ChapterPDF Available

The Liability of Internet Intermediaries and the European Court of Human Rights

Authors:

Abstract

This chapter explores recent developments in the liability of internet intermediaries for user-generated content as promoted by the European Court of Human Rights in the cases Delfi v Estonia (2015) and MTE v Hungary (2016). It offers an in-depth analysis of the legal arguments deployed by the Court to unveil the practical challenges involved in the regulation of internet intermediaries. This chapter, while unfolding the Court’s justificatory narratives for deciding the case, attempts to show how the decision may be inherently connected to the need to preserve the performativity of law in the Internet ecosystem. In this analysis, attention has been paid to how the question of anonymity might challenge the role of law in granting remedies. These issues influenced the Court’s ruling which then opted to design a more regulated environment at the expense of a broader understanding of freedom of expression. Finally, this chapter argues that delegating what counts as freedom of expression to private companies risks reducing freedom of expression to a “technicality”, thus neglecting the normative complexity of freedom of expression with consequences for the Internet as a pluralistic environment.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
The Liability of Internet Intermediaries and the European Court of Human
Rights
This chapter explores recent developments in the liability of internet intermediaries for user-generated
content as promoted by the European Court of Human Rights in the cases Delfi v Estonia (2015) and MTE v
Hungary (2016). It offers an in-depth analysis of the legal arguments deployed by the Court to unveil the
practical challenges involved in the regulation of internet intermediaries. This chapter, while unfolding the
Court’s justificatory narratives for deciding the case, attempts to show how the decision may be inherently
connected to the need to preserve the performativity of law in the internet ecosystem. In this analysis,
attention has been paid to how the question of anonymity might challenge the role of law in granting
remedies. These issues influenced the Court’s ruling which then opted to design a more regulated
environment at the expense of a broader understanding of freedom of expression. Finally, this chapter argues
that delegating what counts as freedom of expression to private companies risks reducing freedom of
expression to a “technicality”, thus neglecting the normative complexity of freedom of expression with
consequences for the Internet as a pluralistic environment.
Bio
Marta Maroni is a PhD candidate in constitutional law at the University of Helsinki. Marta’s doctoral thesis
on the right to access the Internet and internet governance analyses the constitutional questions and
regulatory problems pertaining to the internet infrastructure.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
The Liability of Internet Intermediaries and the European Court of Human
Rights
1. INTRODUCTION
Freedom of expression has gone through a profound transformation since the advent of the Internet.
At the core of this transformation lies the role of a wide set of private actors, known as hosting
providers (social media or platforms), whose policies and actions shape a more or less conducive
ecosystem for freedom of expression. These actors do not produce the content themselves, but they
provide the interface for creating and circulating content. Yet they mediate what information is
visible and what content stays up or is taken down. The regulatory framework under which these
actors operate contributes to the way freedom of expression is managed. Thus far, “strict liability”
for illegal content disseminated through social media services, has been considered undesirable for
two reasons.
First, it would affect the economic freedom/business model of these actors as Internet
intermediaries would constantly be at risk of being sued, this would result in requiring them to
monitor the information flow provided through their services continuously. Hence, a strict model of
liability would also increase the risk of interference with freedom of expression: in order to avoid
“legal problems”, platforms/social media would remove any edgy content - that is, information at
the margins of freedom of expression
1
For these reasons, current legislation and soft law encourage self-regulation and design a secondary/
indirect model of liability: a platform/intermediary generally benefits from the liability exemption,
when it plays a passive function. This means that while providing a space for interaction and
production of content, the platform should have no knowledge of the content disseminated through
its service. However, when it comes to knowing about an illegal activity, the platform must
expeditiously remove or disable access to the information. In addition, it should be mentioned that
Marta Maroni would like to acknowledge the support received from the project Reconfiguring Privacy A Study of the
Political Foundations of Privacy Regulation-funded by the University of Helsinki.
1
Elda Brogi and Marta Maroni, ‘Freedom of expression and the rule of law: the debate on editorial responsibility in the
context of online platforms´ regulation. Elda Brogi and Pier Luigi Parcu (eds), Research Handbook on EU Media Law
and Policy (forthcoming, Edward Elgar Publishing 2020).
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
under EU law, member states are prevented from imposing a general monitoring obligation on
social media platforms.
2
The case law of the European Court of Human Rights (the Court, ECtHR) has also contributed to
the understanding of this general framework and furthered what scholars have termed the “duty of
care” of social media platforms, which roughly means that negligence is not exempted.
3
However,
as illustrated below, this approach has not been well-received because it is considered to encourage
arbitrary removal of online content. This work explores the case law of the ECtHR on the liability
of internet intermediaries. The main purpose is first, to offer an in-depth analysis of the legal
arguments deployed by the Court to unveil the practical challenges intertwined with the regulation
of internet intermediaries, and second, to anchor them to a more theoretically-based analysis. This
work, while unfolding the Court´s justificatory narratives for deciding the case, attempts to show
how the decision may be inherently connected to the need to preserve the performativity of law in
the Internet ecosystem. In this analysis, attention has then been paid to the question of anonymity
and how it might challenge the role of law in granting remedies. These issues influenced the Court´s
ruling which then opted to design a more regulated environment at the expense of a broader
understanding of freedom of expression.
This chapter is organised as follows. There is an overview of the main criticisms raised against the
Court’s ruling in relation to freedom of expression, before introducing ECtHR case law. Subsequent
sections focus on the legal arguments and the narratives of the Court. In the conclusion, I argue that
delegating what counts as freedom of expression to private entities risks reducing freedom of
expression to a technicality, thus neglecting the normative complexity of freedom of expression,
which affects the Internet as a pluralistic environment.
2
Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal aspects of
information society services, in particular electronic commerce, in the Internal Market [2000] OJ L178/1 (hereinafter E-
Commerce Directive), art 15.
3
See Peggy Valcke and Pieter-Jan Ombelet, ‘Did the Romans Get It Right? What Delfi, Google, Ebay, and UPC
Telekabel Wien Have in Common’ in Luciano Floridi and Mariarosaria Taddeo (eds), The Responsibilities of Online
Service Providers (Springer 2016); Giovanni Sartor, ‘Internet Providers Liability: From the Ecommerce Directive to the
Future’ in Elda Brogi and Pier Luigi Parcu (eds), Research Handbook on EU Media Law and Policy (forthcoming,
Edward Elgar Publishing 2020).
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
II. The cases
During 2015-2016, the Court dealt twice with the liability of internet intermediaries for offensive
comments posted by their users and it ruled in two cases: Delfi v Estonia (2015)
4
and MTE v
Hungary (2016).
5
In the first case, Delfi v Estonia (2015), Delfi, an online news portal, alleged that the Estonian Court
had violated its freedom of expression
6
by holding it liable for anonymous and unlawful user
comments. In finding no violation of freedom of expression, the ECtHR stated that in principle,
when run on a commercial basis, internet intermediaries (in the form of content providers) are liable
for third party content.
7
This case provoked strong criticism, as the ruling might encourage content
providers to monitor user activities and act as “censors to avoid liability problems.
8
For this
reason, both legal scholars and civil society have expressed concern about freedom of expression
online.
9
This criticism could be summarised in the following points:
4
Delfi v Estonia ECHR 2015II 319 (hereinafter Delfi).
5
Magyar Tartalomszolgáltatók Egyesülete and Index.hu Zrt v Hungary App no 22947/13 (ECtHR, 2 February 2016)
(hereinafter MTE v Hungary).
6
Convention for the Protection of Human Rights and Fundamental Freedoms (European Convention on Human Rights,
as amended) (hereinafter ECHR), art 10.
7
Third party content or user-generated content refers to any digital content that is produced and shared by end users of
an online service or website. This includes any content that is shared or produced by users that are members or
subscribers of the service, but it is not produced by the website or service itself.” ‘User-Generated Content (UGC)’
(Techopedia.com, 2018) <www.techopedia.com/definition/3138/user-generated-content-ugc> accessed 11 January
2018.
8
See, for example in Delfi (n 4) third party interveners, paras 94, 96; or the joint dissenting opinion of Judges Sajó and
Tsotsoria, para 1.
9
See for example, Article 19, ‘European Court Strikes Serious Blow to Free Speech Online’ (Article 19, 14 October
2013) <www.article19.org/resources.php/resource/37287/en/european-court-strikes-serious-blow-to-free-speech-
online> accessed 15 April 2017; Dirk Voorhoof, ‘Delfi AS v. Estonia: Grand Chamber Confirms Liability of Online
News Portal for Offensive Comments Posted by Its Readers’ (Strasbourg Observers, 18 June 2015)
<https://strasbourgobservers.com/2015/06/18/delfi-as-v-estonia-grand-chamber-confirms-liability-of-online-news-
portal-for-offensive-comments-posted-by-its-readers/> accessed 11 January 2018; Lisl Brunner, ‘The Liability of an
Online Intermediary for Third Party Content : The Watchdog Becomes the Monitor: Intermediary Liability after Delfi v
Estonia’ (2016) 16 Human Rights Law Review 163. Dirk Voorhoof offers more sources on the debate in ‘Qualification
of News Portal as Publisher of Users Comment May Have Far-Reaching Consequences for Online Freedom of
Expression: Delfi AS v. Estonia’ (Strasbourg Observers, 25 October 2013)
<https://strasbourgobservers.com/2013/10/25/qualification-of-news-portal-as-publisher-of-users-comment-may-have-
far-reaching-consequences-for-online-freedom-of-expression-delfi-as-v-estonia/> accessed 11 January 2018. See also
Martin Husovec, ‘General Monitoring of Third-Party Content: Compatible with Freedom of Expression?’ (2016) 11
Journal of Intellectual Property Law & Practice 17.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
Content intermediaries may seek to avoid liability by monitoring and filtering user content,
which could result in unfair blanket censorship;
10
Intermediaries may stop providing services when they cannot afford quick and effective
moderation.
11
This is a concern especially relevant to smaller hosting providers. Whereas there
is an increasing demand to regulate dominant platforms which control most of the information
flow having paramount societal impact (e.g. freedom of expression, elections and so forth),
smaller platforms might not be able to offer a qualitative-based content moderation.
Content may become authored, thereby eroding one of the key aspects of the Internet: the
facilitation of self-expression and participation in online social interactions.
The second case, MTE vs Hungary (2016),
12
which shares several similarities with the previous
case, has generally received more positive comments. The similarities are 1) the role of applicants
as with Delfi, a key question was whether the applicants should have been considered to be media
publishers or intermediaries, 2) the context in both cases the applicants were sued for violating
Article 8 (the right to respect for private and family life) for hosting offensive comments, and,
finally, 3) the reasoning of the ECtHR, which confirmed the principle of the liability of content
providers. However, in MTE vs Hungary (2016), the ECtHR took a softer stance and regardless of
the similarities, ruled that there had been a violation of freedom of expression. The main
explanation for the different results lies in the details, from the nature of the comments themselves
to the type of intermediary.
The similarities of cases lie in the fact that: the applicants in both Delfi and MTE alleged that their
freedom of expression had been violated because they had been found responsible for comments
written below some news items they had published. The ECtHR, in making its assessments, followed
a similar matrix of criteria:
13
a) the context and content of the impugned comments,
b) the liability of the actual authors of the comments,
c) the measures taken by the applicants and the conduct of the injured party,
10
The dissenting opinion defined the ruling as “an invitation to self-censorship at its worst. Delfi (n 4) dissenting
opinion, para 2; Brunner (n 9) 172.
11
Brunner (n 9) 172.
12
MTE v Hungary (n 5).
13
See Christina Angelopoulos, ‘MTE v Hungary: New ECtHR Judgment on Intermediary Liability and Freedom of
Expression’ (Kluwer Copyright Blog, 5 March 2016) <http://MTE v Hungary: New ECtHR Judgment on Intermediary
Liability and Freedom of Expression> accessed 12 January 2018.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
d) the consequences of the comments for the injured party,
e) the consequences for the applicants.
The next section outlines the main facts and arguments of the cases.
Delfi v Estonia
In 2015, the Grand Chamber of the ECtHR heard the landmark case Delfi v Estonia. Cases heard in
the Grand Chamber usually involve important issues, such as case law consistency, ‘high profile’
cases, and cases dealing with ‘new issues’. Furthermore, Grand Chamber Judgments are passed by
those considered to be the most experienced judges and decided by the largest formation of the Court.
Finally, they involve “a very selective jurisdiction”.
14
In addition to this, it should be noted that the
ECtHR’s decisions influence understanding of the legal problem in question in the national courts of
the contracting parties.
15
For these reasons, the ruling is considered to bear high resonance within the
members of the Council of Europe.
Delfi, the biggest news portal in Estonia, reported that the Saaremaa Shipping Company, a public
ferry transporter, had destroyed the roads over the sea ice used in winter to connect Estonian islands
with the mainland.
16
The article was then followed by the comments of anonymous and unregistered
users. There were twenty comments (out of 185) deemed to be hate speech and to threaten the owner
of the shipping company.
17
Some examples include:
2. [T]hey bathe in money anyway thanks to that monopoly and state subsidies, and now
[they] start to fear that cars may drive to the islands for a couple of days without anything
filling their purses. Burn in your own ship, sick Jew!
14
Fiona de Londras and Kanstantsin Dzehtsiarou, ‘Managing Judicial Innovation in The European Court Of Human
Rights’ (2015) 15 Human Rights Law Review 523.
15
See, for example, the work of Janneke Gerards and Jospeh Fleuren (eds), The Implementation of the European
Convention of Human rights and the Judgments of the ECtHR in National Case-Law (Intersentia 2014) 1; “the way the
Court deals with the structure of fundamental rights strongly influences the interpretation and application of the
Convention by national courts. Structural faults and deficits thus may be multiplied in national cases that never reach
Strasbourg. In the end, this may hamper the effectiveness of the Convention system and limit the protection offered to
individual citizens”, in Janneke Gerards and Hanneke Senden, ‘The Structure of Fundamental Rights and the European
Court of Human Rights’ (2009) 7 International Journal of Constitutional Law 619, 621. It is worth mentioning that the
Supreme Court of Cassation (Italy) recently ruled for the first time on the liability of hosting a webpage. The Court
stated that the webpage hosting the comments can be concurrently found liable for defamatory comments posted, the
unofficial document of the ruling Cassazione Penale Sent Sez 5 Num 54946, Anno 2016
<http://download.repubblica.it/pdf/2016/tecnologia/figc.pdf> accessed 12 January 2018.
16
Delfi (n 4) para 16.
17
ibid para 17.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
3. Good that [La.’s] initiative has not broken down the lines of the web flamers. Go ahead,
guys, [L.] into the oven!
4. [little L.] Go and drown yourself
18
Delfi had implemented some safeguards, such as its Rules of Comment, in which the news portal
claimed to be solely a technical medium which did not edit comments, thereby attributing liability to
the authors themselves.
19
The domestic court nevertheless rejected the Delfi argument that it was a
technical intermediary, since it invited users to post comments and thus could determine the rules for
the comment space. Furthermore, it argued that Delfi should have been aware of the unlawful content
of the comments and prevented their publication and in addition it had failed to remove them.
In response, Delfi argued that the decision of the Supreme Court of Estonia constituted a breach
of its freedom of expression and claimed that instead of being treated as a publisher, it should be
considered to be a different type of disseminator, such as a postal service.
20
The ECtHR nonetheless agreed with the Supreme Court of Estonia, mostly focusing on the duty
and responsibilities that intermediaries should bear. The Court considered particularly relevant the
fact that Delfi was a professionally-managed, commercially run internet news portal which sought to
attract many comments on the news articles it published.
21
One of the first arguments that the ECtHR considered was the nature of the comments. The ECtHR
ruled that the finding of the Estonian Supreme Court that Delfi was liable had been a justified and
proportionate restriction of the portal’s freedom of expression, as users of the platform “engaged in
clearly unlawful speech, which infringed the personality rights of others and amounted to hate speech
and incitement to violence against them”.
22
As we will see, the Court also viewed as particularly
problematic the impossibility of bringing a civil claim against the authors of the comments. Moreover,
according to the Court, Delfi’s decision to allow anonymous comments implied that it took
responsibility for such comments.
23
18
ibid para 18.
19
ibid para 32.
20
Delfi (n 4) para 20.
21
ibid para 144.
22
ECtHR, Registrar of the Court, ‘Commercially-Run Internet News Portal Was Liable for the Offensive Online
Comments of Its Readers’ (16 June 2015) Press Release 205, 3.
23
Delfi (n 4) para 65.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
As regards the measures adopted by the applicants in Delfi, in addition to the disclaimer stating
that the writers, and not the applicant company were responsible for the comments, the portal used
an automatic deletion system, based on the presence of certain key words, for removing offensive
remarks. Further, a notice-and-take-down system was also in place through which any user could
report inappropriate comments, at the click of a button. Nevertheless, the comments in question were
only removed by Delfi six weeks after they had been “uploaded to the website, upon notification by
the injured person’s lawyers to the applicant company”.
24
The Court ruled that if accompanied by
effective procedures allowing a rapid response, this system could indeed “function in many cases as
an appropriate tool for balancing the rights and interests of all those involved”.
25
However, in relation
to unlawful speech the Court stated that
the rights and interests of others and of society as a whole may entitle Contracting States
to impose liability on Internet news portals, without contravening Article 10 of the
Convention, if they fail to take measures to remove clearly unlawful comments without
delay, even without notice from the alleged victim or from third parties.
26
Lastly, the ECtHR emphasised that since Delfi had been required to pay a merely symbolic fee of 320
euros, there was no compulsion for it to change its business model as a result of the domestic
proceedings.
The Narrative of the Court
The Court’s reading begins with preliminary remarks and the scope of the assessment, and in the very
first paragraph the Court sets the narrative for the rest of the judgment. Although the Court explains
the challenges that it faced in striking a balance between the beneficial aspects of online
communications and their risks, the language deployed indicates that the emphasis of the ruling in
this case is on the dangers provided by the Internet, rather than the benefits:
27
24
ibid para 152.
25
ibid para 159.
26
ibid.
27
See Sajó’s and Tsotsoria’s remark “The Internet is described as an “unprecedented platform” and while there is
reference to benefits, it is described as posing ‘certain dangers’, the advantages being scarcely mentioned. We disagree.
The Internet is more than a uniquely dangerous novelty.” Delfi (n 4) joint dissenting opinion, para 6. Also, Cox
highlights a certain imbalance in the way the Court described the Internet. He argues that ´the Court significantly
underplayed the realistic threat to freedom of expression that was at stake and overplayed the realistic extent to which
the applicants were exclusively responsible for any harm to L’s reputation which may have occurred´. See Neville Cox,
Delfi AS v Estonia: The Liability of Secondary Internet Publishers for Violation of Reputational Rights under the
European Convention on Human Rights’ (2014) 77 The Modern Law Review 619, 622;
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
The Court notes at the outset that user-generated expressive activity on the Internet
provides an unprecedented platform for the exercise of freedom of expression. That is
undisputed and has been recognised by the Court on previous occasions. . . . However,
alongside these benefits, certain dangers may also arise. Defamatory and other types of
clearly unlawful speech, including hate speech and speech inciting violence, can be
disseminated like never before, worldwide, in a matter of seconds, and sometimes remain
persistently available online.
28
(The Italics are mine)
In this short passage, the Court introduces a narrative pattern which is often found in Delfi: with
benefits come dangers.
29
This construction can be appreciated when the ECtHR highlights the
dilemma between the Internet’s role in enhancing public access to information and the higher risk of
harm posed by internet content and communication in comparison to that posed by the press.
30
By adopting the scheme of mentioning the benefits but focusing on the risks, the Court seems to
emphasise the existence of an environment in which law fails to perform its ordinary functions,
thereby reinforcing the need for law in the internet ecosystem. Nevertheless, while being aware of
the controversies that it might raise (e.g. relating to freedom of expression), the Court also stresses
the contingency of its own decision and proffers a disclaimer, stating that as it was operating for the
first time in an “evolving field of technological innovation”,
31
it was thus “necessary to delineate the
scope of its inquiry in the light of the facts of the present case”.
32
Along this line of thought, the Court
also notes that as a news portal, Delfi should have considered the risks coming from its activities.
The Delfi news portal is one of the largest in Estonia, and a degree of notoriety has been
attributable to comments posted in its commenting area. Thus, the Court considered that the applicant
company was able to assess the risks related to its activities and to foresee the consequences that
come with it.
Oreste Pollicino and Graziella Romeo ‘Concluding Remarks: Internet Law, Protection of Fundamental Rights and the
Role of Constitutional Adjudication’ The Internet and Constitutional law (Routledge 2016) p. 244
28
Delfi (n 4) para 110;
29
Moreover, to read the passage in this way, one should pay attention to the word “however”. As an adverb, it performs
the function of modifying the meaning of the premises (the beneficial aspects of the Internet). The word “however” is
used to introduce a conclusion of the phrase and to create a hierarchy between the conclusion and the premises.
30
ibid para 133.
31
ibid para 111.
32
ibid.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
Although the Delfi case concerned hate/unlawful speech, this does not seem to be the Court’s
primary concern.
33
Instead, as will be argued below, a key element of its reasoning appears to be the
issue of granting remedies.
More precisely, in Delfi the ECtHR avoided exploring what constitutes unlawful speech, as its
attention was directed entirely towards the ‘duties and responsibilities’ of the provider of the service.
In fact, the ECtHR limited itself to agreeing with the Supreme Court of Estonia on the nature of the
comments: “those comments are defamatory since they are of a vulgar nature, degrade human dignity
and contain threats”.
34
The ECtHR further noted that, as the evidence showed them to be unlawful,
the comments “did not require any linguistic or legal analysis”.
35
The reticence of the Court on the nature of the comments was criticised in the dissenting opinion,
which claimed that “it is unfortunate that the characterisation of the comments remains murky.”
36
Moreover, the dissenting opinion remarked that this approach of the Court does not seem correct in
light of the manifestly racist comments that appeared on the intermediary’s website, for which
criminal action would have been appropriate. Furthermore, an investigation into the nature of the
comments would have provided an example of how to assess problematic content to decide which
speech may be kept and which should be removed. Such an example is important because delineating
the scope of freedom of expression demands a sophisticated level of examination given that it is a
limited right. The taken-for-grantedness of the ECtHR’s comments is underscored, as noted in the
dissenting opinion, by the fact that the ECtHR usually sets a high bar “as to what amounts to an
impermissible call for violence”.
37
In both Delfi and MTE, the Court touched on what has been considered to be one of the
characterising features of the Internet, which is the lack of central editorial control, i.e. the idea that,
as a matter of principle, anyone can participate in the public debate. This is far from saying that all
types of content should be available. As mentioned before, freedom of expression is not an absolute
right and there are limitations.
38
The discourse of the Court thus concerns the significance of the
Internet in changing the traditional media structure, as users are now “creators and primary subjects”
33
The argument of the ECtHR in MTE v Hungary (n 5), which will be illustrated below, is slightly different.
34
Delfi (n 4) para 114.
35
ibid para 117.
36
ibid dissenting opinion, para 72.
37
ibid dissenting opinion, fn 10.
38
ECHR, art 10, para 2.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
and not only consumers.
39
Nonetheless, this was also Delfi’s argument in its submission to the Grand
Chamber:
The applicant company argued that in today’s world, Internet media content was
increasingly created by the users themselves. User-generated content was of high
importance comments on news stories and articles often raised serious debates in society
and even informed journalists of issues that were not publicly known, thereby
contributing to the initiation of journalistic investigations. The opportunity of “everyone”
to contribute to public debate advanced democracy and fulfilled the purposes of freedom
of expression in the best possible way. It was a great challenge in this setting to hold
those who infringed the rights of others accountable while avoiding censorship.
40
(The
Italics are mine)
This reflection does not wish to downplay the harm caused by violent or indecent speech. Nor does
it deny the tensions created by entities such as internet intermediaries, which today have increasingly
become gatekeepers of information flow and mediate which content we get to see and how.
41
Most
importantly, this work does not advocate the absence of regulation of harmful content
42
. What it
attempts to highlight is that the remedy adopted by the Court seems somewhat at odds with the idea
of the pluralistic freedom of expression emphasised by the Court itself in 1976, a freedom which is
applicable not only to “information” or “ideas” that are favourably received or regarded
as inoffensive or as a matter of indifference, but also to those that offend, shock or disturb
the State or any sector of the population. Such are the demands of that pluralism, tolerance
and broadmindedness without which there is no “democratic society.”
43
39
Yochai Benkler, Wealth Of Networks (Yale University Press 2008) (as cited in Rikke Frank Jørgensen, Framing The
Net: The Internet and Human Rights (Edward Elgar Publishing 2011) 85). See also Delfi (n 4) para 107.
40
ibid para 66.
41
As a matter of fact I have expressed the same criticism on how platforms affect freedom of expression in Brogi
Maroni (n 1); Marta Maroni, ‘Some Reflections on the Announced Facebook Oversight Board’ (Robert Schuman
Centre for Advanced Studies, Centre for Media Pluralism and Media Freedom (CMPF) Discussion Series, 17 October
2019) <https://cmpf.eui.eu/some-reflections-on-the-announced-facebook-oversight-board/> accessed 7.11.2019
See in particular, Laidlaw Emily B, Regulating Speech in Cyberspace: Gatekeepers, Human Rights and Corporate
Responsibility (Cambridge University Press 2015).
42
see Robert Spano, 'Intermediary Liability For Online User Comments Under The European Convention On Human
Rights' in Mart Susi, Jukka Viljanen, Human Rights Law and Regulating Freedom of Expression in New Media :
Lessons from Nordic Approaches, (Routledge 2018)
43
Handyside v the United Kingdom App no 5493/72 (ECtHR, 7 December 1976).
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
In addition, the ruling clashes with another strand of the Court´s jurisprudence which so far has
protected and promoted the critical role of the Internet in enhancing the public´s access to news
and facilitating the dissemination of information in general”.
44
During the last few years, the
Court has assessed several cases dealing with undue interference to Internet access by public
authorities
45
. In doing so, the Court stated that
[freedom of expression] applies not only to the content of the Information, but also to
the means of dissemination, since any restriction imposed on the latter necessarily
interferes with the right to receive and impart information”.
46
Eventually, the Court also
recognised Internet access to be understood as a right given its role for the enjoyment
of other rights as such.
47
If we read Delfi in conjunction with the above case law, there seems to be a mismatch when
the Court does not consider how private actors could impair the whole set of values that the
Court itself attaches to the Internet.
The early regulatory design that intermediaries should have played a neutral and passive function,
was a construction aimed at preserving internet intermediaries’ economic interests, but also
suggested that an intermediary should not have knowledge or control over content. These private
companies should not have decided which content could be circulated or not. UN documents can
also provide some guidance on the matter. The UN Special Rapporteur for freedom of expression
44
Times Newspapers Ltd (nos 1 and 2) v the United Kingdom ECHR 2009I 377. The case concerns the responsibility
of a publisher for internet archives and violation of personality rights.
45
See Ahmet Yildirim v Turkey ECHR 2012VI 505; and Cengiz and Others v Turkey ECHR 2015VIII 177. In the
case of Cengiz and Others v Turkey the Court held that “The internet has now become one of the principal means to by
which individuals exercise their right to receive and impart information and ideas, providing as it does essential tools
for participation in activities and discussions concerning political issues and issues of general interests” (para 49). In
Kelda v Estonia App no 17429/10 (ECtHR, 19 January 2016) the Court held that “increasing amount of services and
information is available only via the Internet”. The Internet was increasingly recognised as a right in case of Jankovskis
v Lithuania App no 21575/08 (ECtHR, 17 January 2017).
46
Ahmet Yildirim v Turkey (n 44) para 50.
47
See Ahmet Yildirim v Turkey (n 44) para 31; “The right to Internet access is protected in theory by the constitutional
guarantees applicable to freedom of expression and freedom to receive ideas and information. The right to Internet
access is considered to be inherent in the right to access information and communication protected by national
Constitutions and encompasses the right for each individual to participate in the information society and the obligation
for States to guarantee access to the Internet for their citizens. It can therefore be inferred from all the general
guarantees protecting freedom of expression that a right to unhindered Internet access should also be recognised” and in
the Case of Jankovskis v Lithuania (n 44) “The Court is mindful of the fact that in a number of Council of Europe´s and
other international instruments the public-service value of the Internet and its importance for the enjoyment of a range
of human rights has been recognised. Internet access has increasingly been understood as a right, and calls have been
made to develop effective policies to achieve universal access to the Internet and to overcome the digital divide (see
kalda) para 62”.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
has stressed that internet intermediaries are “not best placed to make the determination of whether a
particular content is illegal, which requires careful balancing of competing interests and
consideration of defences”.
48
The role played by Delfi
The functioning of content intermediaries must be considered carefully as their economic revenue
comes from advertisements. They tend to make sensationalistic information more visible because it
gets more engagement in order to increase profits.
This was an important aspect in Delfi as well: the Court repeatedly remarked on the economic nature
of the applicant. For instance, in handing down its judgment, the ECtHR stressed that Delfi was “a
news portal run on commercial basis which sought to attract a large number of comments on news
articles published by it”,
49
and as such it had “duties and responsibilities”.
Here the ECtHR is mindful that it was on a slippery slope and its decision could have deep
repercussions for a vast array of smaller actors. It therefore excluded the notion that social platforms
or private and personal blogs faced the same liability risk.
50
Thus, one of the criteria adopted by the
court concerned the size of the applicant involved. Although this criterion seems reasonable to some
extent, especially considering the role played by platforms, it might be extremely difficult to design
fixed legal categories for such a diversified environment.
As for user-generated content, Delfi submitted that it was sufficient to remove such content as
soon as it became aware of its offensive nature.
51
Thus, the Estonian news portal claimed that it should
be treated as an intermediary in relation to the comments of its readers and that it was entitled to
follow the law limiting the obligation to monitor third party content.
52
The Estonian Government disagreed with this argument, stressing that a) Delfi was the largest
portal in the country; b) it also played an active role in inviting readers to comment; and c) the
applicant was the only actor who could modify the comments.
53
48
UN Human Rights Council, ‘Report of the Special Rapporteur on the Promotion and Protection of Freedom of
Opinion and Expression, Frank La Rue’ (16 May 2011) UN Doc A7/HRC/717/27.
49
Delfi (n 4) paras 115, 144.
50
ibid para 117; dissenting opinion, para 9.
51
E-Commerce Directive, art 14.
52
Delfi (n 4) para 69.
53
ibid para 81.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
Interestingly, Brunner suggests that the ECtHR failed to seek a tailored solution for dealing with
online news portals. The Court treated Delfi as a traditional news medium by relying on its own
case law and it did not acknowledge other interpretations, such as the ones of the Court of Justice of
the European Union (CJEU). Brunner explains that the case law of the CJEU could have provided
insights for assessing the case at hand, and how in the Google France and L´Oreal cases,
commercial gain was considered to be irrelevant for interpreting the E-commerce Directive.
54
In
particular, as Brunner claims,
If the Delfi case were examined in light of the E-commerce Directive and the case law of the CJEU,
the enquiry would turn on whether Delfi should be regarded as an intermediary and, if so, the type of
role that it played with respect to the comments at issue. Delfi could be deemed an intermediary for
the purposes of user-generated content that it hosted, and the principles articulated in the Google
France and L’Oreal cases would apply.
55
However, other scholars have considered the CJEU case law to be confusing when it comes to the
interpretation of who an intermediary is
56
, and there is consensus that the E-Commerce directive is
no longer able to deal with challenges coming from platforms such as Google or Facebook and new
technological developments such as algorithms. Nonetheless it is interesting (and problematic) how
the CJEU keeps holding on to the idea that intermediaries should be treated as neutral.
57
Nonetheless, while one might easily agree with Brunner on the narrow approach adopted by the
ECtHR, which treated Delfi as a traditional medium, the two Courts work in different jurisdictions
and operate with different semantics.
Conceptualising the liability of Internet intermediaries in this way contributes to legitimising a
system in which private companies are more and more in the position of deciding how we
consume content. In addition, when decisions on freedom of expression are delegated to
intermediaries, a democratic problem might arise: the content might be taken down proactively
54
Brunner (n 9); for another comparison between the ECJ approach and ECtHRs, see Valcke and Ombelet (n 3).
55
Brunner (n 9) p. 168.
56
Sophie Stalla-Bourdillon and Robert Thorburn, 'The Scandal Of Intermediary: Acknowledging The Both/And
Dispensation For Regulating Hybrid Actors', in Bilyana Petkova and Tuomas Ojanen (eds) Fundamental Rights
Protection Online: the Future Regulation of Intermediaries (Edward Elgar 2020).
57
This construction has now been put under further pressure because of the use that major platforms make of artificial
Intelligence. The CJEU still constructs its cases based on the idea of neutrality. See Elda Brogi and Marta Maroni, ‘Eva
Glawischnig-Piesczek V Facebook Ireland Limited: A New Layer of Neutrality’ (Robert Schuman Centre for Advanced
Studies, Centre for Media Pluralism and Media Freedom (CMPF) Discussion Series, 17 October 2019)
<https://cmpf.eui.eu/eva-glawischnig-piesczek-v-facebook-ireland-limited-a-new-layer-of-neutrality/> accessed
7.1.2019
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
and there might not be information on how much content is taken down and on what grounds.
Finally, intermediaries might fail to identify the jurisprudential basis guiding complex freedom
of expression cases
58
: whereas some cases are more straightforward to assess (e.g. terrorism)
some other content might not be “proper”, yet it is still protected under freedom of expression.
The lack of qualitative assessment on freedom of expression results in the removal of content.
Not only could this be considered to be censorship, but also it affects the normative dimension
of freedom of expression that is its multidimensional meaning.
As a matter of fact, several categories subsume freedom of expression: political expression,
artistic expression, satire, commercial speech to mention a few, and for each of these categories,
there is a different degree of tolerance. Yet expression is constrained when concerned with
national security, prevention of disorder and crime, and protection of personality rights
59
. The
point I would like to make is that the way these categories and limitations are tested against
each other matters for the substantial understanding of freedom of expression.
60
If companies
bear the primary responsibility for deciding on the content, freedom of expression is reduced to
a technicality/ a quick standard to be implemented: what we lose is the normative tension that
characterises freedom of expression. Again, I am not maintaining an idealised position that
platforms and intermediaries should not bear responsibility for harmful content
61
. Rather, my
argument is that there is the need for a more refined model of regulation, one which does not
give rise to a homologated understanding of freedom of expression online which would have
repercussions on diversity and pluralism.
58
This reflection builds on Andrew D. Murray, 'Regulation And Rights In Networked Space' (2003) 30 Journal of Law
and Society.
59
For example, ECHR, art 10(2) stipulates that restriction on freedom of expression can occur in cases of national
security, territorial integrity or public safety, prevention of disorder or crime, protection of health or morals, protection
of the reputation or rights of others, for preventing the disclosure of information received in confidence and
maintaining the authority and impartiality of the judiciary. These restrictions need to be prescribed by law and
necessary in a democratic society and considered ‘foreseeable’. As Brogi and Bania highlighted, these provisions on
freedom of expression “apply equally online and offline.” See Elda Brogi and Konstantina Bania, Freedom of
Expression Online Report Prepared by the Centre for Media Pluralism and Media Freedom for the Council of Europe
(unpublished).
60
Kohl also remarks that when intermediaries are forced to act as gatekeepers, they must play a “quasi-legal-judgment”
regardless of the fact they don´t have the independence nor the expertise of the judiciary. Uta Kohl, ‘The Rise and Rise
of Online Intermediaries in the Governance of the Internet and Beyond Connectivity Intermediaries’ (2012) 26
International Review of Law, Computers & Technology 185, 191.
61
To be sure, intermediaries and major platforms are often on the news for their involvement in this or that scandal.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
Law without remedies?
At this point, it is worth turning to the thread which runs through the Court’s reasoning: the issue of
remedies. The overarching questions of the case seem to be what would happen if it were impossible
to grant remedies, and what would occur if it were impossible to enforce those basic rights that law
confers on individuals. Fundamental to the case is the apparent threat of law losing terrain in the
online environment. The narrative employed by the ECtHR portrays a scenario in which the function
of law is undermined by the technicalities of the Internet, which restrict the capacity of law to perform
its traditional tasks. Consequently, throughout a persuasive construction, the ECtHR emphasised the
need to build an environment in which justice can be ensured and personal private life protected.
In order to strengthen its argument, the ECtHR quoted the case of K.U v Finland,
62
stressing that
legitimate imperatives, such as the prevention of disorder or crime or the protection of the rights and
freedoms of others”
63
must be borne in mind, as in the case of sexual offences against minors. One
could speculate that this quote has a threefold function. First, it appealed to the idea of not allowing
an environment in which a sexual crime against a minor could remain unpunished, and consequently
(reading between the lines) where law cannot function.
Second, reference to the K.U v Finland case also calls into question the appropriateness and
fairness of the ECtHR decisions - that is, its commitment to proportionality. Throughout K.U v
Finland, the Court demonstrated how it decided differently from Delfi on the opportunity for the
victim to obtain damages from the service providers, as on that occasion it “held that there had to be
a remedy enabling the actual offender to be identified and brought to justice”.
64
Moreover, one of the
more evident contradictions in the Court’s reasoning emerges here: while the Court appeals to the
solidity of its jurisprudential reasoning, it implicitly delegates to a content provider the complicated
task of deciding which language, which register and consequently whose voice can be heard.
However, the question that remains is whether, in such a scenario, intermediaries will act in a
62
K.U. v Finland ECHR 2008V 125. The case concerns a sexual advertisement on a dating site with a picture and full
personal details of a Finnish boy. When the applicant requested the service provider to disclose the identity of the
person who posted the advertisement, the internet service provider refused. The case was brought under the Helsinki
district Court which nonetheless stated that there was no legal provision to oblige the provider to divulge the identity of
the person who posted the advertisement. The ECtHR ruled that the legislature should have provided a framework for
reconciling the confidentiality of Internet services with the prevention of disorder or crime and the protection of the
rights and freedoms of other”.
63
Delfi (n 4) para 149.
64
ibid.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
professional (deontological?) fashion when it comes to third-party-generated content or silence
whatever speech is ‘sensitive’ or politically incorrect simply to avoid liability problems.
Thus, the puzzling aspect of the Court ruling lies in entrusting a private actor with the role of
defining the boundaries of freedom of expression. As noted above, intermediaries might not be best
equipped for this task. Finally, quoting K.U v Finland helped the Court to introduce and underline
one of the more problematic aspects of the internet ecosystem, the issue of anonymity:
Although the case of K.U. v. Finland concerned a breach classified as a criminal offence
under the domestic law and involved a more sweeping intrusion into the victim’s private
life than the present case, it is evident from the Court’s reasoning that anonymity on the
Internet, although an important value, must be balanced against other rights and interests
65
Anonymity
The ECtHR further devoted some space on the issue of anonymity in its arguments and framed its
question in terms of the need to maintain order:
Anonymity has long been a means of avoiding reprisals or unwanted attention. As such,
it is capable of promoting the free flow of ideas and information in an important manner,
including, notably, on the Internet. At the same time, the Court does not lose sight of the
ease, scope and speed of the dissemination of information on the Internet, and the
persistence of the information once disclosed, which may considerably aggravate the
effects of unlawful speech on the Internet compared to traditional media.
66
(The Italics
are mine)
Anonymity is one of the Internet’s more challenging and interesting features. Anonymity is
controversial, as while it allows a greater sense of safety when expressing opinions, and thus can be
considered to promote freedom of expression, it can also be used otherwise to cause harm.
67
However, in this specific case, the anonymity provided by technology appears as a disruptive force
for the legal system. For instance, the Estonian Government emphasised the impossibility of
65
ibid.
66
ibid para 147.
67
The US Supreme Court has held in Doe v 2themart.com that the right of internet users to speak anonymously should
be safeguarded like other constitutional rights. Anonymity, ‘the absence of identity’, disinhibits people in expressing
their thoughts, as they feel free to comment on sensitive or controversial political issues. Doe v United States, 487 US
201 (1988).
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
identifying all the authors of the comments primarily because Delfi “allowed comments by non-
registered users”,
68
but also for several technical reasons, such as the existence of Wi-Fi hotspots,
dynamic IP servers or servers in foreign countries. Delfi nonetheless disagreed with the Estonian
Court’s argument, claiming that the case could have been brought against the authors of the
comments.
69
In engaging with these issues, the Court observed that while anonymity cannot be
absolute, different degrees of anonymity can be allowed. More precisely
an Internet user may be anonymous to the wider public while being identifiable by a
service provider through an account or contact data that may be either unverified or
subject to some kind of verification
70
The Court then continued its reasoning by providing an assurance that “the release of such
information would usually require an injunction by the investigative or judicial authorities and would
be subject to restrictive conditions”.
71
Thus, the message sent by the Court is that authentication must
be secured, when there is a need to “identify and prosecute perpetrators”.
72
This is also important for
its reasoning.
In this paragraph, one can detect the classical dichotomy between private and collective interest in
that absolute anonymity cannot be permitted for the good of the collectivity.
Thus, the passage framed by the Court is again rich with meaning and abstractly revolves around
a security discourse. Anonymity is presented as limiting the capacity of the legal system to operate
in the digital environment with consequences for the security offered by the legal system to the
collectivity. There is also another dimension of security, namely of the legal system itself, which
would not allow access to identity information unless necessary.
73
This latter feature plays a constitutive role in the Court’s reasoning, as it enables the ECtHR to
rely on its own procedures and to make proportionality its hallmark.
74
The sharpness of this framing
68
ibid para 65.
69
ibid para 77.
70
ibid para 148; the Court continues “A service provider may also allow a degree of anonymity for its users, in which
case the users are not required to identify themselves at all and they may only be traceable to a limited extent
through the information retained by Internet access providers”.
71
ibid.
72
ibid.
73
For this broad concept of security see Massimo Fichera, ‘Security Issues as Existential Threat to the Community’ in
Massimo Fichera and Jens Kremer (eds), Law and Security in Europe Reconsidering the Security Constitution
(Intersertia 2013).
74
Proportionality is employed here as “evok(ing) an ideal of harmony, natural (quasi-geometric) order or coherence
founded on reason”, Nicola Lacey, The Metaphor of Proportionality (2016) 43 Journal of Law and Society 27, 31.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
prevents counter arguments related to anonymity from emerging because, after all, the Court is
implying that it is operating in a geographical space committed to democracy
75
and safeguarded by
the Court itself.
76
As a regional Court, the ECtHR must justify itself as a credible institution, and thus
cannot fail or override these democratic values.
Although one may agree that Delfi failed to follow best practice (for instance by allowing
unregistered comments), the strict logic
77
advanced by the Court seems to leave no room for
questioning whether other means existed for achieving the same regulatory aim, without weakening
freedom of expression online. Thus, while the Internet was supposed to provide new opportunities,
the Court simply returned to the status quo: the Internet is treated as a traditional medium, in which
there is a centralisation of distribution of content, and the legal system is “an organ of governance
and regulation responsible for the establishment and current maintaining of a just order”.
78
Today, full anonymity on the Internet is more difficult to achieve because of technical identifiers,
which enable traceable anonymity.
79
However, at the time of the cases, the Estonian Government
justified its failure to attempt to collect pre-trial evidence by stating that “the State’s enacting of a
regulation providing for mandatory identification of commentators on an Internet portal would
constitute excessive interference.
80
Moreover, the Estonian Government claimed that as civil
proceedings were more appropriate in this specific case, it was not possible to rely on surveillance
techniques.
Investigating each individual commentator in this case would have been too invasive and thus
disproportionate and it would have drawn attention to Estonian institutions for using harsh measures.
In this sense, by agreeing with the Estonian Government in finding the content provider liable, the
ECtHR might have achieved a twofold aim: to preserve the function of law by creating a more
disciplined environment (e.g. by overcoming the problem of anonymity and dealing with different
jurisdictions) and avoiding excessively invasive measures (e.g. by investigating the commentators).
75
As stated in ECHR, CETS 5. “as the governments of European countries which are likeminded and have a common
heritage of political traditions, ideals, freedom and the rule of law”.
76
Consequently, what the Court is saying is that full anonymity might not even be needed.
77
In this work legal argumentations are approached from Luhmann’s functionalism perspective, which rejects the idea
that legal arguments work on best reason / or better arguments. In Luhmann´s view, legal arguments should not be
evaluated for their reasonability, which ultimately means autology (the ability to justify themselves), but for their
persuasive power in the process of communication.
78
Jeam Clam, Contingency, Reciprocity, the Other, and the Other in the Other LuhmannLacan, an Encounter in
Anders La Cour and Andreas Philippopoulos-Mihalopoulos (eds), Luhmann Observed (Palgrave Macmillan 2013) 20.
79
Laura DeNardis, The Global War For Internet Governance (Yale University Press 2014).
80
Delfi (n 4) para 64.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
While at face value, this could count as the court´s balancing exercise between security and liberty,
this appears to have been done in a partial way and to be influenced by the need of the court to
preserve the operations of law in the online ecosystem. Previous commentaries have already stressed
how the logic of the Court was driven by the concern to protect reputational rights
81
. As Neville Cox
contends,
“there must be some defendant available from whom the injured party could seek redress, in order
for the state’s positive obligation to support reputational and privacy rights to be fulfilled, and, the
applicants were the only entity capable of providing realistic redress for the injured party, then
liability should be imposed on them”
82
With this ruling, one might ponder why the ECtHR, a court that has historically been at the
forefront of protecting freedom of expression, adopted such a controversial decision. The narrative
constructed by the Court can be considered to be a move enabling the judiciary to preserve the
opportunity to perform in an environment that could potentially be difficult to control (legally
speaking). As shown above, anonymity coupled with internet technicalities and multiple
jurisdictions prevented the law from granting remedies. Hence one might need to ask what the
remedies represent for the legal system. West’s Legal Thesaurus/Dictionary defines remedies “as
means by which a right is enforced; the steps by which the violation of a right is prevented,
redressed, or compensated”. Remedies also intersect the machinery through which law can operate.
As such, remedies are situated between substantive law and procedural law, which means that it
would be pointless to have a rule if it then cannot be enforced. It would have been self-defeating for
the Court to rule in a way that would give up on the possibility of protecting those who expected
protection. A lack of expectations of redress would create a crisis in the legal system, as it would no
longer be necessary to appeal to legal instruments. For these reasons, it seems that with this ruling,
first and foremost the Court attempted to carve a space for the law to operate within the Internet
although as illustrated above, this comes at the expense of a broader understanding of freedom of
expression online. Internet intermediaries operate between the users and the content that is
circulated, hence they exert a great deal of control within the Internet ecosystem. Although turning
to intermediaries is the most immediate solution, there is a subtle and difficult balance to maintain
between devising redresses mechanisms and avoiding a “disciplinary” ecosystem
83
.
81
see Spano (n 42).
82
see Cox ( n 27) p. 623
83
See Kohl (n 60).
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
Magyar T.E. and Index.hu Zrt v Hungary
In February 2016, the ECtHR ruled on its second case, MTE v Hungary. Even though it confirmed
the substantial reasoning of Delfi, it concluded that in this case a violation of freedom of expression
had occurred. In its reasoning, the Court attempted to re-balance the previous approach, setting out a
list of criteria to reduce the chance of the imposition of liability and the restriction of content in the
online environment.
Two companies were involved in the MTE case. The first was Magyar Tartalomszolgáltatók
Egyesülete, a not-for-profit self-regulatory body of Hungarian internet service providers. The second
applicant, Index.hu Zrt, was a major internet news portal. Both published an article on the unethical
commercial conduct of two real-estate-management websites, owned by the same company, and this
news item attracted rather unpleasant comments.
84
In contrast to Delfi, in MTE the ECtHR ruled that
the utterances in question, in spite of their vulgarity (e.g. “People like this should go and shit a
hedgehog and spend all their money on their mothers’ tombs until they drop dead”),
85
could not be
categorised as hate speech or incitement to violence.
The ECtHR simply claimed that the statements in this case “did not constitute clearly unlawful
speech and they certainly did not amount to hate speech or incitement to violence”.
86
The ECtHR, in
handing down its ruling, remarked that the domestic authorities had not investigated whether the
impugned statements could be deemed unlawful. What is more, the Court declared that the domestic
authorities had also failed to consider whether the actual authors could have been found liable. This
lack of investigation contributed to the Hungarian court’s poor proportionality analysis.
As in Delfi, both applicants in the MTE case had a notice-and-take-down system in place in which
any user could flag unlawful comments to the service provider so that they could be removed.
However, in this latter case, the ECtHR observed that “the injured company never requested the
applicants to remove the comments but opted to seek justice directly in court”.
87
The ECtHR
considered that the Hungarian domestic court had failed to take this latter element into consideration.
The Court also remarked that the “notice-and-take-down system could function in many cases as an
84
MTE v Hungary (n 5) para 19.
85
ibid para 14.
86
ibid para 64.
87
ibid para 83.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
appropriate tool for balancing the rights and interests of all those involved”.
88
A relevant factor that
led the Court to decide differently in MTE was that the conduct of the company had already been the
subject of various complaints to the country’s consumer protection organs.
89
For this reason, in
assessing the proportionality of the domestic Hungarian authorities, the ECtHR emphasised that the
comments in question would not have made “any additional and significant impact on the attitude of
the consumers concerned”.
90
Moreover, while in Delfi the ECtHR stressed that the news portal had been sued for breaching the
personality rights of the company owner, in MTE the real estate company had claimed that MTE
and Index.hu Zrt had breached its right to a good reputation.
91
Here, the ECtHR paid particular
attention to the fact that the right to a good reputation does not require the same level of protection
as personality rights. Moreover, the ECtHR emphasised that companies cannot claim to be victims
of personality rights violation.
92
Personality rights (such those involved in the Delfi case) are strictly
intertwined with dignity and as such apply to natural persons, whereas commercial reputation is
devoid of that moral dimension.
93
The ECtHR also reasoned on the role of the plaintiffs and
emphasised that the domestic Court should have considered the role that the applicants played in
generating the comments as MTE is a not-for-profit association with no economic interest in
providing the services. As for the comments, the ECtHR observed that the domestic court had not
investigated whether the comments were serious to the extent of causing prejudice to the
professional reputation of the real estate. Finally, whereas Delfi had been required to pay a
symbolic fee, the MTE applicants had been obliged to pay the court fees, which the ECtHR
acknowledged could have negative consequences, such as the closure of the comment space.
94
In addition, the Court stated that:
[I]t cannot be excluded that the court decision finding against the applicants in the present
case might produce a legal basis for a further legal action resulting in a damage award. In
any event, the Court is of the view that the decisive question when assessing the
consequence for the applicants is not the absence of damages payable, but the manner in
which Internet portals such as theirs can be held liable for third party comments. Such
88
ibid para 91.
89
ibid para 72.
90
ibid para 85.
91
ibid para 15.
92
ibid para 66.
93
ibid.
94
ibid para 86.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
liability may have foreseeable negative consequences on the comment environment of an
Internet portal, for example by impelling it to close the commenting space altogether. For
the Court, these consequences may have, directly or indirectly, a chilling effect on the
freedom of expression on the Internet. This effect could be particularly detrimental for a
non-commercial website such the first applicant.
95
The narrative of the Court
If the concern of the Court in the Delfi case was to safeguard the performativity of law, the MTE case
can be viewed as a refined analysis of its approach. Since the Delfi Case was heard by the Grand
Chamber, it represents a precedent. Therefore, the MTE cases presuppose a departure from the
previous orientation of the Court. As already mentioned, the MTE case does not change the
substantive approach to the liability of content providers,
96
but it does restrict the reasons for such
liability.
After recalling the Delfi framework, the Court concluded that the present case was different, as the
comments did not constitute clearly unlawful speech.
97
This time, the Court paid attention to the
registers used in the new medium, the Internet. The Court not only remarked that vulgarity does not
always qualify as an offence, but also that the style of the speech should be protected as part of the
freedom of expression.
98
This reflection can be seen as an attempt by the Court to remedy the problems deriving from the Delfi
case, which appeared to lower the standards of freedom of expression that the ECtHR had built over
time. Thus, the Court highlighted that
[w]ithout losing sight of the effects of defamation on the Internet, especially given the
ease, scope and speed of the dissemination of information,
99
the Court also considers that
regard must be had to the specificities of the style of communication on certain Internet
portals. The expressions used in the comments, albeit belonging to a low register of style,
are common in communication on many Internet portals a consideration that reduces
the impact that can be attributed to those expressions.
100
(The Italics are mine)
95
ibid.
96
MTE v Hungary (n 5) concurring opinion, para 3.
97
ibid para 64.
98
ibid para 68.
99
Delfi (n 4) para 147.
100
ibid para 77.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
Further, the Court refuted the Hungarian understanding that they should have expected that some
might be in breach of the law, as the applicants had allowed unfiltered comments. This, the Court
claimed, would have required “excessive and impracticable forethought capable of undermining
freedom of the right to impart information on the Internet.
101
Continuing this line of argument, the
Court highlighted the domestic court’s failure to assess the effect of liability on freedom of expression
on the Internet.
102
To sum up, the Court found the Hungarian notion of liability problematic, as it
precluded balancing between competing rights. This case engages in a direct dialogue with the
previous case, as shown in the well-known necessary in a democratic society test, when the Court, in
handing down the ruling, compared the two decisions. In MTE, the Court addressed the questions left
open in Delfi concerning the restriction of freedom of expression, re-elaborating some meanings and
clarifying others in order to restrict the possibility for cases of liability. This second ruling has drawn
more positive reactions, but the scenario on the regulatory approach of internet intermediaries remains
ill-defined.
CONCLUSIONS
In this paper I have provided an analysis of the ECtHR’s approach to the liability of content
providers for user generated comments. Two cases have been considered: Delfi v. Estonia (2015)
and MTE v Hungary (2016). More attention has been paid to Delfi, as it represents the first ruling
on the matter and has been generally criticised for potentially leading to a scenario in which online
providers might filter user content, thereby resulting in arbitral removal of content, in order to avoid
liability. The second case, MTE v. Hungary, is also important as an effort to rebalance the stance
taken in Delfi. The paper started by highlighting what the Court neglected, i.e., the general concern
that is associated with delegating traditional public functions to private actors, and second by
observing, that the Internet creates benefits and not only the risks depicted by the Court in Delfi.
Regarding the first point, it should be borne in mind that whenever it is up to intermediaries to
decide on the content of rights, these are reduced to a “technicality”. This means that rights are
evaluated according to standards which might not consider the tensions that different interests at
stake might generate. Instead, exposing the tensions concerning the interpretation of rights
101
ibid para 82.
102
ibid para 88.
BILYANA PETKOVA & TUOMAS OJANEN (EDS) FUNDAMENTAL RIGHTS PROTECTION ONLINE: THE FUTURE
REGULATION OF INTERMEDIARIES, EDWARD ELGAR PUBLISHING
1
contributes to developing new meanings and dimensions of rights.
103
When intermediaries bear the
responsibility for assessing the scope of application of rights, they might silence these divergences,
thus affecting the configuration of the Internet as a pluralistic environment.
The aim in this paper has been to demonstrate that the concern of the Court exceeded the mere
exercise of balancing between freedom of expression and unlawful speech. In addition, the paper
has relied on a second narrative that underlines the challenges that internet technologies pose for the
legal system. Thus, the emphasis in the paper was that the Court’s reasoning was intimately bound
to the circumstances created by Delfi: anonymous and unregistered comments. Moreover, it further
highlights how the Court attempted to regulate Internet service providers in order to create a more
disciplined environment in which law can function. Nonetheless, in sketching the main passages of
the Delfi judgment, the work reveals the complexity of the arguments and the difficulties the Court
faced in deciding the case.
The second case (MTE) shows how the Court attempted to adjust to a supposedly new environment
where the internet infrastructure itself had become a subject to be taken into account beyond users’
rights.
103
For an extensive and theoretical study on technological disruption and the privatisation of coercion see Riikka
Koulu, Law, Technology and Dispute Resolution: The Privatisation of Coercion (Routledge 2018).
Article
Full-text available
Digital governance is a public concern, yet under private control. After numerous scandals, all stakeholders in the European Union (EU) agreed to establish a “novel constitution for the internet” that would effectively constrain the power of large platforms. Yet the Digital Services Act (DSA) ultimately legitimized and institutionalized their position as the gatekeepers of the internet. Why? We argue that platforms prevailed thanks to their ability as intermediaries to quietly shape the available policy options. Our “ platform power mechanism” combines institutional and ideational sources of business power to show how big tech drew on its entrenched position as an indispensable provider of essential services and promulgated the idea of itself as a responsible and neutral intermediary. We follow the unfolding of platform power through a process-tracing analysis of Google and Meta’s activities with respect to DSA legislation from its announcement (2020) to its adoption (2022). Besides contributing a reconceptualization of the DSA as a regulatory capture, we integrate the notion of platform power into a “regulator–intermediary–target” model and demonstrate how gatekeepers have exploited information asymmetries to share “the public space.” Our analysis thus supplements established approaches that have derived regulators’ deference to platforms from the tacit allegiance of consumers.
Book
Full-text available
This is the second book of the Online Hate Speech Trilogy. The work focuses on the legal challenges of combating toxic language and retaliating against those who spread hate on the Internet. Although the need for fighting violent narratives appears evident, given the role of hate in eroding trust and fragmenting the social fabric, there are many sensitive layers to the matter. The authors analyse the weaknesses of platform self-regulation, the European Union’s legal approach to combating online hate, the use of toxic language as a political weapon in Latin America, and the risks it poses to peace in Africa.
Article
Fundamental rights are at the centre of the proposal for a new Digital Services Act (DSA), an EU Regulation that aims to update the e-Commerce Directive, the nuclear legislation on intermediary service providers. Of special importance among these providers are online platforms, the gatekeepers of the new social, economic and political life of the digital XXI century. Both content management and conflict resolution in the context of online platforms fall under the provisions of the DSA and they are key to an adequate regulation of users’ interaction, both with the online platforms and among themselves and to the protection of fundamental rights. The paper analysis content management and conflict resolution mechanisms under the e-Commerce Directive and the DSA Proposal through the perspective of fundamental rights protection and promotion. The goal is to offer a first approach and appraisal to the conflict resolution mechanism put in place by the Proposal and ascertain if they comply with EU and Member States law and assure the adequate and balanced exercise of fundamental rights.
Article
Full-text available
Článek se zabývá podmínkami odpovědnosti zprostředkovatelů za protiprávní projevy třetích osob na internetu a zároveň i jejich odpovědností za nepřípustné omezování svobody projevu. Těmito dvěma rovinami regulace zprostředkovatelů, jakožto soukromých subjektů, státy fakticky přenáší část svých mezinárodních lidskoprávních závazků, které plynou zejména ze svobody projevu, práva na respektování soukromého a rodinného života, zákazu diskriminace, ale i dalších hodnot. Pro zajištění rovnováhy mezi protichůdnými zájmy spojenými s výkonem práv a svobod na internetu je nezbytné, aby státy přijaly právní úpravu, která bude dostatečně určitě vymezovat povinnosti zprostředkovatelů. V opačném případě se státy mohou jejich prostřednictvím dopouštět tzv. soukromé cenzury. S ohledem na přeshraniční charakter internetu by se povinnosti kladené na zprostředkovatele neměly napříč státy zásadně lišit. Jedním z harmonizačních prvků v evropském kontextu je judikatura Evropského soudu pro lidská práva. Autorka proto analyzuje jeho relevantní rozsudky a zkoumá, zda obsahují konkrétní kritéria, která by měla regulace zprostředkovatelů na vnitrostátní úrovni zohledňovat či zda se jedná o hodnocení ad hoc.
Article
Within the human rights protection system at both international and national system, there are several rights that might be and usually get into a clash of interaction if applied at the same time. One of the common examples is a clash between the right to privacy and freedom of expression. Both are important in relation to the protection of personal identity and autonomy and both concern development of every human being. Nevertheless, if there is a clash, one has to decide which one is given priority. This study aims to analyse the protection of these two rights in case of such a clash between them occurs. Since there has already been a lot of studies dealing with this clash, this study therefore limits its focus on two issues, namely first, specificities of the digital era and second, elected public figures have been identified as a particular subject of research because of a chosen specific case that has been under judicial scrutiny in Slovakia during the period of analysis of the research topic of the right to privacy in digital age. Striking a balance in which both these fundamental rights are protected is challenging, especially in the digital era. The focus is therefore given to the background and case-law of the European Court of Human Rights and Constitutional Court of the Slovak Republic when necessary to point out some specific features because of the stimulating case-law and the influence that these judicial authorities have in relation to the Slovak Republic. Finally, it is submitted that the online human rights protection should meet the same conditions as the offline one, keeping in mind all the circumstances that are typical for the digital world.
ResearchGate has not been able to resolve any references for this publication.