-known figure in the area’s gang scene. The witnesses to his murder did not cooperate with the police, and the case remains unsolved. Williams’s fellow members in the True Money Gang, based in the Johnson Houses, believed that the Air It Out gang, based in the neighboring Taft Houses, was responsible. The Johnson Houses’ True Money Gang formed an alliance with the Whoadey gang, a few blocks south of Taft in the Lehman Houses, to take revenge on Air It Out.

The result would be a plague of violence over the next three years. Gunfire erupted in public thoroughfares; fights broke out in Manhattan’s courthouses; a True Money Gang member was stomped on the head while held underwater in Central Park’s Harlem Meer. In June 2011, Juan Otero, a teenage robber in Taft’s Air It Out, was shot three times at 9:30 PM outside the inaptly named Peaceful Valley community garden. He had been shot seven months earlier at another housing project, but survived. This time, he wasn’t so lucky: a bullet entered his stomach and killed him. That murder, too, remains unsolved. In early 2012, 16-year-old Whoadey member Aubrie Jackson was viciously beaten around the Taft Houses. He died after an eight-month coma in the hospital. Jackson knew he was a target. He sent a Facebook message to one of Whoadey’s ringleaders in March 2011: “AIO niggaz came for me mann,” and asked for a gun: “IHGT wen u gonna let me hold dat. . . . U got one for sale?”

By the time Vance unsealed the indictments against the 62 members of Whoadey, Air It Out, and the True Money Gang, the prosecutors investigating the case felt as though they knew the players intimately. One assistant district attorney had downloaded thousands of hours of inmate calls from Rikers on her iPod. The world revealed by these communications was pathetically narrow; its residents seem to do everything they can to live down to minstrel-show stereotypes. “Imma really kill one of dem niggaz now,” bragged Air It Out member Davon Powell (Day Day) in December 2011. Whoadey member Christopher Harrison (Creamy) announced in August 2011 that he was planning to run through the Taft Houses and “start popin niggas.” Those are the more straightforward postings. Air It Out associate Joshua Crider (Rocky), who would later be shot by a True Money Gang member, wrote on Facebook in June 2012: “Werd tuu chika ding [a.k.a. Juan Otero] next bitch dat violate betta neva show dey face in taft cuss im rocking bitches niggas whoeva.”

The obsession with procuring guns and ammunition was unending. Whoadey ringleader Gabriel Shelton (Grizzy) posted to the late Aubrie Jackson in 2011: “na I got the 25 shit but it don’t flocka because JP, lost the fire pin but gas some one to buy it for 250 and ima buy another one nd let cha lil whoas hold it down.” Though Shelton was willing to have his fellow Whoadey members hold his gun, other confederates were not always so gracious. Whoadey’s McKinley Campbell (Mac Lex) boasted from Rikers in September 2011: “I got a new bitch and all that. . . . I ain’t sharing that bitch with nobody. I ain’t even take her virginity yet.” This was a period when, according to New York’s anti-cop activists, the NYPD was terrorizing minority communities with its out-of-control pedestrian stops. There is little evidence in these gangbangers’ communications, however, that they feared getting stopped and questioned by the police.

Talking to the authorities was severely punished. Air It Out member Deshonn Green (Boogie) told an associate in June 2011 of a rumored “secret indictment” against the gang and vowed to “snuff” the suspected snitcher when he got out of jail.

Despite the daily machinations of their turf war, gang members managed to find time for racial politics. “Ya all like hoodie on for Trayvon,” wrote Air It Out member Darnel Conde (Man Man) on Facebook in March 2012, referring to Trayvon Martin, the black Florida teenager whose fatal shooting by a Hispanic man in February 2012 had been turned into a symbol of blacks’ racial oppression. Of course, the gang members were confirming the racial stereotypes that had allegedly caused the Martin shooting.

The indictments against the three gangs were so watertight that in April 2014, a year after they had been unsealed, all 62 defendants pled guilty—a virtually unheard-of success rate that saved taxpayers the considerable expense of taking the case to trial. Memories of the gang war are still fresh around the three housing projects. Elisa, a young resident of the Johnson Houses, recalls a day in 2012 when “a whole bunch of kids from Taft came running through here with Blood bandannas on and their hands on their waistbands [signaling the presence of a gun].” Elisa had been helping out in a day-care center: “I told my kids to get down on the floor and begged them not to shoot.” They didn’t that time; but another day, a similar posse opened fire on a second-floor window, she says. Such days of horde violence in the area are now largely gone.

But support for the takedown is not unqualified. This Mother’s Day at the Johnson Houses, there was plenty of sympathy for the mothers who had lost their sons—not to violence but to prison. “These mothers, they have to go to jail now to see their sons,” complained Angie, a tattooed young girl with neon scarlet hair. Angie’s half-brother Abdiel Martinez Linares was one of the 62 defendants. “Though it’s quieter now, we don’t get to see their happy faces. We now have to wait for their phone calls” (Martinez Linares had participated in the Harlem Meer stomping; a fellow AIO member reported in a Rikers call in September 2011 that Martinez Linares was “going crazy” with his gun and that he had shot at Whoadey and True Money Gang rivals. Martinez Linares posted on Facebook in January 2012: “[T]errorizen [a]nybody [f]rom [t]rilla [t]orture [a]nybody [f]rom [t]enth.”)

Angie blames the authorities for taking too long to investigate the case. “Why did they let it build up so much rivalry?” she asks. “In 2009, my brother was in the eighth grade. When it started, his friends became his enemies. He was in Taft; he had no choice. It was your life versus everyone else’s, the survival of the fittest.” The district attorney and the police would have been delighted to have concluded their case in 2009, of course, or at any point before 2013. It was the refusal of the “community” to provide evidence that prolonged the investigations, not foot-dragging by law enforcement or indifference to the rising bloodshed.

Other Harlem residents are similarly ambivalent, even if they didn’t have relatives in the takedown. “These [defendants] were someone’s babies; someone’s mothers are left without their child,” says Michelle, a nurse waiting outside a bodega on Lexington Avenue with her two young children. “That’s a real touchy thing.” Such interpenetration of crime into family structure explains in part the animus toward the police in crime-filled neighborhoods.

In 2014, as shootings increased throughout New York City, especially in housing projects, East Harlem remained relatively calm. While Taft had two shooting incidents as of May 25, up from zero at the same point last year, Johnson and Lehman had had no shootings, down from the previous year’s one. In all of the projects in the 23rd Precinct, where Taft, Johnson, and Lehman are located, four shootings had occurred by late May, up from two the year before. While this is a 100 percent increase, the numbers are low enough to begin with, and so much lower than the rates before the takedown, that it would be premature to view the percentage rise as a trend. Shootings in the entire 23rd Precinct are down 50 percent from two years ago. To be sure, fistfights between gang members—sometimes in the middle of traffic—seem to be an almost daily occurrence, but the 62-member indictment appears to have accurately targeted the area’s violent-crime drivers.

Nevertheless, the Crime Strategies Unit continues to monitor the vicinity closely. One night in late May, the head of the unit, Kerry Chicon, got arrest alerts on two East Harlem gang members, 15 and 16 years old, who had been picked up in a large brawl with sticks at the Jefferson Houses, just east of the Johnson Houses. The Crime Strategies Unit attorney responsible for East Harlem followed up with the arresting officers the next day for details about the fight and started scanning Facebook for chatter among the relevant parties. The attorney would also speak to the family-court prosecutor to try to elicit further information from the juveniles. “If there’s a shooting in three months, the worst thing is for our investigator to have to backpedal to figure out what happened,” Chicon explains. “We should know in advance if there’s a spike in violence coming.” Such an expectation, though standard now in police departments, was unknown in district attorney offices before 2010.

In June 2014, Vance and New York police commissioner William Bratton (back in New York on his second tour as police commissioner) announced the largest-ever gang conspiracy indictment in New York history. One hundred and three members of three West Harlem youth gangs based in two housing projects were charged with conspiracy to commit murder and gun possession, among other crimes. The prosecutors and police investigators had scanned more than a million social media pages to map out the web of criminality among the defendants and listened to more than 40,000 calls from jails and prisons. And the collaboration between the Manhattan district attorney and the NYPD that led to the West Harlem gang indictment and the 13 indictments that preceded it is only set to grow. Bratton wants a “seamless web” of information between the prosecutors and the police, he told the New York Times in early June. The Crime Strategies Unit will gain access to more of the police department’s databases on suspects and to its network of security cameras, while police detectives will receive the granular intelligence about criminal conspiracies developed during trial preparation.

Other district attorney offices are following Manhattan’s progress closely. Chicon testified before San Francisco’s Board of Supervisors in May, at Gascon’s invitation, as part of an effort to persuade the supervisors to fund New York–style crime analysts for the San Francisco prosecutors. “We’re envious of Manhattan,” says Cristine DeBerry, Gascon’s chief of staff. “You’re where we’re trying to get to so that we can share information and benefit from it collectively.” Philadelphia is following suit as well. “We stole your arrest-alert system—though we only have the junior varsity version,” says Baraldi. “Nevertheless, our involvement in intelligence gathering has opened up valuable collaboration with other law enforcement players.” The Brooklyn and Staten Island district attorneys are building a unit similar to the Crime Strategies Unit; district attorneys throughout New York State, as well as from Richmond, Virginia, and Rockland County, Maryland, are looking into it.

The Compstat analogy goes only so far. A prosecutor can declare that his mission is to lower crime, but measuring his office’s contribution to a crime drop is, at this point, impossible. There are no Compstat meetings in the Manhattan D.A.’s office where assistant district attorneys are grilled about crime trends and their strategies for improving those trends. But according to Dan Richman, a former assistant U.S. attorney in the Southern District and currently a Columbia University law professor, even if you can’t figure out the value-added of prosecution to lowering crime, “acting as if you can is a huge step forward. The prosecutors start conceiving of felony arrests as information.”

The rethinking of prosecution has only begun. Gascon is exploring the idea of predictive prosecution, echoing the nascent predictive policing concept. “We want to create tools to project crime patterns several years out” by mapping the connections between victims and offenders in a neighborhood, he says. In addition, he wants improved means of measuring whether his office’s court filings are targeted efficiently.

New York’s record-breaking crime decline since the 1990s was the product of a deliberate revolution in policing. Maintaining that crime drop will require every component of the criminal-justice system to be at the top of its game, especially after New York’s city council and a federal judge imposed superfluous and intrusive new bureaucracy on the New York Police Department last year. Intelligence-driven prosecution is coming not a moment too soon.


City Journal is a publication of the Manhattan Institute for Policy Research (MI), a leading free-market think tank. Are you interested in supporting the magazine? As a 501(c)(3) nonprofit, donations in support of MI and City Journal are fully tax-deductible as provided by law (EIN #13-2912529).

Further Reading

Up Next