EP Committee adopts short-sighted anti-”radicalisation” report

On 19 October 2015, after months of discussion, the European Parliament Committee on Civil Liberties, Justice and Home Affairs (LIBE) adopted an extremely incoherent and short-sighted anti-“radicalisation” report. The text is very-wide ranging, covering everything from prisons and schools to the Internet. Regarding the measures related to the Internet, there is a complete lack of evidence that they would be effective. Worse still, no effort has been spent on establishing if any of the Internet-related measures would be counter-productive.

One of the paragraphs on Internet regulation is a good case study for the failings of the report. It reads as follows:

“[1] Recalls that Internet companies and service providers have a legal responsibility to cooperate with Member States’ authorities by deleting any illegal content that spreads violent extremism expeditiously and with full respect for the rule of law, fundamental rights, including the freedom of expression; [2] believes that the Member States should consider legal actions, against Internet companies which refuse to comply with an administrative or judicial request to delete illegal content on their internet platforms; [3] believes that Internet platforms’ refusal or deliberate failure to cooperate allowing such illegal content to circulate should be considered an act of complicity that can be equated to criminal intent or neglect and should in those cases be brought to justice to that effect”.

With regard to the first sentence, what are the companies that the text is referring to? They are either “Internet companies” or “service providers” or “Internet platforms”. All of them, any of them? The text talks about legal responsibility, yet neither “Internet companies” nor “Internet platforms” is defined in relevant EU law. The term used by the relevant law, the E-commerce Directive, is “hosting provider”, which is not mentioned anywhere in the text. Furthermore, sadly, there is no clear law saying that companies have the legal obligation to delete content “in full respect of the rule of law, fundamental rights, including the freedom of expression”. In the world of excessive Internet provider liability, law is replaced with unpredictable terms of service. We welcome the inclusion of fundamental rights, but the text should have said that companies should respect the rule of law and fundamental rights in this context, as difficult as this is when the law pushes them towards imposing arbitrary restrictions.

With regard to the second sentence of the highlighted paragraph, the LIBE Committee thankfully followed our advice and deleted its original call for criminal sanctions. In the drafting process for this text, we had urged parliamentarians to replace the words “administrative request” with “judicial order”, but they consciously kept those words. But what is a “request” from an administrative authority? On what basis is it respectful of the rule of law to make a company liable for failing to react to a non-judicial “request”? Are there European hosting providers that are not acting expeditiously? Is there any evidence showing that EU Member States are not taking action in such circumstances? What is the problem that this text is trying to solve?

Third sentence: This makes the second sentence entirely redundant. It confuses the notion of a hosting provider (“platform”) which hosts messages from subscribers, and network providers, through which messages circulate. It completely abandons any notion of the rule of law, by calling for criminal sanctions against companies whose “failure to cooperate” (with what or whom, we don’t know) to stop “such” messages from circulating. “Such messages” means a category of messages – which somehow would need to be read, monitored and arbitrarily deleted in order to stop them from being posted. The hosting provider would need to do this in order to avoid criminal liability. And this, somehow “in full respect for the rule of law, fundamental rights, including (the) freedom of expression”.

But what if all of these extreme restrictions on our freedom of expression and our rule of law make us safer? Maybe a more urgent question is: What if they make us less safe? On the same day as the report was adopted, the Chief Constable of Greater Manchester Police warned, in the Guardian, that “there is a concern that efforts to control extremist narratives will limit free speech and backfire if we don’t get the balance right”, while a study of the Quillam Foundation on online extremism warned that “though governments are increasingly relying on censorship and filtering methods to counter online extremism, [their] report found that negative measures, or censorship in general, were not only ineffective and costly but potentially counter-productive.”

Legally inaccurate. Factually inaccurate. Probably reckless. Is this really the best we can expect from the European Parliament’s Civil Liberties Committee, that voted with a large majority in favour of this flawed text?

The final vote on this report will take place in the European Parliament plenary session in November.

Quillam Foundation: Jihad trending: A comprehensive analysis of online extremism and how to counter it (05.2014)
http://www.quilliamfoundation.org/wp/wp-content/uploads/publications/free/jihad-trending-quilliam-report.pdf

Anti-extremism drive puts British values at risk, says police chief (19.10.2015)
http://www.theguardian.com/uk-news/2015/oct/19/government-extremism-crackdown-hurt-uk-values-peter-fahy-manchester

EDRi-gram: ENDitorial: European Parliament – translating freedoms into Chinese (29.07.2015)
https://edri.org/enditorial-european-parliament-translating-freedoms-into-chinese/

EDRi-gram: EU Parliament’s “radicalisation” draft report – lost in translation (23.09.2015)
https://edri.org/eu-parliaments-radicalisation-draft-report-lost-in-translation/

Report on the prevention of radicalisation and recruitment of European citizens by terrorist organisations (03.11.2015)
http://www.europarl.europa.eu/sides/getDoc.do?pubRef=-%2f%2fEP%2f%2fNONSGML%2bREPORT%2bA8-2015-0316%2b0%2bDOC%2bPDF%2bV0%2f%2fEN

(Contribution by Joe McNamee, EDRi)