2013-07-27 08:45:00

When the Tail Wags the Dog: Dangers of Crowdsourcing Justice

Crowdsourcing is relying on the public to solve a particular problem. When it comes to catching dangerous terrorists who are planting bombs at large public events, there is an even greater responsibility that comes with crowdsourcing.

Apparently, this was overlooked by every element involved in the 3-ring circus following Boston: from the irresponsible news outlets that published certified trash and then patted themselves on the back, to the FBI who decided to crowdsource information it should have already known in the first place, to the online users who quit their day jobs for new detective gigs. There was plenty of blame to go around, writes attorney Lalit Kundani.

 As a trial lawyer, evidence is the only thing that I care about in a courtroom. Sure — speculation, rumors, and baseless hearsay are all engaging storytellers, but they make for rather lousy tour guides. When it comes to the business of getting it right, there are no prizes awarded for being the first to get it wrong, unless, of course, your name is Christopher Columbus. Or, unless your name is Reddit. Or the New York Post. Or CNN.

In the wake of the Boston bombings, each of these “news outlets” was trending with more viewers, but for all the wrong reasons. Having grossly misread the age-old parable of “The Tortoise and the Hare,” each chose to swiftly report the “news” based only on speculation, rumors, and baseless hearsay from people they had never interviewed.

It’s called “crowdsourcing” – relying on the public to solve a particular problem. As a concept, it’s been around for ages. And in the virtual world of clicks and tweets, posts and swipes, crowdsourcing enjoys permanent job security. Its impressive resume includes fund-raising startup companies, building online dictionaries, and distributing complex problem solving. Those who swear by it are able to mine collective intelligence while still assessing quality and productivity.

But those who helped coin the phrase also warned of its limits. It’s not for every situation. “If you are on a plane, and the pilot faints, your next move should be to ask if there is another pilot on board, not, I repeat not, to crowdsource what people think should happen next and then average that out.” And presumably, when it comes to catching dangerous terrorists who are planting bombs at large public events, there is an even greater responsibility that comes with crowdsourcing. Apparently, this was overlooked by every element involved in the 3-ring circus following Boston: from the irresponsible news outlets that published certified trash and then patted themselves on the back, to the FBI who decided to crowdsource information it should have already known in the first place, to the online users who quit their day jobs for new detective gigs. There was plenty of blame to go around.

The News Outlets

Darwin was only half right. He should have called it “survival of the quickest.” With a constant barrage of 24-hour media coverage, eight reporters on the ground, sister stations, mobile apps, i-reporters, and Twitter accounts, the “news” is constantly racing against its own world record. Hours after the bombings in Boston, those reporting the “news” initially reported that the “suspect” was a 21-year old Saudi national because members of the public saw him running away from the bombs. This changed the next morning when the “news” plastered on its front page a photograph of two other “suspects,” this time a 17-year old Moroccan teenager and a 24-year old Middle Eastern male, because crowd-generated photographs showed both men wearing backpacks and standing near the finish line. By the end of the following day, the “news” changed yet again and proclaimed that one of the “suspects” was a missing 22-year old Indian college student because those on social media thought he resembled a grainy surveillance photograph of the actual bomber.

Nancy Drew would have been proud. In less than three days, the “news” already had publicly identified four suspicious males standing near the bomb site in an investigation where the FBI suspected only two. The only problem was that the four accused males, all of whom were young minorities, were each completely innocent. The evidence revealed that the 21-year old accused Saudi national was running away from the bombs because he was an innocent and injured bystander. The evidence showed that the Moroccan and Middle Eastern men wearing backpacks were a high school track star and coach who had gone to the marathon to support the runners. And the evidence demonstrated that Sunil Tripathi, the missing Indian college student from Brown University, was already dead and face down underwater some 60 miles away in Rhode Island at the time he was publicly accused of blowing up Boston. While “survival of the quickest” ensured ratings, it did not provide for accuracy.


Alongside the “news” outlets that published unverified reports from the crowd, it was law enforcement that retained the crowd in the first place. Evidence shows that the FBI failed miserably. Not because they couldn’t predict when and where sociopathy would next occur. But because they were so quick after the bombings to rely on outside information even though their own department had all the information their agents needed. In 2011, after a tip from Russian intelligence, the FBI knew of the two bombers, their names, their ages, their faces, their home, and their address. They interviewed the brothers, visited their residence, and even placed them on a watch-list. No doubt, in 2011, the Tsarnaev family was on the FBI radar. However, when two bombs suddenly exploded three miles away from the Tsarnaev home during the Boston Marathon, the FBI radar went down. Despite having interviewed the very two suspects depicted in the surveillance photographs about claims of Islamic extremism just two years earlier, and just three miles down the street, the FBI stared blankly at the photographs as though they had no clue who these two young men were. Had they checked their own watch list or had they bothered to first show the photographs to their own agents, perhaps MIT police officer Sean Collier would be alive today. Instead, as soon as they obtained the photographs, the FBI immediately turned to crowdsourcing. Who were these two men?

The FBI plan to crowdsource was two-fold. First, to obtain more photographs and videos of the marathon that may provide better images of the suspects. This part of crowdsourcing worked well, as thousands of photographs were submitted to authorities, which resulted in even better photographs of the two suspects. Secondly, the “crowd” was asked if they could identify the two suspects. If yes, the “crowd” would then contact the authorities, who would then investigate the leads.

This second part backfired tragically. What was supposed to happen never did. The individuals who intimately knew the Tsarnaev brothers, and recognized them in the photographs, never reported them. One friend of the younger Tsarnaev stated, “I saw the photograph of Suspect 2 and immediately thought it looked like Dzhokhar but then I thought it couldn’t be him because he would never do something like this, so I didn’t call it in.”

Meanwhile, the individuals who never met Sunil Tripathi thought he resembled Suspect 2 and were never more positive. Rather than calling in their suspicions to the FBI, the users took it upon themselves to become the FBI.

Online Users

Kami Mattioli had gone to high school with Sunil Tripathi but had not seen him in three years. After the FBI released photographs, she tweeted that Sunil resembled Suspect 2. When Reddit user “pizzatime” (who saw Mattioli’s tweet but who never knew Tripathi) confirmed that Tripathi looked exactly like Suspect 2, a subreddit devoted to the bombing confirmed it was Tripathi. Twitter followers then re-tweeted Reddit posts. Within minutes, collages of side-by-side comparisons were posted on the web comparing Tripathi to Suspect 2.

At 2:14 a.m. Eastern, an official Boston PD scanner said “Last name: Mulugeta, M-U-L-U-G-E-T-A, M as in Mike, Mulugeta.” Even though the scanner never stated Mr. Mulugeta’s first name, a twitter user named Carcel Mousineau misunderstood the announcement and tweeted: “Just read the name Mike Mulugeta on the scanner.” By 2:42 a.m., Greg Hughes tweeted: “This is the Internet’s test of ‘be right, not first’ with the reporting of this story. So far, people are doing a great job. #Watertown”. Then, a minute later, for no apparent reason, he tweeted: “BPD has identified the names: Suspect 1: Mike Mulugeta. Suspect 2: Sunil Tripathi.” And from there, they were off to the races (albeit, traveling in the wrong direction).

Alexis Madrigal at The Atlantic provides the best rundown of the sequence of events that night but reports definitively that Sunil Tripathi’s name is nowhere on the police scanner. “I’ve listened to it a dozen times and there’s nothing there even remotely resembling Tripathi’s name.”

The misinformation grew like wildfire. Dozens after dozens of notable news agencies, reporters, and investigative journalists, re-tweeted the misinformation thousands of times. Redditors were proclaiming an early victory: “If Sunil Tripathi did indeed commit this #BostonBombing, Reddit has scored a significant, game-changing victory.” Greg Hughes gave advice: “Journalism students take note: tonight, the best reporting was crowdsourced, digital and done by bystanders. #Watertown.” Others added: “This is historic Internet sleuthing.” “Reddit solved the bombing. Before the Feds Solved the Bombing.”

“Solved”? In just under 60 minutes – much like a TV crime drama – Internet users had solved what the FBI had outsourced them to do. At least that was the thinking of Kami Mattioli, when she justified herself in another tweet saying: “The FBI’s motto has been ‘See something, say something’ and that’s all I intended to do by posting the comparison photo.” (No, Kami. What they wanted you to do was to pick up a phone and “say something” to them, so that they could send experts to investigate the lead). Rather than calling the FBI and reporting her suspicion, Mattioli instead publicly posted her tweet regarding the possible identity of an outstanding terror suspect, and then claimed that she intended that someone over at the FBI would just happen to stumble across her post. This idea of “tweeting in a crime” in no way amounts to helping authorities. Instead, it amounts to gossip of the worst sort, which in real time can have dangerous consequences not only for the accused person but for any innocent persons nearby. It’s not a matter of free speech, it’s a matter of common sense.

Just in case the FBI was in need of even further “evidence” from its impressive “crowd” of sleuths that Suspect 2 was Sunil Tripathi, Twitter and Reddit were closing in with compelling, game-changing evidence: a single photograph taken of Sunil while he was wearing a Che Guevara t-shirt. The political implication was obvious. If a dark skinned Indian college student from Rhode Island wore a Che Guevara t-shirt, he must be blowing up bombs in Boston. In other forums, the Twitter “crowd” turned to hate speech: “The alleged suspect for Boston bombings was a Hindu Sunil Tripathi. We urge Western media to now showcase all Hindus as Terrorists.” Game, set, match. Checkmate.

In an irony too painful to script, history had repeated itself. Much like the FBI had turned to crowdsourcing to locate the bombers, the Tripathi family had turned to crowdsourcing to locate their son and brother who went missing on March 16. Indeed, the “crowd” had done both: located Sunil and identified the bomber all in one stroke. For his parents, the experience of learning (amidst hundreds of calls and knocks on their door in the middle of the night) that their 22-year old son had finally been located in Boston came at a shockingly high price: the realization that the very “crowd” that they had been relying on for help for so long had deemed him a domestic terrorist. Refusing to accept the finding of the “crowd” only made more painful the reality that their son was still missing.

Taming the 3-ring Circus

In short, the FBI’s hasty decision to crowdsource invited speculation by online armchair quarterbacks which, in turn, was published by even hastier news outlets who dumped accuracy out the window. Garbage in-garbage out.

If Sunil’s parents were searching for comfort in the middle of the night, as media vans knocked on their door for a comment, CNN was not helping. For those few hours, the news giant displayed in its running ticker that Twitter was reporting that Suspect 2 was Sunil Tripathi. A CNN expert made vague references, without naming names, that one of the bombing suspects lived out of state and authorities were probably contacting his family members who may be shocked to hear the news. For those few hours, according to the “news,” Sunil Tripathi was no longer missing. He was found in a city where he never was, perpetrating atrocities which he never dreamed, condemned by those he never knew. It was not until NBC’s Pete Williams confirmed the identity of the Tsarnaev brothers that the rest of the “news” followed in step.

Crowdsourcing can be an absolutely wonderful tool

In the days following the devastation, crowdsourcing efforts helped Boston bombing victims pay for prosthetic limbs and their medical costs. But when applied to the high stakes context of apprehending a fugitive terror suspect, crowdsourcing was not executed responsibly. This starts with authorities like the FBI who, prior to seeking “crowd” assistance, should define and regulate the scope of the request, decide early on what information they truly need, the best methods to obtain that information, and only then release the data to the public. In turn, the public has a duty to act responsibly as well. If the “crowd” has helpful information, it should relay it to the authorities in charge rather than try to fly a plane with no experience (and then tweet as you go along). Finally, unlike the public, “news” agencies have a duty, and the resources, to be accountable and report facts accurately rather than sell gossip on a wholesale level.

In the wake of mass killings, enough innocent names have been sacrificed to the public appetite. Richard Jewell, a police officer, was wrongly accused of committing the Atlanta Olympic bombings after news agencies played judge and jury. Ryan Lanza, the brother of suspected Newtown killer Adam Lanza, received death threats after he was wrongfully labeled as the murderer of Sandy Hook. Sunil Tripathi, the Saudi national, the Moroccan track star, and the Middle Eastern coach have now joined that list.

To ignore these lessons is to add Boston to the larger broken record that we all have heard too many times before.

Lalit Kundani is an attorney with Panish, Shea & Boyle LLP in Los Angeles and is a former prosecutor with the LA County District Attorney’s Office where he prosecuted cases alongside the FBI. He can be reached at [email protected]

Proud Partners