Facebook depends on increasingly sophisticated ad targeting to make money, so it’s unsurprising to learn that Facebook’s new “reactions” feature will benefit advertisers in a variety of ways. But what about that other social-media scanning, sentiment-analyzing monolith, American law enforcement? A recent item from Reveal News argues that Facebook reactions will give law enforcement “a column in a spreadsheet that corresponds to emotion…. With the six new icons, determining emotion and sentiment through Facebook could now be exponentially simpler.” We talked to some digital-privacy-rights experts to find out more.
Lee Rowland, a senior staff attorney at the ACLU’s Speech, Privacy, and Technology Project, said that while there’s no proof that analysis of Facebook reactions will provide any actual assistance to law enforcement in building a legitimate case against a suspect, the ACLU has “concerns about police attitude toward collecting and monitoring online speech. What’s most important is that because people don’t have a formal expectation of privacy on social media, when they’re answering a quiz or using a reaction, they’re adding more information for law enforcement and advertisers.”
Police monitoring of social media is at an all-time high. A 2014 LexisNexis survey of law-enforcement officers found that over 80 percent use social media in investigations, and that Facebook is the platform they monitor most, a full 26 percentage points more than the next-most-surveilled platform, YouTube.
There are already multiple cases in which police have used similar social-media actions to prosecute people who have turned out to be innocent. In 2011, then-19-year-old Jelani Henry was jailed for 19 months on Rikers Island because he matched the description of someone who had shot and killed two people in East Harlem — tall, light-skinned black male — and police had seen him in pictures with friends of his from the neighborhood, some of whom had been arrested for violent crimes and liked Facebook pages associated with his crew. “Because of them pictures, the DA said I was affiliated, that I know what’s going on in the hood,” he told the Verge in 2014. “Those are people I would call my friends, but what they was doing, I wasn’t doing. To her, I’m part of them. I’m a monster.”
His case was dropped in the 19th month of his imprisonment (nine of which were in solitary confinement) with no comment or apology from Cyrus Vance, the Manhattan district attorney, who, coincidentally, has recently said that he has some 175 locked iPhones he’d like Apple to unlock for him. Henry’s mother, Alethia, told the Verge, “People don’t understand why it’s so dangerous to put yourself out there on social media. You know what my son is guilty of? Being born on 129th Street.”
Additionally, multiple law enforcement and security forces, from Fresno, California, police to the Oregon Department of Justice to the Department of Homeland Security to Mall of America security guards, have been caught surveilling and, in some cases, targeting black activists over social media in recent months. These aren’t rogue organizations, this is top-down policy; a 2013 report from the Department of Justice’s Office of Community Oriented Policing Services recommends heavy law-enforcement monitoring at all levels, as does a 2013 report from the International Association of Police Chiefs’ Center for Social Media, which concluded, bafflingly, that social-media policies for law-enforcement organizations should follow the exact “same principles” that are used when physically monitoring citizens. (The IACP’s Center for Social Media did not respond to a request for comment.)
Desmond Patton, an assistant professor at Columbia University’s School of Social Work and a data researcher who focuses on the intersection of social media, policing, and youth violence, said that the root of the issue comes down to law enforcement inaccurately interpreting the language young people of color use on social media. “The ways people are using these forms of communications are not widely understood,” he said. “They’re pretty clear on some patterns of communication online, but even young people in the same neighborhood are not sure what different emoji mean. For the police to be using it as a data point for criminalization is quite nerve-racking for me … The issue is how social-media platforms are determining what is criminal or violent or threatening. It’s important to think about racialization of communications. Things that appear uniquely ‘urban’ or black or Latino become the very things that are categorized as being criminal.”
There’s even precedent for police and prosecutors attempting to use emoji to convict someone of terrorist threats against police. The case against 17-year-old Brooklyn resident Osiris Aristy centered on a Facebook status that read, “Nigga run up on me, he gunna get blown down 👮🔫🔫🔫.” The charges were dropped after a grand jury decided not to indict him. Jeffrey Lane, an urban ethnographer and assistant professor at the School of Communication and Information at Rutgers University, said, “When I’ve seen teens in my research post emoji of a gun at a cop, it’s without intention or planning to shoot an officer but with deep hatred and frustration toward police following an arrest or prosecution of a close peer or peers. This is part of how teens and police in poor neighborhoods come to see each other.”
All of this has implications for Facebook’s reactions. At the moment, they’re treated exactly the same as a “like” — a binary, yes/no data point. But, as pointed out by Reveal, the point of the reactions is to gather more accurate data about our emotions, and Facebook has said they will begin providing this more detailed analysis of us to advertisers. And though Facebook’s law-enforcement guidelines say they only give information out to the government when it comes knocking with a warrant, law-enforcement agencies already do their own scraping of Facebook’s data, with or without a judge’s go-ahead — something Facebook has acknowledged in the past.
“Emoji are expressive, and just like speech, you could use emoji to express threats,” said the ACLU’s Rowland. “However, there is nothing illegal, ever, about expressing your like or love for something online. It never a crime to say that you like something, whether it be the KKK or ISIS. An emoticon itself is never going to constitute a criminal action.” A Facebook spokesperson declined to comment.