ECHR ruling might have a chilling effect on data journalism

As I was doing research on the definition of journalism by the European Court of Human Rights, I found a ruling from 21 July 2015 which has tremendous implications for the future of journalism in Europe.

Out-law.com, the only media outlet that covered the ruling, summarizes the case as follows [1] (emphasis mine):

[A Finnish company] has been publishing Veropörssi magazine since 1994, giving information about Finnish people’s taxable income and assets – information that is publicly available under Finnish law.

In 2003 [its sister company] started a text message service where interested parties could request tax information on individuals, using data that had already been published in the magazine.

The Finnish data protection ombudsman contacted the companies and advised them to stop publishing in this way [as it was infringing on the privacy of taxpayers]. The companies refused, because they felt that the request violated their freedom of expression, the ruling said.

After many appeals in Finland, the case landed in Strasbourg, where the judges of the European Court of Human Rights agreed with the Finnish state 6 to 1.

For the judges, an automated treatment of publicly available data could be considered journalism. For them, journalism is the activity of “disclos[ing] to the public information, opinions or ideas”. However, if the data set contains personal information, it can be argued that the need for privacy trumps the need for freedom of expression and that the latter takes precedence over the former.

What is interesting for datajournalists is to be found in §48. There, the Finnish government argues that publication of data “without journalistic comments” and in bulk made a piece of information unworthy of the protections given to journalism (the judges did not address this particular point of the argument, they only reiterated their definition of journalism). This means that automated publication of data, or robot-journalism, will be considered data processing, not journalism.

The most troubling aspect of the ruling comes in judge’s Nicolaou’s opinion. For him, the balance between the need for privacy and the need for openness lied in the journalistic character of the Finnish company. Because they were not “solely” journalists, he refused to follow the traditionally broad definition of journalism of the court’s jurisprudence. The status of the applicant played a role in the protection it enjoyed.

The only dissenting judge, Nona Tsotsoria, is right to say that the ruling “is incompatible with democracy”. That Mr Nicolaou is 70 and that Ms Tsotsoria is 42 might play a role in the matter.


This ruling would be legitimate if the data being published by the Finnish companies had been private. After all, Europe places a higher price on privacy than the United States do. But the information was already publicly accessible. The judges ruled that making public data on individuals more accessible and easier to use is illegal.

The Fair-Play Alliance, a Slovak NGO, face a similar lawsuit in 2011. It had released zNašichDaní.sk, a search engine that associated physical persons to public tenders their companies won using two public data sets. One of the persons in the database, Ms Považanová, complained that her being in the zNašichDaní database was a violation of her privacy, using the exact same arguments as the Finnish state [2]. A court of first instance ordered the website to remove information about Ms Považanová. The decision was quashed a few weeks later. That was in 2011. Today, the ECHR ruling would make it very hard for a court to reach the same conclusion and bring zNašichDaní.sk back up.


If you prepare a database on individuals (e.g of the indictments of German politicians, or a list of the properties owned by the richest Greeks) based on public data, you might be sued, all the more so if you are not working in a newsroom.

Notes

1. Read the full post here.

2. Read the analysis of the ePSI Platform.


Comments

Want to raise an issue about this post? Please open a new one on Github or make a pull request directly, but make sure to read the rationale behind this blog first.