The Gods of Facebook controlling life and death
Meta’s decision to permit calls for violence towards Russian invaders is a shift that could shake free speech, says Ed Cumming
Can a death threat ever be justified? According to Facebook – from this weekend – it can. Yesterday, Meta, the name of the company which runs Facebook and Instagram, did something unexpected. In a reversal of its usual “hate speech” policy, the organisation announced that users of the social platforms in certain countries would be allowed to advocate violence against Ukraine’s Russian invaders. Like a Caesar in an amphitheatre, whose capricious thumbs up or down would decide a slave’s fate, or a Greek god treating cruelty as a public sport, the social media platform took control of Europe’s free-speech destiny.
“As a result of the Russian invasion of Ukraine we have temporarily made allowances for forms of political expression that would normally violate our rules, like violent speech such as ‘death to the Russian invaders’,” Meta said in a statement, adding however that it still wouldn’t allow credible calls for violence against Russian civilians.
Thou shalt not kill thy brother, but thou shalt chat about it online if it doesn’t seem like you’ll actually do it. And only in Armenia, Azerbaijan, Estonia, Georgia, Hungary, Latvia, Lithuania, Poland, Romania, Russia, Slovakia and Ukraine.
It’s no surprise that such announcements have the stentorian tones of divine intervention. In 2022, Meta is a supranational power; an empire ruled over by a single man, which influences voters, proscribes language and provides meeting places for cranks and perverts – as well as somewhere to moan about bin collections. A church, in other words. Yet unlike a church, Meta and its subsidiary businesses operate for the financial benefit of their shareholders. They must make decisions along ethical, political and cultural lines while in thrall to the bottom one.
In recent years, the company has talked tough about the language it permits on its platforms, with firm rules on obscenity and violence. It would like you to believe that a sophisticated Ai-powered anti-hate algorithm, as well as a battery of human moderators, is protecting you from harmful words. Posting nudity, threats or discriminatory language can land you a warning or temporary ban. In extreme situations, for example if you have recently been US President and your supporters have stormed the Capitol building, they’ll remove you altogether. They’re all about free speech, just not too much.
On the face of things Meta’s decision seems unobjectionable, says Ruth Smeeth, a former Labour MP who is now the CEO of Index on Censorship. “It looks like a pragmatic decision,” she says. “The idea that you couldn’t have a call to arms within Ukraine is perverse. Would First World War poetry have survived Facebook’s original approach? But there is a bigger issue here, about who is the owner of speech, and determining what is and isn’t acceptable. I’m not sure Meta should be the sole arbiter.”
Facebook’s verdict on what is and what isn’t iffy has rarely been consistent. In 2021, a Wall Street Journal story revealed some of the firm’s confidential moderation guidelines. In places it reads like a bullies’ charter: “We do not remove content like “frizzy hair,” “lanky arms,” “broad shoulders,” etc. since “frizzy,” “lanky,” and “broad,” are not deficient or inferior, and therefore not degrading.” A history teacher was banned for 30 days for telling a friend he was “spewing crazy.” A writer, Alex Gendler, was censored for sharing a story about tribes in New Guinea.
Even in Ukraine, the cases are far from obvious. Where is the line drawn? “Is ‘drive the invaders out of our land’ allowable? How about the old lady telling a Russian soldier to put sunflower seeds in his pockets so that when he is killed they’ll mark his grave. Is that a call to violence?” asks Ben Evans, a tech writer. “If [Ukrainian president] Zelensky says that every Russian who comes to Ukraine to kill civilians will go home dead – is that allowed? And if Facebook was taking those down, what would we think?”
Besides, having guidelines is not the same as enforcing them. Facebook moderates more than two million posts a day, using a mix of algorithms and humans. The AI is good at detecting specific things – naked bodies, for example – but struggles with context. A humanitarian group and terrorist organisation might post the same image with a very different meaning.
“The idea of Facebook loosening its rules is spin, because they don’t enforce their rules,” says Imran Ahmed, CEO of the Centre for Countering Digital Hate. “It’s designed to reinforce Facebook’s framing that they have to take down the pleas of the Ukrainian people. In fact they’ve done a terrible job at dealing with any of the problems in that space. There are lots of examples of people calling for violence against politicians on their platform. They basically give it a free pass. It suits the company to make it look like they’ve got a crack team on the case, with Nick Clegg standing over them. But they barely have any Ukrainian language moderators.”
And it’s hardly as though Facebook has a perfect record when it comes to the spread of honest information. “It has monetised disinformation,” says Ahmed. “Putin has used Facebook to pump disinformation into his country for years. It can influence the case for war, it can influence the outcome of war. It’s been used to affect elections. Zuckerberg wields unbelievable power. Nobody who isn’t democratically elected should have that degree of power.”
The obvious counter to all this is social media’s power for good. In theory, it provides a platform for solidarity that ought to be especially useful in times of war. But since the first hopeful days of the Arab Spring, that rosy vision has looked naive.
“Every time I see anything about the Russia-ukraine war posted on social media I discount it,” says Alex Kradosomski-jones, director of the Centre for the Analysis of Social Media at Demos. “I’ve never felt that before. The quality of information on those platforms is just so low. This isn’t new, but there is a uniquely Russian aspect to the current conflict: this idea, which has been attributed to Putin, that you can just make a full-frontal assault on the truth.”
For Kradosomski-jones, this week’s
‘They have to accept they’re a political entity. Long gone are the days of just being a platform’
news is more evidence that Facebook has found itself in a position for which it does not have the tools to cope. “They have to accept that they are a political entity,” he says. “Long gone are the days when they could say they were just a platform.”
While Meta’s move has the appearance of solidarity with Ukraine, it is also a business calculation. They hope the 12 countries affected by the ruling will see this as a sign of ally-ship. Without their custom, Facebook will cease to exist in this region.
Closer to home, Smeeth has advocated for greater accountability for social media companies in the UK, but is worried we are going in the wrong direction. “With the upcoming Online Safety Bill, we’re giving even more power to the platforms to determine what is and isn’t acceptable speech,” she says. “There has to be democratic oversight.”
Still, she is confident social media can play a positive role. “We’re living the experiences of those innocent civilians in Ukraine, in no small part because of social media. It can’t be covered up. Last year, people were found guilty of war crimes in Syria based solely on evidence gathered online. This is a tool for huge good.”
Conditions will be trickier for Meta in Russia: the prosecutor general has now asked for it to be listed as an extremist organisation and for its services, except for Whatsapp, to be banned. Other platforms are available, but it won’t be a popular call. It rarely ends well when people are asked to choose between their country and their religion.