Internet content regulation and freedom of expression: three questions legislators should consider

Internet content regulators should take care: Not
everything that lookslike hate speech, is hate speech.

I am currently busy co-authoring a book for the Council of Europe on freedom of epxression on the Internet. In the process of preparing the materials I was amazed by how many documents we have by now expressly pointing out the importance of freedom of expression online. 

A very informative list can be found in what is probably the most important judgment on freedom of expression on the Internet we've yet seen by an international court: the European Court of Human RIghts's decision in Yildirim v. Turkey, §§19 et seq. (You can find a very good analysis here at the UK Human Rights Blog).

The Court chided Turkey for allowing an interference that does not meet the 

"requirement of foreseeability required by the Convention and did not allow the applicant to enjoy the sufficient degree of protection required by the rule of law in a democratic society. Moreover, this law appears to be directly contrary to the wording of paragraph 1 of Article 10 of the Convention, that the rights recognized therein apply “regardless of frontiers”(§ 67; Einglish translaten by UK Human Rights Blog)

States thus have a duty not ensure freedom of expression online just as offline. 

But legistic shortcomings are representative of a larger phenomenon. Some states have issues with developing a coherent framework of laws applicable to the Internet and or of refining existing laws to extend to the Internet. Such a duty has  been explicitly confirmed by the ECtHR in Editorial Board of Pravoye Delo and Shtekel v. Ukraine (in this case: extension of protection of journalists offline to online publications).

This made me think of an article I wrote in 2011 on the three key questions each legislator must ask when it comes to normative action in the Internet age. It's published in the collection of the 14th International Symposium on Legal Informatics, Salzburg, but I will republish my main arguments here:

• First, legislators have to consider whether to regulate or not. It is always difficult to pinpoint exactly when a social situation demands legal norms and up to when social norms are enough. If a state introduces too strict a rule too early in time this might hamper technological development. Then again, not ruling at all might lead to sectoral anarchy and human rights violations. 
• Second, legislators need to decide which entity should regulate. Often the self-regulatory powers of stakeholders will be enough but sometimes governments as the traditional rule-making authority come into play. 
• Third, the technical question of how to regulate needs to be answered. Depending on the normative goal, the normative and technological means will differ. 

The example of regulating hate speech is a good case in point. 

First, a state has to consider whether to fight (certain) online hate speech or allow it. International law helps answer that question as it sets down certain standards that all states must obey, such as the prohibition of incitement to genocide. But with regard to other hate speech, such as negationism, a country with a different historical from, say, Germany, may decide not to penalize holocaust denial. Some states, such as France or Switzerland, may also decide to penalize denial of other genocides will other states consider these denials covered by freedom of expression. 

On the second level, states have to consider which actor is best suited to implement the prohibition of hate speech. States may consider censorship by Internet Service Providers, relying on unofficial but widely circulated blacklists established by non-governmental organizations to be must effective or their own criminal law or a combination of both. 

On the third level, may also consider whether it makes more sense to blacklist sites or to have servers physically remove them. In assessing the different approaches states have to keep in mind that they should always choose the least invasive regulatory alternative to fight hate speech.

Taking this into account will help legislators decide whether to regulate or not, who to empower with regulatiory attempts and how to regulate.