Online disinformation and freedom of expression in the electoral context:
the European and Italian responses
Oreste Pollicino*
Laura Somaini**
1. Introduction - 1.1 Defining the object: why “disinformation”? - 1.2 Defining the
problem - 1.3 Disinformation in electoral processes and specificities of referenda -
2. Tackling disinformation practices v. upholding freedom of expression - 2.1
Theoretical justifications - 2.2 The US perspective - 2.3 The European perspective -
2.4 Freedom of expression in the electoral context - 3. Legislators’ responses - 3.1
The EU level - 3.2 Italy’s (failed) attempt to legislate “fake news” - 4. The Italian
Referendum case - 4.1 The Italian framework on referenda - 4.2 Disinformation and
interference with the 2016 referendum - 4.3 The shift towards an integrated approach
in electoral monitoring - 5. Conclusions
1. Introduction
This contribution addresses the relevance of the sub-category “disinformation” under a
constitutional law lens, i.e. where possible tensions with fundamental rights and alteration of
dynamics of the democratic process are created, with specific regards to the electoral context.
Through the comparison of the constitutional paradigms of freedom of expression across the Atlantic,
the regulatory responses attempted by the European Union are explored to set the background of the
Italian experience. Section 1 presents the issues relating to disinformation practices, translating them
into the specific electoral context. Section 2 provides a comparative view of the constitutional
framework, including the US and European perspectives. Section 3 provides an overview of the
different responses of legislators at the EU level and Italy. Section 4 focuses on the case of the 2016
Italian constitutional referendum and the adopted approach with a forward-looking view to
subsequent elections. Section 5 concludes.
1.1 Defining the object: why “disinformation”?
Online disinformation has rapidly become a topical interest, with terms such as “fake news”
and “online propaganda” becoming worldwide buzzwords. As reported by many studies (Bayer et
al., 2019; De Gregorio, Perotti, 2019),1
to adequately frame the problem, an ambiguous use of the
*
Full Professor of Constitutional Law and Media Law, Bocconi University. Author of paragraphs 1-2.3.
**
Research Assistant, Bocconi University. Author of paragraphs 2.4-4.3. The authors kindly thank Giovanni De Gregorio
for his valuable comments.
Electronic copy available at: https://ssrn.com/abstract=3552680
terms must be rejected. This contribution relies on the definition of “disinformation” given by the
Commission’s High-Level Expert Group (“HLEG”) encompassing “false, inaccurate, or misleading
information designed, presented and promoted to intentionally cause public harm or for profit”.2
The
distinguishing factor from the broader concept of “misinformation” is the intentional character with
which dissemination occurs, defining the latter as “misleading or inaccurate information shared by
people who do not recognize it as such”3
(Pollicino, Bietti, 2019, 47-49). Given the preponderant role
of the concept of “disinformation” in the European agenda and the Italian framework, as will be
discussed infra, this contribution will focus on the particular subcategory of “disinformation”.
Its characteristics display a risk of threatening democratic political processes and values in different
sectors and is driven by “the production and promotion of disinformation for economic gains or for
political or ideological goals”,4
which may vary depending on how it is received by audiences and
communities. This definition includes harmful speech which does not fall under other categories
already identified as illegal, i.e. hate speech, defamation or incitement to violence (ibid.).
1.2 Defining the problem
From a constitutional perspective, disinformation practices become relevant where they cause
tensions to the exercise of citizens’ rights, or where other dynamics of the democratic process are
altered. In recent years they have affected the functioning of foundational democratic processes. As
a matter of fact, the debate exploded in the wake of the 2016 US Presidential elections and the UK’s
Brexit referendum, as well as major privacy scandals such as that involving the company Cambridge
Analytica (Pollicino, Bietti, 2019, 63; Bayer et al., 2019, 15) which, in turn, raise another line of
concerns on citizens’ personal data collection and use. These events triggered fear across the Atlantic
that democracy itself was at risk at the hand of technology and its harmful uses (Benkler et al., 2018,
4). Whereas it is consensual that disinformation, “fake news”, propaganda and political persuasion
do not constitute new phenomena (Martens et al., 2018, 8), today’s novelty resides in the
‘empowering’ character of the Internet in terms of potential scale and reach, velocity and
pervasiveness (Sunstein, 2017).
As is well known, social media have evolved into a major source of information for Internet users.
Evidence shows that social networks’ algorithms have been used to purposefully distribute
manipulated contents with the view of polarizing and influencing public perception (Flore et al.,
1
See also European Commission, A multi-dimensional approach to disinformation, Report of the independent High-level
Group on fake news and online disinformation, 2018.
2
European Commission, cit., A multi-dimensional approach to disinformation, 10.
3
Ibid.
4
Ibid.
Electronic copy available at: https://ssrn.com/abstract=3552680
2019, 29). Whilst trust in mainstream media decreased, alternative news ecosystems have risen to
spread “extremist, sensationalist and conspirational” content (Del Vicario et al., 2016). Indeed, the
exploitative distribution of polarizing political content has been facilitated by highly personalized
algorithms able to, based on preferences and behavior, select visible information, create “filter
bubble” effects characterized by the imperceptibility of their inherent biases (Pariser, 2011). This, in
turn, allows for greater and faster dissemination of content (Flore et al., 2019, 29) and group
polarization leading to higher degrees of societal fragmentation (Sunstein, 2017, 76-77).
1.3 Disinformation in electoral processes and specificities of referenda
Undeniably, technology has improved the opportunities to search, receive and impart
information and ideas in the political debate and the Internet has allowed politicians, political
organizations and parties to communicate with citizens and exchange information in an
unprecedented way. Nevertheless, serious concerns are cast as to the effects that such information
flow generates on public debates and society at large (Tucker et al. 2017). Institutions like the
European Commission have recognized in an official capacity that disinformation is capable of
undermining trust in institutions and in the media, both traditional and digital outlets, and may harm
democracy by affecting citizens’ ability to take informed decisions.5
In fact, according to the HLEG
report, “special attention should be paid to the threat represented by disinformation aimed at
undermining the integrity of elections”,6
as accurate information is necessary to ensure full enjoyment
of the right to vote. The element of information in connection with the public sphere illustrates a
number of relations that are typically recognized and protected in liberal constitutional orders. The
interplay between freedom of expression, the right to receive information and informed participation
in a democracy is an example of the interdependency and ‘triangular’ nature of the relationship
involving democracy, the rule of law and fundamental rights (Carrera et al., 2013). As notably held,
the freedom of speech paradigm is wider than the notion of democracy solely understood in the frame
of electoral and public deliberation processes, as it broadly embraces the notion of “democratic
culture” (Balkin, 2004, 21-22). Nevertheless, the role of information and the ability to freely express
and receive it remains undeniably essential in that timeframe when citizens are called to express their
political views through the election of representatives. Factual information and knowledge are
necessary to ensure the genuine ability of a citizen to take informed decisions, as well as to participate
5
Communication from the Commission, Tackling online disinformation: a European Approach, COM/2018/236 final, 1,
(“Disinformation Communication”).
6
European Commission, cit., A multi-dimensional approach to disinformation, 12.
Electronic copy available at: https://ssrn.com/abstract=3552680
in public debates (Bayer et al., 2019, 61). Accordingly, freedom of speech is not an end in itself but
is instrumental to the self-government and determination of a people (Meiklejohn, 1961).
As free participation in the public sphere during electoral periods presupposes awareness on public
policies, their effects and alternatives,7
the lack of freedom of information, or being subject to
information that is systematically deceitful, may cause a distortion of the opinion-forming process,
which may ultimately reflect in the election ballot. Along these lines, if disinformation practices were
to attain a degree of intensity as to alter or misrepresent public discourse, the exercise of citizens’
voting rights would be at risk in that it could, ultimately, lead to electoral results distorted by a
perverted public discourse (Bayer et al., 2019, 77-78). Analogically, this reflection may be translated
into a particular kind of voting exercise: the referendum. Broadly speaking, the referendum endows
the electorate with the right to express its opinion directly on the constitutional or legislative questions
of a State, typically with the view of confirming, amending or abrogating laws adopted through
parliamentary processes.8
As is well known, referendum consultations are often used as a tool to call
for the electorate’s opinion on contentious, divisive issues, or as a mandate to back political choices
which affect the polity. Along these lines, this instrument can only closely and reliably reflect the
will and preferences of a citizenry at a given time where there is, on the one hand, wide participation
in terms of turnover, and on the other hand, where this position is genuinely expressed on the basis
of an informed choice. It follows that factual information on referendum questions and related public
debates are necessary to ensure that the vote authentically expresses the electorate’s stand. For the
purposes of this contribution, the heart of the problem lies in investigating the latter element, which
translated into the digital sphere amounts to understanding whether and how disinformation practices
may affect the informative process prior to electoral consultations.
2. Tackling disinformation practices v. upholding freedom of expression
2.1 Theoretical justifications
This section will briefly present the theoretical frame of free speech and its development
across the Atlantic, through the constitutional paradigms of the US and European traditions.
The possibility of regulatory intervention to tackle disinformation practices and other forms of illegal
speech display the risk of affecting fundamental rights in a way that may not be compatible with the
liberal constitutional framework. Disinformation is not a new, “digital age” phenomenon. Its
occurrence in today’s online environment, however, highlights particular features linked to the
7
Confirmed in a number of decisions of the ECtHR, see LCB v United Kingdom (1998), paras 38-9 and Roche v. United
Kingdom (2006), paras. 165-6.
8
A more detailed discussion will follow on the Italian legal framework.
Electronic copy available at: https://ssrn.com/abstract=3552680
borderless, pervasiveness and velocity of its dissemination. Indeed, attempts to regulate such
phenomenon raise concerns relating to the freedoms of the parties involved and break the traditional
vertical State-individual relationship. A clear example would be intervention to regulate platforms’
online content, opposing free speech and privacy of individuals and platform owners, and the latter’s
freedom to conduct business. Regulatory measures would likely intervene to defend certain freedoms
at the cost of sacrificing other equally protected ones. Silencing certain forms of speech causes
evident tension, among others, with freedom of expression, at the heart of this inquiry.
Freedom of expression and opinion, along with the right to information, are recognized as a
component of citizens’ political rights,9
the lack of which is typically associated with authoritarian
and illiberal states. Those rights are considered worthy of protection in that they are necessary to
safeguard citizens’ ability to participate in democratic decision-making processes (Bayer et al., 2019,
76). As is well known, different theoretical arguments have emphasized the justifications on which
the freedom may rely. Already in the XVII century, Milton argued that freedom of expression should
suffer no limitations, as a free and open confrontation of ideas would eventually lead to the emergence
of the truth (Milton, 1644). Moreover, Mill (1863, 50-58) understood the flourishing of speech as a
community’s necessary pursuit of the truth, an exercise admitting falsehood as a means to achieve
this goal. Barendt’s (2005, 19-20) instrumental theory places the focus on citizens’ ability to engage
in public debates and participate in a polity; on the other hand, Dworkin’s constitutive argument
frames the freedom as enabling “responsible moral agents of a just political society” (Dworkin, 1999,
200). In light of these brief remarks, the following sections will, firstly, present the constitutional
traditions on freedom of expression and their application to the Internet from the US and European
perspectives and secondly, frame the European dimension of the freedom in the electoral context.
2.2 The US perspective
Freedom of speech is a fundamental tenet of the United States’ constitutional tradition. Its
protection, grounded in the First Amendment to the US Constitution, provides a peremptory
prohibition for Congress to interfere with free speech, enjoying an unparalleled scope of protection
(Bollinger, 1988; Sedler, 2007; Pollicino, 2017a, 24) and a pivotal role in connection with democratic
processes (Sunstein, 2017, 204). In time, this broad scope has been construed by the US Supreme
Court relying on the paradigm formulated by Justice Holmes in his Dissenting Opinion in Abrams v
United States of 1919,10
known as the origin of the celebrated metaphor of the “marketplace of ideas”.
According to this rationale:
9
Articles 19(1) and (2) of the International Covenant on Civil and Political Rights of 1966.
10
US Supreme Court, Abrams v United States, 250 U.S. 616 (1919).
Electronic copy available at: https://ssrn.com/abstract=3552680
“[T]he ultimate good desired is better reached by free trade in ideas – that the best test of truth
is the power of the thought to get itself accepted in the competition of the market, and that truth
is the only ground upon which their wishes safely can be carried out […] I think that we should
be eternally vigilant against attempts to check the expression of opinions that we loathe and
believe to be fraught with death, unless they so imminently threatened immediate interference
with the lawful and pressing purposes of the law that an immediate check is required to save the
country.”11
This evocative model has acted time and again as a shield against State intrusion on a range of forms
of speech. US constitutionalism, nevertheless, does not deny per se that certain exercises of free
speech may be subject to legislative limitations, i.e., in those cases where speech determines a “clear
and present danger”12
assessed on the basis of a strict scrutiny (Pollicino, 2017a, 23-27).
Justice Holmes’ rationale was first analogically applied to the Internet context in Reno v ACLU,13
where the Supreme Court found that the contested provisions, aiming to restrict obscene online
speech, were unconstitutional as they exercised a censoring force spilling over to other forms of nonobscene
speech (Pollicino, Bietti, 2019, 53). The view expressed by the Supreme Court is that:
“The record demonstrates that the growth of the Internet has been and continues to be
phenomenal. […] we presume that governmental regulation of the content of speech is more likely
to interfere with the free exchange of ideas than to encourage it. The interest in encouraging
freedom of expression in a democratic society outweighs any theoretical but unproven benefit of
censorship.”14
This landmark judgment effectively elected the Internet as the “new” marketplace of ideas and
awarded online speech the strongest level of protection, setting accordingly the highest parameter for
judicial scrutiny on possible content-based Internet restrictions that may be established in the future;
which may be upheld to the extent that a compelling state interest is promoted by the contested
regulation and only where this special interest successfully fulfills the proportionality test (Fraleigh,
Tuman, 2011, 309).15
More recently, the Supreme Court has recognized the importance of
cyberspace, and particularly of social media, as a key forum for exchanging views,16
limitations to
11
Ibid., para 630.
12
US Supreme Court, Schenck v United States, 249 U.S. 47 (1919). Nuanced to an “imminent lawless action” test in the
case of political speech in Brandenburg v Ohio, 395 U.S. 444 (1969).
13
US Supreme Court, Reno v ACLU, 521 U.S. 844 (1997).
14
Ibid., 885.
15
This strict test was subsequently endorsed by lower US Courts in ACLU v Johnson, 194 F.3d 1149 (10th
Cir. 1999),
para 1156; Mainstream Loudoun v Board of Trustees of the Loudoun County Library, 24 F. Supp. 552 (E.D. Va, 1998),
para 563; Cyberspace Comm. Inc. v Engler, 142 F. Supp. 2d 827 (E.D. Mich., 2001), para 830.
16
US Supreme Court, Packingham v North Carolina, 582 U.S. (2017).
Electronic copy available at: https://ssrn.com/abstract=3552680
which may only be imposed – even for the most serious crimes – where the provision is “narrowly
tailored to serve a significant governmental interest”.17
In less suspicious times, Balkin (2004, 4-6) highlighted how the Internet was bringing a
‘democratizing’ effect on speech; nowadays, at a more advanced stage of development of the social
web, a new pluralist model of speech emerges, where the threat of speech limitation and censorship
do not derive solely, nor prevalently, from the traditional State powers but rather, from a range of
private actors (Balkin, 2017). Fundamental changes in online speech regulation created a new
governance model characterized by collateral censorship, public/private cooperation or cooptation
and private governance (Balkin, 2014; Id., 2017, 25, 29).
Despite negative consequences following from State intervention to contrast online speech,
government regulation in a free society should not be opposed as such, nor should the approaches to
regulation be confronted as an aut-aut (Sunstein, 2017, 189-190). Whereas US constitutional tradition
recognizing that “there is no such thing as a false idea”18
and that First Amendment law “requires
that we protect some falsehood in order to protect speech that matters”,19
a more cautious handling
of the “new” marketplace of ideas metaphor was already suggested (Pollicino, 2017b). Not only has
the factual context in which the metaphor was devised fundamentally changed, but also doubts on the
self-correcting ability of the marketplace may be raised (ibid.). Accordingly, the US free speech
paradigm may be in need of a reformed construction, mindful of other democratic values such as
equality and pluralism, whilst providing protection vis-à-vis private parties exerting uncontrolled
power. As traditionally applying in the bilateral, vertical State-individual relationship, the doctrine
could be tailored to intercept speech expressed within a “highly privatized digital public sphere”
(Pollicino, Bietti, 2019, 55) as to allow tackling disinformation perpetrated by a mix of public and
private agents. Indeed, Balkin (2017, 66-67) suggested online platforms change their self-conception,
calling for a “new set of social responsibilities” to defend democratic values in the digital sphere.
Specifically concerning disinformation, the question is whether or not patently false forms of speech
are worthy of protection in a democratic society. As illustrated by Pollicino and Bietti (2019, 51),
false forms of speech are often awarded protection with the view of safeguarding other equally worthy
values, including diversity of opinion, with a broad scope for public debates and freedom of the press.
Along these lines, false speech displays “instrumental rather than intrinsic value” and, to the extent
that it serves to promote values such as plurality, its worthiness of constitutional protection may be
17
Ibid., 1736.
18
US Supreme Court, Gertz v Welch, 418 U.S. 323 (1974).
19
Ibid., para 341.
Electronic copy available at: https://ssrn.com/abstract=3552680
justified (ibid.). However, in light of “overgenerous constitutional protection”, an overprotective
approach could be questioned, in favor of a more careful consideration of counter claims (ibid., 52).
2.3 The European perspective
The suitability of the US free speech paradigm to the European context is not evident, as it
finds a narrower scope of protection and different focus (Pollicino, 2017b; De Gregorio, 2018, 4-5).
Unlike notably put by Meikeljohn (1967) for the United States, in Europe, freedom of expression is
not “an absolute”. US First Amendment law refers to the active dimension of the freedom, i.e. a right
to express oneself, whereas the European Convention of Human Rights (“ECHR”), in its article 10
highlights the freedom’s passive dimension,20
incorporating a right to be “pluralistically informed”
(Pollicino, 2017b). At the EU level, the provision is reflected in the EU Charter of Fundamental
Rights (“EU Charter”),21
where the fundamental tenets of media freedom and pluralism are explicitly
enshrined in the freedom’s design.22
The EU Court of Justice has developed its caselaw on freedom
of expression consistently with the construction of its constitutional framework of fundamental rights
(Pollicino, Bassini, 2014, 14), broadly fashioned along the ECHR paradigm through the interpretative
parameter of article 52(3) of the Charter.
The European approach does not automatically assume that all forms of speech enjoy the same degree
of immunity from State intervention by relying openly on the self-correcting force of the information
market. European courts have placed at the forefront of their analysis values such as human dignity
and pluralism (Pollicino, Bietti, 2019, 55). In the nature of obiter dictum, the European Court
(“ECtHR”) had maintained that:
“Freedom of expression […] 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’.”23
Broadly speaking, article 10 ECHR caselaw has enabled the construction of a common standard of
protection within the European national legal orders and has acted as a binding interpretative
parameter vis-à-vis EU law.24
Whereas article 10(1) of the ECHR frames the fundamental character
20
ECtHR, Bladet Tromsø and Stensaas v. Norway (1999), para 62.
21
Article 11 EU Charter.
22
Article 11(2) EU Charter.
23
ECtHR, Handyside v. United Kingdom (1976), in obiter dictum.
24
Resulting from article 6(3) of the Treaty on the European Union and article 52(3) EU Charter.
Electronic copy available at: https://ssrn.com/abstract=3552680
of the right, the second paragraph clearly asserts its relative nature, susceptible of meeting certain
limitations as may be “necessary in a democratic society” – provided the requirements set by the
Convention are met. As noted, the highlight on the right’s passive dimension, encompassing a right
to receive sources in a pluralistic context, would suggest that, translated into the online setting, false,
misleading or deceitful information do not enjoy unfettered protection under the chapeau of article
10 ECHR (Pollicino, Bietti, 57). A further containment is enabled by the so-called “abuse clause”
provided by article 17 ECHR, which protects Convention rights from depletion through abusive
exercise of other equally protected rights.25
On this basis, legislation to contrast hate speech has been
adopted (ibid., 58), and may serve as an enabler if article 10 were to be abused through other forms
of unprotected speech.
The balancing act between harm and freedom of the press is no less complex when relating to online
speech to define the contours of protection of false news. The ECtHR’s approach seems to be that of
protecting fundamental rights from the perils of the Internet, rather than treating it as an empowering
tool. Despite a strict interpretation of article 10(2), cases brought before the European Court have
emphasized the possibility of limiting freedom of expression. The Court appears to rely on the
presumption that the Internet is fundamentally different from the traditional media environment, with
the consequence that the rules governing the latter are not automatically fit for the former, and in need
of a new balance (Pollicino, Bietti, 59-60). In fact, concerning freedom of the press and the Internet,
the Strasbourg Court expressed the view that:
“The risk of harm posed by content and communications on the Internet to the exercise and
enjoyment of human rights and freedoms […] is certainly higher than that posed by the press.
Therefore, the policies governing reproduction of material from the printed media and the
Internet may differ. The latter undeniably have to be adjusted according to technology’s specific
features in order to secure the protection and promotion of the rights and freedoms concerned”.26
In this case, the balancing test was tipped in favor of allowing greater limitation under article 10
ECHR. The Court has further maintained the relativity of guarantees underpinning Convention rights,
in the sense that they may give way to ensure that other legitimate imperatives are preserved, further
highlighting that it would be a task for the legislator to “provide the framework for reconciling the
various claims which compete for protection in this context”.27
25
For instance, disputing the existence of “clearly established historical facts” constitutes an abuse of freedom of
expression, breaching article 10 ECHR, see Lehideux and Isorni v. France (1998). The practice may therefore be
legitimately restricted in accordance with the Convention, provided the applicable conditions are met.
26
ECtHR, Editorial Board of Pravoye Delo and Shtekel v. Ukraine (2011), para 63.
27
ECtHR, K.U. v. Finland (2008), para 49. In this case, namely, the prevention of disorder or crime and competing thirdparty
rights were at stake.
Electronic copy available at: https://ssrn.com/abstract=3552680
The means through which content is distributed rise to a crucial consideration within the
proportionality test performed by the Court (ibid., 62-63). Presumably, this balancing act continues
to prove consequential with the view of assessing competing interests in the management of online
content and assessment of prevailing rights in each particular case. Accordingly, issues relating to the
role of Internet intermediaries and their liability (or lack thereof) remain inextricably linked to
possible interventions covering content regulation.
2.4 Freedom of expression in the electoral context
Freedom of expression and opinion is particularly salient in the electoral phase and involves
another Convention-protected right, i.e. the right to free elections.28
In fact, the Strasbourg Court
notably held:
Free elections and freedom of expression, particularly freedom of political debate, together form
the bedrock of any democratic system […] The two rights are inter-related and operate to
reinforce each other: […] freedom of expression is one of the “conditions” necessary to “ensure
the free expression of the opinion of the people in the choice of the legislature” […] For this
reason, it is particularly important in the period preceding an election that opinions and
information of all kinds are permitted to circulate freely.29
Although freedom of expression is not absolute – one clear limitation is that imposed to defamatory
allegations and speech violating personality rights – the reference to “all kinds of information”
suggests a broad scope of protection from what would otherwise constitute undue censorship.30
Nevertheless, the ECtHR also recognized that in case of conflict between the two rights, Contracting
States have a margin of appreciation in balancing the opportunity of a restriction on speech during
the electoral period if necessary to protect the “free expression of the opinion of the people in the
choice of the legislature”.31
The work of the Council of Europe’s platform on media freedom32
commits to promoting and
facilitating pluralistic expression of opinions through regulatory frameworks, providing an obligation
to cover electoral campaigns in a “fair, balanced and impartial manner in the overall programme
services of broadcasters”,33
for both public service media and private broadcasters. As illustrated by
28
Article 3 of Protocol no. 1 to the ECHR.
29
ECtHR, Bowman v. The United Kingdom (1998), para 42.
30
ECtHR, Bladet Tromsø and Stensaas v. Norway (1999), paras 66, 73.
31
Ibid., para 43.
32
See https://www.coe.int/en/web/media-freedom.
33
Council of Europe, Freedom of expression and elections, July 2018, 1, https://rm.coe.int/factsheet-on-media-and-
elections-july2018-pdf/16808c5ee0.
Electronic copy available at: https://ssrn.com/abstract=3552680
the Strasbourg Court’s caselaw involving freedom of expression, the media and the right to free
elections, the interplay between these rights may touch upon different angles. Inter alia, cases have
related to electoral law restrictions on print media during electoral campaigns,34
injunctions against
print media for criticism in the electoral period,35
reporting on candidates’ private life,36
convictions
for disseminating false information in the period preceding elections,37
as well as different issues
concerning broadcasting. With specific reference to online media outlets, the European Court found
a violation of article 10 ECHR, in the case of an editor of an Internet media website which had been
fined for publishing allegations of child abuse against an Icelandic candidate, amounting, according
to the Icelandic Supreme Court, to defamatory conduct.38
Interestingly, the candidate had acted in
proceedings for defamation only against the editor of the media website and not against his accusers,
who had already previously made available the same statements on which the contested article was
based. Safe the candidate’s possibility to act against his accusers, the Court took the view that the
decision against the editor, albeit only civil in nature and regardless of the fine’s amount, that:“[a]ny
undue restriction on freedom of expression effectively entails a risk of obstructing or paralysing
future media coverage of similar questions”.39
This approach shows the importance of the media’s
role in public discourse and opinion formation for the electorate voting their representatives, i.e. their
freedom to inform citizens can only be restricted in exceptional circumstances.
3. Legislators’ responses
In the wake of increasing concerns of the effects of online disinformation on democratic
discourse, legislators have attempted to tackle the phenomenon. This section will address the efforts
currently underway within the EU and subsequently, Italy’s unsuccessful attempt to legislate.
3.1 EU level
The European Union was first confronted with the growing concern of disinformation in
March 2015, when the European Council invited the development of an action plan by the High
Representative to address Russian ongoing disinformation campaigns,40
leading to the establishment
34
ECtHR, Orlovskaya Iskra v. Russia (2017).
35
ECtHR, Cumhuriyet Vakfı and Others v. Turkey (2013).
36
ECtHR, Saaristo and Others v. Finland (2010).
37
ECtHR, Salov v. Ukraine (2005).
38
ECtHR, Ólafsson v. Iceland (2017).
39
Ibid., para 61.
40
European Council conclusions, 19-20 March 2015.
Electronic copy available at: https://ssrn.com/abstract=3552680
of the East Strategic Communication Task Force of the European External Action Service.
Responding to the European Parliament’s mandate, 41
inviting the Commission “to verify the
possibility of legislative intervention to limit the dissemination and spreading of fake content”;42
a
range of exploratory actions were undertaken and finally resulted in the Communication on
disinformation of April 2018. Based on the specifically European constitutional commitment to the
protection of the freedom, to receive and impart information, enshrined in both the ECHR and the
EU Charter, the HLEG set out the groundwork for possible action. This includes general and specific
objectives to tackle forms of speech which are not illegal in themselves, yet, still harmful for society
and citizens. The HLEG presumed that all responses should avoid any interference with the freedoms
involved.43
Indeed, according to the “multidimensional approach” required by the multi-faceted
nature of disinformation practices,44
a number of lines of action, combining both short-term and longterm
solutions was set forth in the following form:
(1) a legal and regulatory effort to enhance the transparency of online news, including on data
practices; (2) an educational effort to promote digital media literacy; (3) a technical effort to
develop tools that empower readers and journalists and allow them to engage in positive public
discourse; (4) a cultural effort to preserve and enhance the diversity and sustainability of the
European news media ecosystem, and finally (5) an effort keep promoting research on and
monitoring of disinformation in Europe (Pollicino, Bietti, 2019, 82-83).
Along those lines, the “European approach” illustrated by the Commission relied on four main
principles: (i) improvement of transparency as to how information is produced or sponsored;
promotion of (ii) diversity of information and (iii) credibility of information; (iv) fostering inclusive
solutions with broad stakeholder involvement.
Among parallel measures, in October 2018, major online platforms signed a Code of Practice on
Disinformation,45
voluntarily committing to a set of standards to fight disinformation practices
ranging from efforts to enhance transparency in political advertising, to those concerning the take
down of fake accounts. Whereas the Commission had initially expressed a positive view in its follow-
41
European Parliament resolution of 15 June 2017 on online platforms and the digital single market (2016/2276(INI)).
42
Ibid., pt. 36.
43
European Commission, A multi-dimensional approach to disinformation, cit., 19.
44
Ibid., 12.
45
See https://ec.europa.eu/digital-single-market/en/news/code-practice-disinformation. The Code is accompanied by an
annex setting best practices.
Electronic copy available at: https://ssrn.com/abstract=3552680
up report,46
the results following the baseline report on the implementation of the Code submitted by
signatories to the Code were unsatisfactory.47
The years 2018 and 2019 saw the fight against disinformation practices and hybrid threats at the top
of the EU agenda,48
as the then upcoming European elections raised concerns to ensure the integrity
of the EU Parliament’s IX legislature’s formation.49
In that period, improvements in actions to
contrast inauthentic behavior and limit the scope of spam and disinformation were reported.50
Before the end of 2019, an assessment on the first year of implementation of the Code is expected by
the Commission, with the caveat that: “[s]hould the results of this assessment not be satisfactory, the
Commission may propose further initiatives, including of a regulatory nature”.51
As illustrated
above, the adequacy of regulatory intervention at the EU level is still being evaluated.
Against this backdrop, some remarks can be made. Firstly, while the EU legislator’s cautious and
incremental approach is generally to be welcomed, yet, the appropriateness of intervention may be
questioned altogether. On the one hand, while a number of soft law instruments are being adopted in
the field, hard regulation would require the identification of a legal basis in the EU Treaties, which is
not evident in itself. Although the internal market basis52
has served to construct the digital single
market and was used as the foundation of data protection legislation – at the time devised as an
instrument for the internal market’s completion –53
it is hard to imagine how it could justify, alone, a
measure whose connection to the internal market appears loose. Even if the internal market basis was
proposed and not contested by the Member States on a matter of competence, it could still face
46
Report from the Commission on the implementation of the Communication "Tackling online disinformation: a
European Approach", COM(2018) 794 final, 2.
47
See https://ec.europa.eu/digital-single-market/en/news/first-results-eu-code-practice-against-disinformation.
48
See Press Release of 5 December 2018, http://europa.eu/rapid/press-release_IP-18-6647_en.htm.
49
An “Election Package” was adopted and includes: Commission Communication on securing free and fair European
elections, COM (2018), 637; Commission Recommendation on election cooperation networks, online transparency,
protection against cybersecurity incidents and fighting disinformation campaigns in the context of elections to the
European Parliament, COM (2018) 5949; Commission Guidance on the application of Union data protection law in the
electoral context, COM (2018) 638; Regulation (EU, Euratom) No 2019/493 of the European Parliament and of the
Council of 25 March 2019 amending Regulation (EU, Euratom) No 1141/2014 as regards a verification procedure related
to infringements of rules on the protection of personal data in the context of elections to the European Parliament, OJ L
85I , 27.3.2019.
50
Joint Communication to the European Parliament, the European Council, the Council, the European Economic and
Social Committee and the Committee of the Regions Report on the implementation of the Action Plan Against
Disinformation, JOIN(2019) 12 final, 4.
51
Ibid., 5.
52
Article 114 TFEU.
53
Directive 95/46/EC on the protection of individuals with regard to the processing of personal data and on the free
movement of such data, was founded on the equivalent of today’s article 114 TFEU and despite the Lisbon reform
establishing a novel basis to protect the fundamental right – article 16 TFEU – it remains broadly fashioned along the
mixed internal market / human rights rationale underlying EU data protection law.
Electronic copy available at: https://ssrn.com/abstract=3552680
judicial scrutiny as regards to its compliance with fundamental rights.54
Secondly, the appropriateness
of some of these soft law instruments could be questioned. Whereas co- and self-regulatory
instruments certainly display benefits in terms of efficiency and practical expertise,55
and may well
serve as trial-and-error groundwork in a context that is still yet to be entirely understood, serious
doubts may be cast on their use where fundamental rights and major political choices are at stake.56
Well-known problems also concern the effectiveness of their implementation. In the particular case
of the Code on disinformation, the EU institutions’ efforts to monitor, evaluate and complement is
clear. Nevertheless, effective control and implementation remains with the signatories, leaving no
redress mechanisms nor possibility of judicial scrutiny. The report on the Code’s first year of
implementation expected by the end of 2019 may perhaps show whether the threat of regulatory
intervention has pushed for strong implementation.
Finally, the fact that different national legislations are emerging in Member States, such France57
and
Germany,58
cannot be overlooked from a coherence perspective of possible EU-wide action.
3.2 Italy’s (failed) attempt to legislate “fake news”
In February 2017, a draft law was filed at the Italian Senate titled “Provisions to prevent the
manipulation of online information, to ensure transparency on the web and to encourage media
literacy” (“DDL Gambaro”).59
The report to the draft law opens with an address referring to the
cornerstones of any democratic system, including: “the freedom and credibility of information, which
in turn represent the essence of journalism, whose first duty is to the truth”.60
Admittedly, the
initiative stems from the concerns expressed by the Council of Europe’s assembly resolution on
online media and journalism61
regarding online campaigns devised to “misguide sectors of the public
54
A parallel may be drawn with the Data Retention Directive 2006/24/EC saga. In Case C-301/06 Ireland v European
Parliament and Council of the European Union [2009] I-00593, paras 28-33, 58, Ireland questioned the suitability of
article 114 TFEU as a legal basis for the Directive vis-à-vis the mixed nature of surveillance-related requirements and
economic activity. Despite the Court of Justice ruling the legitimacy of the legal basis, subsequent litigation, as is well
known, struck down the Directive due to its serious interference with fundamental rights in Case C-293/12 Digital Rights
Ireland Ltd, [2014] ECLI:EU:C:2014:238.
55
See https://ec.europa.eu/digital-single-market/en/cop-principles-better-self-and-co-regulation.
56
White Paper on European Governance, COM/2001/0428 final, 12.10.2001; implemented by European Parliament,
Council, Commission “Interinstitutional Agreement on better law-making”, 16.12.2003 (cfr. para. 18), no longer in force
and substituted by a new “Interinstitutional Agreement on Better Law-Making”, 13.04.2016, dropping the reference to
co- and self-regulation.
57
Loi n° 2018-1202, 22 décembre 2018 relative à la lutte contre la manipulation de l'information.
58
Netzdurchsetzunggesetz, Law of 30 June 2017, Federal Law Gazette I, p. 3352 ff (“NetzDG”).
59
DDL no. 2688 – XVII Legislature. All translations from Italian are by the author.
60
See http://www.senato.it/japp/bgt/showdoc/17/DDLPRES/0/1006504/index.html?part=ddlpres_ddlpres1.
61
Parliamentary Assembly, “Online media and journalism: challenges and accountability”, Resolution 2143 (2017).
Electronic copy available at: https://ssrn.com/abstract=3552680
through intentionally biased or false information, hate campaigns against individuals and also
personal attacks, often in a political context, with the objective of harming democratic political
processes”.62
DDL Gambaro introduced amendments to the Italian criminal code concerning on the
one hand, the “publication or dissemination of false, exaggerated or biased news likely to disrupt
public order, through computer platforms”,63
and on the other, the “dissemination of false news that
can raise public alarm, mislead sectors of public opinion or concerning hate campaigns and
campaigns aimed at undermining the democratic process”.64
Subsequently, a short set of provisions
deal with communication and rectification requirements,65
as well as intermediary liability in case of
publication or dissemination of “fake news”.66
The debate around DDL Gambaro was united in
opposing it, identifying violations of the principles of determination, the exhaustiveness and
offensiveness of the types of offence, likewise a patent violation of freedom of expression (Lehner,
2019, 100-101). Furthermore, as the monitoring requirements were entrusted to the platform provider,
the possible consequences of private censoring activities were heavily criticized (Magnani, 2019, 4).
As remarked by the opening of the report to the draft law, a “duty of truth” appears to underlie the
ratio legis, preoccupying in that it appears to assume falsehood as an illegal element in itself (Bassini,
Vigevani, 2017, 15) and not in relation to its aptness to violate constitutionally protected values
(Allegri, 2018, 201). The draft law did not proceed in the parliamentary procedure. Later that year,
in December 2017, another draft law,67
promoted by different first signatories, was announced but
never lodged, titled “General rules on social networks and to combat the dissemination of illegal
content and fake news on the Internet” (“DDL Zanda and Filippin”). The new draft, largely inspired
by the German NetzDG,68
targeted specifically social networks – rather than “computer platforms” –
and relies on existing provisions of the Italian criminal code. The draft provides a seven day-period
for removal of “non manifestly illegal” content imposed upon social networks, avoidable in two
limited circumstances: firstly, where the decision on the illegal nature of the content depends on false
allegation or examination of other circumstances; secondly, where, before the seven-day term, the
social network delegates the decision to a self-regulatory authority established in accordance with the
law (Lehner, 2019, 99).69
As Monti (2017) observed, the rationale of the second initiative frames the
62
Ibid., pt. 6.
63
Article 1 DDL no. 2688.
64
Ibid., Article 2.
65
Ibid., Articles 3-4.
66
Ibid., Article 7.
67
DDL no. 3001 – XVII Legislature.
68
Supra fn. 58.
69
This provision reproduces verbatim article 1(3)(2) of NetzDG.
Electronic copy available at: https://ssrn.com/abstract=3552680
problem of interference with public opinion solely from the political dimension, not considering the
distortion of free information and the press, essential for democratic accountability of political power.
Accordingly, Monti (ibid.) concludes the ineffectiveness of a solution disregarding its context, nor
considering social networks’ role in the information chain.
The assessment of whether and how similar interventions are admissible in Italy must be searched in
relation to constitutional principles and the interpretation that freedom of expression in the Italian
constitutional order. Distinguished commentators (Esposito, 1958; Pace, Manetti, 2006) originally
excluded that consciously spread falsehood deserved protection. This allowed legislative intervention
to protect public faith from the spread of the patently false, insofar as the objective and subjective
elements of expression and inner thought could be separated (Esposito, 1958, 36-37). According to
Esposito (1958, 48) freedom of expression cannot jeopardize constitutional principles, nor social
peace. Yet, forms of speech including “propaganda, apology, public exaltation and manifestation
instigating the realization of the thought expressed” are legitimately protected (ibid., 49).
Pace (2006, 88-96) considered that false information may be worthy of protection to the extent that
it originated from bona fide conduct. As illustrated by Barile (1975), freedom of expression
encounters the sole limitation of illegal “purposeful deception”. Therefore, spreading falsehood
cannot be deemed illegal in and of itself (ibid., 229). As Allegri points out (2018, 191) the question
whether to regulate “fake news” with the view of suppressing their dissemination, assuming its
constitutional compatibility, depends on preferring a “functional approach” relying on the passive
dimension of the freedom. Conversely, an “individualistic approach” would protect false news, as
contributor of public opinion (Allegri, 2018, 192). Assuming the harmfulness of the “fake news”
phenomenon, rather than regulatory intervention, a constitutionally sounder solution would focus on
increasing information production and dissemination, as well as promoting tools for critical
information consumption (Bassini, Vigevani, 2017, 18-19). This second approach reflects more
closely the known marketplace of ideas rationale, as able to autonomously select worthy information.
Given the way disinformation practices spread, fragment and polarize audiences, emphasis should
therefore be placed on enhancing transparency and pluralistic values in the online media system,
rather than attempting to suppress it altogether (Allegri, 2018, 202).
In any case, any legislation imposing duties on Internet intermediaries in terms of monitoring,
filtering and content removal would have to comply with the legal standards excluding, as a rule,
general monitoring and liability of neutral intermediaries with regards to user-generated content.70
70
Articles 12-15 of 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, OJ L 178, 17.7.2000.
Electronic copy available at: https://ssrn.com/abstract=3552680
4. The Italian Referendum case
This section will present the disinformation debate surrounding the Italian constitutional
referendum of 2016. Against brief remarks on the referendum, this section focuses on the approaches
and responsive actions taken by the Italian authorities.
4.1 The Italian framework on referenda
As an instrument of direct democracy, the referendum consists of a consultation of the
electorate producing legal effects. The Italian Constitution provides for different types of referenda.71
According to article 138, laws promoting revision of the constitutional text and other constitutional
laws must be adopted by each Chamber of Parliament with two successive resolutions, occurring at
least three months from one another. In the second voting session, the reforming laws may be
approved by absolute majority of the members of each Chamber.72
Where the law is approved by a
qualified majority below two-thirds of the members of each Chamber, a referendum procedure may
be initiated within three months of the law’s publication, upon request of either: (i) one-fifth of the
members of a Chamber; (ii) five hundred thousand voters; (iii) five regional councils. Once subject
to consultation, the law may only be promulgated if approved by a majority of valid votes.73
Whereas
if the law is passed with a two-thirds majority, resort to a referendum is excluded.74
One distinguishing procedural element is the lack of a participation quorum for the validity of the
referendum. 75
On the one hand, the rationale was devised to allow, upon minority action, the
electorate to express its view on a revision likely promoted by a governmental majority. On the other,
the lack of a minimum quorum may lead to the slimmest part of the electorate blocking reform, for
instance, in case of general disinterest by the majority (Bin, Pitruzzella, 2017, 362-365).
A participatory quorum for the referendum’s validity would emphasize the protection of
constitutional rigidity, inasmuch as a proportional or majority representation could in principle adopt
constitutional laws that do not correspond to the majorities expressed by the electorate (Canepa, 2001,
303). Nevertheless, Canepa (2001, 306) illustrates how prevailing scholarship, relying on the
constituent debate, highlighted the opposition character of this institution, safeguarding political
71
See articles 138, 75, 132 and 133 Constitution.
72
Article 138(1).
73
Article 138(2).
74
Article 138(3).
75
For instance, provided for the abrogative referendum of article 75 Constitution.
Electronic copy available at: https://ssrn.com/abstract=3552680
minorities and the constitutional text, in consideration of the optional nature of the procedure and the
entitlement of a parliamentary minority to launch the referendum initiative.76
It must be noted that the instrument has seen rare application, as the 2016 consultation was the third
in Republican history.77
The text of the reform, approved in second vote by absolute majority below
two thirds, provided to overcome perfect bicameralism, reduce the number of MPs, contain the
operating costs of the institutions, abolish the National Council of Economy and Labour and revise
Title V, Part II of the Constitution.78
During the consultation, the electorate rejected the reform, with
59.11% of valid votes against and 40.89% in favor, upon an overall turnover of 65.47%.79
4.2 Disinformation and interference with the 2016 referendum
Allegations that undue interferences had influenced the political debate and results of the
referendum emerged. Notably, Biden and Carpenter80
held that following the US presidential
elections, other “steps to sway political campaigns”81
were taken to influence consultations in
Europe, including the Italian referendum. According to the US State Department, efforts to influence
elections and referenda in Europe include “overt and covert support for far left and right political
parties, funding front groups and NGOs, and making small, low-profile investments in key economic
sectors to build political influence over time,”82
and that its tactics “focus on exploiting internal
discord in an effort to break centrist consensus on the importance of core institutions.”83
Concerning Italy specifically, Russian electoral interference is reported to have occurred through
disinformation and “fake news” campaigns promoting anti-establishment and extremist political
parties,84
widely shared on social media, with alarm for the then upcoming national elections of
March 2018.85
Manipulation of social media, in the Italian case, has been described as an “ecosystem
76
In contrast, the abrogative referendum of article 75 Constitution does not provide entitlement to launch the initiative
upon MPs: this power lies with the electorate – upon request of 500.000 rightholders – or five regional Councils.
77
See http://elezionistorico.interno.it/index.php?tpel=F.
78
GU Serie Generale n.88, 15.04.2016.
79
See
https://elezionistorico.interno.gov.it/index.php?tpel=F&dtel=04/12/2016&tpa=Y&tpe=A&lev0=0&levsut0=0&es0=N&
ms=S.
80
J. R. Biden, M. Carpenter, “How to Stand Up to the Kremlin; Defending Democracy Against Its Enemies” Foreign
Affairs, January 2, 2018.
81
Ibid.
82
Committee on Foreign Relations of the USA, Putin's Asymmetric Assault on Democracy in Russia and Europe:
Implications for US National Security, 10 January 2018, 38.
83
Ibid.
84
Ibid., 137-139.
85
A vast number of reported disinformation cases in that period is available at “EUvsDisinfo”, see
https://euvsdisinfo.eu/disinformation-cases/?text=italy&disinfo_issue=&date=.
Electronic copy available at: https://ssrn.com/abstract=3552680
coordinating different types of initiatives mostly affiliated with populist forces” (Bradshaw, Howard,
2018, 35). Moreover, false news is shared by leading politicians, shifting to platforms the burden to
moderate content (Bayer et al., 2019, 45). Analysis has shown how foreign media have played a
polarizing role in debates on salient issues by “reinforcing local media narratives and supporting
communities with abnormal levels of engagement, creating an imbalance in narratives being
published and therefore consumed by the public”86
and confirming that historical data preceding 2017
points to specific narratives’ construction and systematic dissemination long before the electoral
process.87
In fact, it was found that half of the most shared stories on social media covering the
constitutional referendum were fabricated.88
In 2016, the Italian independent administrative authority responsible for the communications sector
(“AGCOM”) adopted safeguards including: (i) dispositions implementing the framework on
“political communication and equal access to the media relating to the campaign recommendation”89
to give effect to the principles of pluralism, impartiality, independence, objectivity and completeness
of the radio and television system, and rights granted to political subjects;90
and (ii) recommendations
issued for the respect of pluralism in view of the constitutional referendum.91
These instruments,
directed at public and private broadcasters and print media outlets, promote traditional measures in
the political campaigning context. Following the referendum, in the wake of allegations and evidence
of disinformation practices’ interference, the AGCOM rapidly responded to evolve its policies.
Accordingly, a technical working group on the safeguard of “pluralism and fairness of information
on digital platforms”92
was established with the objective of promoting self-regulation of online
platforms and the exchange of good practices for the identification and combating of online
disinformation resulting from targeted strategies, an example of institutionalized cooperation between
independent media, information market regulators and platforms. In its first technical report, the
working group analyzed online disinformation strategies and the fake content chain.93
What can be
noted is that rather than relying on “truthfulness” of the information, the report focuses on information
distortion and public opinion formation, based on analysis of the underlying socio-economic
86
Alto Data Analytics, The construction of anti-immigration electoral messages in Italy, 2018.
87
Ibid.
88
I. Kottasova, “Did Fake News Influence Italy’s Referendum?” CNNMoney, December 5, 2016.
https://money.cnn.com/2016/12/05/media/fake-news-italy-referendum/index.html.
89
AGCOM, Delibera N. 448/16/CONS, 4 October 2016.
90
Ibid., see article 1.
91
AGCOM, Delibera N. 327/16/CONS, 7 July 2016.
92
AGCOM, Delibera N. 423/17/CONS, 6 November 2017.
93
Rapporto tecnico - Le strategie di disinformazione online e la filiera dei contenuti fake, 9 November 2018.
Electronic copy available at: https://ssrn.com/abstract=3552680
characteristics of supply and demand for online information (Magnani, 2019, 23). The report
identifies a particular disinformation strategy created by stable organizations, or temporary
organizations with specific interests, driven by precise economic and/or political-ideological
objectives, supported by financial, technological and organizational endowments and with identified
targets.94
These strategies include actions constituting fully-fledged campaigns, as a series of
publications and/or republications of fake contents, not single sporadic episodes.95
Alongside, a
number of studies have been investigating disinformation,96
online news consumption,97
their
application to the electoral context,98
as well as monthly reviews on online disinformation patterns.99
As noted by Magnani (2019, 22), the AGCOM’s active role is conceived in close relation with market
operators, pushing its specific competences to “a role of impulse and coordination between the
different actors operating in the field of online information to promote self-regulation on a voluntary
basis”.100
A clear distinction from the failed legislative attempts emerges from these brief remarks,
departing reliance on any conception of “truth” as the object of protection, pursuing instead the
protection of the fundamental tenet of pluralism.
4.3 The shift towards an integrated approach in electoral monitoring
Furthermore, prior to the national elections held on March 4th
2018, a more tailored approach
to disinformation can be noticed on part of the supervisory authority, as specific “Guidelines on for
equal access to online platforms during the 2018 general election campaign”101
were issued to adapt
the general principles to all media,102
including digital platforms. The Guidelines provide “prevention
strategies, detection methods, tools for blocking or removing online content that can be qualified as
damaging to the correctness, impartiality and pluralism of information”.103
This exemplifies a choice
to unify under the AGCOM’s umbrella different facets of mass communications and technology,
striving for a comprehensive understanding and increasingly integrated approach (Magnani, 2019,
94
Ibid., 51.
95
Ibid.
96
AGCOM, News vs. Fake nel Sistema dell’Informazione, Interim Report, Indagine Conoscitiva, Delibera N.
309/16/Cons, November 2018.
97
AGCOM, Rapporto sul consumo di informazione, February 2018.
98
AGCOM, Il consumo di informazione e la comunicazione politica in campagna elettorale.
99
See https://www.agcom.it/documenti-tppo.
100
AGCOM, Delibera n. 423/17/CONS, 3.
101
Delibera n. 423/17/CONS.
102
Legge 22 febbraio 2000, n. 28 - Disposizioni per la parita' di accesso ai mezzi di informazione durante le campagne
elettorali e referendarie e per la comunicazione politica, GU n.43, 22-2-2000.
103
Allegato A alla delibera n. 423/17/CONS.
Electronic copy available at: https://ssrn.com/abstract=3552680
21). As illustrated above, a shift in the authority’s approach is noticeable since the experience of the
2016 referendum. Despite concerns on disinformation being already salient, the Italian approach was
still fairly cautious. After 2016, action sought to frame the phenomenon, with an eye specifically on
preparing for the then upcoming national and European elections. This has been, so far, mostly of an
aware, cautious self-regulatory nature, relying on concerted actions involving stakeholders. In parallel
to the EU approach, it promotes, similarly, multi-faceted responses. With a forward-looking view,
the tools include, as envisaged at EU level and generally supported in scholarship, the promotion of
media literacy, fact-checking, measures supporting quality journalism. In the longer term, these
solutions constitute the least invasive, relying on citizens’ resilience to informative overload and
malicious practices.
5. Conclusions
This contribution explored from a constitutional and comparative perspective the question of
disinformation practices’ capability to affect democratic processes, specifically, with regards to the
electoral context, focusing on the possible tensions to the exercise of citizens’ rights. Secondly,
through an analysis of the US and European constitutional traditions, doubt is cast on the
appropriateness of applying to the Internet an uncontrolled paradigm of “new marketplace of ideas”.
Accordingly, the need for a construction mindful of other democratic values and the media’s role in
public opinion, whilst providing protection vis-à-vis private parties is suggested. Thirdly, this
contribution explored the legislative responses underway at the EU level and in Italy, based on a
cautious approach reliant on stakeholder cooperation. The effectiveness of the Code of Practice on
disinformation signed by major online platforms and the very choice to rely on soft law are
questioned. Moreover, the Italian attempts to regulate “fake news” have been presented and criticized
in their framework. It is maintained that interventions should emphasize the principles of transparency
and pluralism in the online media system, rather than attempt to suppress disinformation altogether.
Fourthly, focusing on the case of online disinformation practices during the 2016 Italian constitutional
referendum, a shift in approaches was presented, as a number of initiatives to counter online
disinformation have been launched and the AGCOM promoted increasingly integrated action. Indeed,
going forward, an approach not invasive of citizens’ fundamental rights constitutes the sounder
solution, as emphasis should be placed on tools promoting informative resilience.
Electronic copy available at: https://ssrn.com/abstract=3552680
Bibliography
Allegri, M. R., 2018. Ubi Social, Ibi Ius. Fondamenti costituzionali dei social network e profili
giuridici della responsabilità dei provider, Franco Angeli, Milano.
Balkin, J. M., Commentary: Digital Speech and Democratic Culture: A Theory of Freedom of
Expression for the Information Society, 2004, 79(1) NYUL Rev.
Balkin, J. M., Old School/New School Speech Regulation, 2014, 127 Harvard Law Review.
Balkin, J. M., Free speech in the algorithmic society: big data, private governance, and new school
speech regulation, 2017, 51 UCDL Rev.
Barendt, E., 2005. Freedom of Speech, Oxford.
Barile, P. 1975. Libertà di manifestazione del pensiero, Milano, Giuffrè.
Bassini, M., Vigevani, G. E., Primi appunti su fake news e dintorni, 2017(1) Medialaws – Rivista di
diritto dei media.
Bayer et al, Disinformation and propaganda – impact on the functioning of the rule of law in the EU
and its Member States, 2019, Study for the LIBE Committee, European Parliament.
Benkler, Y., Faris, R., Roberts, H., 2018. Network Propaganda, Oxford.
Bollinger, L., 1988. The Tolerant Society: Freedom of Speech and Extremist Speech in America,
Oxford.
Bradshaw, S., Howard, P. N., Challenging Truth and Trust: A Global Inventory of Organized Social
Media Manipulation, Computational Propaganda Research Project, Online Supplement to Working
Paper, 2018.1.
Canepa, A., Referendum Costituzionale e Quorum di Partecipazione, 2001, 2, Quaderni
costituzionali.
Carrera, S., Guild, E., Hernanz, N., The Triangular Relationship between Fundamental Rights,
Democracy and the Rule of Law in the EU, Towards an EU Copenhagen Mechanism, CEPS, 2013.
De Gregorio, G., Expressions on Platforms: Freedom of Expression and ISP Liability in the European
Digital Single Market, 2018, 3(2) European Competition and Regulatory Law Review.
De Gregorio, Perotta, E., Tackling disinformation around the world, Public Affairs Media Policy,
WAN-IFRA, 2019.
Del Vicario, M., Bessi, A., Zollo, F., Petroni, F., Scala, A., The Spreading of Misinformation Online,
2016 113(3) PNAS.
Dworkin, R. 1999. Freedom's Law: The Moral Reading of the American Constitution, Oxford.
Esposito, C., 1958, La libertà di manifestazione del pensiero nell’ordinamento italiano, Milano,
Giuffrè.
Flauss, J. F., The European Court of Human Rights and the freedom of expression, 2009, 84(3)
Indiana Law Journal.
Flore M. et al, Understanding Citizens' Vulnerabilities to Disinformation and Data-Driven
Propaganda Case Study: The 2018 Italian General Election, JRC Technical Report, 2019.
Fraleigh, D. M., Tuman, J. S., 2011. Freedom of Expression in the Marketplace of Ideas, 2011, Sage.
Electronic copy available at: https://ssrn.com/abstract=3552680
Lehner, E., Fake news e democrazia, 2019(1) Medialaws – Rivista di diritto dei media.
Magnani, C., Libertà d’informazione online e fake news: vera emergenza? Appunti sul contrasto alla
disinformazione tra legislatori statali e politiche europee, 2019, Forum di Quaderni Costituzionali.
Martens, B., Aguiar L., Gomez-Herrera, E., Mueller-Langer, F., The digital transformation of news
media and the rise of disinformation and fake news, JRC Technical Report, JRC Digital Economy
Working Paper 2018-02.
Meiklejohn, A., The First Amendment is an Absolute, 1961 Sup. Ct. Rev. 245.
Mill, J. S. 1863. On Liberty, Boston.
Milton, J., 1644, Aeropagitica, London.
Monti, M., La proposta del ddl Zanda-Filippin sul contrasto alle fake news sui social network: profili
problematici, in Diritti Comparati, December 7, 2017.
Pace A., Manetti, M. 2006. Art. 21. La libertà di manifestazione del proprio pensiero, in Branca G.
(a cura di), Commentario della Costituzione, Bologna, Zanichelli.
Pollicino, O., 2017a. La prospettiva costituzionale sulla libertà di espressione nell’era di Internet, in
(eds.) Pitruzzella, G., Pollicino, O., Quintarelli, S. 2017. Parole e potere, Libertà d’espressione, hate
speech e fake news, Milano, Egea.
Pollicino, O., 2017b, Editorial, Fake News, Internet and Metaphors (to be handled carefully), 9(1)
Italian Journal of Public Law.
Pollicino, O., Bassini, M., Free Speech, Defamation and the Limits to Freedom of Expression in the
EU: A Comparative Analysis, in (eds.) Savin, A., Trzaskowski, J. 2014. Research Handbook on EU
Internet Law, Edward Elgar.
Pollicino, O., Bietti, E., Truth and Deception Across the Atlantic: a Roadmap of Disinformation in
the US and Europe, 2019, 11(1) Italian Journal of Public Law.
Sedler, R. A., Freedom of Speech: The United States versus the Rest of the World, Wayne State
University Law School Research Paper No. 07-21.
Sunstein, C. R. 2017. #Republic: Divided Democracy in the Age of Social Media, Princeton.
Tucker, J. A. et al., From liberation to turmoil: social media and democracy, 2017, 28(4) Journal of
democracy.
Electronic copy available at: https://ssrn.com/abstract=3552680