The Path Dependence of European Copyright

Stefan Larsson*

1. Introduction – legal path dependence and social norms

The development of law is generally conservative and retrospective. Values embedded are long lasting and consequent upon the main principle of predictability.1 At the same time, the problem addressed in this article is not that legal development follows a path, but, in relation to a historically relatively sudden shift in society, that the dependence on this path within law has become too strong, and hence, too retrospective, in the sense that it has failed to incorporate the social changes now at hand. Law is often prone to falling behind social change, and this gap causes conflict between the social and the legal spheres.2

That there is something very inconsistent and discordant between online behaviour and copyright regulation has been well documented and widely discussed.3 There is a growing amount of research that portrays the problems of applying unrevised copyright regulation in a digitalised society, in terms of creativity, cultural aspects and privacy as well as a dominant industry’s struggle for power.4 SCRIPTed contributes to the debate on different aspects of the regulatory dilemmas of digital practices, or on social change resulting from an increasingly connected society.5 I link to this by closing in on the question of why European legal development with regard to copyright and related legislation during the growth of the Internet have failed to encompass the changes in behaviour and social norms that have followed. The argument concerns law and the historical and retrospect aspects of legal development in relation to social changes. In order to outline how path dependent European copyright is, and in what way, along with those consequences that derive from this dependency, I undertake a detailed analysis of the most recent directives that amend explicit copyright legislation as well as the most important ones that are affected to some degree by copyright. Their implementation in Sweden has been chosen as a case study since the copyright dilemma in terms of illegal file sharing is highly active and at stake there and also to illustrate the gap between an EU directive and its implementation.

I proceed in three stages. Firstly, I develop a theoretical framework for my conceptualisation of path dependence. Secondly, I show the most important regulatory bodies of interest on a European supranational level, i.e. InfoSoc, the IPRED, the Data Retention Directive, the Telecoms Reform Package/ACTA, and their implementations in Sweden where relevant, and I also point out the applicable aspects that contribute to forming “the path”. Thirdly, I elaborate the lock-in effects of copyright development in Europe, and I conclude with the main general and also specific consequences of the path dependence at hand.

2. Path dependence of law

Although much of it has concerned technological development, the literature on path dependence may apply with equal force to legal development.6 Regulatory regimes provide obvious analogies to technological standards.7 Legal developments have been analysed in terms of path dependence, especially by American scholars, with reference to the classic text The Path of the Law by Oliver Wendell Holmes.8 Predictability, as described by the legal scholar Peczenik, is “one of the basic values in democracy and a state governed by law”, and many legal theorists hold that the norm of “jurisdiction and the actions of public authorities in a democratic state should be predictable”.9 The Norwegian sociologist of law Vilhelm Aubert speaks of law as something that serves to safeguard expectations, one of five main tasks of law, and as Niklas Luhmann has argued, its most important one.10 This predictability can also account for the often-incremental development of law, the minor steering towards a retrospective activity that changes mostly in evolutionary rather than revolutionary terms. This retrospectivity, this “past-dependency” of law, in the terms of the American Judge Richard A Posner, is probably not a problem when society changes according to stable curves.11

In terms of technologies, there are huge advantages to standardisation, which makes it lock in certain conditions.12 These standards solve coordination problems among users, allowing them to constitute a network beneficiary for those who use the same technology.13 Gillette adds an element of power to the otherwise relatively cold equation of transactions costs, in much of the literature on path dependence. The dominant interest groups can, supported by the regulatory standards, use their powers to ward off emerging attempts for regulatory change. This is the reason why, if one seeks to understand the whole picture, one has to include those actors who depend on regulation, in order to understand its development.

Furthermore, a factor that perpetuates lock-in effects and has to do with metaphors and conceptions of legal development, somewhat similar to the “rhetorical repertoires” of international organisations studied by Halliday et al., but which mostly draw on the findings of the conceptual metaphor school founded by George Lakoff and Mark Johnson in the 1980s.14 Gillette speaks as if the lock-in effects are always conscious and are therefore an outcome of power struggles and transaction costs. These struggles for power, to a high degree, probably are, but there is also an element of the language-based legal means that include metaphors, categorisations and labels used to make certain conceptions that are supported in law.15 Although these metaphors can be used very much consciously, as in copyright education efforts, they may just as well be part of an unconscious but language-based pattern that functions as a type of standardisation.16 It is just not as visible as other standards. Gillette does not see these conceptual lock-in effects that emanate from the retrospect practices of a law-making nature, but they have been shown to be relevant when it comes to copyright.17

Mahoney, who analyses the use of path dependence as an analytical tool in historical sociology, divides the types of path dependencies into self-reinforcing sequences and reactive sequences.18 Mahoney broadly defines path dependence as something that occurs when a “contingent historical event triggers a subsequent sequence that follows a relatively deterministic pattern”. In relation to the two categories of path dependence, Mahoney concludes that in the case of a self-reinforcing sequence this means that “the contingent period corresponds with the initial adoption of a particular institutional arrangement, while the deterministic pattern corresponds to the stable reproduction of this institution over time”. Mahoney contrasts this with the reactive sequence, where the contingent period “corresponds with a key breakpoint in history, while the deterministic pattern corresponds with a series of reactions that logically follow from this breakpoint”.19 As we will see, self-reinforcing path dependency is very much relevant to the development of the European copyright regime.

3. European Copyright in the days of the Internet

The development of copyright is directly connected to contemporary technological development. It is the networking technologies that have challenged the legislation, thus leading to amendments in order to cope with the new technical and social changes. A series of legislative initiatives have been taken to strengthen online compliance with copyright regulation. This article presents four of the most important recent regulatory initiatives in the European Union that either have had an explicit focus on copyright, or an indirect, but important effect, as well as a brief outtake about future development.

• The European Community Directive on Copyright in the Information Society (“InfoSoc Directive”) was tabled in December 1997,20 and the directive was passed in 2001.21 It was implemented in Sweden on 1 July 2005.

• The Intellectual Property Rights Enforcement Directive 2004/48/EC (“IPRED”) was approved by the European Parliament on 9 March 2004, and implemented in Sweden on 1 April 2009.

• The Data Retention Directive aims at harmonising the regulation of the Member States, who require telephone operators and Internet Service Providers to retain personal data.22 This will play a role in the enforcement of copyright, and how this will happen is explained in this article. It has yet to be implemented in Sweden.

• The European Telecoms Reform Package was widely debated in the Swedish press in 2009. It was presented to the European Parliament in Strasbourg on 13 November 2007 but not voted on until 6 May 2009. This is a cluster of directives that are being prepared (COM [2007] 697),23 which includes aspects of the role of the Internet Service Providers and which will also play a role in the enforcement of copyright violations.

• The imminent future developments can be interpreted from the outcome of the international negotiations on the Anti-Counterfeiting Trade Agreement (“ACTA”), a multilateral agreement negotiated outside WTO processes and protections, and to some extent the Green Paper – Copyright in the Knowledge Economy from July 2008.

3.1. Stronger Copyright: The InfoSoc Directive

The initial proposal for the European Community Directive on Copyright in the Information Society was tabled in December 199724 and the directive was passed in 2001.25 This followed the Green Paper on Copyright and Related Rights in the Information Society of July 1995.26 One of the original two purposes of the directive was to bring the laws on copyright and related rights in the European Union into line with the WIPO Internet Treaties, in order to set the stage for joint ratification of the treaties by the member states and the European Community.

The second goal of the InfoSoc Directive was to harmonise certain aspects of substantive copyright law across the European Union. The Directive states the importance of legal protection of copyright and related rights with regard to the “information society” in article 1.27 The Directive has been criticised for focusing on the aggregators’ rights rather than the creators’, and that it is “primarily geared towards protecting the rights and interests of the ‘main players’ in the information industry (producers, broadcasters and institutional users), not of the creators that provide the invaluable ‘content’ that drives the industry”.28 During the almost eight years from the first proposal in 1997 to Sweden’s implementation in 2005, much happened on line in terms of techniques and technology for communicating in general on the “information superhighway” mentioned in the Green Paper from 1995. In terms of organised file sharing initiatives for music and films etc., the architecture went from the centralised unstructured peer-to-peer system of Napster to the first decentralised file-sharing network, Gnutella, in 2000 and then to Kazaa in 2001. From 2002 through 2003, a number of popular BitTorrent services were established, including The Pirate Bay.29

The InfoSoc Directive includes protection for “technological measures” which often are referred to as Digital Rights Management, DRM (article 6). This criminalisation of the circumvention of technological measures has been seen by critics as a way to authorise copyright more powerfully than ownership in terms of consumers buying music for example, but at the same time being restricted as to what they are allowed to do with the purchased product (for example, the owner of a music CD who cannot copy it in order to play it in his or her car).30 The protection of technological measures is not new to the Swedish Copyright Act, but the version prior to the InfoSoc Directive implementation applied only to computer programmes.31The directive was implemented among the Member States, mainly between 2003 and 2004, by Denmark, the Czech Republic and Greece at an early date and by Sweden, Finland, Spain and France at a later one.32 The original last implementation date for the InfoSoc Directive was 22 December 2002, but only Denmark and Greece had implemented it by then. In Sweden, the proposal from the governmental commission (the SOU) was presented in 2003.33 In the following government draft bill 2004/05:110, legal changes were accepted by the Parliament and came into force on 1 July 2005.34

The digital technologies from the mid-1990s provoked worldwide and interdisciplinary debates on their potential impact on the non-digital world. It is in this context that the European Council called for a report on “the problems” of the information society. The report often referred to as the Bangemann Report, after the chair of the group that produced it, concluded that the protection of intellectual property was of the greatest importance. The InfoSoc Directive has meant a wider scope for copyright and a criminalisation of more actions.35 For the legal concepts that once described the reproduction and protection of pieces of vinyl and other plastic materials, a sudden enlargement had to be undertaken to include digital formats in increasing use. The legal concepts that describe and regulate the analogue practices became metaphorical in the sense that they were held to regulate emerging practices of a different format, with new (digital) restraints and possibilities, unlike their existing analogue counterparts.36 An emphasis was placed on the control of an environment, which at the time lacked this very feature.

3.2. Enforcing Copyright: IPRED and its implementation in Sweden

Looking back on the origins of the Enforcement Directive, the Commission presented a Communication in November 2000, announcing a series of practical measures intended to improve and intensify countermeasures against “counterfeiting and piracy in the single market”. As part of these measures, the Commission forwarded a proposal for a directive harmonising the legislation of Member States so as to strengthen the means of enforcing intellectual property rights.37 Even then, around the time the IPRED was approved by the European Parliament (9 March 2004),38 it caused a stir amongst civil rights groups in the United States and Europe.39 In April 2004, the EU passed the Directive on the Enforcement of Intellectual Property Rights, the so-called IPRED Directive. It was established because it was “necessary to ensure that the substantive law on intellectual property….is applied effectively in the Community” (Recital 3). Recital 4 of the Directive explicitly relates it to copyright legislation according to the TRIPS Agreement.40 Although its scope covers the entire IP spectrum, the Directive has generally been discussed in terms of copyright enforcement. Central to the debate is the fact that the directive gives the copyright holders the right, by virtue of a court decision, to retrieve the identity information behind an IP address at a certain time, when they “have presented reasonably available evidence sufficient to support its claims” (article 6.1). The “competent judicial authorities” may then requisition such information.The IPRED is a minimum directive, meaning that the member states can stipulate national conditions that are even more favourable to the rights holders than the directive prescribes (article 2).

Sweden did not meet the requirements of article 8 of the Directive. To do so required that the Swedish law introduce provisions that give holders of intellectual property rights, a right to information about the infringer.41 This was one of the most widely debated issues in Sweden, because copyright holders’ representatives, such as the IFPI and the Antipiratbyrån, could apply to the courts to approve the release of identity information from ISPs (S 53 c of the Swedish Copyright Act). For such an injunction to be issued by the court, requires neither that the applicant identify the infringers nor that infringement has been intentional or grossly negligent. It is enough that probable cause has been shown that a person has committed an infringement or a violation, according to s 53 c Swedish Copyright Act.42 It requires the court to have found that an infringement has occurred, which means that in implementing the directive, Sweden lowered the requirement below that contained in it the Directive and also departed from the proposed level in the preparatory memorandum.43

The injunction is aimed at ISPs and describes the relationship between them and the copyright holders (their representatives). This is an expansion of rights linked to IPR, in the name of enforcement of the latter. Implementation meant that the majority of the provisions in the IPRED were in force by 1 April 2009.44 To date, this legislation has led to only a few court cases in Sweden, of which no more than two are of interest here. This is despite the initial reports in the media of “hundreds” of cases being prepared by copyright holders’ interest groups and the rough estimate of preparatory legal work on an estimated 400 to 800 cases per year.45

In general, the IPRED highlights the issue of the ISPs’ position as being increasingly targeted because they are the key to identifying information behind the IP numbers, on the one hand, and are also the guardian of their customers’ privacy, on the other. The two aforementioned Swedish cases underline this. An interesting aspect of both cases is the defendants’ claims that the Data Retention Directive, although not yet implemented in Sweden, is applicable in a way that would hinder the legal use of the rights that the IPRED law grants the copyright holders. No court agreed until the Ephone case was granted a review permit for a hearing in the Supreme Court in September 2010, which acknowledged the legal uncertainty of the non-implemented Data Retention Directive with regard to the application of the IPRED implemented in Sweden by asking the European Court of Justice for a preliminary ruling in the matter, and its response is still pending. This underlines the complex and uncertain but applicable role of the Data Retention Directive with regard to European copyright.

3.3. Combating “Serious Crime”: The Data Retention Directive

The Data Retention Directive amends Directive 2002/58/EC in order to force operators of public telephone services and Internet Service Providers, ISPs to keep data such as calling number, user ID and identity of a user of an IP address for a period of between six months and two years. The aim is to “ensure that the data are available for the purpose of the investigation, detection and prosecution of serious crime, as defined by each Member State in its national law”.46 Though the scope of government-mandated data retention may vary, at its core is the requirement that ISPs collect and store data that track the Internet activity of their customers.47 In December 2005, the European Parliament passed the Data Retention Directive.48 The origin of the directive was the fight against terrorism in response to the Madrid and London bombings in 2004 and 2005.

We may ask why this is relevant in the copyright context. A first step in answering this question is to conclude that the Data Retention Directive lacks clarity with regard to three main issues. It does not define what is meant by ‘‘serious crime’’ and instead leaves this task to each Member State’s national law; it does not limit access to retained data to specifically designated law enforcement authorities, as it refers only to ‘‘competent national authorities’’; and it leaves it up to each Member State’s national law as to when access to data is permitted, all of which are relevant to copyright enforcement, depending on where the line is drawn in each of the three cases.

Furthermore, the directive targets the role of the ISPs and their relationship to the customer by weakening the latter’s privacy rights. The former Directive, 2002/58/EC, established a principle that traffic data must be erased as soon as storage is no longer needed for purposes relating to the communication itself (including billing). The Data Retention Directive entails a breach of this principle. The European Data Protection Supervisor (“EDPS”),49 was harsh in its criticism of the proposal, and actually termed it illegal.50 Criticism of this kind was also voiced by the Article 29 Data Protection Working Party, which is an independent EU body for the data protection of privacy.51 It concluded with regard to the directive that it “encroaches into the daily life of every citizen and may endanger the fundamental values and freedoms all European citizens enjoy and cherish”.52 The Article 29 Data Protection Working Party continues by stating that it is “of utmost importance that the Directive is accompanied and implemented in each Member State by measures curtailing the impact on privacy”.53

The important choices to be made by the Member States in implementing the directive related to the data storage period (6-24 months), exactly what data should be stored, and who should be obliged to retain them (should small operators have to?). The question of who should pay for data retention and data delivery when data have been requested has also been very widely debated, and different Member States have adopted different solutions. Some commentators believe that media companies will seek to lobby national governments to include file sharing in the definition of ‘‘serious crime’’ so that data can be accessed for this purpose.54 In general, the directive touches on the fundamental concern about what extended access to traffic data in a digitalised society will bring, irrespective of the intentions behind it. Specifically, the directive aims to harmonise the regulation by the Member States of telephone operators and ISPs, with a view to retaining personal traffic data.

In examining the case of the Swedish implementation of the Directive, there are a few questions about data retention that are of relevance to the copyright path. An important question, as previously mentioned, is whether or not the Directive will override the IPRED. Secondly, when it comes to the “serious crimes”, it is not only about the data storage period, but also about what levels of criminal penalties will allow the police to obtain and use subscriber identification.55 In late 2010, the media reported that a proposal was being prepared that would diminish the necessary severity of the crime, so that the police could also retrieve traffic data from the ISPs for crimes punishable only by a fine. This would include copyright crime, often referred to as illegal file sharing.

The Data Retention Directive aims at aiding the prosecution of criminal cases, while the IPRED relates to civil cases. This probably means that the traffic data stored under the provisions of the Data Retention Directive cannot be used to aid copyright holders’ representatives in a civil law case for damages for copyright infringement. This will in turn probably require some sort of dual database for the ISPs. The type of data that the copyright holders have the right to obtain from the ISPs, as a result of the implementation of the IPRED, is often already erased by the ISPs in accordance with the principle of consumer privacy. These data will probably still be available to some extent as a result of the implementation of the Data Retention Directive. This means that the Data Retention Directive may aid the IPRED and the copyright holder’s case against illegal file sharing, a consequence of the Data Retention Directive that never was mentioned in its draft stages.

The copyright owners’ interest groups collect IP numbers that they believe violate the rights of their clients. In order to link the IP number to the persons behind the actions, these groups need to approach those who have access to this link, the ISPs. Since present legislation on ISP responsibilities focuses on the integrity of the subscribers, the ISPs generally do not store the data for a long period. Some ISPs even stated that they store traffic data for the minimum time possible, when the IPRED was implemented in Sweden in 2009. Implementing the Data Retention Directive therefore helps the case of the copyright holders in that changing the responsibilities of the ISPs from being prohibited from storing the data for an unnecessarily long period, to being obliged to store the data for a longer time.56 The ISPs will no longer be able to lawfully choose to discard the data logs as soon as the billing purposes have been fulfilled when the Data Retention Directive is implemented. This was according to the draft bill set to be done 1 July 2011 in Sweden but was postponed in a vote in the Swedish Parliament 16 March 2011 for at least a year from the original date of implementation.57

3.4. The Telecoms Reform Package, ACTA and the future

In order to see some of the future outcomes of this regulatory path a few on-going or otherwise related processes must be presented: The European Telecoms Reform Package, the Anti-Counterfeiting Trade Agreement (ACTA), and the Green Paper – Copyright in the Knowledge Economy from July 2008.

The Telecoms Reform Package was presented to the European Parliament in Strasbourg on 13 November 2007, voted upon 6 May 2009, and finalised on 25 November 2009. The reform package originated from a non-legislative resolution on “Cultural industries in Europe”, generally referred to as the “Bono Report” after the French Socialist MEP responsible for drafting the resolution. The reform package is a cluster of directives (COM [2007] 697) that to a great extent focus on the role of ISPs.58 It comprises five different EU directives, and its total scope is vast and only of limited relevance here.59 Much of this regulation has already been implemented in Swedish law, through the Electronic Communications Act (2003:396) and a few sections in a law on standards for broadcasting of radio and television.60 The Commission’s initial proposal for the Telecom Package was presented to the European Parliament and Council as the “Better Regulation Directive” and “Citizen Rights Directive”, 16 November 2007.61 During the Parliament’s examination of the Commission’s proposals, a number of amendment proposals were produced. Of the 126 amendment proposals to the Better regulation directive and 155 amendment proposals to the Citizens rights directive (in that first reading on 24 September 2008), it was proposal 138 for the Better regulation directive and proposal 166 on the Citizen rights directive that were the most widely debated, in media, on blogs, and in the EU Parliament and the Council. These stated that users’ access may not be restricted in any way that infringes their fundamental rights, and (166) that any sanctions should be proportionate and (138) require a court order. In May 2009, the French representatives wanted to withdraw amendment 138, which ensures that court proceedings precede a possible disconnection. At the same time, the issue of disconnecting Internet users for suspected copyright violations before they are proven guilty in court has been highlighted in France through the three-strikes HADOPI-law.62 A compromise version of the amendment was eventually adopted by the European Parliament in November 2009 replacing the requirement for a “prior ruling by the judicial authorities” with the requirement for a “prior fair and impartial procedure”. The Telecoms reform package reaffirms how the copyright path may colonise a variety of types of regulations. This time it was highlighted by the ISPs’ key position in the battle over copyright enforcement and the attempts made by strong forces to disassemble consumer protection in the case of disconnecting copyright violators from Internet access.

When it comes to the future development of copyright within the EU, there are conflicting initiatives. On one hand, the EU takes part in the confidential negotiations of the Anti-Counterfeit Trade Agreement (“ACTA”) which will significantly elevate the level of copyright protection on a global level, and, on the other, the EU speaks of the importance and need to “promote free movement of knowledge and innovation as the ‘Fifth Freedom’ in the single market”, presented in the Green Paper on Copyright in the Knowledge Economy 2008, a document inviting participation in these issues.

This begins with the ACTA, a plurilateral agreement negotiated secretly by around a dozen countries. Although the negotiations have been hidden from public view, an ACTA text dated 18 January 2010 was leaked in March 2010, followed by an official EU version published 21 April 2010 and the final version of 3 December 2010.63 One difference between the leaked document and the official ones is that the parties’ opinions are not stated in the latter. The leaked version dated 18 January 2010 reveals that the U.S., unsurprisingly, is an IP maximalist here, pushing for strong provisions. The question of privacy interests is of very great importance when analysing the ACTA, since the agreement seems to increase data sharing with both other countries and with rights holders.64 The official document of 21 April, in short, seems to picture an active, pro-rights holder role for ISPs and other online intermediaries. The ACTA may also limit the type of services that can fall into a “mere conduit” exception to notice-and-takedown. Once again, the ISPs are identified as the key target. Although the draft versions to a high degree targeted ISP liability, the most controversial enforcement measures proposed in the initial stages of the negotiations of the ACTA have been narrowed down in its final version.65 All in all, the ACTA however shows that there are strong international forces that seek to extend the means of enforcing copyright undemocratically, at the expense of ISP neutrality.

The purpose of the Green Paper on copyright in the knowledge economy is to “foster a debate on how knowledge for research, science and education can best be disseminated in the online environment”. The Green Paper aims to set out a number of issues connected with “the role of copyright in the ‛knowledge economy”’.66 The problem here is the lock-in effects of the legislative path that EU has taken with regard to copyright. The way that it proposes to “unlock” some of the aspects is to broaden the exemptions under copyright. The Green Paper has been said to highlight the need for serious research and a dialogue on the future of the InfoSoc Directive, especially as regards competing policies in the areas of “consumer protection, telecoms regulation and electronic commerce”.67 The Green Paper states that “a high level of copyright protection is crucial for intellectual creation.” “Copyright ensures the maintenance and development of creativity in the interests of authors, producers, consumers and the public at large”, and at the same time it seeks to acknowledge the essence of no protection for certain groups as well as types of creativity, for instance “user-created content”.

It is the legal heritage of a strongly path-dependent copyright that creates this contradictory stance, in its attempt to grasp the online flow without being able to diverge from the legally locked-in path.

4. Analysis and discussion

There are at least five main findings in a path dependence analysis of European copyright development. Most of them have already been pointed out, but here they will be collated and further elaborated. These are presented in the following subchapters, where the first, the legitimacy issue, deals with the fundamental conflict between social and legal norms, making the path dependence analysis important in the first place; the second deals with how the preservation of the path is being undertaken, the third describes how the copyright path and its inherent conceptions colonise other legal paths, at the expense of other values; the fourth fairly succinctly highlights the targeting of the ISPs in this development; and the fifth, on the other hand, elaborates how the increased use of data traceability in legal enforcement has implications for how privacy should be handled. Finally, a brief section suggests future possible research in line with this article.

4.1. The legitimacy issue

As mentioned, the gaps between regulation, social norms and conduct have been widely discussed and stated in the literature.68 This stems in part from the fact that the global copyright construction is a legal complex that in general is based on ideas of the conditions of an analogue world for distribution and production of copies, but it is armed with increasingly protective measures when faced with human conduct in the context of digital networks. Around this regulation an industry emerged mainly during the second half of the twentieth century that is dependent on the ability to enforce this type of law against those who attempt to benefit from the immaterial work of others. This dependency is expressed in terms of control over copies and distribution. Every copy was inevitably connected to a physical object, which demanded an investment. In this analogue context, this construction is functional, without too many anomalies in terms of breach of this control. Although, in a digital context, where the distribution costs are close to zero and making copies does not need any investment, this control is fundamentally breached, with the consequence that law and practice does not correspond well. This non-digital versus digital divide plays an important role in the legitimacy of copyright laws. The legal answer to the Read/Write environment of the Internet technologies, to borrow terminology from Lawrence Lessig, has been a constant increase in regulatory efforts to maintain the prevailing constraints of a Read Only conception of copyright.69 This is where the concepts of copyright law becomes the metaphors of copyright, where these standardised modes of conceptualising how copyright best serves its purpose are preserved in law, regardless of the external changes in society. This is where the transaction costs of legal change become manifest, to the extent that the cost becomes remarkably high for shifting the path for copyright regulation. This kind of change seemingly provokes a political struggle, and the legal domain cannot abandon certain legally embedded conceptions; the law lags behind. There are transaction costs attached to such a fundamental change of ideas embedded in law, especially for one so locked into national, international and supranational law and treaties.

The media industry’s struggle to push illegal file sharing of copyrighted content into a legitimate market in conformity with copyright laws has been described by many in terms of a “war”.70 This metaphor not only strongly emphasises how common illegal file sharing is, but the aggressive legal and other measures the industry (especially in the United States) has taken in the battle as well. This “war” has led copyright owners’ interest groups to employ a very active strategy in collecting evidence against violators of the regulation,71 which highlights the distance and the clash between the behaviour and social norms of some groups on the one hand, and the laws that try to regulate this behaviour and change these social norms on the other hand.72

4.2. Preserving the path

The path dependence of European copyright serves as a strong argument for those who benefit from its conservation. Appeals to tradition impede change by privileging the status quo in terms of an increased protection. The reason why these appeals still prevail as dominant ones is probably a consequence of the linkage to a strong industry protecting and voicing them, thereby complementing the internal functions of path dependence. As has been mentioned, the “adoption of a particular institutional arrangement” in copyright had already been undertaken before Internet came into being, and the deterministic pattern of this “self-reinforcing sequence” over the last few years had corresponded with the stable reproduction of this institution over time.73 It had already been self-reinforced, reproduced and legitimised when it became fundamentally challenged by the development of digitalised networks.

The legal path, when it becomes strongly dependent, is related to internal legal “legitimating processes” in the sense that increasing legitimating processes are marked by a positive feedback cycle in which an initial precedent about what is appropriate forms a basis for making future decisions about what is appropriate. As a result, a familiar cycle of self-reinforcement occurs: the institution that is initially favoured sets a standard for it own legitimacy; this institution is reproduced because it is seen as legitimate at this stage; and the reproduction of the institution reinforces its internal legitimacy.74 This should in the perspective of social change relating to Internet and similar technologies be seen as an internal legal legitimacy, not necessarily connected to external social norms and the behaviour of ordinary citizens today, when the institution has reached an own momentum, albeit not without support from the actors benefitting from its continued dominance.

A negative implication of this “standardisation”, according to Gillette, is that the incentive to produce an improved system is discouraged, because no single user within the existing network can be induced to shift to the new system without assurances that a critical mass of potential users will do likewise. As an analogy, Gillette claims, regulatory regimes that share the characteristics of lock-in could be as vulnerable to path dependence as technological standards:

Where those who are favoured by the status quo can organise with ease relative to those who are disfavoured (that is, where those who favour the status quo can form a dominant interest group), they are likely to take advantage of that position whenever an inkling of change arises.75

The revisions of copyright the last ten years shows an interesting incremental approach. In terms of what analysis is perceived as necessary when developing copyright regulation, it merely reinforces values already in position. There are no signs of deregulation or decreasing protection, indeed the opposite is the case, assuming that the concepts and metaphors used will also function in a digital environment under proper control. It is a “disjointed incrementalism”, that in the words of Etzioni seeks to “adapt decision-making strategies to the limited cognitive abilities of decision-makers and to reduce the scope and cost of information collection and computation”.76 This decision-making focuses on close goals instead of comprehensive ones, through a limited analysis. The successive composition reduces the need for theory, or for a detailed understanding of social changes outside the control of the law. Minor adjustments to a model that has been working fairly well for a considerable time are a decision-making model used by policy makers. This is where the (paradigmatic) shift between an analogically anchored way of communicating (in the broadest sense) and a digital one inevitably leads to policy-related problems.77

4.3. Colonising other paths

As has been mentioned, there are power structures that contribute to make this legal path colonise other legal paths. When concerned interests, relying on the power balances of the regulation drafted in non-digital times, seek to maintain their position, other values that the law protects become secondary. Not only does the path dependency incorporate a broadened criminalisation of types of actions, but as more actions become criminal, it also affects the hierarchy of rights, as with the non-circumvention of technical measures.78 Furthermore, if the Data Retention Directive describes how copyright enforcement can become embroiled in legal efforts against terrorism, the Telecoms reform package shows how it can get tangled up in telecommunications market issues. For that matter, the ACTA shows how copyright can increasingly be understood in terms of trade, and hence, be part of trade agreements that can circumvent more democratic legislative processes on a national or supranational level. The effect of copyright distribution and the formulations for how copyright is constructed and conceptualised are reproduced and strengthened in various related and sometimes only remotely related legislative efforts. A directive that is drafted to fight terrorism, such as the one on data retention, which combats an activity with extremely low legitimacy in social norms, can end up in including the struggle against illegal file sharing of copyrighted content, an activity with extremely high legitimacy in terms of social norms.

4.4. Targeting the ISPs

From the strong path dependence of copyright there derives a clear tendency to target the ISPs and other intermediaries in an attempt to keep the copyright path intact. The IPRED is a clear example of this, and also the Telecoms Reform Package and, the ACTA further emphasise this fact. There are plans to revise the IPRED, and a recent report from the Commission discusses that the currently available legislative and non-legislative instruments are not powerful enough to combat online infringements of intellectual property rights effectively, which leads to the conclusion that ISPs could be further targeted and involved:

Given intermediaries’ favourable position to contribute to the prevention and termination of online infringements, the Commission could explore how to involve them more closely.79

In addition, the exact implications of the Data Retention Directive for copyright are not clear, but it will probably affect copyright enforcement, as outlined above, by ensuring that identification information is stored. The key role of the ISPs is, however, also part of a bigger issue that concerns the character of the Internet as we know it and the features and possibilities for the online enforcement of the law.

4.5. Looking ahead: Pushing the limits of effective legal action

Law is in many aspects very dependent on its history, in the sense that history matters. Concepts and principles create paths that also lock in future legal directions. The problem here is not that legal developments relate to its past, or lock in standardised modes of prescribed conduct. On the contrary, these elements serve as parts of the strong legal principle of predictability. However, problems occur when they relates to the past in such a manner that it fails to include or to grasp important changes in society, and it is so locked in that it cannot even consider alternatives that might be more efficient, given the new conditions in society. In short, problems occur when law is too path dependent in relation to social change.

This development shows that the fight against file sharing risks being drawn into legislative contexts of fundamentally different origin and legitimacy. A significant predicament regards the fact that a directive that is drafted to fight terrorism, an activity with extremely low legitimacy in social norms, can end up in including the struggle against illegal file sharing of copyrighted content, an activity with extremely high legitimacy in social norms.

The development of European copyright, in its broad sense, not only re-builds the Internet in terms of traceability (the IPRED and possibly the Data Retention Directive) but also legal enforcement in terms of mass-surveillance. Since this is a strong claim, the analysis has to be elaborated further. This paper shows the legislative and supranational responses to the problems that copyright faces in the digital milieu and the legislative developments where the authoritarian trend in copyright-related European legislation is striking. This cluster of legislation that seeks to harmonise the national legislations of the European Union are all part of a trend for increasing control over the flows on the Internet. More data are being generated and retained in order to support the copyright owners’ fight against the illegal file sharing of protected content. At the same time, the copyright holders’ representatives are given easier access to identification data through regulation assigning greater responsibility to the ISPs for content that is being trafficked through their infrastructure. This is one of the reasons that the debate around net neutrality has widened.80

A relevant question here is whether the price for the enforcement of copyright is acceptable in terms of decreased privacy for all. This shows that following the copyright path in legal development is not an independent trend but one that also affects other very important values, which is something that is not always considered in the process. Thus file sharing and copyright are closely linked to issues of privacy and the character of the Internet, and. there are different values that oppose each other. It is of interest to note that copyright legislation, seemingly, is path dependent and inviolable while privacy regulation is not, or at least not as strongly.

The Internet can be used by citizens to circumvent authority and governments use the Internet and related technologies to respond to this. There is, however, always a limit to the effectiveness of legal action, although this is hard to draft in detail. Roscoe Pound seeks to lay down principles suggested by a consideration of basic characteristics of modern law, and advances the fact that law can only deal with “the outside”, the law cannot attempt to control social norms and beliefs but only observable behaviour.81 This is even more interesting in the context of Internet-related behaviour, considering the newly emerging opportunities to “observe behaviour” on such a vast scale. Never before has there been so much collectable information revealing the inner thoughts and every-day habits of an increasingly connected society. This poses new questions as to who has the rights to do what with this information, it shapes the strategies of the copyright protectors, it magnifies the question of the role of the ISPs, and it certainly asks to what extent the law should stipulate a centrally located collection of, for instance, traffic data in order to enforce whatever laws the legislator seeks to enforce. As the Internet develops, the effectiveness of legal regulation meets new obstacles with respect to controlling actions on line. On the one hand, the possibilities that the medium provides have an impact on the prerequisites for social norms, which also affects compliance with copyright legislation, as well as give new tools for resistance to legal enforcement.82 On the other hand, central to this article is that the digital networks that form the “new social morphologies” impose completely new ways of legal enforcement and mass-surveillance over the multitude of habits and secrets of our everyday lives. The “long arm of the law” has acquired an extensive reach. It now has a new potential to discover and control everyday behaviour in a way that forces us to ask questions about how far we want it to extend.

Law enforcement is expanding in terms of the possibilities for control, surveillance and identification that the digitalised and networked society can offer. The possible “nodality” of legal enforcement is greater than ever before. The IPRED is one such European Union response to the circumvention of copyright legislation that the Internet has brought about. One must ask who can, for instance, really guarantee the legitimate uses of the massive traffic data collection in which the Data Retention Directive results.

While describing the “generativity” (and citing Zittrain) of the Internet technologies, the Internet policy researcher Margetts asks in what way the Internet serves as a platform for policy innovation.83 One answer can be found in the growing responsibility of balancing integrity concerns and new and extreme possibilities for recording behaviour by means of data logs and digital supervision. The potential of technology and its embeddedness in all aspects of social life test the limits on the effectiveness of legal action in determining the borders of legitimacy. Where to draw this line is, hopefully, a political question, and, ideally, this question will be decided in a democratic manner. It is important to be clear about the fact that the development of a general mass surveillance of the entire population is not an issue to be taken lightly or a development that should be allowed to pass unscrutinised.

4.6. Future research

This article contributes to a knowledge base pertaining to the legal aspects of a dilemma relating to illegal file sharing, copyright regulation and its role in social and societal developments. A detailed and deeper understanding of these developments probably requires a more comprehensive research approach including studies of social norms and a greater understanding of what is happening online, or socially, in conjunction with network lifestyles. One could also imagine that a more detailed study of the origins of the directives would tell us more about transnational law-making; the interests that have been able to influence this process, the kind of forces are at play, and those who have the power to influence the process, etc. Questions of further interest are, e.g., to what extent there are hidden aspects of international considerations, perhaps involving global politics related to trade and “strong” versus “weak” countries that shape the regulatory formulations.

In terms of path dependence in copyright, this study has focused on the days of the Internet, meaning the last decade or so, but it would be of interest to see a more historical study of the development of the copyright complex, how it grew, and evolved its “particular institutional arrangement” throughout the twentieth century, in the words of Mahoney.

5. Conclusion

Copyright’s strong path dependence has been elucidated by describing how copyright is legally constructed and harmonised through international treaties as well as recent European regulatory efforts, in the form of InfoSoc Directive and the IPRED. The Data Retention Directive and the Telecommunications Reform package highlights the ISPs as being increasingly conceptualised as the key to identity information about copyright violators, on the one hand, and as guardians of subscriber privacy on the other hand, while they themselves advocate neutrality and a “mere conduit” as their leading principle. Furthermore, the “secretly” negotiated ACTA agreement may result in imposing stronger copyright on Member States, which the US is seemingly striving to achieve. The underlying formulations of how copyright is constructed and conceptualised is reproduced and strengthened in various related and sometimes only tenuously related legislative efforts.

From a socio-legal perspective, copyright regulation suffers from legitimacy issues. The global copyright construction is a legal complex that in general is based on ideas of the conditions of an analogue world for distribution and production of copies- but armed with increasingly protective measures when faced with human conduct in the context of digital networks. To some extent, this most probably involves expansion of the concepts and metaphors that once described only non-digital practice but now form standardised ways of creating new law. The trend in European copyright is protectionist, through the expanding and strengthening of rights and their enforcement, and in that it is self-reinforcing and locked in to certain standards. The path dependence of European copyright serves as a strong argument for those who benefit from its preservation, signalling that there are power structures that support the colonisation by this legal path of other legal paths that protect conflicting rights.

There is a clearly visible tendency in targeting the ISPs and other intermediaries in an attempt to keep the copyright path intact. The development of European copyright, in its broad sense, not only re-designs the Internet in terms of traceability but also law enforcement in terms of mass-surveillance. The digitalisation of society requires that new questions be asked as to how legal enforcement is or can be performed in terms of mass-surveillance of the multitude of habits and secrets in our everyday lives. This means that there is a growing political responsibility for balancing integrity concerns and new and extreme possibilities for recording behaviour by means of data logs and digital supervision. Thus, the path dependence of copyright leads to an imbalance of principal importance between the interests at stake. The imbalance lies in that a special interest is allowed to modify methods of legal enforcement from the reactive and particular to the pre-emptive and general. The special copyright interest gains at the expense of the privacy of everyone.

---