EU should not make platforms the judges of free speech

Meanwhile, sites beyond the reach of UK law, including the Washington Post, republished the video, leaving everyone—including British viewers—free to draw their own conclusions.

Whatever one thinks of the conviction or the law behind it, at least Meechan can say he had his day in court.

But a recent threat by the European Commission to hold platforms responsible for their users' posts would preclude such due process by pushing platforms to remove anything they are unsure about before any court has ruled it illegal, lest they find themselves on trial as well.

On 1 March, the commission issued a recommendation (a form of non-binding soft law that does not have to go through the legislature) demanding that online platforms take "proactive measures" to remove all varieties of illegal content posted by users, which it defines as "any information which is not in compliance with Union law or the law of the member state concerned."

The commission threatened regulation if platforms do not comply.

But increasing platforms' liability would drastically alter the effects of the different national laws on free speech.

The threat of fines, combined with the grey areas inherent to laws limiting what people can say, would push platforms to remove content whenever in doubt, even where a court might let it remain online—either because the court decides no crime has been committed, or because the law punishes the author without requiring any intermediary to remove the content.

The legal limits on free expression vary tremendously between European countries.

'Blasphemy' laws still apply

Nearly half of EU member states have laws against lese majeste (insulting the head of state) and about half have some form of law against blasphemy (insulting religion).

Several criminalise holocaust denial, while Poland has outlawed statements that implicate "the Polish nation" in the crimes of Nazi Germany, limiting debate about Polish collaborators.

UK communications law prohibits "grossly offensive" messages, and its courts have produced several controversial convictions, such as that of a man who joked on Twitter he would blow an airport "sky high" if it did not reopen to let him visit his girlfriend.

The judge presiding at Meechan's trial found him guilty of violating the same law.

Countries that maintain such laws also decide how to enforce them, and some do so with a lighter touch than others.

For example, the Danish penal code criminalizes lese majeste, but there are no recorded convictions.

Where the authorities do enforce laws against particular types of speech, they usually limit online platforms' liability.

Platforms are not always obliged to remove content even when courts convict someone of having committed a crime by uploading it. Whether the relevant content must come down depends on the judgement of the court and the powers given to it by the law.

German exception

Germany is an exception, where new legislation requires platforms to remove hate speech within 24 hours of a user having complained about it, rather than after a court has ruled on the matter.

Naturally, platforms should comply with court orders and be punished when they do not, but to safeguard free speech, claims of hate speech, illegally "offensive" messages, or similar allegations should be judged in court on a case-by-case basis.

If the EU makes platforms liable for what their users post, then throughout Europe, the threat of fines will pressure platforms to delete all dubious content, raising the likelihood that legal content will be removed too, limiting opportunities to challenge allegations of hate speech, and stifling public debate.

Of course, online platforms are entitled to their own rules about legal content they prohibit, such as graphic violence or pornography, but Internet users can at least choose between platforms with different terms of service, whereas they cannot pick and choose laws.

Another problem with the commission's recommendation is that it provides identical rules for almost all illegal content.

But illegal content is clearly not all the same.

Platforms can easily identify some types of obviously illegal content, such as pirated movies or child abuse imagery, whereas hateful or offensive comments are open to subjective interpretation.

Few people would mistake obscene images of abused children for protected opinion.

The recommendation does suggest tougher rules for terrorist propaganda, but even that is not always easy to separate from religious fundamentalism or political radicalism.

Thus, platforms should remove indisputably illegal content, but forcing them to do that for less obviously illegal content would threaten lawful free speech.

The major online platforms already for the most part adhere to the commission's voluntary code of conduct for illegal hate speech, which uses a common definition of hate speech that holds in all member states.

But making platforms liable for when users might violate different national laws would undermine the Digital Single Market and harm European Internet users.

The commission should withdraw both the recommendation its threat to regulate, and work with industry on voluntary practices to address clearly illegal activity, while leaving decisions about the limits to free speech to the courts.

Nick Wallace is a Brussels-based senior policy analyst at the Centre for Data Innovation. His Brussels Bytes column deals with the digital single market and data-related policy issues in the European Union