Tuesday 31 January 2017 4:59 am

Fake news is troubling – but censorship is far worse

We're in the grip of a fake news epidemic that is now poisoning our politics. That is the impression one would get from observing the tweets of the great and good.

In the aftermath of Brexit in the UK and Donald Trump’s victory in the United States, it has been tempting for those who could not conceive of the results to attribute the unexpected outcomes to the malign impact of lies, untruths and false stories. If only the media had held Trump or the Brexiteers to account, some say, then all this could have been avoided. Parliament’s Culture, Media and Sport Committee has even launched an inquiry into whether fake news is a “threat to democracy”.

The irony of this whole narrative is that it is itself exaggerated. The explosion of online news has no doubt led to a host of click bait stories that are totally false. A claim that the Pope had endorsed Trump for President went viral last July. There are regularly viral pictures in the UK which create news but turn out to be fakes too – the supposed London Underground sign with the #YouAintNoMuslimBruv hashtag that was widely shared, for example.

But many people are broadening the definition of fake news to include stories which merely exaggerate an issue or contain half-truths as a means of encouraging self or external censorship. This is all based on the false premise that the effect of fake news on public opinion is significant.

In the UK, it is not even clear fake news is an issue. Despite claims that lies and fake news contributed to the Brexit victory, a Buzzfeed investigation found that, compared to other countries, far from being outright lies, most of the popular dubious stories shared on social media in the UK were ones with kernels of truth that were editorialised by the partisan press. In order to clamp down on supposed fake news, one would have to regulate what is considered important within a story, or how it is presented – a very slippery slope towards censorship.

Indeed, the real issue here is that media outlets have their biases. This is an almost inevitable consequence of a producer or editor having to decide what is important, who to quote and how those within a story are presented. In a recent Institute of Economic Affairs book on the BBC, I examined certain programmes and found that they historically had been relatively biased against views in favour of Britain exiting the EU and free market capitalism, for example.

This “framing”, whether deliberate or inadvertent, is not unique to the BBC of course. But recently it was announced that the BBC is working with Facebook to clamp down on fake news. How broad that definition will be is unclear, but it is surely a bigger risk in the long term to install the idea of an established “truth” on some highly contested issue than to allow the occasional exaggerated story to creep through.

Consider the debate before and during the referendum campaign about whether the EU was in the process of developing its own army. In a public debate, Nick Clegg categorically denied that this was the case, but it has subsequently become clear that senior officials in the EU do desire a joint-fighting force.

The point here is that both sides have a point. Conflating differences of opinion with the very different phenomenon of genuinely fake news is a real danger once one installs a supposed neutral body to be the independent arbiter.

There is much more evidence that genuinely fake news filters through to social media in the United States. Just this weekend, in response to Trump’s executive order to temporarily suspend people from certain nations from coming to the US, a fake story flared up on social media that Dubai was threatening to deport US citizens.

Social media gives a medium to these sorts of stories. But a recent economic analysis suggests they may not actually have a big impact on people’s political views. Using data from audiences, fact-checking websites and an online survey, Hunt Allcott and Matthew Gentzkow’s new paper suggests that 14 per cent of Americans used social media for their primary source of news during the election.

Even though there were more pro-Trump fake news stories, the average American only remembered just under one of them. For fake news to have swung the election, coming across these stories would have had to be as persuasive as watching 36 television campaign ads – which seems highly unlikely indeed.

Honest people of all political persuasions should desire decent journalism and seek truth, of course. But reacting to evidence of the very limited phenomenon of genuine fake news with heavy-handed censorship or regulation would be a cure worse than the disease.

Most issues, particularly in politics and economics, are contested, complex and difficult. Tasking one individual or body with truth-telling, or determining what is fake or real in such an environment, is fraught with danger and inherently illiberal.