Draft: 2010/9/9 (do not quote)

Internet Creativity, Communicative Freedom and a Constitutional Rights Theory

Response to Code is Law
Christoph B. Graber1

Abstract
The code that regulates cyberspace empowers private bodies to set standards of Internet
access and use, which are often not visible. Content filtering, as a response to copyright
infringement, or models differentiating between various data transmissions are examples of
measures that have been undertaken by Internet intermediaries. Arguably, they are necessary
to protect intellectual property and digital business. Emanating from private bodies, such
measures are beyond the reach of constitutional rights, although they may strongly impact
conditions of communicative freedom and creativity on the Internet. This paper endeavours to
explore whether a theory of constitutional rights in the private sphere may have a case in a
digital networked ecology.

1. Introduction
Code is law is the famous formula coined by Lawrence Lessig to describe that the
technological architecture of the Internet functions as a regulator in addition to state law,
social norms and the market.2 Joel Reidenberg was one of the first to emphasise that the
technological architecture of the network imposes rules on access and use of information.

Technological architectures may prohibit certain actions on the network, such as

access without security clearances, or may impose certain flows, such as mandatory
address routing data for electronic messages. Technology may also offer policymakers
a choice of information flow rules through configuration decisions.3

Reidenberg called these rules Lex Informatica. Accordingly, Lex Informatica is a rule system
that is embedded in technological standards and that exists parallel to the law of the state.4

The author would like to thank William Gallagher, Shubha Ghosh, Peter Yu and Thomas Steiner for comments
on an earlier drafts of the paper.
2
Lessig, 1999b; Lessig, 2000.
3
Reidenberg, 1998, 568. See also Reidenberg, 1996.
4
Reidenberg, 1998, 568-9.

2
For many lawyers it is difficult to accept that code fulfils regulatory functions which are
similar to the law of the state. This is due to an understanding of regulation that is based on
John Austins state-centred concept of the law.5 For Austin, all law derives directly or
indirectly from the state.6 According to this concept, the law can be recognised by the way it
is enacted and enforced. Law is enacted by an elected legislative body that observes
prescribed procedures and it is enforced through the threat of state sanctions.7 This theory of
law as commands backed by threats is not of much use on the Internet. In cyberspace, an
Austinian concept of law is too limited and a two-step enlargement would be necessary.

First, it should be expanded from narrow state-centrism towards the inclusion of social norms
and other private forms of regulation. Such a concept of lebendes Recht (living law) was
postulated by Eugen Ehrlich already at the beginning of the 20th century.8 Similarly, Niklas
Luhmanns concept of law is not limited to state law but includes regulations that emanate
from private entities as well.9 On the Internet, living law is omnipresent, for instance in selfregulatory activities of net communities such as the exclusion of misbehaving participants
from a chat forum or a multiplayer online game,10 or where Internet Service Providers (ISPs)
administer blacklists of servers disseminating spam-containing electronic mail.11 More recent
examples that will be further discussed below include Internet intermediaries throttling
peer-to-peer (P2P) traffic in order to reserve bandwidth for the distribution of more lucrative
content12 or agreements between the entertainment industry and Internet intermediaries to
fight copyright piracy by the means of filtering technology.13

Second, a more thorough understanding of how technology regulates behaviour on the

Internet would be necessary. Lessigs formula code is law suggesting that law and code
are identical is more appealing than precise. A close reading of Lessigs writings reveals
however, that he does not equalise law and code as the formula would suggest. Lessig
rather defines code as a form of co-action between software and hardware on the Internet
constituting an architecture of technology.

Austin, 1954 (1832).

Cotterrell, 1992, 26.
7
Austin, 1954 (1832), 13-20.
8
Ehrlich, 2002 (1913), 486-498.
9
Luhmann, 2004, 291-296.
10
See Lastowka and Hunter, 2006; Karavas, 2009a, 194.
11
See Lessig, 2006, 112.
12
Katyal, 2009, 420-421.
13
Infra, note 35, and accompanying text.
6

3
We are coming to understand a newly powerful regulator in cyberspace. () This
regulator is what I call code the instructions embedded in the software or hardware
that makes cyberspace what it is. This code is the built environment of social life in
cyberspace. It is its architecture.14

This architecture of technology is a structure that conditions regulation on the Internet. Hence,
code does not directly regulate the Internet but pre-structures the form that regulation on the
Internet may take to be effective in determining social behaviour. As Vaios Karavas observed,
the main purpose for Lessig in using the formula code is law was to sensitise his readers
towards the fact that in the digital environment nothing is natural, as it appears to be, but
instead, everything is the result of technological solutions, in other words, of the code.15
Code is so powerful because it may fulfil legislative and executive functions at the same time
without being constrained by any type of a digital separation of powers.

With regard to its legislative functions the code of cyberspace is different from the law of real
space in at least one important manner. In real space, the law is a form of communication. It is
deliberated over in a discursive political process and is enacted by the constitutionally
competent legislative body. This is different for the code of cyberspace. Here, the actor who
reigns over the architecture of technology also defines the rights and constraints existing
within this architecture. Since on the Internet this actor is often a private company, this raises
serious concerns from a constitutional perspective.

With regard to its executive functions code is different from law since it is self-executive. In
real space, compliance with (state) law depends on both enforcement by state power and
acceptance by the addressees of the regulation. Acceptance is a precondition for compliance
not only for state law but also for social norms.16 For Lessig, [l]aw and norms are more
efficient the more subjective they are, but they need some minimal subjectivity to be effective
at all.17 In cyberspace, however, the situation is different. Effects of code are not dependent
on enforcement at all, nor is there any subjective constraint. The code of cyberspace even

14

Lessig, 2006, 121 (citation omitted).

Karavas, 2010a, 157.
16
Locus classicus is Eugen Ehrlichs recognition theory of law (Anerkennungstheorie). See Ehrlich, 2002
(1913), 164-167. See also Habermas, 1996, 448.
17
Lessig, 2006, 344.
15

4
works when the person constrained does not know that the constraint exists. This is due to
self-execution, which is a key feature of the technological architecture of the Internet.18

Lessig uses the example of copyright protection in Second Life to better explain how code
executes itself independently of any threats of state sanctions or any subjective constraints.19
In real space, intellectual property (IP) legislation regulates any use a person may make of a
copyright-protected work. For instance, IP legislation will decide whether a third person is
allowed to take a picture of the copyright-protected sculpture in the garden of my house. In
cyberspace, however, the constraint is not in the law, it is in the code. If the sculpture is in the
garden of my Second Life house, then taking a picture will simply not be possible for
anybody who is not in possession of a licence to do so.

Since the technological infrastructure of the Internet can be as easily manipulated by private
persons as by states/governments, this poses the question of whether such private action
should be subject to constitutional scrutiny.20 This paper will focus on threats to free-speech
values arising from code that is controlled by private actors. More specifically, it will look at
technologies that are used by private companies to filter content on the Internet in order to
fight copyright piracy. Often, filter technologies rely on automated methods to detect
copyright infringements and, in some cases, such interference is even not visible to the user.
These are concerns that advocates of communicative freedom on the Internet should take
seriously.

Filtering technologies are used by the entertainment industry in its battle against copyright
infringements. The industry argues that such technologies are necessary to protect creativity
on the Internet. Critics object that filtering endangers free-speech values, including user
creativity. Hence, creativity is used by both sides as a catchword to defend their respective
ideological standpoint. What is hidden behind ideology is a collision between economic and
non-economic values. In this paper, we will argue that such a conflict of values should be
discussed within a constitutional framework. Finally, the paper will explore from a
transnational legal perspective whether a theory of constitutional rights in the private
sphere, as developed in a European context, might have a case in a digital networked
ecology.
18

Lessig, 2006, 342-344.

Lessig, 2006, 110-111.
20
Berman, 2004b, 1271.
19

5
2. Private Regulation of Internet Speech
Recent developments show that threats to free-speech values on the Internet arise more often
from private regulation rather than from state intervention.21 According to Hans-Jrgen
Papier, president of the German Constitutional Court until March 2010 and renowned
guardian of civil liberties, the basic rights of the German Grundgesetz (i.e. the German
Constitution) related to the integrity and confidence in computer systems (including the
Internet) today are not only threatened by the state, but also by private persons and
corporations.22 What holds true for network and PC infiltration applies even more to content
filtering, which is increasingly used by private entities particularly to fight copyright piracy.

2.1 Paradigm Change in Copyright Enforcement

Content filtering executed by private entities is on the rise.23 According to observers, this is a
consequence of a new global trend in copyright enforcement, shifting the lead management in
the combat against piracy from the content owners to Internet intermediaries. The new trend
started in the second half of 2007.

As a broad survey by De Beer and Clemmer shows, many industrial countries worldwide have
been tightening the conditions under which ISPs and other Internet intermediaries may benefit
from copyright safe harbours.24 Judicial authorities, policymakers and government lawmakers
alike have been reconsidering intermediaries responsibilities in copyright enforcement and
now require them to actively fight copyright infringement in order to escape liability. The
consequence is a shift in the model of copyright enforcement. Whereas, under the old model,
intermediaries were only required to react ex post to a right holders notice of copyright
infringements, the new model puts them into an active role of ex ante filtering copyright
infringements. This policy shift from passive-reactive to active-preventive schemes for
communication intermediaries is mainly a result of strong lobbying by the entertainment
industries, which apparently convinced many governments of the importance of creative

21

For examples of nominally private actors wielding regulatory power on the Internet see Bellia, Berman and
Post, 2007, 308-327.
22
See interview, Sddeutsche Zeitung, 6 March 2010 <[Link]
23
Content filtering relies on a technology that is known as Deep Packet Inspection (DPI). DPI allows the
network to make decisions about how to handle a certain type of content in network data streams, including
filtering or slowing down certain packets based on the recognition of a specific pattern of the datagram. For
technical details see Mueller, forthcoming.
24
De Beer and Clemmer, 2009, 376. The survey covers twelve diverse jurisdictions including Australia, Canada,
China, the European Union, France, Germany, Japan, New Zealand, Singapore, South Korea, the United
Kingdom, and the United States.

6
industries to their nations cultural and economic well-being.25 Since the recording
industries global lawsuits have proved ineffective to fight piracy,26 [c]opyright holders and
governments realize that Internet intermediaries are technologically and commercially well
positioned to assist with copyright enforcement.27
The notice and takedown framework28 of the US Digital Millennium Copyright Act
(DMCA) serves well to exemplify the functioning of the old model and its eventual shift from
a low standard of liability for ISPs towards enhanced liability.29 Under the DMCA, as crafted
by Congress, it was the copyrights holders responsibility to search for infringing behaviour.
In order to escape from liability, intermediaries were only obliged to either identify the
subscriber or take down the illicit content after having been noticed by the right holder.30
Hence, under the DMCA, intermediaries were in a passive-reactive role and were not required
to actively detect copyright-infringing behaviour. However, as Sonia Katyal has shown, a
common-law based reinterpretation of ISPs liability by US courts has asked ISPs to
increasingly play a more active role in copyright enforcement.31 Although the DMCA does
not require intermediaries to actively detect copyright infringements, a common law trend
[] tends to indirectly expand the boundaries of secondary liability for ISPs.32 As a result,
this new case law has been prompting Internet intermediaries to actively employ content
filtering methods to affirmatively search for evidence of infringement in order to protect
themselves from secondary liability.

In addition to judicial decisions, three-strike type legislative enactments and industry selfregulation emanating from voluntary collaboration have been major drivers of the paradigm
shift in copyright enforcement. Three-strike laws are being debated almost everywhere in the
developed world.33 The exception is the United States, where Internet intermediaries are

25

De Beer and Clemmer, 2009, 404.

According to Katyal, 2009, 420, the Recording Industry Association of America (RIAA) in the last months of
2008 changed its copyright enforcement strategy from law suits (35000 altogether) to notifying ISPs and calling
them to act in their respect:
27
De Beer and Clemmer, 2009, 405.
28
According to De Beer and Clemmer, 2009, 386, a notice-and-takedown system exists in almost every
jurisdiction where legislation details ISPs obligations when hosting information.
29
Katyal, 2009, 405.
30
De Beer and Clemmer, 2009, 385.
31
Katyal, 2009, 408.
32
Katyal, 2009, 408.
33
The most famous legislative enactment under the new paradigm is the French law for the distribution and
protection of creative content on the Internet that entered into force on 1 January 2010. This law, known also as
three strikes law, graduate response or loi HADOPI, is designed to protect business models of the creative
economy by sanctioning copyright violations with Internet access cut-offs. HADOPI is the French acronym for
26

7
cooperating quietly with the entertainment industry. A telling example to illustrate the new
collaboration between the entertainment industry and Internet intermediaries in the US are the
Principles for User-Generated Content, which were negotiated between major players of the
entertainment industry (including several Hollywood studios and broadcasting networks) and
intermediaries (including Crackle, Dailymotion, Microsoft, MySpace, and Veoh).34 According
to the drafters, the Principles are designed to foster [] creativity while respecting the rights
of copyright owners.35 The Principles require the operators of sites hosting user-created
content (UCC) to utilise filtering and identification technology to block copyright-infringing
uploads before they are made public.36 Intermediaries subscribing to the Principles are exempt
from secondary liability. From a legal perspective, the Principles are non-binding guidelines.
Despite their voluntary character, they are regulations that may strongly interfere with users
communicative freedom on the Internet. They are followed by content owners and
intermediaries alike because they are the expression of the shared antipathy of both parties
toward P2P activities on the Internet. Whereas the content industry perceives file sharing in
social networks as a major source of copyright piracy, many intermediaries feel that UCC
congests the tubes of the net in a way that will ultimately prevent more lucrative economic
development.37 According to De Beer and Clemmer [i]ntermediaries care about network
efficiency; copyright holders care about online infringements. File sharing is perceived as a
threat to both.38 Since certain customers would be willing to pay higher rates for preferential
distribution, commercial reasons are the main incentive for intermediaries to shape traffic on
the Internet.39 Accordingly, filtering P2P traffic and distinguishing between commercial and
non-commercial content is a shared goal of Internet intermediaries and the entertainment
industries.

Self-regulatory activities emanating from collaboration between copyright owners and

Internet intermediaries seem to be very dynamic beyond the Principles. Comcast, for

the Higher Authority for the Distribution of Works and the Protection of Copyright on the Internet that is in
charge for implementing the new law. For a reference to the French three strikes model and HADOPI see
Winn and Jondet, 2009, at 560-1. The first three strikes laws were introduced in spring of 2009 in Taiwan and
South Korea. See also De Beer and Clemmer, 2009, 389-390 and 393, and Katyal, 2009, 409.
34
De Beer and Clemmer, 2009, 399.
35
See CBS Corporation, DailyMotion, Disney et al., Internet and Media Industry Leaders Unveil Principles to
Foster Online Innovation While Protecting Copyrights, Press Release, Oct. 18, 2007, available at
[Link]
36
See Principles for User-Generated Content Services as quoted by De Beer and Clemmer, 2009, 399 and
Katyal, 2009, 421.
37
Katyal, 2009, 420-421.
38
De Beer and Clemmer, 2009, 405-6.
39
De Beer and Clemmer, 2009, 406.

8
example, has been covertly constraining the flow of data from customers using BitTorrent, a
file-sharing service. According to Comcast, this step was necessary because BitTorrent was
used to illegally communicate huge files of data, including pirated films.40 Since this made the
Internet slower, Comcasts other customers had complained to the service provider.
Slowdown practices have been used by other ISPs as a reaction to video portals, such as
Youtube, and social networks, such as Facebook, who swell the amount of data in the net.41

Slowdown practices and other Internet traffic shaping strategies are in violation of the
principle of net neutrality. Net neutrality (sometimes also network neutrality) is a term
originally coined by Tim Wu,42 which is used to defend the openness and freedom of the
Internet architecture. However, the principle of net neutrality stands on shaky legal ground, as
recent case law in the US demonstrates. In 2008, the US Federal Communications
Commission (FCC) required Comcast to respect net neutrality and prohibited the intermediary
from throttling P2P traffic on its network. However, the FCC decision was reversed by the
April 6, 2010 decision by the United States Court of Appeals for the District of Columbia
Circuit.43 The Court of Appeals found that the FCC acted beyond its authority to regulate the
Internet and was not allowed to sanction Comcasts network management practices.

2.2 Consequences of the New Paradigm from a Free Speech Perspective

When people think of dangers to free speech they usually think of government censorship.
What may be right in the offline world is wrong on the Internet. On the Internet, as we have
demonstrated above, threats to communicative freedom often stem from private rather than
state action. Filtering and traffic management practices are normally a consequence of
decisions taken at the level of the architecture of the Internet and this architecture is mostly in
the hands of private actors.

From the perspective of free-speech values, the most important implication of the paradigm
change in copyright enforcement is certainly the shift from ex post to ex ante methods of
surveillance. In the offline world, ex ante surveillance of communication exerted by
governments would be considered as censureship and, thus, as something that rips into the
heart of communicative freedom. On the Internet, however, where ex ante surveillance is

40

Katyal, 2009, 416-7.

For further information on notice and slowdown practices see De Beer and Clemmer, 2009, 391-393.
42
See Wu, 2003.
43
Comcast v. FCC, US Court of Appeals for the District of Columbia Circuit (6.4.2010).
41

9
practised by private parties, policymakers and regulators do not seem to be fully aware of the
problem and there exists very little substantive judicial oversight or intervention.44

Intermediaries surveying the Internet ex ante rely on automated methods to detect copyright
infringements.45 Because it is difficult to assess whether certain content is fair use, such
automated methods are often overbroad.46 Automated control tends to over-block and
conflicts with the principle of proportionality.47 This amounts to a particular problem for
communicative freedom, especially if one considers the impact on UCC of a non-commercial
origin. In cases of appropriation of content, including mash-up or remix creations, this
uncertainty of what is allowed by fair use often risks chilling the transformative work of
artists who seek to incorporate the work of others.48

Under the new paradigm, users will often not know that their movements on an Internet
intermediarys platform are filtered and hence no response against this interference will be
available. This is a critical issue from a transparency perspective. In cases where possibilities
for consumers to respond to an intervention exist, there is a change in the burden of proof.
Whereas under the old paradigm it was the copyright owner who had to prove that a consumer
had violated copyright, ex ante surveillance requires consumers wanting to oppose blocking,
slowing down or deletion to show that certain content posted on the Internet is legal.

44

Katyal, 2009, 416.

Often, automated filtering will take place without the users knowledge. However, as the German
Constitutional Court held in its recent data retention judgment (see infra, n. 123) non-transparent interferences
with data flows may harm users expectations of the networks confidentiality and integrity and leave users
with a diffusely threatening feeling of being observed that may impede an unbiased exercise of fundamental
rights in many areas. (Para 212, German in the original).
46
For a distinction between manual and automated review in its impact on UCC see Katyal, 2009, 422.
47
McIntyre, 2009, 7. The principle of proportionality is well established as a general principle of law in the
European countries, in the European Union and in the framework of the European Convention on Human Rights.
In EU law it is enshrined as a general principle in Article 5 of the Treaty on European Union (Official Journal of
the European Union, C-/83/15, 30.3.2010) and provides that action by the Union shall not go beyond what is
necessary to achieve the objectives of the Treaty. The proportionality principle has been further fleshed out in
the case law of the European Court of Justice (ECJ). Before the ECJ, the proportionality principle allows
applicants to challenge Union action as well as action by EU Member States that falls within the sphere of
application of EU law. It normally requires a Union or government measure to be 1) suitable and 2) necessary to
achieve the desired end and 3) to respect reasonable relationship between the measure taken and the objective
sought to be achieved. See Craig/de Brca, 2008, 544-551, and Ueda, 2003. On the proportionality principle in
the realm of the European Convention on Human Rights see n. 122, below.
48
As Katyal, 2009, 412, notes, an overinclusive approach to piracy surveillance risks not only chilling some
forms of valuable speech, but it also risks having a deleterious effect on the technologies that distribute content
as well, making it even more costly for new technologies to develop unless they devote substantial resources to
the perfection of such strategies.
45

10
Any shift of piracy surveillance from copyright owners to the intermediaries is particularly
dangerous from a free-speech perspective since the interests of the entertainment industry to
prompt filtering, as a means for copyright enforcement, converge with the interests of
intermediaries to discriminate between types of content on their networks. As we have
highlighted above, commercial interests may induce intermediaries to employ even more
restrictive controls of P2P traffic. Accordingly, such an alignment of interests directed against
P2P and UCC may raise deeper structural concerns about net neutrality and creativity on the
Internet.49

3. Creativity, Commercial Interests and Communicative Freedom in a Constitutional

Setting
Creativity has been the magic word in any discussion regarding the protection of intellectual
property rights (IPRs) on the Internet in recent years. Amazingly, creativity has been the
battle cry on the sides of proponents and opponents of a strong institutionalisation of IPRs
alike. On the one side, the entertainment industries successfully lobbied governments to raise
the standard of protection of IPRs at the national50 and international level.51 The argument
was consistently that this is necessary for the sake of creativity and the creative industries on
the Internet.52 Creativity has also been a central term in the entertainment industries rhetoric
in the context of copyright enforcement be it in the Recording Industry Association of
Americas (RIAA) 35,000 law suits53 or in the debate related to the creation of ACTA, the
planned new international Anti-Counterfeiting Trade Agreement.54

On the other side, critics of an overbroad protection of IPRs have been arguing that, in a
digital networked environment, creativity depends on an IP system providing for a fair
balance between the private interests of right holders and the public interest in enjoining
broad access to their productions. According to these views, networked digital technologies
49

Katyal, 2009, 416.

In the United States, Congress lengthened the copyright protection term several times over the last decades.
Section 8 of the US Constitution empowers Congress to promote the progress of science and useful arts, by
securing for limited times to authors and inventors the exclusive right to their respective writings and
discoveries. What started in the 18th century as a renewable 14-year term now lasts for life of the artist plus 70
years. For company copyrights it has even been extended to 95 years. See Towse, 2006, 88.
51
Internationally copyright has been strengthened above all in the framework of the World Intellectual Property
Organization (WIPO) and the Agreement on Trade-related Aspects of Intellectual Property Rights (TRIPS) of
the World Trade Organization (WTO).
52
For references see Towse, 2006, 87.
53
Katyal, 2009, 419-420.
54
See De Beer and Clemmer, 2009, 404.
50

11
have radically changed not only the way in which content is produced, disseminated and
consumed, but have also lead to the development of participatory technologies, including
blogs, social networks, wikis etc. Hence, IPR systems must be flexible enough not to
criminalise or suffocate these new expressions of creativity on the Internet.

3.1 The Participative Web and the Change of Free Speech

The participative web blurred the frontiers between creators and consumers and democratised
the production of art. In the view of Jonathan Zittrain:

Today, thanks to networked information technology and the recursively generative code
produced in large part by amateurs, art can be produced and shared by people other than
professional artists, citizens can engage in far-ranging dialogues with others whom they
would not otherwise encounter, and people can work together from the four corners of the
globe to produce intellectual projects of social and economic significance.55

Jack Balkin argues that the participative web changed the right of free speech. In his view,
the point of free speech is to promote a democratic culture. Democratic culture he
understands as

a culture in which people can participate actively in the creation of cultural meanings that
in turn constitute them. A democratic culture is democratic not in the sense that everyone
gets to vote on what is in culture. It is democratic in the sense that everyone gets to
participate in the production of culture. People are free to express their individuality
through creativity and through participation in the forms of meaning-making that, in turn,
constitute them and other people in society.56

The value of free speech must be interpreted in a new light because of technological
developments in the digital ecosystem. Although the social institutionalisation of this value is
similar in most Western societies, there exist differences with regard to its legal
institutionalisation as a constitutional right. Dependent on the respective legal order, this right
has different names, including e.g. free speech (US context) or freedom of expression and
information (European and international human rights law). For the purpose of the
transnational research approach underlying this paper, we will use the term communicative
55
56

Zittrain, 2006, 2028.

Balkin, 2004a, 8.

12
freedom as a general term capturing the central features of the right at issue across the
mentioned legal orders.

Whereas, in the age of broadcasting, communicative freedom was interpreted by courts as

something passive, the participative forms of meaning-making on the Internet would arguably
require communicative freedom to be interpreted as an active concept. As an active concept,
communicative freedom would acknowledge UCC being a typical new phenomenon of
creativity on the Internet. Digital technologies on the Internet offer consumers the possibility
of appropriating content produced by mass media, to mash up and remix pre-existing works
and to create new forms of content as collages.57 Although there is no generally accepted
definition of UCC,58 the central feature of the concept for our context is that amateur users
express their creativity by adding something new to pre-existing works, without necessarily
respecting the narrow fair-use limits of copyright law. The possibility of UCC, thus, requires
redefining the constitutional relationship between the values of communicative freedom and
intellectual property.

As Balkins concept of democratic culture suggests, there is a need for the right to free
speech on the Internet, not only to protect political communication, but also non-political
communication, including popular culture. Judicial authorities have traditionally preferred
political speech to cultural speech. This is true not only for US First Amendment doctrine59
but also for European categorisations in the interpretation of the freedom of expression and
information. The reasons for these value judgements date back to the emancipation of the
bourgeois society in the 18th century.60 In the age of the participatory web, however, political
speech should no longer be preferred to cultural speech of netizens. The Internet is a public
space61 that must not be propertised62 by the entertainment industries to control by the

57

Balkin, 2004b, 9.
The OECDs definition of UCC provides three criteria: (i) content made publicly available over the Internet,
(ii) which reflects a certain amount of creative effort; and (iii) which is created outside of professional
routines and practices (OECD, Participative Web: User-created Content, DSTI/ICCP/IE(2006)7/FINAL, 12
April 2007, 9). This definition is also used by the European Commission, see e.g. Green Paper of the
Commission of the European Union, Copyright in the Knowledge Economy, COM(2008) 466/3, 19.
59
Balkin, 2004b,
60
Discuss Habermas, 2007.
61
According to Balkin, 2004b, 50: Communications networks are public in nature even if their technological
infrastructure is privately owned. Therefore they must grant fair access to their networks.
62
Propertisation has been used in recent socio-legal literature as a term to describe and criticise processes
limiting access to commons, public goods and public domains by the means of property-like legal tools in the
areas of economics, technology, culture and communication. For an overview see the various contributions in
Siegrist, 2007.
58

13
means of IPRs how ordinary people use digital content.63 Balkin is right that constitutional
rights must adapt to this value change and establish a new balance between creativity, as a
free-speech value, and the exclusive commercial interests of digital business and the
entertainment industries. Beyond the borders of the nation state, arguments for a reconsideration of the relationship between IP values and free speech are supported by the new
UNESCO Convention on the Protection and Promotion of the Diversity of Cultural
Expressions (CCD), which stands for a stronger institutionalisation of cultural values at the
international level.64 Although the CCD suffers from considerable normative weaknesses,65 its
almost unanimous adoption and the rapid and broad ratification66 is a political signal from a
large group of governments worldwide, that economic values should no longer be preferred to
values of creative expression.67 Since the CCD affects not only the external relations of a
State Party but also its internal ones, it needs to undertake measures contributing to the
attainment of the goal of protecting and promoting cultural diversity in order to fulfil its
obligations under the Convention.

3.2 Communicative Freedom and the Network Grid

It is an important step for constitutional theory to postulate that the technological changes of
the participative web bring particular features of communicative freedom to the forefront,
requiring a re-evaluation of the relationship between political and cultural speech within a
constitutional setting. This, however, is not sufficient. As we have argued above, on the
Internet, implications of code must be considered as potential threats to free-speech values in
addition to implications of an overbroad IP system. Accordingly, considerations over
communicative freedom on the Internet must also encompass the specific technological
implications of networked digital communication. As the discussion on network neutrality
shows, the openness and freedom of the Internet directly depend on decisions taken at the
level of the technological architecture.68 Technology is used to differentiate between various

63

Balkin, 2004b, 21, 26.

Although the CCD does not address the relationship between IP and values of cultural diversity explicitly,
Article 7(1) CCD obliges parties to create [] an environment which encourages individuals and social groups:
(a) to create, produce, disseminate, distribute and have access to their own cultural expressions [and] (b) to have
access to diverse cultural expressions from within their territory as well as from other countries of the world.
On UCC from a cultural diversity perspective see Burri-Nenova, 2010.
65
For an analysis see Graber, 2008.
66
The CCD, which entered into force on 18 March 2007, has been ratified as of 1 August 2010 by more than 110
countries (<[Link]
67
Graber, 2010.
68
According to Balkin, 2004b, 50, communication networks must grant fair access to their networks, they must
not act as chokepoints or bottlenecks, and they must not unfairly discriminate against content from other
sources.
64

14
data transmissions and to make the quality and speed of the transmitted information
contingent on financial rates of return or other unilaterally set criteria. Network neutrality has
been used during recent years as a slogan to combat any such models and to convince
regulators that the network should stay neutral to the passing content.69 Whereas the legal
basis of the concept has been contested by a district court in the US,70 the European Union has
included network neutrality as a policy objective and regulatory principle in Directive
2009/140/EC,71 amending the European Unions package of communication law, including
the framework, access and authorisation directives. In an annex, Directive 2009/140/EC
provides for a Commission declaration on net neutrality that reads as follows:

The Commission attaches high importance to preserving the open and neutral character
of the Internet, taking full account of the will of the co-legislators now to enshrine net
neutrality as a policy objective and regulatory principle to be promoted by national
regulatory authorities, alongside the strengthening of related transparency requirements
and the creation of safeguard powers for national regulatory authorities to prevent the
degradation of services and the hindering or slowing down of traffic over public networks.
The Commission will monitor closely the implementation of these provisions in the
Member States, introducing a particular focus on how the net freedoms of European
citizens are being safeguarded in its annual Progress Report to the European Parliament
and the Council. In the meantime, the Commission will monitor the impact of market and
technological developments on net freedoms reporting to the European Parliament and
Council before the end of 2010 on whether additional guidance is required, and will
invoke its existing competition law powers to deal with any anti-competitive practices that
may emerge.

Since Directives must be implemented by EU Member States at the level of domestic

legislation, the principle is not self-executing. The Commission made it clear, however, that it
will closely monitor how Member States implement this principle. In addition, the
Commission has announced its willingness to use its competition law powers to ensure
adherence to the principle in competitive relationships between market players.

69

See Crawford, 2007, at 51; Wu, 2003, at 141.

See above, footnote 43, and accompanying text.
71
See recitals 34-38 and 40 (recital 37 specifically addresses the relationship between net neutrality and cultural
diversity!). Article 8 (4) (g) framework directive (as amended) provides: promoting the ability of end-users to
access and distribute information or run applications and services of their choice. See also Articles 20(1)(b),
21(3)(c) and (d) and 22(3) of the Universal Service Directive.
70

15

Zittrain has criticised the concept of network neutrality as still not being sufficiently
comprehensive.72 He argues that the discussion on freedom of the Internet must not be limited
to constraints of network connectivity and protocols, but must include any information
technology systems involved between PC and PC.73 The PC and the Internet are so
inextricably intertwined that it is not enough for network engineers to worry only about
network openness and assume that the endpoints can take care of themselves.74 Hence, in
addition to the network itself, it is important to look also at the endpoints.75 In his view, endto-end neutrality does not fully capture all aspects that safeguarding Internet freedom would
require to be taken into account. Consequently, the distinction between middle and endpoints,
which underlies the rhetoric of end-to-end neutrality, should be replaced with the concept of
Internet generativity:

It remains correct that from a network standpoint, protocol designs and the ISPs that
implement them are the middle of the network, as distinct from PC endpoints. But
the true import of a vernacular of middle and endpoint for policy purposes relates
to individuals power to control their experiences on the network. Now that the
networks endpoints are controllable by faraway entities, abandoning the end-to-end
debates simplistic divide between middle and endpoint will enable us to identify and
respond better to the emerging threats to the Internets generativity.76

The term Internet generativity stands for the potential of individual Internet users to
creatively tinker with the technology of both their PC and the network. Internet generativity
can be endangered by so-called trusted systems, that is, by decisions taken at the level of
code.77 Hence, in the networked digital environment, social spheres of action are determined
by technology. As Vaios Karavas argues, technological media pre-structure the social sphere
of action, thus transforming the whole social sphere into a hybrid entity of technosocial
character.78 He argues that on the Internet basic constitutional rights can no longer be
conceived as existing only in a sphere of normative communication independent from
technology. Arguably, the laws dependency on technological media paves the way for the
72

Zittrain, 2006, 2029.

Zittrain, 2006, 1978.
74
Zittrain, 2006, 2031.
75
Zittrain, 2008, 8.
76
Zittrain, 2006, 2031.
77
Zittrain, 2008, 105.
78
Karavas, 2009b, 471; see also Karavas, 2007, 155.
73

16
emergence of what Karavas calls technodigital normativity, that is the amalgamation of
normative and digital expectations inside the digital medium.79 For Karavas, the specific
function of basic rights on the Internet consists of safeguarding the integrity of digital
systems. The German Constitutional Court has come to a similar conclusion in a recent
leading case. In the judgement, resulting in the annulment of legislation regulating online
investigations, the Court developed unwritten constitutional guarantees protecting the
integrity and confidentiality of information technology systems.80 Inspired by Bruno Latour,
Karavas has lately argued that protecting the integrity of information technology systems
would also encompass an associative link between creative user activities (so-called Web
2.0 activities) and the information technology system.81

To conclude, communicative freedom must also protect the integrity of the technological
architecture. The technodigital normativity of the Internet requires conceiving the
possibility to creatively communicate on the Internet as being inextricably intertwined with
the technology of the network grid.

4. Constitutional Rights in the Private Sphere of the Internet

So far we have discussed how communicative freedom has changed in order to fully protect
new aspects that come to the surface of the constitutional right. What remains is the problem
that many of the threats to free-speech values that we emphasised originate in private action,
whereas constitutional rights can only be applied in cases of state action. It is a common
feature of classic constitutional doctrine, both in the US and in European legal orders, that
constitutional rights are perceived as applying exclusively in relationships between a private
person and the state. In Europe, the classic formulation of constitutional rights views these
rights as individual defences against the intrusive power of the state. Although the German
Constitutional Court has been developing, step by step, an institutional formulation of
constitutional rights, since the end of World War II, there has been a strong countermovement
defending the classic doctrine.82 In the United States, the possibility of a constitutionally
based oversight in the private sphere is limited by the so-called state action doctrine.
Although the state action doctrine has shifted over time since its first formulation in an 1883
79

Karavas, 2009b, 478.

Bundesverfassungsgericht, 1 BvR 370/07, 1 BvR 595/07, 27 February 2008, at paras 201, 203, 207 (German
in the original).
81
Karavas, 2010b (German in the original). See also Steiner, 2010, 69.
82
Graber and Teubner, 1998, 63; Ladeur, 2009, 543-558.
80

17
Supreme Court decision, it still rests on an interpretation of the US Constitution to proscribe
only the conduct of government agencies in most of its commandments.83 In Shelley v.
Kraemer, the Supreme Court held that the Constitution erects no shield against merely
private conduct, however discriminatory or wrongful.84 In cases where a clear demarcation
between private and public conduct is difficult, courts have often looked at either whether
the private party is performing a public function or whether the action of a private party is so
closely linked to the government that it is attributable to the state for constitutional purposes.85

Although scholarship in the US has been busy attacking the public/private distinction
underlying the state action doctrine for decades,86 it has been unable to dismantle it. Rather
than adding to these criticisms, Paul Schiff Berman proposed an alternative approach that he
calls constitutive constitutionalism. According to his view, the debate should focus on the
Constitutions constitutive role in our cultural life, regardless of whether that life is lived in
the public or private sphere.87 When adjudicating conflicts between private parties, courts
should transform into a forum for debating core societal values.88 In cases where such values
are endangered by private action, courts would be required to conduct a broader discussion
over these values. We agree with Berman, that requiring an element of state action misses the
point of why constitutional rights should play a role in the private sphere. Rather than asking
whether a state act was involved directly or indirectly, the relevant criterion should be
effective infringement of individual or social autonomy whatever its source.89 Berman uses
examples of private powers employing code to regulate behaviour on the Internet to support
his argument.90 We agree that these examples demonstrate convincingly that subjecting
private standard-setting powers to constitutional review is necessary for securing individual
and social autonomy on the Internet. However, we join Karavas in his critique, that Bermans
approach is too vague and that it would make people believe that constitutional law and
constitutional courts can resolve any societal conflict.91 Berman concedes that courts are not
the only possible forum for articulating fundamental societal values.92 Nonetheless, he is

83

Bellia, Berman and Post, 2007, 328-9.

Shelley v. Kraemer, 334 U.S. 1, 13 (1948).
85
Bellia, Berman and Post, 2007, 329.
86
For a general critique of the public/private distinction see Chinkin, 1999, 387-; Kennedy, 1982, 1349-.
87
Berman, 2000, 1289.
88
Berman, 2000, 1296-8.
89
Graber and Teubner, 1998, 70. For a more comprehensive elaboration of this argument see chapter 5.1 below.
90
Berman, 2000, 1307.
91
Karavas, 2007, 64.
92
Berman, 2000, 1268 ( discussing drawbacks of an extension of the scope of state action as a strategy to subject
more private activities under constitutional scrutiny).
84

18
convinced that they may sometimes provide a more satisfying resolution than the political
process can.93

From a transnational law perspective it is interesting to note that in Germany a direct effect of
basic rights of the Grundgesetz has been rejected mainly because it does not fit in with the
dominant doctrines view of basic rights as tools designed to fight asymmetries of power
(which, arguably, only exist in relationships between individual and state). 94 Similarly, a
direct horizontal effect of basic rights of the constitution has also been continually and
unanimously rejected by courts and academia in Switzerland. The main concern here was that
a direct effect would lead to a shift of power from the political authorities to the judiciary.
However, as a result of a general constitutional reform in 1999, Article 35(3) of the Swiss
Constitution requires authorities to ensure that fundamental rights, where appropriate, apply
to relationships among private persons.95 This provision has consistently been interpreted as
establishing an indirect horizontal effect of basic rights of the Constitution.96

As Graber and Teubner have argued, the major shortcoming in the European debate on basic
constitutional rights in the private sphere is the narrow view that these rights are constituted
exclusively in a triad of individualpowerState. This is due to an unholy alliance between
liberal political philosophy and the procedures of public law.97 Liberal political philosophy is
based on a concept of the state as a (social) contract between citizens. As a central element of
this contract, citizens confer on the state the monopoly in the exercise of power. At the same
time, they reserve basic individual rights defining a quasi-spatial exclusion zone against state
intervention. The relationship of state and society is understood as a dichotomy. Cooriginality of private and public autonomy is the term Jrgen Habermas coined to describe the
sociological observation that basic rights and the state have the same normative origin.98 The
state-centrism of this concept was bolstered up by the procedures of public law, placing

93

Berman, 2000, 1298.

According to Karavas, 2007, 64, Bermans approach closely resembles the doctrine of the German
Constitutional Court stating that the basic rights of the constitution include an objective dimension and must be
observed throughout the legal order. Karavas has systematically reconstructed the debate in Germany and
Switzerland on constitutional rights in the private sphere (also called horizontal effect of constitutional rights)
and compared it to the US debate on the state action doctrine.
95
Federal Constitution of the Swiss Confederation of 18 April 1999 (Status as of 27 September 2009), available
at <[Link]
96
Schweizer, 2008, notes 35-37 on Article 35 of the Swiss Constitution.
97
Graber and Teubner, 1998, 63.
98
Habermas, 1996, 104. This formula leaves no space for forms of legal autonomy that are neither exclusively
public nor exclusively private. For a critique see Calliess, 1999, 186-7.
94

19
particular procedures at the individuals disposal, which are necessary for the enforcement of
his/her basic rights against the state.

The view that basic rights of the constitution are directed against the state was challenged by
Graber and Teubner using a socio-legal approach based on Niklas Luhmanns system theory.
If one analyses the coming into being of basic rights from a perspective of sociological
jurisprudence, one first has to note that basic rights are a typical phenomenon of the modern,
i.e. functionally differentiated, society. In pre-modern segmented or stratified societies, the
idea of an autonomous individual did not exist.99 Until the end of the 18th century, the formula
for social relations was status and not (individual rights based) contract.100 Consequently,
rather than a constitution vesting basic rights in a person, it was the clan, the family or the
social stratum constituting rights, obligations and opportunities of its members. In the course
of the process of social differentiation eventually leading to modernity, distinct spheres of
social autonomy emerged. The autonomy of the individual was only one among many newly
differentiated social spheres, including the functionally differentiated systems of law, politics,
economics, science, art etc. In Luhmanns theory, a system can be understood as a distinct
discourse where a specific language game (i.e. law, politics, economics etc.) is practised
according to autonomously generated grammatical rules. To protect the autonomy of these
spheres of action, modern society invented basic rights as social institutions. Basic rights are,
thus, first and foremost social institutions and must be distinguished from basic rights as legal
institutions. As social institutions, the function of basic rights is to protect functional
differentiation against self-destroying tendencies of modern society.101 Hence, it is not only
the autonomous sphere of the individual, but rather the autonomous spheres of all functionally
differentiated systems of modern society that must be protected by basic rights. This
distinction between basic rights as social institutions and basic rights as legal institutions
represents the main difference between a Luhmanian institutional and a classic concept of
basic rights. Although it is the social institution of constitutional rights that protects social
autonomy, it is the role of the law to stabilise these rights and to shape them to a certain
degree.102

99

Luhmann, 1965, 33-37; Luhmann, 2004, 412-419.

According to Sir Henry Maines famous thesis, expressed in 1861, the history of progressive societies had
hitherto been one of a movement from social relations based on status to relations of contract. Quoted from
Cotterrell, 1992, 119.
101
Graber and Teubner, 1998, 65.
102
Graber and Teubner, 1998, 72.
100

20
The focus on both individual and trans-individual (discursive) situations is the first of two
important implications of an institutional understanding of basic rights. The second
implication is that basic rights are not only directed against the state, since freedom in modern
society can be endangered by any expansionist tendencies of a social system. The fact,
thereby, that in the 19th and in the first half of the 20th century, expansionist tendencies in
Western societies mainly originated in activities from the political system (i.e. the state) is a
mere historic coincident. Experiences from the late 20th and the beginning of the 21st century
show that colonising tendencies more often emanate from the economy and from science.

5. Implementing Communicative Freedom on the Internet

5.1 Theoretical Reflection
We have so far been arguing that, due to a curtailed concept of basic rights, classic
constitutional doctrine is unable to properly respond to threats to communicative freedom
emanating from networked digital technology in the hands of private actors. Our analysis
revealed two reasons for this limited perspective: First, blindness of interdependencies
between speech, creativity and technology; and, second, blindness with regard to the function
of basic rights as guarantees of communicative chances.103 As a response, we first suggested
that communicative freedoms of the constitution be interpreted as extending to creative Web
2.0 activities, while taking account of their specific technosocial character. Second, we
suggested overcoming the state-centrism of the classic constitutional rights doctrine by
introducing a distinction between basic rights as social institutions and basic rights as
institutions of the law. An institutional approach to basic rights allows protecting transindividual discursive values of communicative freedom, even where it is not moulded into the
legal form of a right.

The question now is how these ideas could be brought into effect. We suggest a procedural
implementation strategy consisting in a proceduralisation of basic rights a logical
consequence of an institutional approach. As Graber and Teubner have suggested, a
proceduralisation of basic rights would result in introducing legal procedures effectively
securing a diversity of autonomous discursive spheres.104 This argument is based on
Luhmanns above outlined theoretical assumption that basic rights as institutions are

103
104

Luhmann, 1965, 23.

Graber and Teubner, 1998, 72-73.

21
guarantees for the autonomy of the various subsystems of society.105 Whereas from an
analytical perspective, basic rights and society exist and develop in a relationship of mutual
observation and co-evolution,106 from a normative perspective, basic rights protect societal
differentiation.107 Referring to our discussion in chapter 3.1 above, this is an argument
supporting Balkins postulate for a better constitutional balance between commercial values
and free-speech values and against a property-tied interpretation of communicative freedom
on the Internet.108

Sure enough, a proceduralisation of basic rights on the Internet goes beyond a rebalancing of
key social values in a constitutional order. If basic rights have an institutional effect then the
law must provide remedies to protect the autonomy of the various individual and social
spheres involved. In reality, autonomous spheres of action are often neither exclusively
private nor public. Boundaries between private and public autonomies are regularly blurred
where public duties are performed by private actors.109 Internet intermediaries are a good
example of this. Although they are legally organised as private companies they clearly
perform public duties, and with regard to content filtering or speed throttling there can be no
doubt that both practices interfere with individual and social spheres.

The classic remedies that are made available by the procedures of public law for bringing
basic rights into effect mostly rely on the judge. This is problematic on the Internet. Jack
Balkin rightly emphasised that the free-speech values he postulated for the Internet wont
necessarily be protected and enforced through judicial creation of constitutional rights.110
This statement is confirmed by the cases of invisible content filtering and traffic management
practised by Internet intermediaries. Here, a decision by a judge would either not be available
because of the limitations of the state action doctrine or would not provide a sufficient remedy
to effectively protect the private and public interests involved. On the Internet, moreover, the
specific features of its technodigital normativity must be protected. These features require
procedural responses that take action at the level of the technological infrastructure. Hence,
beyond classic remedies securing due process and judicial response, legislative and
administrative schemes of regulation would be necessary that protect the integrity of the

105

Luhmann, 1965.
Graber, 1994, 190.
107
Graber and Teubner, 1998, 65.
108
For a similar argument, see Balkin, 2004b, 46.
109
Calliess, 1999, 188.
110
Balkin, 2004b, 51.
106

22
networked digital environment through framework requirements regulating the design of
technological infrastructure.

With respect to the specific constitutional rights implications of technology-based filtering

and discrimination of data transmission, protecting the netizens confidence in the integrity of
the network grid must be a central postulate. Because it is essential that the user knows when
his/her communication is regulated, the law should require transparency and prohibit any
filtering and content discrimination technologies that come invisibly.111 From a due process
perspective, feedback and judicial review mechanisms would be important to correct technical
and other failures.112 Regarding the relationship between IP and technology, we have
observed above a widening gap between what the law permits and what a technology
intermediated content market tolerates.113 Since it is often difficult to determine what falls
under fair use, from a constitutional perspective, it is particularly problematic if piracy
surveillance is delegated from the content owners to Internet intermediaries using automated
technologies to filter Internet platforms ex ante.114 Automated filtering technologies are able
to cope with rules but unable to cope with standards such as fair use.115 This creates
uncertainties that as we have argued above in many ways have detrimental effects on Web
2.0 activities and creativity on the Internet.116 To enhance legal certainty and to better protect
fair-use rights, procedures which liberate the consumer from the burden of proof would, thus,
be important when issues related to distinguishing between legal and illegal use of content
arise.

5.2 Practical Solutions

With regard to practical implementation it seems that some kind of government-assisted
regulation would be necessary to secure the public policy goals that are at stake. In this
context, it is interesting to note that the Electronic Frontier Foundation (EFF) called for a
"three strikes before blocking" rule, arguing that such a model appears to be something more

111

See Berman, 200, 1275.

Katyal, 2009, 425.
113
Katyal, 2009, 411, addressing the problem of excluding legal mash-ups through filter technology.
114
For the problems of ex ante methods of surveillance from a free speech perspective see above, footnote 44,
and accompanying text.
115
Katyal, 2009, 411-413.
116
Katyal, 2009, 412, arguing that uncertainties over what constitutes fair use risks to chill of transformative
work of artists who want to incorporate the work of others.
112

23
consumer-friendly than the self-regulatory instruments.117 In our view, this claim must be
understood against the background of the specific situation in the US where voluntary
copyright enforcement agreements between content owners and Internet intermediaries
proliferate.118 In Europe, three strikes and you are out solutions are less acclaimed as the
heated public debate surrounding the French loi HADOPI and the ruling of the Conseil
Constitutionnel against a first draft of the act demonstrated.119

Under the influence of the debate in France, the European Union was more sensitive to
communicative freedom when it adopted at the end of 2009 the new EC Directive
2009/140/EC (Umbrella Directive),120 amending the framework, access and authorisation
directives of the Unions communications framework. Article 1 (3a) Framework Directive (as
amended by the Umbrella Directive)121 provides:

Measures taken by Member States regarding end-users access to, or use of, services and
applications through electronic communications networks shall respect the fundamental
rights and freedoms of natural persons, as guaranteed by the European Convention for the
Protection of Human Rights and Fundamental Freedoms and general principles of
Community law. Any of these measures regarding end-users access to, or use of, services
and applications through electronic communications networks liable to restrict those
fundamental rights or freedoms may only be imposed if they are appropriate,
proportionate and necessary within a democratic society, and their implementation shall
be subject to adequate procedural safeguards in conformity with the European Convention
for the Protection of Human Rights and Fundamental Freedoms and with general
principles of Community law, including effective judicial protection and due process.
Accordingly, these measures may only be taken with due respect for the principle of the
presumption of innocence and the right to privacy. A prior, fair and impartial procedure
shall be guaranteed, including the right to be heard of the person or persons concerned,
subject to the need for appropriate conditions and procedural arrangements in duly
substantiated cases of urgency in conformity with the European Convention for the
117

Katyal, 2009, 421-2.

See above, footnote 36, and accompanying text.
119
Insert full quotation of the ruling
120
Directive 2009/140/EC of the European Parliament and of the Council of 25 November 2009 amending
Directives 2002/21/EC on a common regulatory framework for electronic communications networks and
services, 2002/19/EC on access to, and interconnection of, electronic communications networks and associated
facilities, and 2002/20/EC on the authorisation of electronic communications networks and services, OJ L 337,
18.12.2009, p. 3769.
121
See Article 1(1)(b) Directive 2009/140/EC (Umbrella Directive).
118

24
Protection of Human Rights and Fundamental Freedoms. The right to effective and timely
judicial review shall be guaranteed.

As far as access to and use of networks is concerned, this provision takes account of several
of our postulates for effectively protecting communicative freedom on the Internet, including
the presumption of innocence, feedback procedures, due process and judicial review. We
emphasise that the freedom provision opts for a procedural framework, which must be
implemented by Member States, rather than providing for a substantive rule. With a view to
the sometimes rather competitive relationship between the EU and the Council of Europe in
matters of human rights, it is interesting to note that the freedom provision directly refers to
the adequate procedural safeguards and the proportionality test provided by the European
Convention of Human Rights (ECHR).122 Concerns of proportionality were the major reason
for many to object to the graduated response of the French loi HADOPI providing for an
Internet cut-off after the third copyright-infringing activity on the network. Since Internet
access is of extreme importance for most individuals today, a definitive Internet cut-off seems
to be an overly severe response to a copyright infringement.

A further interesting feature of this clause is its reference to the right to privacy. Private
filtering as a strategy for copyright enforcement requires the collection of huge amounts of
sensitive consumer data. Courts have recently become more sensitive with regard to basicrights implications of such data collection, as a March 2, 2010 landmark judgement of the
German Constitutional Court demonstrates.123 In this case, the German Constitutional Court
found that certain provisions of the German Telecommunications Act (GTA) allowing
preventive data retention by government authorities for a period of six months violated the
right to confidentiality of telecommunications as protected by the German Grundgesetz. The
contested provisions were enacted by the German legislator to implement the EU Data

122

The proportionality test as applied by the European Court of Human Rights in the context of restrictions to
fundamental rights generally consists of three stages: 1) Is the challenged measure suitable to achieve a
legitimate objective? 2) Is the measure necessary in order to achieve that objective? 3) Is there a reasonable
relationship between the measure and the desired aim (proportionality stricto sensu)? Article 10 ECHR (freedom
of expression and information) provides in paragraph 2 an exhaustive list of legitimate objectives, including
national security, territorial integrity, public safety, the prevention of disorder or crime, the protection of health
or morals, the protection of the reputation and the rights of others. See Cremona, 1995, 323-330; McBride, 1999,
23-36. On the proportionality test as applied by the European Court of Human Rights when adjudicating
intellectual property disputes under the ECHRs property rights clause see Helfer, 2008.
123
Bundesverfassungsgericht, BVerfG, 1 BvR 256/08, 2 March 2010, available at
[Link] For an appraisal of the case in context of
the EU Data Retention Directive that was implemented by the contested provision of the German
Telecommunications Act see De Simone, 2010.

25
Retention Directive into national law.124 It is particularly noteworthy that the data retention
allowed by the GTA was limited to traffic data125 and did not extend to content data.

In its reasoning, the Court clarified that under any circumstances a statute allowing for
preventive data retention would only be constitutional if the purposes of that law would be
limited to the prosecution of crimes that threaten overwhelmingly high-ordered interests of
public welfare or avert threats to such interests.126 According to the Court, such high-ordered
interests of public welfare include physical safety, life, or liberty of persons, security of the
federal or state governments and prevention of considerable public danger.127 Furthermore the
danger to these interests must be sufficiently concrete in order to justify preventive data
collection.128 This ruling is an unequivocal signal that any massive data retention as implied
by a three strikes and you are out type of law would not be constitutional in Germany.

[concluding remarks to be added]

124

Directive 2006/24/EC of the European Parliament and of the Council of 15 March 2006 on the retention of
data generated or processed in connection with the provision of publicly available electronic communications
services or of public communications networks and amending Directive 2002/58/EC, OJ L 105/54, 13.4.2006.
125
Traffic data included information necessary to identify the originator and recipient of communications,
information identifying the communication equipment used, the time, duration, and date of communications, and
real time geographic coordinates of mobile devices during mobile communications. Bundesverfassungsgericht,
supra note 123, at para 192.
126
Bundesverfassungsgericht, supra note 123, at para 227.
127
Bundesverfassungsgericht, supra note 123, at para 231.
128
Bundesverfassungsgericht, supra note 123, at para 232.