Delfi vs Estonia in the Court of Human Rights
International media coverage of the adjudication:
Forbes: “every website that accepts comments now has a European problem”
Guardian: “the judgment which should send a shiver of fear down any website operator’s spine”.
Index on Censorship: “truly troubling judgment for website operators”.
Article 19: “the upshot of the Court’s judgment is that news portals should remove their comments sections in order to avoid liability”
On 10.10.2013 the European Court of Human Rights found that the convicting judgement of Estonian Supreme Court in the case of Vjatšeslav Leedo v Delfi is justified and proportional infringement on the freedom of expression of an Internet portal.
As legal counsel to Delfi I reserve my dissenting opinion for the following reasons:
Every webportal operator is wilfully blind?
The ECHR found that the story “SLK broke the planned ice route” published in Delfi in the year 2006 was a proper work of journalism. Unfortunately, the article’s subject matter had a negative impact on many people living on the Estonian islands, who that winter had to use only the payable ferry transport instead of the ice road free of charge. Therefore, Delfi should have realised that the article might cause negative reactions against the shipping company and its managers and that therefore there was a higher-than-average risk that the negative comments could go beyond the boundaries of acceptable criticism and reach the level of gratuitous insult.
Such an institute is called “wilful blindness” in legal terminology. A colourful example of the standard of wilful blindness is a 1992 court case in the United States which weighed whether a shopping centre’s owner is liable for the actions of a store operating in rental premises there. In that case, the store named HARD ROCK, operating in the shopping entire, sold trademarked T-shirts at prices reaching just a few tens of percent of the original prices, often having removed the original price tags and neckline labels. The court found that the shopping centre’s owner could not have overlooked such an activity. So, in legal terms – wilful blindness is a situation where a person should suspect a violation of the law but wilfully decides not to investigate. The same principle has been applied in the digital society to chatrooms, commentariums, e-stores and e-markets.
But until now, when applying that principle, the approach has been that the so-called party storing the information must have provoked the third party’s unlawful activity with its own behaviour. In Delfi’s case – the article should have had provocative or other kind of nature that would have provoked “bad comments”, in order to apply the heightened diligence standard. However, in this case, the provoker was not Delfi who plainly reported an event, but the victimised party itself who broke the ice route.
Padraig Reidy mentions with irony that inter alia, the ECHR judgement is curious because “any moderator will tell you that controversial comments can appear in the unlikeliest of places”. I agree. Perhaps the only thing not provoking a single negative emotion is a forecast of sunny weather in July. Although – even this could not be said in absolute certainty since it could cause lots of “negative emotions” to someone waiting for rain to irrigate their potato field. Thus the ECHR’s judgement, while grammatically referring to “some articles” where heightened diligence applies, means in practice that liability extends to every article.
Insults would only be prevented by … abolishing commentariums?
The ECHR identified inter alia that Delfi has not been entirely negligent. Namely, commenters were informed about their liability for what they publish, via the commenting terms and conditions. Also, an automatic word filtering system was in use to identify certain vulgar word stems. Moreover, a notice-and-take-down notification system was in place which the court assessed to be an “easy-to-use measure by simply clicking on a button”.
Having identified the foregoing, the court still found that the measures implemented by Delfi were insufficient to avoid causing harm to persons’ reputation.
What would be the measure that would prevent causing harm to persons’ reputation? I think an absolute guarantee would be abolishing commentariums. But is this really necessary and reasonable in a democratic society? I doubt it.
Liability would be prevented by … personal identification?
In its adjudication, the ECHR acknowledges the Internet users’ wish to remain anonymous. This is reasonable! Especially considering that the ECHR is a human rights body of the European Council and that the most important declaration of the European Council’s Committee of Ministers regarding the Internet states that in order to ensure protection against online surveillance and to enhance the free expression of information and ideas, member states should respect the will of users of the Internet not to disclose their identity (principle of anonymity).
At the same time, the court considers that the liability of an Internet portal must be acknowledged inter alia when the portal has decided to allow comments by non-registered users. The CHR seems to agree that commenting should be allowed for persons having authenticated themselves by their real name or by registration in some form. I dissent. Namely:
1. Obligation to identify by real name as a prerequisite for commenting would be naïve and silly, to say it mundanely.
An example supporting the argument is in order here. In 2007, to restrict cyber-libel, South Korea adopted a law establishing a compulsory system of identification by real names in Internet portals with high traffic (above 100,000 users per day). According to the law’s impact analysis, the total number of comments decreased significantly, but the number of abusive comments did not really decrease and international companies found ways to avoid the system. Consequently, the Constitutional Review Court of South Korea found on 23.08.2012 that the aforementioned law was unconstitutional due to being effectively a pre-censorship measure and also proven to be inefficient in achieving its intended aim;
2. An obligation to register in some other form (e.g. via e-mail) is insubstantial and would not fulfil any goals whatsoever. The reason is that the portal already has the commentator’s IP address. If we wanted to use registration to “catch” a commenter who took the trouble to use a virtual private network (VPN) or a proxy to hide his IP address, we would not be able to catch them via the registration either. After all, creating a false identity in Internet is just a single Gmail account away.
I’m convinced that “anonymity” does not equal “bad comments”. This is also confirmed by several studies – according to the research of Disqus, a provider of commenting and storage service, bad comments are divided as follows: commenters using a pseudonym and commenters posting anonymously both posted “bad comments” in 11% of cases; commenters posting under their real name posted “bad comments” in 9% of cases. So – any “pre-censorship method” would net us a reduction of “bad comments” by a whopping few percent. Is it worth it? I sincerely believe it is not.
And then there’s the hairy beast called “censorship”, of course… I allege that every measure having the consequence of precluding the opinions of certain people can be considered censorship. Real name policy would pretty effectively preclude any criticism of employees in comments to an article analysing their employer’s activities, similar to any criticism of members of a political party in comments to an article discussing the party, etc. I will not even bother to explain here what anonymity on Internet means to people around the world living under repressive regimes. The Constitutional Review Court of South Korea also realised all of this when abolishing country-wide real name policy.
So what next?
There is a reason why major European publications worry whether commentariums will survive the judgement of Delfi v Estonia. It’s the reason why Estonian internet portals should also be worried. Blogs and other providers of new media allowing interactivity should be concerned as well.
A warning bell sounds in the words of writer Aldous Huxley: “Technological progress has merely provided us with more efficient means for going backwards.”