Fake News Legislation: A Human Rights Problem? Assessing the Compatibility of Fake News Legislation with Article 10 of the European Convention on Human Rights

(1)

University of Amsterdam

Faculty of Law

Fake News Legislation: A Human Rights Problem?

Assessing the Compatibility of Fake News Legislation with

Article 10 of the European Convention on Human Rights

Anamarija Primorac

Master‟s Thesis LL.M. International and European Law

Public International Law Track

Supervised by Dr. Nataša Nedeski

Submitted 19 July 2019

(2)

Anamarija Primorac

anamarija.primorac@uzh.ch

12269484

(3)

In light of a recent trend in Europe to legislate with respect to the phenomenon of fake news online, this paper sets out to add to the current debate on the human rights problems of fake news legislation. It presumes that recent legislation could inspire applications to the European Court of Human Rights and asks with respect to negative state obligations if fake news legis-lation is compatible with the right to freedom of expression as enshrined in Article 10 of the European Convention on Human Rights.

The paper finds that an act of implementation of a fake news law constitutes an inter-ference with the right to freedom of expression of internet users and platform operators. It holds that the Court is unlikely to find fake news legislation incompatible with the right to freedom of expression with respect to the criteria „prescribed by law‟ and „pursuit of a legiti-mate aim‟, but that it is likely to find fake news legislation not necessary in a democratic so-ciety and thus incompatible with the right to freedom of expression as enshrined in Article 10 of the Convention. On the basis of this prognosis of the Court‟s likely findings, the paper con-cludes that fake news legislation is incompatible with the right to freedom of expression. The paper suggests that while fake news may present a danger to the quality of democratic debate, the greater danger to democracy may lie in exaggerated state responses to the phenomenon.

(4)

Bibliography

Alemanno A, „Editorial: How to Counter Fake News? A Taxonomy of Anti-fake News Approaches‟ (2018) 9 European Journal of Risk Regulation 1.

Aswad E. M, „Are Recent Governmental Initiatives to Combat Online Hate Speech, Extrem-ism and Fraudulent News Consistent with the International Human Rights Regime?‟ in Donahoe E and Hampson F. O (eds), Governance Innovation for a Connected World, Special Report 2018.

Baumbach T, „Chilling Effect as a European Court of Human Rights‟ Concept in Media Law Cases‟ (2018) 6(1) Bergen Journal of Criminal Law and Criminal Justice 92.

Benedek W and Kettemann M. C, Freedom of Expression and the Internet (Council of Europe Publishing 2013).

Benkler Y, Faris R and Roberts H, Network Propaganda: Manipulation, Disinformation, and

Radicalization in American Politics (OUP 2018).

De Londras F and Dzehtsiarou K, Great Debates on the European Convention on Human

Rights (Palgrave 2018).

de Morree P. E, Rights and Wrongs under the ECHR: The prohibition of abuse of rights in

Article 17 of the European Convention on Human Rights (Intersentia 2016).

Eskens S, Helberger N and Moeller J, „Challenged by news personalisation: five perspectives on the right to receive information‟ (2017) 9(2) Journal of Media Law 259.

Farkas J and Schou J, Post-Truth, Fake News and Democracy: Mapping the Politics of

False-hood (Routledge forthcoming 2019).

Fathaigh R Ó, „The Chilling Effect of Liability for Online Reader Comments‟ (2017) 4

Euro-pean Human Rights Law Review 387.

Gonzalez A and Schulz D, „Helping Truth with Its Boots: Accreditation as an Antidote to Fake News‟ (2017) 126 Yale Law Journal 315.

Grabenwarter C, European Convention on Human Rights: Commentary (Beck 2014).

Hollyer J. R, Rosendorff B. P and Vreeland J. R, Information, Democracy and Autocracy:

Economic Transparency and Political (In)Stability (Cambridge University Press 2018).

Humprecht E, „Where „fake news„ flourishes: a comparison across four Western democracies‟ (2018) Information, Communication & Society 1.

Katsirea I, „”Fake News”: reconsidering the value of untruthful expression in the face of regu-latory uncertainty‟ (2018) Journal of Media Law 1.

(6)

Lavrysen L, „System of restrictions‟ in Pieter van Dijk et al. (eds), Theory and Practice of the

European Convention on Human Rights (5th ed. Intersentia 2018).

Leerssen P, „Cut Out By The Middle Man: The Free Speech Implications Of Social Network Blocking and Banning In The EU‟ (2015) 6(2) Jipitec 89.

MacKinnon R and Pakzad R, „Private Sector Roles and Responsibilities: Protecting Quality of Discourse, Diversity of Content and Civic Engagement on Digital Platforms and Social Media‟ in Donahoe E and Hampson F. O (eds), Governance Innovation for a Connected

World, Special Report 2018.

McGonagle T, „Interventarisatie methodes om „nepnieuws“ tegen te gaan‟ (2018) Institute for Information Law, University of Amsterdam.

McGonagle T, „”Fake News”: False fears or real concerns?‟ (2017) 35(4) Netherlands Quar-terly of Human Rights 203.

Mégret F, „Nature of Obligations‟ in Daniel Moeckli et al. (eds), International Human Rights

Law (3rd ed. OUP 2018).

Murphy T and Cuinn G. Ó, „Works in Progress: New Technologies and the European Court of Human Rights‟ (2011) 10 Human Rights Law Review 601.

Pöschl M, „Neuvermessung der Meinungsfreiheit?„ in Koziol H (ed.), Tatsachenmitteilungen

und Werturteile: Freiheit und Verantwortung (Jan Sramek 2018).

Rowbottom J, „To Rant, Vent and Converse: Protecting Low Level Digital Speech‟ (2012) 71(2) Cambridge Law Journal 355.

Syed N, „Real Talk About Fake News: Towards a Better Theory for Platform Governance‟ (2017) 127 Yale Law Journal 337.

Tandoc E. C, Lim Z. W and Ling R, „Defining “Fake News”‟ (2018) 6(2) Digital Journalism 137.

van Rijn A, „Freedom of Expression‟ in van Dijk P et al. (eds), Theory and Practice of the

European Convention on Human Rights (5th ed. Intersentia 2018).

Verstraete M, Bambauer D. E and Yakowitz Bambauer J. R, „Identifying and Countering Fake News„ (2017) Arizona Legal Studies Discussion Paper No. 17-15.

Woods L, „Raising Questions About Fake News‟ (2017) 45(3) InterMEDIA 9.

Zysset A, „Freedom of expression, the right to vote, and proportionality at the European Court of Human Rights: An internal critique‟ (2019) 17(1) International Journal of Constitu-tional Law 230.

(7)

A. Introduction

“Everyone is entitled to his own opinion, but not his own facts.”1

In December 2018, an article in New York Magazine reads: “What is gone from the internet, after all, is not „truth‟, but trust: the sense that the people and things we en-counter are what they represent themselves to be.”2

The author paints a gloomy pic-ture of the internet as a sphere for communication: what was once hailed for its de-mocratising potential3 is now the object of profound distrust.

The above exemplifies a discussion about the flaws of the online communica-tive sphere that has permeated global public discourse in the past years. The debate boils down to one prominent term: fake news. Its prominence is easily demonstrated: A Google search of the term yields over a billion entries.4 A new term has been coined to describe an era in which facts are less influential in shaping public opinion: „post-truth‟, which was named word of the year 2016.5

Fake news is referred to so frequently the term is said to have lost all meaning.6

The term has also inspired a large number of scholarly contributions. Com-mon narratives in the academic literature are that firstly, fake news is not a new phe-nomenon, but that there are features of the online sphere that exacerbate its negative effects7 and secondly, that the primary concern with fake news is its adverse effect on opinion formation and therefore democratic debate.8 Having acknowledged the existence of the phenomenon and its problematic character, scholarly contributions mostly advise on what would be effective regulatory responses.9

1_{Daniel Patrick Moynihan, former Harvard professor and United States senator from New York.} 2_{Max Read, „How Much of the Internet Is Fake? Turns Out, a Lot of It, Actually.‟ New York}

Maga-zine (New York, 26 December 2018) <http://nymag.com/intelligencer/2018/12/how-much-of-the-internet-is-fake.html>. All internet sources were last accessed 18 July 2019.

3_{Wolfgang Benedek and Matthias C. Kettemann, Freedom of Expression and the Internet (Council}

of Europe Publishing 2013) 73.

4

The search yields 1.02 billion entries.

5_{<https://global.oup.com/academic/content/word-of-the-year/?cc=nl&lang=en&>.}

6_{Mark Verstraete, Derek E. Bambauer and Jane R. Yakowitz Bambauer, „Identifying and}

Counter-ing Fake News„ (2017) Arizona Legal Studies Discussion Paper No. 17-15, <https://papers.ssrn. com/sol3/papers.cfm?abstract_id=3007971> 4.

7

E.g. Alberto Alemanno, „Editorial: How to Counter Fake News? A Taxonomy of Anti-fake News Approaches‟ (2018) 9 European Journal of Risk Regulation 1, 1f.; Tarlach McGonagle, „”Fake News”: False fears or real concerns?‟ (2017) 35(4) Netherlands Quarterly of Human Rights 203, 205f.; Nabiha Syed, „Real Talk About Fake News: Towards a Better Theory for Platform Govern-ance‟ (2017) 127 Yale Law Journal 337, 337.

8_{E.g. Irini Katsirea, „”Fake News”: reconsidering the value of untruthful expression in the face of}

regulatory uncertainty‟ (2018) Journal of Media Law <https://doi.org/10.1080/17577632.2019. 1573569> 9ff.; McGonagle (n 7) 204.

9_{E.g. Alemanno (n 7); Anna Gonzalez and David Schulz, „Helping Truth with Its Boots:}

(8)

In light of the vast amount of information available on the topic, what merit is there to yet another paper on fake news? Is there anything left to say? I contend that there is.

B. Context and research question

The premise of the argument that follows is that the debate on fake news has evolved since it emerged in 2016 following the US presidential election.10 While in an early first stage, the debate was marked by the effort to understand the phenomenon and its adverse effects,11 today‟s second stage is informed by concerns regarding potential human rights problems arising out of regulatory responses to fake news.12 This shift in the debate has come about as a number of countries have introduced laws to tackle fake news online,13 causing public outcry over the free speech ramifications of these laws and concerns about the fight against fake news being used as pretext to stifle dissent.14

The present paper aims to contribute to the current debate. It presumes that today‟s debate on the potential human rights problems of fake news laws can be en-riched by looking at the European Court of Human Rights‟ jurisprudence on the right to freedom of expression, and explores the insights that can be gained from the Court‟s jurisprudence.

The Court has not yet had to decide on the compatibility of fake news laws with the Convention.15 Consequently, the paper attempts to answer the question: Is

Roya Pakzad, „Private Sector Roles and Responsibilities: Protecting Quality of Discourse, Diver-sity of Content and Civic Engagement on Digital Platforms and Social Media‟ in Eileen Donahoe and Fen Osler Hampson (eds), Governance Innovation for a Connected World, Special Report 2018; Verstraete et al. (n 6).

10

The term has become popular following the presidential election, Syad (n 7) 337.

11_{E.g. Yochai Benkler, Robert Faris and Hal Roberts, Network Propaganda: Manipulation,}

Disin-formation, and Radicalization in American Politics (OUP 2018).

12

E.g. Adam Satariano, „Europe Is Reining In Tech Giants. But Some Say It‟s Going Too Far.‟ The New York Times (New York, 6 May 2019) <www.nytimes.com/2019/05/06/technology/europe -tech-censorship.html> and Evelyn Mary Aswad, „Are Recent Governmental Initiatives to Combat Online Hate Speech, Extremism and Fraudulent News Consistent with the International Human Rights Regime?‟ in Donahoe and Hampson (n 9).

13

See for an overview of global anti fake-news action Daniel Funke, „A guide to anti-misinformation actions around the world‟ Poynter <www.poynter.org/ifcn/anti-misinformation-actions/>.

14_{E.g. „Legislation against fake news is open to abuses‟ Financial Times (London, 7 April 2019)}

<www.ft.com/content/b1d78fc2-57b4-11e9-a3db-1fe89bedc16e>; „Singapore fake news law a „disaster‟ for freedom of speech, says rights group‟ The Guardian (London, 9 May 2019) <www.theguardian.com/world/2019/may/09/singapore-fake-news-law-a-disaster-for-freedom-of-speech-says-rights-group>.

15

The case Dareskizb Ltd v. Armenia concerned the compatibility with Article 10 of a ban on “pub-lications or dissemination by mass media outlets of obviously false or destabilizing information on State and internal issues” during a state of emergency. It led to the submission of written

(9)

observa-fake news legislation compatible with the right to freedom of expression as enshrined in Article 10 of the European Convention on Human Rights?

The paper focuses on negative state obligations arising from the right to free-dom of expression, requiring states to abstain from unlawful interferences with the rights of individuals.16 It approaches the research question in the legal framework used by the Court when assessing potential violations of the right to freedom of ex-pression,17 namely by way of the following four subquestions: Has there been an

interference with the applicant’s right to freedom of expression? Is the interference prescribed by law? Does the interference pursue a legitimate aim? Is the interfer-ence necessary in a democratic society?

The paper does not aim to assess the compatibility of a specific piece of legis-lation with the Convention, but rather to point to potential problems arising out of this type of legislation at each stage of the Court‟s legal assessment. In order to an-swer the research question, it takes note of existing jurisprudence on the right to freedom of expression and attempts to predict how the Court would apply the exist-ing jurisprudence and the legal framework (interference with a right, prescribed by law, pursuit of a legitimate aim, and necessity in a democratic society) to fake news legislation. On the basis of a prediction of the Court‟s likely findings, the paper will hold whether fake news legislation is compatible with Article 10 of the Convention.

The goal of the paper is twofold: firstly, to outline how the Convention‟s le-gal framework might be applied to fake news legislation in potential future cases before the Court and secondly, to highlight the problems of fake news legislation with respect to the right to freedom of expression more generally, in an attempt to inform the making and the application of fake news legislation in the future.

tions of a non-governmental organization, in which the incompatibility of false news provisions with the right to freedom of expression in international human rights law was outlined. See Dare-skizb Ltd v. Armenia (lodged on 16 December 2008), Statement of Facts and Questions to the Parties, available at <http://inforrm.org/wp-content/uploads/2012/03/61737-08-statement-of-facts-and-questions-to-the-parties.doc> and Media Legal Defence Initiative (MLDI), Submission before the European Court of Human Rights <www.inforrm.files.wordpress.com/2012/03/mldi-sub mission-final-as-sent.pdf>. At the time of writing, the case is pending.

16_{Frédéric Mégret, „Nature of Obligations‟ in Daniel Moeckli et al. (eds), International Human}

Rights Law (3rd ed. OUP 2018).

17_{Arjen van Rijn, „Freedom of Expression‟ in Pieter van Dijk et al. (eds), Theory and Practice of the}

(10)

C. Definitions

Although any reader presumably has an understanding of what is meant when one speaks of “fake news” and “fake news laws”, it is helpful to define these terms. Any definitions are necessarily working definitions, given that a large number of publica-tions explore how the term fake news is most accurately defined.18

The term “fake news law” shall cover any adopted piece of legislation in a Council of Europe member state which seeks to tackle the problem of fake news online, irrespective of the strategy employed to do so. The term shall also cover pieces of legislation that are not primarily advocated as fake news laws, but as laws targeting harmful online content more generally.19 Defined this way, fake news laws exist in the following Council of Europe member states: France,20 Germany,21 Ire-land22 and Russia.23

As regards a definition of the term fake news, it would seem reasonable to look into the definitions in the fake news laws enumerated above. However, the laws offer little guidance. The term is either not explicitly defined,24 is defined with regard to a specific context, e.g. elections25 or is primarily defined with reference to its con-sequences.26 Moreover, as the vagueness of legal definitions of the term is in itself a point of criticism,27 this paper will draw its working definition from academic litera-ture instead.

18_{E.g. Katsirea (n 8) 4ff.; Edson C. Tandoc Jr., Zheng Wei Lim and Richard Ling, „Defining “Fake}

News”‟ (2018) 6(2) Digital Journalism 137; Verstraete (n 6) 5ff.; Loorna Woods, „Raising Ques-tions About Fake News‟ (2017) 45(3) InterMEDIA 9, 9ff.

19

E.g. the German law targets „illegal content‟. For a definition, see Article 1 Paragraph 3 Gesetz zur Verbesserung der Rechtsdurchsetzung in sozialen Netzwerken, <www.gesetze-im-internet.de/ netzdg/__1.html>.

20

Loi n° 2018-1202 du 22 décembre 2018 relative à la lutte contre la manipulation de l‟information, <www.legifrance.gouv.fr/affichTexte.do?cidTexte=LEGITEXT000037849750&dateTexte=20190 519>.

21_{(n 19).} 22

Online Advertising and Social Media Transparency Act 2017, <https://www.oireachtas.ie/en/bills/ bill/2017/150/>.

23_{Federal Law on Amending Article 15-3 of the Federal Law on Information, Information}

Tech-nologies and Protection of Information and Federal Law on Amending the Code of Administrative Violations (18 March 2019) <www.loc.gov/law/foreign-news/article/russia-russian-president-signs -anti-fake-news-laws/>.

24_{The German law refers to illegal content, see (n 19).}

25_{Article L163-2 of the French law (n 20), which refers to “inaccurate or misleading allegations}

or imputations of fact of a kind to affect the sincerity of of the upcoming vote” (translation my own).

26_{See the Russian definition drawn from (n 13): “Untruthful socially significant information}

dis-guised as authentic reports, which poses a threat to people‟s lives and health and is fraught with mass violations of public order and security, disruption in the operation of crucial life support facilities, transport, and social infrastructure or other grave consequences”.

(11)

The paper will work with the following definition: “Fake news is information that has been deliberately fabricated and disseminated with the intention to deceive and mislead others into believing falsehoods or doubting verifiable facts; it is disin-formation that is presented as, or is likely to be perceived as, news.”28

The definition points to three elements: deliberate fabrication of information, dissemination with the intention to deceive the recipient into believing the false information and news char-acter. This definition is adopted in the present paper because it is able to capture the contemporary meaning of the term. To illustrate this, consider how information that meets the above criteria was at the heart of a recent fake news headline. The headline read: “Fake news sparks anti-Roma violence in France”.29

The article reported that acts of violence had been committed against members of the Roma ethnic, after de-liberately false information had been circulated on social media about child abduc-tions conducted by the Roma.

One point to keep in mind is that speech that meets the present criteria may meet other definitional criteria and thus overlap with other categories of speech. Most notably, fake news can overlap with hate speech, and the implications of this overlap will be explored throughout the paper.30

A related category of speech that is discussed in the fake news literature, and will hence be discussed in the present paper, is the category of „misleading news‟.31 This category of speech takes into account that nuances exist with respect to two of the defining elements of fake news, namely „deliberate fabrication of information‟ and „dissemination with the intention to deceive‟. Firstly, information must not al-ways be deliberately fabricated: truthful and untruthful elements may be combined; information may be selectively withheld to fit a certain narrative; truthful facts may be subjected to ideological interpretation. There exist varying degrees of falsity of information, as fact and truth are not easily separated.32 Moreover, facts and opinions might overlap, as the two are also not easily separated.33 Secondly, there is not al-ways an underlying intention to deceive the recipient into believing falsehoods. It is conceivable that the individual disseminating the information is itself “blind” to its

28_{McGonagle (n 7) 203.} 29

Jack Guy, „Fake news sparks anti-Roma violence in France‟ (27 March 2019) CNN <www.edition .cnn.com/2019/03/27/europe/paris-fake-kidnapping-scli-intl/index.html>.

30_{Pages 12, 14f., 18, 20f and 23f.}

31_{See e.g. European Association for Viewers Interest, „Beyond Fake News‟,}

<www.eavi.eu/beyond-fake-news-10-types-misleading-info/>. McGonagle (n 7) 204 refers to this infographic.

32_{MLDI (n 15) para. 19.} 33_{ibid para. 18.}

(12)

falsity, for instance because the information coincides with his or her political views. In this paper, misleading news shall thus stand for speech where truth, falsity and opinion are present to varying degrees.

Lastly, fake news has been called an „Internet phenomenon‟,34

as the primary ground for the flourishing of fake news is the online sphere. It follows that whenever the paper mentions fake or misleading news, it is referring to news online.

D. Method

I. Focus on a regional human rights system

A few remarks are in order as to why the present paper focuses on the European Convention on Human Rights and the European Court of Human Rights.

Firstly, a number of Council of Europe member states are among the coun-tries that have introduced laws targeting fake news online.35 Indeed, it has been noted that there is a growing trend in Europe to regulate harmful online content.36 Most recently, Russia has passed a law inter alia allowing for the imposition of fines on individuals who circulate false information online, which has been criticized for be-ing a disguised censorship law.37 This demonstrates that the debate on the human rights compatibility of fake news laws is highly relevant in Europe. The European Convention on Human Rights provides a common legal framework in which to ground the debate, without having to address the details of individual pieces of legis-lation.

Secondly, while the Court has not yet decided on the compatibility of laws targeting fake news with the Convention, it is not unlikely that it will have to do so in the future, in light of both the growing body of European fake news legislation and recent strategic litigation in the area of digital freedoms.38 In any case, it seems a good point in time to take note of the existing jurisprudence of the Court on the right

34_{Edda Humprecht, „Where „fake news‟ flourishes: a comparison across four Western democracies‟}

(2018) Information, Communication & Society <https://doi.org/10.1080/1369118X.2018.14742 41> 2.

35

France, Germany, Ireland and Russia, page 4.

36_{Satariano (n 12).}

37_{„Putin Signs „Fake News‟, „Internet Insults‟ Bills Into Law‟ The Moscow Times (18 March 2019)}

<www.themoscowtimes.com/2019/03/18/putin-signs-fake-news-internet-insults-bills-into-law-a64 850> and Human Rights Watch, „Joint Statement on Russia‟s “Sovereign Internet Bill”‟ (24 April 2019) <www.hrw.org/news/2019/04/24/joint-statement-russias-sovereign-internet-bill>.

38_{A Polish NGO has filed suit against Facebook for banning content, possibly indicating a trend}

of bold strategic litigation in the area of digital freedoms. See Samuel Stolton, „Facebook hit by landmark censorship lawsuit in Poland‟ (7 May 2019) <www.euractiv.com/section/digital/news /facebook-hit-by-landmark-censorship-lawsuit-in-poland/>.

(13)

to freedom of expression and discuss its implications with respect to fake news legis-lation.

Thirdly, the Court‟s „living instrument‟ and „effective protection‟ doctrines – guiding the interpretion of the Convention in light of present-day conditions to en-sure the effective protection of rights39 – suggest that the Court holds the necessary tools to address issues of human rights protection in the online sphere.40 Moreover, as will become evident in the following, much of the Court‟s existing case law is of relevance to laws targeting fake news.

II. Fake news and democratic debate

It was hinted at earlier that a common argument in the fake news debate was the ad-verse effect of fake news on the quality of democratic debate.41 The argument is straightforward: the circulation of false information in a democratic society, the fal-sity of which is deliberately concealed, tampers with the information basis on which individuals form their political opinions.42 To illustrate this argument, consider the following fabricated quote of a left-wing German politician that was circulated online after an asylum seeker had raped and murdered a German student: “The trau-matised young refugee may have killed someone, but we must still help him.”43

Be-lieving that this statement was indeed made may prompt individuals to support po-litical parties with a more restrictive immigration agenda. Granted, the argument is somewhat simplistic, and the effects of fake news on democratic debate are the sub-ject of complex studies.44 Nonetheless, assuming that an adverse effect indeed exists – in that the ideal conditions for democratic deliberation between informed citizens are not present – adds a new layer of complexity to the discussion.

To illustrate this new layer of complexity, consider the following: Democracy is a common theme in the Court‟s case law, including in its case law on Article 10.

39_{Fiona de Londras and Kanstantsin Dzehtsiarou, Great Debates on the European Convention on}

Human Rights (Palgrave 2018) 71ff.

40_{Thérèse Murphy and Gearóid Ó Cuinn, „Works in Progress: New Technologies and the European}

Court of Human Rights‟ (2011) 10 Human Rights Law Review 601, 635.

41_{Page 1.}

42_{McGonagle (n 7) 203f.} 43

Guy Chazan, „Risk of refugee „fake news‟ rattles German politics‟ Financial Times (London, 15 February 2017) <www.ft.com/content/11410abc-ef6e-11e6-ba01-119a44939bb6>.

44_{For studies on the wider democratic and societal impact of fake news, see e.g. Benkler, Faris and}

Roberts (n 11); Johan Farkas and Jannick Schou, Post-Truth, Fake News and Democracy: Map-ping the Politics of Falsehood (Routledge forthcoming 2019); James R. Hollyer, B. Peter Rosen dorff, James Raymond Vreeland, Information, Democracy and Autocracy: Economic Transpar-ency and Political (In)Stability (Cambridge University Press 2018).

(14)

For instance, the criterion “necessary in a democratic society” is present in every assessment of a potential violation of Article 10 of the Convention45 and the Court has held that the Convention “is designed to maintain and promote the ideals and values of a democratic society”.46 One author, having studied the Court‟s notion of a “democratic society”, concludes that the Court‟s underlying normative ideal is that of an “informed, plural and contradictory debate on issues of public interest.”47

If the Convention is designed to maintain and promote the ideals of a democ-ratic society, the Court‟s normative ideal of a democdemoc-ratic society is one in which an informed debate between informed citizens takes place, and fake news undermines this ideal (in that the optimal conditions for democratic deliberation are not present), then one would assume that the adverse effect of fake news on democratic debate would be reflected in the assessment of the compatibility of fake news legislation with Article 10 of the Convention. Consequently, the paper will make suggestions how the adverse effect of fake news on democratic debate could be incorporated in the application of the Convention‟s legal framework to fake news legislation.

E. Literature review

As regards the existing literature on the topic, it can be noted that a relatively small number of publications address fake news in connection with the European Conven-tion on Human Rights.48 When the Convention is mentioned with respect to fake news, it is in the context of setting out positive state obligations with regard to the media landscape, followed by regulatory advice on fake news.49 As regards fake news and negative state obligations, one publication focuses on the compatibility of fake news legislation with international human rights law.50 Presently, it seems that no publication comprehensively addresses the question the present paper seeks to answer.

45_{Laurens Lavrysen, „System of restrictions‟ in Pieter van Dijk et al. (n 17) 315ff.}

46_{Kjeldsen, Busk Madsen and Pedersen v. Denmark App nos 5095/71, 5920/72 and 5926/72 (ECtHR}

7 December 1976) para. 53.

47

Alain Zysset, „Freedom of expression, the right to vote, and proportionality at the European Court of Human Rights: An internal critique‟ (2019) 17(1) International Journal of Constitutional Law 230, 235.

48

E.g. McGonagle (n 7); Katsirea (n 8); Patrick Leerssen, „Cut Out By The Middle Man: The Free Speech Implications Of Social Network Blocking and Banning In The EU” (2015) 6(2) Jipitec 89. For non-English literature, see Tarlach McGonagle, Eugénie Coche and Melanie Klus, „Inventari satie methodes om „nepnieuws“ tegen te gaan„ (2018) Institute for Information Law, University of Amsterdam.

49_{The publications in (n 48) mention positive obligations.} 50_{Aswad (n 12).}

(15)

F. Has there been an interference with the right to freedom of expression?

It was mentioned earlier that the introduction of fake news legislation in various countries had evoked concerns with respect to the right to freedom of expression.51 But whose right to freedom of expression is at stake?

The answer to this question first and foremost depends on which actors are targeted by fake news laws. The laws adopted in Council of Europe member states can be grouped into two categories: those targeting platform operators (defined in the legislation and encompassing large tech companies like Google, Facebook etc.) and those targeting internet users. The German and French law target platform operators and thus fall within the former category, the Irish law targets internet users and thus falls within the latter. The Russian law falls within both categories. By way of illus-tration, consider the following simplified descriptions of the laws:52 The German law requires platform operators to remove illegal content within a short timeframe or be fined. The French law imposes transparency obligations on platform operators in the three months preceding elections, namely to disclose who has purchased campaign ads and for what price or be fined, and authorizes judges to halt the dissemination of misinformation before elections on request. The Irish law makes using a bot to create multiple fake accounts and spreading political messages a criminal offense. The Rus-sian law requires social networks to remove inaccurate comments of their users or be fined and states that anyone who shares false information online can be fined.

It follows from the above that the right to freedom of expression of platform operators and internet users could be affected by fake news laws.

A next question to consider is the following: Would an act implementing a fake news law legally be considered an interference with the right to freedom of ex-pression of internet users and platform operators?

I. Interference with the right to freedom of expression of internet users

The right to freedom of expression as enshrined in Article 10 of the Convention ap-plies to both natural and legal persons.53 The Court has recognized the importance of the internet as a medium for expressive activity. It has namely held that “user-generated expressive activity on the Internet provides an unprecedented platform for

51

Page 2.

52_{The descriptions are drawn from (n 13).} 53_{van Rijn (n 17) 767.}

(16)

the exercise of freedom of expression”.54

Moreover, it has held that “the Internet plays an important role in enhancing the public‟s access to news and facilitating the dissemination of information in general.”55

On a more general note, the Court‟s free-dom of expression jurisprudence is equally applicable to the internet.56 It is irrelevant what type of message is communicated online and the use for commercial purposes falls within the scope of the provision.57 It follows that internet users‟ expressive activity online – the sharing of opinions and information – falls under the wording of “the right to impart information and ideas” as enshrined in Article 10 of the Conven-tion and is thus prima facie protected under the provision.

A follow-up question is whether the activity of disseminating fake news online also falls within the scope of protection of Article 10. Is there a right to un-truthful speech?58 After all, if fake speech masked as news was not protected under the provision, no human rights issues would arise with fake news laws.

Although the Court has not explicitly dealt with fake news as defined in this paper, it has held with regard to presumably untruthful speech: “Article 10 of the Convention as such does not prohibit discussion or dissemination of information re-ceived even if it is strongly suspected that this information might not be truthful.”59 To suggest otherwise, the Court held, “would deprive persons of the right to express their views and opinions about statements made in the mass media.”60

It follows that the discussion of or dissemination of presumably false information falls within the scope of Article 10. What does this imply for fake news and the related category of misleading news?

Firstly, the Court‟s rationale for protecting the circulation of false information is that this allows for the expression of views and opinions on statements in the mass media, which could be rephrased as: allows for public debate. Following this logic, the category of misleading news – where truth, falsity and opinion are present to varying degrees – certainly stimulates public debate. Moreover, some degree of false information circulates in every democratic society, and the presence of falsities as

54

Delfi AS v. Estonia App no 64569/09 (ECtHR 16 June 2015) para. 110.

55_{Ahmet Yildirim v. Turkey App no 3111/10 (ECtHR 18 December 2012 dec.) para. 48.} 56_{Benedek and Kettemann (n 3) 54.}

57

van Rijn (n 17) 772.

58

This question has been debated with respect to the domestic legal order, see Magdalena Pöschl, „Neuvermessung der Meinungsfreiheit?‟ in Helmut Koziol (ed.), Tatsachenmitteilungen und Wert-urteile: Freiheit und Verantwortung (Jan Sramek 2018) 41ff.

59

Salov v. Ukraine App no 65518/01 (ECtHR 6 September 2005 dec.) para. 113. The case is drawn from MLDI (n 15) para. 7.

(17)

such does not hinder that an “informed public debate” takes place, as the Court ide-ally envisions is. Misleading news can thus be said to “allow” for public debate and consequently falls within the scope of protection of Article 10.

As regards fake news, the question is whether deliberately fabricated informa-tion with the inteninforma-tion to deceive can be said to stimulate public debate and thus “al-low” for it. Given the earlier assumption that fake news has an adverse effect on in-formed democratic debate, the answer would have to be that no, fake news does not stimulate or allow for an informed public debate as the Court ideally envisions it. The presumption of an adverse effect of fake news on democratic debate works on a large-scale, societal level. At the same time, it is certainly conceivable that individual fake news stories spark a public debate. Consider the following two examples: In 2015, a Belgian public service broadcaster interrupted the normal television program with the fabricated “breaking news” of a declaration of independence of the Flanders region.61 The contenders of the action argued that the story had sparked an important debate about the sensitive question of Flemish independence and the ethics of jour-nalism.62 The story meets the three-element definition of fake news: deliberate fabri-cation, dissemination with the intention to deceive the recipient into believing false-hoods (deception in this case being a short-term goal, the ultimate goal being the triggering of a public debate) and news character. As a second example, recall the fabricated quote of the German politician on the „murderer refugee‟ and how Ger-many would help him. The large number of similar false stories sparked a public debate about the necessity and the modalities of regulation of online platforms.63 It follows that applying the Court‟s rationale for protecting the circulation of false in-formation to fake news does not yield a conclusive answer to the question whether fake news is protected under Article 10 of the Convention.

Instead, three different arguments can be advanced as to why fake news falls within the scope of protection of Article 10.64

A first argument can be made by relying on the Court‟s notion of “democratic society”. It was mentioned earlier that one author had described the Court‟s ideal of a democratic society as one in which an “informed, plural and contradictory debate on

61

Drawn from McGonagle (n 7) 205. See this publication for further information.

62_ibid. 63_ibid. 64

The following section comes to the same conclusion as Christoph Grabenwarter, European Con-vention on Human Rights: Commentary (Beck 2014) 254, which states that intentionally untrue statements are protected under Article 10 (without elaborating).

(18)

issues of public interest” took place.65

The same author has noted that this notion has informed an expansive reading of the scope of the right to freedom of expression,66 namely in that the Court protects “‟information‟ or „ideas‟ (...) that offend, shock or disturb” because “such are the demands of (...) pluralism (...) without which there is no democratic society.”67

Relying on a notion that pluralism is necessary in a democ-ratic society, the Court thus adopts a wide reading of the scope of the right to free-dom of expression. Applying the argument of pluralism to fake news suggests that fake news would be protected under Article 10 as well. This is because there is an inherent danger in not protecting speech on the basis of its falsity, namely in that upholding notions of “objective truths” may lead to the silencing of alternative views,68 thus diminishing the circulation of plural views in a society.

A second argument relies on the common element of falsity to both fake news and hate speech. Like fake news, hate speech frequently has an element of falsity to it, namely in that it paints complex issues as simple truths. Consider for instance the following two statements regarding the immigration debate and the causes for the persistence of HIV in society: “The day there are 25 million Muslims in France, they will be in command.”69

“The promiscuous lifestyle of homosexuals is one of the main reasons why HIV and AIDS have gained a foothold in society.”70

Given that the Court has protected these expressions under Article 10,71 one can expect that it would also be inclined to protect fake news as a similar category of false speech.

A third argument for an expansive reading of the scope of Article 10 is the framework in which the Court assesses claims of violations of the provision. The Court applies a nuanced, threefold test of whether the interference was prescribed by law, whether it pursued one of the legitimate aims enumerated in paragraph two of Article 10 and whether it was necessary in a democratic society.72 It is only in excep-tional circumstances73 that an expression is not protected under Article 10, namely when the speech is considered to fall under Article 17 of the Convention. Given that finding fake news to be protected under the provision allows for the application of

65_{Zysset (n 47).} 66_{ibid 234.} 67

Handyside v. the UK App no 5493/72 (ECtHR 7 December 1976) para. 49.

68

MLDI (n 15) para. 18.

69_{Le Pen c. France App no 18788/09 (ECtHR 20 April 2010) facts, translation my own.} 70_{Vejdeland and others v. Sweden App no 1813/07 (ECtHR 9 February 2012 dec.) para. 8.} 71

(n 69) and (n 70).

72_{van Rijn (n 17) 783ff.}

(19)

the nuanced, three-element test and the fact that speech is rarely denied protection under the right to freedom of expression, fake news would have to be considered to fall within the scope of protection of Article 10.

One can thus answer the question with regard to internet users as follows: An act of implementation of a fake news law with respect to an internet user‟s fake or misleading news constitutes an interference with the individual internet user‟s right to freedom of expression under the Convention.

II. Interference with the right to freedom of expression of platform operators

As regards the right to freedom of expression of platform operators, one can note that the right applies to legal persons as well.74 The Court has decided a number of cases with regard to internet businesses. For instance, it has decided on the Article 10 compatibility of Internet news portals‟ civil liability for unlawful comments of its users75 and for hyperlinking defamatory third-party content.76 In these cases, the question of interference was not disputed between the parties and the Court generally limited itself to stating that there had been an interference with the right to freedom of expression of the respective applicant.77 In one case, the Court explicitly noted that it was assessing whether there had been a violation of the news portal‟s freedom to impart information as guaranteed by Article 10 of the Convention.78 In a different case regarding a news portal, the Court held that the applicant provided a forum for the exercise of expression rights and thus enabled the public to impart information and ideas.79 It is reasonable to assume that any platform operator targeted by fake news laws fulfils similar functions and either imparts information or ideas on users or provides a platform for the exchange of information and ideas.

The question can thus be answered with regard to platform operators as fol-lows: An act of implementation of a fake news law with respect to a platform opera-tor‟s fake or misleading news constitutes an interference with the platform operaopera-tor‟s right to freedom of expression under the Convention.

74_{van Rijn (n 53).} 75

Delfi AS v. Estonia (n 54); Magyar Tartalomszolgáltatók Egyesülete and Index.hu Zrt v. Hungary App no 22947/13 (ECtHR 2 February 2016 dec.).

76_{Magyar Jeti Zrt v. Hungary App no 11257/16 (ECtHR 4 December 2019 dec.).}

77_{Delfi AS v. Estonia (n 54) para. 118; Magyar Tartalomszolgáltatók Egyesülete and Index.hu Zrt v.}

Hungary (n 75) para. 45; Magyar Jeti Zrt v. Hungary (n 76) para. 56.

78_{Delfi AS v. Estonia (n 54) para. 140.}

(20)

III. Relevance of Article 17 of the Convention

It was hinted at earlier that speech may exceptionally fall outside the scope of protec-tion of Article 10.80 The relevant provision in this regard is Article 17 of the Conven-tion. The provision reads: “Nothing in this Convention may be interpreted as imply-ing for any State, group or persons any right to engage in any activity or perform any act aimed at the destruction of any of the rights and freedoms set forth herein or at their limitation to a greater extent than is provided for in the Convention.” The article serves to prevent applicants from relying on the Convention to justify activities con-trary to democratic values.81 Applications considered to fall under the provision are declared inadmissible.82

Is Article 17 likely to be applied in future fake news cases before the Court? To answer this question, it is helpful to briefly review some of the main lines of the Court‟s jurisprudence on the provision.83

One category of speech to which the Court has frequently applied Article 17 is revisionist speech.84 A prominent line of jurisprudence concerns expressions of Holocaust denial, which the Court has found to “run counter to the fundamental val-ues of the Convention, as expressed in its Preamble, namely justice and peace”85

and “counter to the text and spirit of the Convention”.86

A related category of speech prevalent in the case law on Article 17 is speech that promotes totalitarian ideologies.87 For instance, the Court has held with respect to National Socialism: “National Socialism is a totalitarian doctrine incompatible with democracy and human rights and (...) its adherents undoubtedly pursue aims of the kind referred to in Article 17.”88

Another important category is hate speech. The Court has held that “as a mat-ter of principle it may be considered necessary in certain democratic societies to sanction or even prevent all forms of expression which spread, incite, promote or

80_{Pages 12f.} 81

Paulien Elsbeth de Morree, Rights and Wrongs under the ECHR: The prohibition of abuse of rights in Article 17 of the European Convention on Human Rights (Intersentia 2016) 27.

82_{Pursuant to Article 35 paras 3 and 4 ECHR.} 83

The categories and the case law that follow are drawn from (n 81). See this publication for further information.

84_{de Morree (n 81) 28.}

85_{Garaudy v. France App no 65831/01 (ECtHR 24 June 2003 dec.) para. 1.} 86

ibid.

87_{de Morree (n 81) 42ff.}

(21)

justify hatred based on intolerance.”89

Notable hate speech cases where Article 17 was applied include the display of a poster with the burning Twin Towers and the words „Islam out of Britain – Protect the British People‟90

and the dissemination of an article calling for the exclusion of Jews from social life.91 A similar category that has been found to fall within the scope of Article 17 is speech that incites to vio-lence.92 For instance, the Court has held with regard to a radical Islamic organization that “a political organisation whose leaders incite to violence or put forward a policy which fails to respect democracy (...) cannot lay claim to the Convention‟s protection against penalties imposed on those grounds”.93

Overall, the Court‟s jurisprudence on Article 17 has been described as highly inconsistent and unpredictable.94 How, if at all, is the jurisprudence relevant for fake and misleading news? Given the Court‟s inconsistent application of Article 17, the following remarks about the applicability of the provision to fake and misleading news are speculative. They are guided by the above-mentioned rationale of Article 17, namely to prevent applicants from relying on the Convention to justify activities contrary to democratic values.

The pertinent question in the context of fake news legislation is whether the dissemination of fake news displays a similar level of disregard for democratic val-ues as revisionist speech and the other categories of speech that the Court has, at times, found to fall under Article 17. In my view, this is not the case. A speaker who deliberately disseminates fabricated information is not disparaging the value of de-mocracy or democratic debate per se – unlike a revisionist speaker, one that calls for the exclusion of a particular societal group or one that propagates radical Islamism, which are all indirectly expressing support for anti-democratic movements – but is rather attempting to influence the recipient‟s political views and thus influence de-mocratic debate in his favour. Moreover, fake news can be valuable to dede-mocratic deliberation in that it sparks a public debate (recall the example of Flemish inde-pendence) or highlights existing socio-economic tensions (recall the example of al-leged child abductions conducted by the Roma). The above arguments are equally

89_{Erbakan v. Turkey App no 59405/00 (ECtHR 6 July 2006) para. 56; Gündüz v. Turkey App no}

35071/97 (ECtHR 4 December 2003) para. 40.

90

Norwood v. the United Kingdom App no 23131/03 (ECtHR 16 November 2004 dec.).

91_{Pavel Ivanov v. Russia App no 35222/04 (ECtHR 20 February 2007 dec.).} 92_{de Morree (n 81) 56ff.}

93

Kasymakhunov and Saybatalov v. Russia App nos 26261/05 and 26377/06 (ECtHR 14 March 2013) para. 105.

(22)

true for misleading news. It follows that fake and misleading news are unlikely to be considered to fall under Article 17 of the Convention. It is thus unlikely that the Court will apply the provision to future cases concerning fake news legislation.

G. Is the interference prescribed by law?

In order to be lawful, an interference with the right to freedom of expression must be prescribed by law. With respect to this criterion, the Court has noted that this “not only requires that the impugned measure should have a legal basis in domestic law, but also refers to the quality of the law in question, which should be accessible to the person concerned and foreseeable as to its effects.”95

Moreover, the Court has speci-fied its standard for foreseeability as one where citizens “must be able (...) to foresee, to a degree that is reasonable in the circumstances, the consequences which a given action may entail.”96

Absolute certainty is not required, given that “many laws are inevitably couched in terms which, to a greater or lesser extent, are vague, and whose interpretation and application are questions of practice.”97

Having noted the applicable general principles, and without assessing whether definitions in existing fake news legislation meet these criteria, a few gen-eral observations can be made.

Firstly, legislators must reflect on which phenomenon they are in fact target-ing. Striving for absolute truth of information in a democratic society is not realistic and comes with the danger of mistakenly silencing “false” minority views.98

The “real danger”, with respect to which the adverse effect on democratic debate is being discussed, is information that intends to deceive the recipient into believing false-hoods. In my view, legislators should beware of proscribing falsities of any kind and should instead focus on „fake news‟, as it was narrowly defined for the purposes of the present paper. Legal definitions of fake news should thus include the element „intent to deceive‟.

Secondly, it is certainly difficult to come up with a precise definition of fake news in light of the multifaceted nature of the phenomenon. It is not surprising that in the public debate on fake news legislation, existing definitions have been criticized

95_{E.g. Delfi AS v. Estonia (n 54) para. 120, referring to further case law.} 96

ibid para. 121.

97_{ibid, referring to further case law.} 98_{MLDI (n 15) para. 18.}

(23)

for being too vague.99 However, it seems that the real concern expressed in the public debate is not necessarily that fake news legislation is too vague, but rather that it is sufficiently vague to be used for illegitimate purposes. Potentially abusive use of fake news legislation is a frequently named concern,100 and it will be explored at a later stage101 if the Court‟s jurisprudence can inform this debate in any way.

Lastly, the Court rarely finds that a right has been violated because the inter-ference was not “prescribed by law”.102

Consequently, it is unlikely that the Court would find a fake news law incompatible with Article 10 of the Convention at this stage of the legal assessment.

H. Does the interference pursue a legitimate aim?

An interference with the right to freedom of expression must pursue one of the le-gitimate aims enshrined in paragraph two of Article 10, which are namely: national security, territorial integrity, public safety, the prevention of disorder or crime, the protection of health or morals, the protection of the reputation or rights of others, the prevention of the disclosure of information received in confidence and the mainte-nance of authority and impartiality of the judiciary.

A preliminary point to make with respect to the criterion is that the existence of a legitimate aim is rarely problematic.103 Accordingly, it is likewise unlikely that the Court would find a fake news law incompatible with the Convention at this stage of the legal assessment. Given that a case regarding fake news legislation has not been decided by the Court yet,104 this section of the paper will focus on identifying relevant aims that could be invoked in future fake news cases before the Court.

I. What is the harm of fake news?

To identify these aims, it is helpful to reflect on why fake news is considered harm-ful, since this can provide indications as to why legislators would decide to regulate with respect to this phenomenon. One harm has been referred to already, namely the presumed adverse effect of fake news on democratic debate.105

99

(n 14).

100

E.g. McGonagle (n 7) 204; MLDI (n 15) para. 20.

101_{Pages 20f.} 102_{Implied in Lavrysen (n 45) 314.} 103 ibid. 104_{Pages 2f.} 105_{Pages 7f.}

(24)

A second harmful effect one can envision is that fake news incites to hatred or violence. This is the case when fake news overlaps with hate speech. Recall for instance the two examples that were mentioned in a different context earlier,106 namely the fabricated quote of a German politician about the „murderer refugee‟ and how Germany would help him and the fabricated story on child abductions con-ducted by the Roma. The latter indeed led to acts of violence against individuals of the Roma ethnic, as did other fake news stories circulated on social media.107 It is evident how fake news that exploits preconceived notions about already disadvan-taged groups (e.g. immigrants, certain ethnic groups) can spark new waves of hatred and violence against these groups.

II. Legitimate aims of fake news legislation

If a case concerning a piece of fake news legislation ended up before the Court, the Contracting Party in question would likely argue that the aim of the law (and the ap-plication of the law to the particular applicant at hand) was to counter the harmful effects of fake news. How could fighting the above-mentioned harmful effects be framed as legitimate aims?

As regards the presumed adverse effect of fake news on democratic debate, it could be argued that fake news legislation (seeking to counter this effect) is protect-ing the rights of others, namely the right of individuals to access truthful information online. The Court has not formally recognized the existence of a “right to access truthful information online” to date, but an argument for the recognition of such a right could be made on the basis of existing case law on the right to receive informa-tion. The argument presents one way in which the adverse effect of fake news on democratic debate could be incorporated in the Convention‟s legal framework and will be sketched out in the following paragraphs.

The right to receive information is enshrined in paragraph one of Article 10 of the Convention, stating that everyone has the right to receive information and ideas without interference by public authorities and regardless of frontiers. The right pro-tects the passive freedom to receive information and the effort to acquire or access information.108 Positive state obligations are derived from the provision only in

106_{Pages 5 and 7.} 107

E.g. Marcos Martinez, „Burned to death because of a rumour on WhatsApp‟ BBC (London, 12 November 2018) <https://www.bbc.com/news/world-latin-america-46145986>.

(25)

ited circumstances.109 Nonetheless, the Court has recognized a positive obligation “to ensure (...) that the public has access through television and radio to impartial and accurate information and a range of opinion and comment, reflecting inter alia the diversity of political outlook within the country.”110

The obligation is confined to the audiovisual media due to its “immediate and powerful effect.”111

The obligation is reminiscient of the Court‟s notion of a democratic society as one in which an “informed and plural debate on issues of public interest” takes place, and indeed the obligation seems to seek to implement the Court‟s notion of an ideal democratic debate,112 in that it requires states to ensure access to impartial and accu-rate information (“informed debate”) and a diverse range of opinion and comment (“plural debate”). It could be argued that the obligation extends to the internet, as the internet has just as an “immediate and powerful effect”. It could likewise be argued that the obligation extends to the private sphere, e.g. to the relationship of internet news consumers vis-à-vis private news media outlets (or vis-à-vis any other news publishers, including individuals publishing news online). Indeed, existing case law on positive obligations with respect to the right to freedom of expression in the sphere between individuals has been interpreted to possibly imply a positive obliga-tion to “adopt rules to ensure media organisaobliga-tions respect the right to receive infor-mation of news consumers”.113

Combining these arguments results in the recognition of a right to access truthful information online: It is directed towards the state and entails the right to adoption of rules which ensure that private actors respect the right of individuals to access truthful information online. Construed this way and in justi-fying the application of a fake news law in a case at hand, the protection of internet users‟ right to access truthful information could function as a legitimate aim for the interference with the right to freedom of expression of internet users and online plat-forms.

Given that fake news is often hateful speech that targets a particular group of society,114 another right that could be framed under the aim of „protection of the rights of others‟ is the right to private life under paragraph one of Article 8 of the Convention. The Court has held that the negative stereotyping of an ethnic group,

109

Sarah Eskens, Natali Helberger and Judith Moeller, „Challenged by news personalisation: five perspectives on the right to receive information‟ (2017) 9(2) Journal of Media Law 259, 263.

110_{Manole and others App no 13936/02 (ECtHR 17 September 2009 dec.) para. 100.} 111_{Eskens et al. (n 109) 267, referring to Manole and others (n 110) para. 97.}

112

Pages 7f.

113_{Eskens et al. (n 109) 263.} 114_{Page 18.}

(26)

capable of “having an impact on the group‟s sense of identity and on its members‟ feelings of self-worth and self-confidence” could affect their „private life‟ within the meaning of Article 8 of the Convention.115

As regards the potential of fake news to incite to hatred or violence,116 likely aims to be argued are public safety and the prevention of disorder or crime (hereinaf-ter summarized as the legitimate aim of „public safety‟).

III. The problem of illegitimate aims

A commonly voiced concern with respect to fake news legislation is that it is adopted under the veil of pursuing legitimate aims, but that the legislation and namely the term „fake news‟ will in practice be interpreted and applied in a way to silence dis-sent.117 This raises the question how the Court would deal with fake news legislation that was in fact used for illegitimate purposes, if this were apparent from the way the law was applied to a particular case brought before the Court.

An indication can be found in a recent Court case on a Russian criminal law provision on hate speech.118 The applicant, a journalist, had been sentenced for hav-ing committed “actions aimed at incithav-ing hatred (...) through the mass media” pursu-ant to the Russian Criminal Code after he had published a newsletter discussing his views on the Russian war in the Chechen Republic.119 The Court ultimately found a violation of Article 10, holding that the speech in question had (partly) not been hate speech, but merely speech critical of the government.120 Interestingly, it also noted the following: “It (the Court) stresses in this connection that it is vitally important that the domestic authorities adopt a cautious approach in determining the scope of “hate speech” crimes and strictly construe the relevant legal provisions in order to avoid excessive interference under the guise of action taken against “hate speech”, where such charges are brought for a mere criticism of the Government, State institu-tions and their policies and practices.”121

In more general terms: If states have adopted laws aimed at the restriction of a particular kind of harmful speech, they should refrain from using the law as pretext to restrict speech critical of the government. This „guideline‟ is of direct relevance to

115

Aksu v. Turkey App nos 4149/04 and 41029/04 (ECtHR 15 March 2012) para. 58.

116

Page 18.

117_{(n 100).}

118_{Stomakhin v. Russia App no 52273/07 (ECtHR 9 May 2018 dec.).} 119

ibid paras 5ff.

120_{ibid paras 103ff.} 121_{ibid para. 117.}

(27)

fake news laws, which may potentially be abused for illegitimate purposes. It sug-gests that the Court would also object to a politically motivated interpretation of the term „fake news‟ in fake news legislation that in reality serves illegitimate purposes, for instance the silencing of dissent.

I. Is the interference necessary in a democratic society? I. The margin of appreciation

The margin of appreciation is the “space for manoeuvre” the Court awards to Con-tracting Parties in fulfilling their obligations under the Convention.122 It impacts how strictly the Court scrutinizes Contracting Parties‟ compliance with the Convention: if the margin of appreciation is wide, the Court is more lenient in applying the “neces-sity in a democratic society” test.123

The following sections will discuss three factors that influence the width of the margin of appreciation in the context of fake news legislation. Additional factors could evidently influence the width of the margin, but the paper will focus on three factors that lend themselves to discussion.124 The analy-sis of these factors does not yield a conclusive finding that the margin in fake news legislation cases is always narrow or always wide. Accordingly, the section of this paper focusing on the “necessity in a democratic society” will highlight problems of fake news legislation with respect to this criterion independently of the width of the margin of appreciation.

1. The level of protection of internet users’ fake and misleading news (a) A hierarchy of protection of speech

The Court assigns different kinds of speech different levels of protection125 and sub-jects interferences with speech that enjoys high protection under the Convention to strict scrutiny.126 While the margin of appreciation is the formal figure the Court in-vokes in this regard, it has been argued that the substance of the Court‟s approach is

122

de Londras and Dzehtsiarou (n 39) 93f.

123_{ibid 98.}

124_{The section on the level of protection of fake and misleading news relies heavily on Katsirea (n 8)}

13 ff.

125_{McGonagle (n 7) 204.}

(28)

a consideration of the value of particular types of speech.127 Underlying this catego-rization, so the argument, is the idea that the potential harm caused by a particular type of speech is to be tolerated if the speech is of high value to democratic debate.128

A high level of protection is awarded to political speech. The Court has held that “there is little scope (...) for restrictions on political speech or on debate on ques-tions of public interest”,129

that “interferences with the freedom of expression of an opposition member of parliament (...) calls for the closest scrutiny on the part of the Court”130

and that “freedom of political debate is at the very core of the concept of a democratic society which prevails throughout the Convention. The limits of accept-able criticism are accordingly wider as regards a politician as such than as regards a private individual.”131

As regards other types of speech, it has been noted that the Court‟s jurisprudence suggests the following categorization:132

Artistic and commer-cial speech attract intermediate protection, with artistic speech ranking higher than commercial speech. Celebrity gossip attracts a lower level of protection. Even lower rank pornography, gratuitous personal attacks and hate speech, which enjoys very little protection or no protection at all.133

(b) Locating internet users’ fake and misleading news in the hierarchy

How do fake and misleading news fit into the picture? A good starting point is to consider in which of the above categories online speech would fall. One author notes that much “everyday speech” posted online would be awarded a low level of protec-tion in the Court‟s framework, with the excepprotec-tion of online speech relating to politi-cal matters.134 As regards a categorization of fake news in the Court‟s framework, one author notes that fake news often aims at commercial profit and would thus fall within the low-protection category of commercial speech. However, the author strongly cautions against equating fake news with low-protection categories of speech: Due to the uncertainty revolving around the term, fake news and political speech are not easily separated and there is a risk of “contaminating” political speech

127_{Jacob Rowbottom, „To Rant, Vent and Converse: Protecting Low Level Digital Speech‟ (2012)}

71(2) Cambridge Law Journal 355, 368.

128

ibid 369.

129

E.g. Ceylan v. Turkey App no 23556/94 (ECtHR 8 July 1999) para. 34.

130_{E.g. Castells v. Spain App no 11798/85 (ECtHR 23 April 1992) para. 42.} 131_{E.g. Lingens v. Austria App no 9815/82 (ECtHR 8 July 1986) para. 42.} 132

Drawn from Rowbottom (n 127) 369.

133_ibid.

Fake News Legislation: A Human Rights Problem? Assessing the Compatibility of Fake News Legislation with Article 10 of the European Convention on Human Rights

University of Amsterdam

Faculty of Law