Why the Internet Needs a Hate Spell-Checker
WHY YOU SHOULD CARE
Because the internet is a #$@&%.
What with the hateful hashtags and menacing memes, logging onto the internet can feel like stumbling onto a gladiator scene. Or a sewer in hell.
And while Twitter and Facebook have launched attacks on hate speech and fake news, it won’t be enough to prevent our digital ears from going deaf. Deactivate an account, and a dozen more like it sprout up, more vitriolic than before. Surely, technology can do better; indeed, who says the internet can’t be picked and pruned like a garden? Introducing the hate spell-checker — a tool that can swiftly detect content that incites violence or malice in every nook and cranny of the internet.
This (yet imaginary) plug-in works much like a normal spell-checker, but it’s programmed to autocorrect hate and harassment instead. For example, an ISIS recruiter typing “murder the Americans” might have their tweet auto-corrected to “give an American a hug.” Or a troll calling someone “Ms. Piggy” could have their post changed to “as beautiful as a rose.” Hate would be stopped in its tracks, even before someone with ill will hits Enter.
We can’t take credit for this concept. Former Google CEO Eric Schmidt floated the idea of a hate spell-checker last year as a way to “help de-escalate tensions on social media,” he wrote in The New York Times. Schmidt didn’t respond to requests for comment, and plans for a hate spell-checker never materialized. But a hate-busting weapon of this caliber may be exactly what we need these days. “I think many of us would like to see more civility both online and offline,” says Patti Agatston, president of the International Bullying Prevention Association.
Many companies already monitor their platforms, at least to some extent. Twitter already rooted out hundreds of thousands of terrorism-related accounts this year, and, just a couple of weeks ago, suspended high-profile alt-right accounts. But it’s unclear how effective shutting down accounts is. Cries of censorship can just galvanize the haters, and it’s relatively easy to register a new account or use another platform. A hate spell-checker, on the other hand, could quell sinister acts before they take root and keep trolls from breathing down everyone’s backs. Besides, what better way to troll the trolls than to silence them?
To be sure, this kind of policing — making a hateful speech act literally impossible — is a more insidious form of censorship than merely shutting down accounts. Language is always evolving, and we expect that the users would just create new code words and dog whistles to express the hate. Technologically, a hate spell-checker would be hard to pull off: Pinning down subtle differences between incendiary comments that are contrary to the mainstream and speech that genuinely incites violence and inspires hatred is tricky, says University of Texas psychology professor Art Markman. Sameer Hinduja, codirector of the Cyberbullying Research Center, doubts that code, no matter how sophisticated, can recognize the language of hate. “We are just not there yet,” he says.
Maybe not. But who cares, right? Anyone who disagrees with us is a bleepity-bleep blankety-blank.