Why should we engage with online ‘trolls’?

With the growth of the Internet, countless new ideas are spread online. But hate speech also spreads with ease on the Web, allowing hurtful and inflammatory language to reach wide audiences.

Governments often use censorship and punishment to counter hate speech, yet, oppressive tactics not only limit free expression, but also have little impact on reducing hate speech.

What’s more, censorship and punishment “work less-well online," according to Susan Benesch, a professor at American University’s School of International Service and founder of the Dangerous Speech Project.

But old ideas are not just part of how governments confront hate speech.

With much of our communication online, people tend to believe that ignoring hate speech is the best way to deal with Internet “trolls,” in my definition, people who generally make hateful or anti-social remarks online.

What may be needed instead, Benesch, a fellow at Harvard’s Berkman Center for Internet and Society, said in a recorded talk at Harvard’s Berkman Center last month, is a new approach for reducing hate speech online.

A vast majority of Internet users describe their interactions on the Web as positive, according to a Pew Internet survey. Online hate speech nevertheless has both real and dangerous consequences.

This was particularly evident at Kenya’s presidential election in 2007, when, “after months of inflammatory speech online,” the elections “fell part,” Benesch said.

Hateful rhetoric led to violence in Kenya surrounding the vote, leaving more than 1,000 dead and over 1/2 million people displaced, with many still displaced today, Benesch said.

While Kenya’s Constitution protects freedom of expression, it does not protect hate speech, or what Benesch called “speech with a special capacity to incite mass violence.”

Benesch said researchers have used a speech-tracking system called “Umati” — Swahili for “crowd” — looking for hateful, inflammatory, or dangerous social media posts with Kenya's last elections in 2013.

What might be better at reducing hate speech online is what Benesch called “counter speech," or encouraging free speech, not restricting it.

Counter speech, she explained, means speaking out against people who use hate speech, relying on contrasting ideas rather than oppression.

Parody and humor can sometimes be effective forms of counter speech, she said, although there are many different kinds of counter speech.

Whereas before the Internet, Benesch said, people would generally expect to know their audience — both whom they were speaking to and whether anyone could overhear them — now the Internet allows us to virtually overhear more.

Now, one person’s bad day can change the mood of an entire online community, she said.

Sheltered and ignorant people create hate speech, but “non-trolls,” people who behave “normally” in most circumstances, can also be hateful online, Benesch said.

Counter speech can potentially influence a large part of the roughly 80 percent of people whose views are malleable to community norms, she said, though there is a minority who will not be persuaded by reasoned debate.

Benesch pointed to examples on Twitter with the Miss America nomination last year, showing how thoughtful responses might calm a hateful commentator, and in this case, warrant an apology.

Nina Davuluri, the Miss America winner, who has Indian parents, was attacked online for being Arab.

“I am literally soo mad right now a ARAB won. #MissAmerica,” said a post on Twitter from the user @DallasRobinson8.

Many responses condemned the post as racist -- the incident was picked up by the mainstream media -- and there was a much different response:

@MissAmerica sorry for being rude and "racist" and calling you a Arab please tweet back so everyone will know its real
— Dallas Robinson (@DallasRobinson8) September 18, 2013

Social networks play an important role in exposing different communities to new ideas, especially in places where there is limited Internet penetration.

In Myanmar, where the government recently opened the Web after strict online censorship, Facebook is increasingly popular. For many in Myanmar who use Facebook, Benesch said, according to a colleague of hers, Facebook and the Internet are essentially the same.

However, Facebook is particularly fertile ground for hate speech, Benesch’s research has found.

Memories of life under the country's former military regime mean Facebook’s “reporting” function is feared by many in Myanmar, though the reporting function is useful in other contexts, Benesch said.

When governments use heavy-handed tactics to silence certain voices online — even those that are legitimate threats — it punishes the majority of Internet users who can benefit from unrestricted access to the Web.

Seventy percent of Internet users have been treated positively online, and most think that the Internet has improved their relationships with friends and family, according to the Pew Internet survey.

Regardless of their age, race, or gender, the survey said, 90 percent of Internet users believe that the Internet has benefited them personally, while three-quarters think that the Web is good for society.

Though governments’ oppressive tactics for limiting hate speech often suppress freedom of speech, hate speech itself has the capacity to alter someone else's speech, Benesch said, which can also be damaging to free expression.

In conclusion, with more communities using the Web, we need more to research understand whether hate speech online could be a kind of venting, which takes hate out of the real world, Benesch said.