When thousands, perhaps millions, of people use social networks to spread hate speech, online harassment and abuse, the problem might often seem insurmountable. But whatever approaches are being used to address the problem, something, somewhere is clearly not working. YouTube videos by extremists and terrorists abound. Targeted harassment and trolling on Twitter remains to this day, years and years after you might reasonably have thought they might get it right. Facebook almost automatically removes women who breast feed their children and yet seems to shrug its shoulders over fake news, which might move an entire election, spreading on its network.
It’s no wonder politicians feel justified in saying enough is enough.
The news last week that the German government had proposed fining Facebook and Twitter up to $53m (€50m) for failing to remove abuse, slander, fake news and hate speech within 24 hours once again threw into sharp relief the differences in attitudes between Europe and the US to online content. If followed through, such a law would make it the most draconian clampdown by a European country on an online network.
In the US, freedom of speech — any speech — is sacrosanct. In Europe, which experienced genuine fascism and extreme politics in its relatively recent history, hate speech is usually recognised in law and has long-defined sanctions on those who spread it.
The German government bill, introduced by interior minister Heiko Maas, will be designed to “combat hate crime and criminal offences on social networks more effectively,” and would cover “defamation, slander, public prosecution, crimes and threats.”
“Too little criminal content is being deleted, and it’s not being deleted sufficiently quickly,” he told journalists. “The biggest problem is and remains that the networks don’t take the complaints of their own users seriously enough,” he added. Germany also has long been pressuring US tech companies to more aggressively combat hate material and ‘fake news’.
Social networks would be obliged to delete “obvious” criminal content inside 24 hours, and across an entire platform, not just a single instance of the content. Other criminal content would have to be blocked or deleted within a week and companies would have to inform complainants of any decision. Companies would also have to publish quarterly reports on the complaints in terms of number, how the complaints were handled and staffing of their complaints team. Failure to comply would mean a €5m fine against the individual responsible, and up to €50m against the company.
If a company takes issue with the execution of the law, the issue would go to the German courts for deliberation. Maas has also suggested that Facebook should be treated as a media company under German law.
It’s not an idle threat. Germany’s interior ministry not only published an explanation of the draft law, Maas also published a fuller explanation of the law and gave a speech on the issue. He said “freedom of expression also protects repulsive and ugly utterances … Even a lie can be covered by freedom of expression. But freedom of expression ends where criminal law begins.” He argues that “verbal radicalization is often the precursor to physical violence.” Let’s face it, Germany should know.
In Germany it is illegal to promote Nazi ideology (or to deny the Holocaust) and many tech companies have previously agreed to work with the German officials to remove xenophobic and racist messages.
Maas’s point is that social networks are not ‘dumb’ carriers of content, but are equally responsible if their platform “is abused to spread criminal hate.”
In his pronouncements, Maas referenced a new report by the Jugendschutz organisation, a youth protection watchdog, which showed that social platforms were often achingly slow to act on hate speech, and ineffectual in deletions. It said social networks including Twitter, Facebook and YouTube are removing only a small portion of the criminal content reported to them by their one users. The report said that Twitter removed only 1% of content reported to it as criminal within 24 hours; Facebook deleted 39%; and YouTube removed 90% of user-reported criminal content and 82% of the deletions occurred 24 hours after the notification. The German government is going to set a target of 70%.
The appallingly slow rate of deletion of hate speech on these networks shows either that these networks do not take the complaints of their own users seriously enough, or that the problem of sifting through billions of posts a day is far larger than any of the companies can deal with. Whatever the case, the practical realities are that they must at least be seen to be trying.
But the tech giants are either saying nothing, or fighting back.
Facebook disputes the result of the Jugendschutz report. It pointed to another report from the FSM organisation which found 65% of illegal content was removed from Facebook within 24 hours, and overall it had a deletion rate of 80%.)
And in it’s defence, Facebook said that, by the end of 2017, it will have more than 700 employees working with content moderation partner Arvato, based in Berlin. It’s also been piloting measures to fight fake news in Germany since January, working with local third-party fact-checking organization Correctiv to try to identify and flag dubious content.
Against these findings, Google’s YouTube is looking much better. And admittedly, it may be easier to discern problematic content from video instead of text. But it is far easier to take down content when it’s being matched to a database or, say, copyrighted material, than when you have to determine whether something is hate speech.
Twitter is not commenting on the proposed draft law, but, independently, has made recent moves to try to address the problems of hate speech on its network. This includes: Identifying abusive accounts; adding extra filtering options; identifying persistent abusers and preventing them from opening new accounts; and creating a “safe search” option.
The matter is not going to stop in Germany, however. The government intends to build a Europe-wide consensus on dealing with abusive and criminal content on social networks, and take its fight EU-wide.
And the idea seems to be spreading. British Conservative MP Tim Loughton, a member of the powerful Home Affairs Select Committee, heavily criticised the “woeful” excuses by the UK heads of Google, Twitter and Facebook to MPs in a gruelling three-hour grilling in the House of Commons. He said it defied belief that they could make billions from online advertising but do so little to protect the public from the effects of online abuse.
In particular, Facebook has come under fire in the UK after a BBC investigation found secret Facebook groups being used to share child abuse imagery. (And it didn’t exactly cover itself in glory when it reported the journalists who found the content to the police).
“German proposals for proactive fines must be the way for us to go if these companies cannot clean up their act,” Loughton said.
In other moves, the European Commission is threatening to fine Facebook, Twitter, Google and other social networks unless they overhaul their terms and conditions to “comply with EU consumer rules.” It could be just the first wave of moves which may go on to address the content they also carry on their platforms.
Indeed, they’ve already started. In May last year, the European Commission set forth an anti-hate speech code of conduct, and it enlisted Facebook, YouTube, Twitter and Microsoft to take part in the fight. This was after moves in 2015 which saw Facebook, Google and Microsoft agree to a voluntary code of conduct in Europe to fight the spread of hate speech online. It’s not like any of this is a surprise to them.
And let’s not be naive: The laws Germany is proposing are not suddenly being dreamt up in a vacuum.
Seeing the effects of how fake news may have influenced the US election to favour a demagogic populist like Donald Trump, there is huge concern in German political circles about the potential influence fake news and hate speech could have on the country’s federal elections later this year.
Angela Merkel’s ruling conservatives are facing a strong challenge from the populist, anti-immigration Alternative for Germany (AFD), which has effectively used social networks to get across many of its more extremist and intolerant ideas.
Hate speech is on the rise on social networks in Germany after Mrs Merkel allowed in a million refugees fleeing civil war and economic impoverishment in the Middle East and North Africa.
And an increasing number of policy makers in Europe, the United States and elsewhere, argue that social media companies have a responsibility to block harmful content and respect national rules.
But the implications are obvious. An EU-wide, or even global regulatory push would end up limiting how individuals communicated online and allow governments to exert control over the content on social media.
Working out what exactly is hate speech and what might be, for instance, simple democratic opposition to a differing point of view, is a complex issue. As the phrase goes, one man’s terrorist is another’s freedom fighter.
There are no easy answers.
But Mark Zuckerberg cannot seriously say one the one hand that Facebook could not do anything to stop fake news spreading, when it can do plenty to serve us targeted advertising based on our interactions with the platform. The two notions are just not compatible.
A way out may be to give users far more control to weed out the hate speech themselves, with AI-driven algorithms applied to the problem. In the background, however, lurks a dark question for the social networks — if their platforms become fluffier and more soporific, will their traffic, engagement and therefore revenues go down? There can be no doubt that this question will have been raised at some point, even in the rarefied atmosphere of Facebook’s inner circle.
Whatever the case, until the problem is better addressed, politicians and governments will continue to come up with largely unworkable solutions like fines, so long these huge social platforms look like they are sitting on their hands.
Asking private sector companies to police their networks is a double-edged sword. Like the disappearing breast-feeders, if faced with a huge fine for allowing an instance of hate speech on its network, social networks would be more likely to ‘delete first and ask questions later’ since deleting content in the first place (whether incorrectly or not) is a lot cheaper than getting the legal position wrong, and hence fined.
And if democratic governments get to police online content, they will likely end up being heavy-handed also. Plus, that policy will give succour to all those tin-pot dictatorships around the world, who will enjoy seeing democracies turn to their own methods of censorship.
We can all agree however that, yes, platforms need to allow for free speech, but not to the point where extremism and misinformation drowns out the truth.
What is as stake here? The freedoms and rights enjoyed by companies which organically developed in democratic countries could well end up being eroded. How many global, truly innovative tech companies (not counting clones) have come out of Putin’s Russia, or Erdogan’s Turkey? Tech innovation depends on free, democratic societies. The Enlightenment of the 19th Century was only possible when freedom of thought was able to flourish. It did not emerge from the fug of the Dark Ages, where conspiracy and superstition abounded (not unlike the fake news era of today).
To not put in place a mechanism where abuse and untruths smother science and factual reporting is to put at stake any progression in innovation and entrepreneurialism. It is surely in the interests of the giant online and social platforms to pay attention to the erosion of trust in genuine media sources and professional journalism. Without it, the rights and freedoms they enjoy under democracy, and from who’s breast they themselves were succoured, will likely disappear, and populist politicians will take advantage of a misinformed and inflamed public discourse to the detriment of a free society.
Featured Image: Bryce Durbin