Facebook attempts to walk the tightrope on censorship

Having criticized Twitter for poking the bear, Facebook seems to be adopting a more nuanced approach to policing its platform.

Twitter’s decision to censor President Trump was an astounding mistake. Of course nobody, no matter how powerful, should be exempt from its policies, but if you’re going to single out one of the most powerful people in the world, you had better make sure you have all your bases covered. Twitter didn’t.

Facebook boss Mark Zuckerberg recognised Twitter’s mistake immediately and announced during an interview with Fox News that Facebook shouldn’t be the arbiter of truth of everything people say online. Even his choice of news outlet was telling, as Fox seems to be the only one not despised by Trump. Zuckerberg was effectively saying ‘leave us out of this’.

Twitter boss Jack Dorsey responded directly with the following tweet thread, which at first attempted to isolate the decision to censor Trump to him alone, but then proceeded to talk in the first person plural.

Within a couple of days Zuckerberg posted further clarification of his position on, of course, Facebook. He noted the current violent public response to a man dying in US police custody served as a further reminder of the importance of getting these decisions right.

“Unlike Twitter, we do not have a policy of putting a warning in front of posts that may incite violence because we believe that if a post incites violence, it should be removed regardless of whether it is newsworthy, even if it comes from a politician,” wrote Zuckerberg. “We have been in touch with the White House today to explain these policies as well.”

From that post we can see that Zuckerberg is still in favour of censorship, but sets the bar higher than Twitter and doesn’t see the point in half measures. Worryingly for Zuckerberg, many Facebook employees have taken to Twitter to voice their displeasure at this policy, apparently demanding Facebook does censor the President.

It’s worth reflecting on the two forms of censorship Twitter has imposed on Trump. The first was simply to fact-check a claim he made about postal voting, which contained a hyperlink to a statement on Twitter saying his claim was ‘unsubstantiated’ according to select US media consistently hostile to Trump.

The second superimposed a warning label over the top of a Trump tweet which promised repercussions for rioting. The label reads: “This Tweet violated the Twitter Rules about glorifying violence. However, Twitter has determined that it may be in the public’s interest for the Tweet to remain accessible.” Note the capitalization of ‘Twitter Rules’ and the clear admission that Twitter considers itself the arbiter of what is in the public interest. Clicking on the label reveals Trump’s hidden tweet, which features the phrase “when the looting starts, the shooting starts.”

That was apparently the bit that was interpreted as glorifying violence, and yet a subsequent Trump tweet, using exactly the same phrase, has not been subject to any censorious action by Twitter. That discrepancy alone (not to mention the fact that the labels don’t survive the embedding process) illustrates the impossible position Twitter has put itself in. There are presumably millions of other examples of borderline glorifications of violence, let alone direct threats, that it has also let pass. Such inconsistent censoring can easily be viewed as simple bias, seeking to tip the scales of public conversation in your favour.

For many people censorship is a simple matter of harm reduction. Why would anyone want to allow speech that could cause harm? The mistake they make is to view harm as an objective, absolute concept on which there is unilateral consensus. As Zuckerberg’s post shows, the perception of harm is often highly subjective, and the threshold at which to censor harmful speech is entirely arbitrary.

There is clearly a lot of demand for extensive policing of internet speech nonetheless, but social media companies have to resist it if they want to be able to claim they’re impartial. There’s just no way to keep bias out of the censorship process. If they don’t, they risk being designated as publishers and thus legally responsible for every piece of content they host. This would be calamitous for their entire business model, which makes it all the more baffling that Dorsey would so openly risk such an outcome.