Predatory adults are rare on social media, compared to mean girls and crude boys, writes Emily Bazelon in How to Stop the Bullies in The Atlantic.

Facebook gets millions of complaints a week about cyberbullying, she finds. Employees are expected to decide in a few seconds which have merit.

Henry Lieberman, an MIT computer scientist, is working on a program to spot nasty posts immediately. In middle school, he  was a “fat kid with the nickname Hank the Tank,” he tells Bazelon.

. . . he and his graduate students built a “commonsense knowledge base” called BullySpace—essentially a repository of words and phrases that could be paired with an algorithm to comb through text and spot bullying situations. Yes, BullySpace can be used to recognize words like fat and slut (and all their text-speak misspellings), but also to determine when the use of common words varies from the norm in a way that suggests they’re meant to wound.

In tests, BullySpace caught 80 percent of the insults flagged by human testers.

Lieberman also hopes to use “ladders of reflection” to persuade kids not to harass others.

Think about the kid who posted “Because he’s a fag! ROTFL [rolling on the floor laughing]!!!” What if, when he pushed the button to submit, a box popped up saying “Waiting 60 seconds to post,” next to another box that read “I don’t want to post” and offered a big X to click on? Or what if the message read “That sounds harsh! Are you sure you want to send that?” Or what if it simply reminded the poster that his comment was about to go to thousands of people?

“Ash” and “Katherine,” members of the hacker group, Anonymous, publicized the identities and vicious tweets of four high school boys who were harassing a 12-year-old girl with rape threats and suggestions she commit suicide. She’d followed one of the boys on Twitter, then angered him by un-following him.

At first the boys railed against Ash on Twitter, and one played down his involvement, denying that he had ever threatened to rape the girl. But after a while, two of the boys began sending remorseful messages. “For two solid days, every time we logged on, we had another apology from them,” Ash said. . . . Katherine thought the boys hadn’t understood what impact their tweets would have on the girl receiving them—they hadn’t thought of her as a real person. “They were actually shocked,” she said. . . . we started talking to them about anti-bullying initiatives they could bring to their schools.”

“When i found out she was hurt by it i had felt horrible,” one of the boys e-mailed Bazelon. Perhaps a few seconds of reflection would have helped.

Anti-bullying laws can conflict with free-speech rights, argues Eugene Volokh, a law professor. A proposed Minnesota law bans “interfering” with an individual’s ability “to participate in a safe and supportive learning environment.”

Say that students are talking over lunch about how a classmate committed a crime, cheated, said racist things, treated his girlfriend cruelly, or whatever else, which causes people to feel hostile towards the classmate. That interferes with his ability “to participate in a … supportive learning environment.”

Bullying may include speech or conduct that “relates to the actual or perceived race, ethnicity, color, creed, religion, national origin, immigration status, sex, age, marital status, familial status, socioeconomic status, physical appearance, sexual orientation, gender identity and expression, academic status, disability, or status with regard to public assistance, age” of the individual or his/her associates.