Teisė ISSN 1392-1274 eISSN 2424-6050

2024, Vol. 130, pp. 20–32 DOI: https://doi.org/10.15388/Teise.2024.130.2

Utilitarian and Deontological Approaches to Liability of Internet Intermediaries for Infringements of Intellectual Property Rights

Justinas Drakšas
PhD student of Vilnius University Law Faculty
Department of Private Law
Saulėtekio al. 9, Building I, LT-10222 Vilnius, Lithuania
Phone. +370 5 236 6170
E-mail: Justinas.draksas@tf.vu.lt

Utilitarian and Deontological Approaches to Liability of Internet Intermediaries for Infringements of Intellectual Property Rights

Justinas Drakšas
(Vilnius University (Lithuania))

Summary. Internet intermediaries are the central figures through which the majority of informational flows go through on the Internet. It is therefore only natural to wonder, whether they should be held responsible, or even liable, for facilitating the exchange of illegal information being done by their users. Currently, academic literature focuses on two main views that could help to justify the imposition of liability on these subjects – these are the utilitarian and deontological theories of Internet intermediary liability. This paper attempts to deepen the understanding of these theories, by, firstly, identifying the main ideas behind the utilitarian and deontological approaches to the liability of Internet intermediaries; and secondly, looking at the main examples of regulation that are consistent with the utilitarian and deontological approaches, which could help to describe the overall evolution of intermediary liability models in the European Union.
Keywords: Internet intermediaries, secondary liability, platform liability, gatekeepers, utilitarianism, deontology, moral theories.

Utilitarinis ir deontologinis požiūriai į interneto tarpininkų atsakomybę už intelektinės nuosavybės teisių pažeidimus

Justinas Drakšas
(Vilniaus universitetas (Lietuva))

Santrauka. Šiuo straipsniu bandoma identifikuoti pagrindines filosofines idėjas, kuriomis yra grindžiamas interneto tarpininkų atsakomybės už intelektinės nuosavybės teisių pažeidimus reguliavimas. Daugiausia dėmesio yra skiriama utilitarinei ir deontologinei tarpininkų atsakomybės teorijoms, pirma, nustatant pagrindines šių teorijų nuostatas; antra, nagrinėjant pagrindinius reguliavimo pavyzdžius, atitinkančius utilitarinį ir deontologinį požiūrius, kurie galėtų padėti apibūdinti bendrą tarpininkų atsakomybės modelių raidą Europos Sąjungoje. Iliustraciniais tikslais dėmesio taip pat skiriama „internetinio anarchizmo“ idėjoms, sudarančioms bendrą kontekstą kartu su utilitarinėmis bei deontologinėmis reguliavimo teorijomis ir leidžiančioms atskleisti tarpininkų atsakomybės reguliavimo raidos paveikslą per šių idėjų santykį.
Pagrindiniai žodžiai: interneto tarpininkai, netiesioginė atsakomybė, interneto platformų atsakomybė, vartų sargai, utilitarizmas, deontologija, moralinės teorijos.

_______

Received: 14/03/2024. Accepted: 19/03/2024
Copyright © 2024 Justinas Drakšas. Published by Vilnius University Press
This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Introduction

For more than three decades, regulation of Internet intermediary liability has been a battleground for different interests and economic, ethical and political beliefs about innovation, business, human rights, competition and incentives for behaviour of various actors. In order to better understand these factors, academic literature tries to derive generalised theories for describing the objectives of Internet intermediary liability regulation, attempting to explain individual regulatory ideas and grouping them into specific categories. Of these, two main approaches have recently begun to emerge in the doctrine, focusing on the ethical aspects of intermediary regulation, but also encompassing a wide range of economic, legal, sociological, and technological issues. These are the utilitarian and deontological approaches to Internet intermediary liability.

This article will attempt to deepen the understanding of each of these approaches, which may be useful for further researchers when dealing with the issue of the liability of Internet intermediaries in several ways. Firstly, it could help to properly understand the objectives of Internet intermediary liability regulation and, consequently, provide a point of reference for assessing whether a particular proposed or already adopted legislation is in line with the raised expectations. Secondly, it would attempt to enable research on questions of whether there is a need for regulation of the liability of Internet intermediaries at all and, if so, what form such regulation should take. And thirdly, it would provide an additional tool for the interpretation of the relevant legal rules on intermediaries.

In addition, the paper will also propose to identify a third approach to the liability of Internet intermediaries, which is sometimes referred to in academic literature as “anarchic” approach. In contrast to the theories mentioned above, ideas related to this approach do not focus so much on the ethical evaluation of intermediaries’ activities, although moral questions regarding Internet intermediaries also remain relevant for them. Given the historical importance of the “anarchic” approach to the development of the regulation of the Internet, it will be discussed alongside the other approaches identified in the paper, although this will be done mainly for contextual purposes, to illustrate the evolution of Internet regulatory ideas that eventually developed into the theories dominating the sphere of intermediary legislation today.

In the light of the aim of the study, the objectives of the paper are: firstly, to identify the main ideas behind the “anarchic”, utilitarian and deontological approaches to the liability of Internet intermediaries; and secondly, to identify the main examples of regulation that are consistent with the utilitarian and deontological approaches, which could help to describe the development of attitudes to intermediary regulation that took place overtime.

The object of the study is the “anarchic”, utilitarian and deontological approaches to liability (and where relevant, responsibility) of Internet intermediaries, their founding attitudes and development. For the purposes of the study, it will be limited to the regulation adopted at the regulatory level of the European Union, using it as an example to illustrate the operation of the different theories, without the ambition to classify all legislation on intermediary liability in different jurisdictions according to these approaches.

Although there are many works in the academic literature that analyse the objectives of the liability of Internet intermediaries, analysis is usually carried out through a slightly different prism than the one proposed by this paper. First of all, the current doctrine mostly focuses on identifying the interests of subjects involved in online information exchange system (rightsholders, Internet users, content platforms, etc.) and the analysis of the relevant legislation that reflects those interests. The objective of regulation in this case is seen as protecting one or another of these interests, redefining their balance and moving towards a more favourable position for one or another subject. A discussion of the utilitarian and deontological approaches could open up an alternative way of looking at the objectives of a given regulation, by examining issues of liability through the spectrum of the action-consequence dichotomy. In this way, this paper will contribute to the debate fostered by authors such as Giancarlo Frosio (2017), Martin Husovec (2020), Marcelo Thompson (2016), Jaani Riordan (2016), whose works have respectively been of particular importance for the preparation of the present paper.

1. Internet anarchism

Internet intermediaries were not always as heavily regulated as we know them to be today. In fact, when discussing the history of the regulation of the liability of Internet intermediaries, academic literature often identifies as many as four stages in the development of the regulation of the liability of Internet intermediaries. These are: first, the disintermediation phase, in which “cyberspace” and its actors were treated as an unregulated anarchy of open communication; second, the protection phase, in which territorial content regulation coagulated and intermediaries acquired immunity from certain forms of liability through a patchwork of safe harbours; thirdly, expansion period, in which new liability rules developed in national and international institutions; and, fourthly, balancing stage, in which the clamour for stronger enforcement encountered limitations in the fundamental rights of intermediaries and users (Riordan, 2016, p. 4–5). These stages mark the actual development of the regulation of Internet intermediaries, moving from a period of total non-regulation to a period of review and improvement of the already established liability framework for Internet intermediaries. This evolution, however, could also be looked at in a different way, viewing it through the respective theories that prevailed at one time or another, which could indicate not the factual state of legislation at a particular period, but rather the prevailing attitudes or ideas in it, which were the basis for adopting or modifying regulation at the time. In particular, the regulatory development of Internet intermediaries could be proposed to be divided into three phases through this prism: first, “Internet anarchism”, a period in which the uniqueness of the Internet as a largely unregulated space and the possibility of an online decentralisation that would leave little room for intermediaries to operate at all was believed in; second, “Internet utilitarianism”, a period in which it was observed that, once various Internet intermediaries had begun their activities, it was necessary to impose a set of rules of liability (or to grant immunity) in order to achieve certain benefits for the society; and, thirdly, the spread of deontological regulation, which has led to a more widespread declaration of the moral obligation of Internet intermediaries not to facilitate online infringements. Just like the actual stages of regulation, which, as Jaani Riordan points out, do not have clear dividing lines (Riordan, 2016, p. 4–5), these conceptual phases overlap and cannot be completely separated from each other. Even though some ideas held relative dominance during one stage or another, they did not erase others and were often combined to achieve specific regulatory results.

Despite the blurring of the boundaries between these different periods, it is fairly safe to say that the evolution of the regulation of the liability of Internet intermediaries began, quite obviously, with a period of non-regulation. This first phase is often referred to as the period of decentralisation or “Internet anarchism”, because it is characterised by a particularly low degree of legislative intervention and a widely held view that online activities could be largely uncontrolled. This phase of “online anarchy” was only natural, as, after the emergence of the Internet, it took a long time for the legislators, doctrinal authors and various interest groups (as well as ordinary Internet users) to understand this new technological phenomenon and its impact on society, and then to decide to enact (or to suggest to enact) legislation of some kind, that would target entities providing services linked to Internet networks. Some authors writing during this period questioned the right of national legislators to intervene in the Internet as a space without national borders (Post, 2001), while others questioned how existing law could apply at all to the Internet as a completely unique and previously unseen intangible phenomenon, pointing out that law will persist, but it will not, could not, and should not be the same law as that applicable to physical, geographically-defined territories (Johnson, Post, 1996, p. 1402). Still others have argued that the Internet could potentially not require legal regulation (at least in its traditional form), as regulatory outcomes can be achieved through software code (code, too, can be law), which in turn would be regulated, at most, by free market forces (Lessig, 2006, p. 122). Separate mention can also be made of declarations that traditional legislators did not have the “moral” right to regulate the Internet, which was seen as a space free of “tyranny”, dedicated to completely free self-expression and information dissemination (Barlow, 1996). Such views arose because, at the dawn of the Internet, many believed that the purpose of the Internet was to help combat censorship, centralised authoritarian monitoring, control or surveillance (Kovacs, 2001, p. 756). And even if it could somehow have been agreed that the legislators had the moral right to regulate the Internet, it was assumed that they would have no means of ensuring that the legal norms governing it would be enforced (Barlow, 1996).

The argument that the legislator cannot be justified in interfering in people’s online activities on ethical or moral grounds will be the focus of the second and third parts of this paper. However, with regard to the arguments concerning the “incapability” of legislators to regulate the Internet through legal norms (i.e. arguments that legislators not only should not but, in fact, could not regulate the Internet), it can be mentioned now that, despite the opinions or concerns expressed, the authors and individual interest groups who had an “anarchic” vision of the Internet must have been disappointed. In the long run, it has become clear that the Internet can be regulated easily enough and is not as unique, as some commentators had thought, to be immune from the regulatory effects of law. This was mainly due to the emergence of a large number of intermediaries on the Internet that could provide services to individual users in a centralised way, without the users having to traverse through the vastness of the Internet to find the information or services they want to acquire (referring, of course, to services provided by search engines, content storage platforms, online libraries or marketplaces, etc.), thus reducing the effort and cost of exchanging information. These intermediaries operated like any other normal business entities, having operation centres in physical space from which they could host, maintain and develop their online data centres. Accordingly, they could also be subjected to the coercive apparatus of the state to enforce compliance with one or another regulation1, which is why most of the assertions that the Internet is not subject to state regulation (or at least, its enforcement) have been proven false2. By controlling the flow of information exchanges, Internet intermediaries could be at the hands of legislators, effectively influencing the activities of the users of their services. Moreover, it has been observed that the Internet has not only enabled the achievement of previously lauded and cherished goals such as the rapid exchange of information and the development of self-expression, but has also opened the gates to the rapid spread of infringements of various kinds. This also made it possible to speak of a greater need to regulate the activities of Internet intermediaries, which was an important influencing factor in the adoption of national rules on Internet intermediary liability (or application of older rules trough case-law), or, on the EU level, adoption of legislation such as Directive 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal aspects of information society services, in particular e-commerce, in the internal market (“the E-commerce Directive”), which establishes the conditions under which Internet intermediaries can avoid liability for third-party infringements3.

However, even after the initial optimism about the existence of the Internet as a space beyond the reach of legal regulation has disappeared, the view that the Internet should remain untouchable by legislators because of their deliberate choice not to regulate relations linked to the use of Internet networks has persisted. Some authors still express frustration with the centralisation of the Internet, pointing out that, in the beginning, the Internet was a ripe environment for invention and innovation and generated tremendous benefits for the entire world, but over time, the influx of money and power began to reward selfish behaviour more and more, breaking open the Internet’s utopia and leading to crime, censorship, and fights over control (Riley, 2013, p. 1). Many individual commentators have warned of the potential censorship that is (or could be) exercised by Internet intermediaries, thus limiting freedom of information and self-expression (Kreimer, 2006, p. 100), and have often shared warnings of the oppression of power exercised through the restriction of information exchange (Anderson, Rainie, 2020). In light of these warnings, and driven by the desire to resist control (which in individual cases is also mixed with the private interests of particular groups), calls are being made to decentralise the Internet again (Chohan, 2022, p. 10). To this end, new technical tools are being developed and new projects are being launched using technologies such as Blockchain and Tor networks (see, for example, Farmer, 2018). As a result, there are frequent predictions that the days of centralised Internet will soon come to an end (Ormandy, 2018), and that, accordingly, the capability of legislators to regulate the Internet will also be over, or, at least, their powers in this regard will be extremely limited.

Of course, at least for the moment, such predictions intuitively still sound quite bold. While the possibilities of being untraceable online and sharing information through non-centralised channels are certainly expanding, it is undeniable that the vast majority of users still access information through Internet intermediaries (such as online content-sharing service providers, online libraries, Internet marketplaces, etc.) (Stjernfelt, Lauritzen, 2020, p. 217). The promoted changes of decentralisation would thus require a particularly significant change in consumeristic habits of everyday Internet users, providing sufficient incentives to move away from the convenience of centralized services, by perhaps placing privacy or free flow of information above the benefits provided by intermediation and content mediation safety. This shift, however, still has not occurred and, at least at the present day, convenience seems to triumph the promoted benefits of “anarchy”, even as significant as they might seem to be. It thus renders the ambition to make the Internet once again a decentralised “wild west” and place where “total anarchy” could take control not yet fully realisable. Although the future remains, as always, open.

2. Utilitarianism as basis for the regulation of liability of Internet intermediaries

In order to curb online infringements, policymakers nowadays often turn to Internet intermediaries that enable their service users to share information in various Internet mediums and infringe the rights of others. These intermediaries, while not directly contributing to (or initiating) infringements, are often held responsible for the facilitation of their service users’ actions, and are consequently held at least partially liable for mismanaging the prevention of infringements and for failing to control the flow of information online. Very often, this shift of responsibility (and according legal liability) from the direct infringers to intermediaries is based on the view that Internet intermediaries - such as certain content sharing platforms - have the best tools available to monitor the activities of their users, to track illegally shared information and to, finally, remove it, if this seems necessary. After all, they are the designers of the heart valves through which the lifeblood of our information environment flows and actions they take or refrain from taking can fundamentally alter medium and message, structure and content of information we impart and receive (Thompson, 2016, p. 798). Thus, it seems only natural when the regulatory policy and scholarly literature on the regulation of intermediaries’ activities display particular concern with the factual outcomes that these actions enable (Ibidem). This sort of consequentialist, or, more specifically, utilitarian approach is based on the notion that liability should be imposed only as a result of a cost-benefit analysis, which is especially relevant in case of dual-use technologies that are deployed both to infringe others’ rights and facilitate socially beneficial uses (Frosio, 2017, p. 25). As would be the utilitarian maxim, under this view Internet intermediary liability should be tackled in a way that supports the generation of the maximum quantity of well-being, happiness and utility, or, in the famous words of Jeremy Bentham, the greatest good for the greatest number. The proponents of such approach would suggest to balance and recognize different societal interests, such as incentivising intermediaries for innovation, promotion of business, protecting intellectual property, free speech and self-expression, while trying to find the golden point in the middle out of all of these interests (which might be in conflict) that would allow the maximum good for the maximum number in a given society. Needless to say, this sort of balancing is incredibly difficult, as it is near impossible to satisfy all the possible interests of different groups at once, and the golden point can seldom be found. If some values, such as author’s rights, are protected too rigorously, though the means of, for example, automatic blocking mechanisms, achieving other virtuous goals, such as free dissemination of information, might be less possible. The same is true from the opposite spectrum as well, when, if no protection through liability is imposed, authors might not be as incentivised to create new works that may be considered original, trademark owners would suffer economic losses due to unlawful uses of their trademarks and patent owners might not want to invest into further inventions. As such, even though it is commonly understood that the goal of regulation should be the maximum good, it is hard to define which values should be “more valuable” and be given priority when discussing regulation. Accordingly, there exist many different approaches to Internet liability regulation that fall under the large umbrella of utilitarianism, which can hold different priorities when considering these questions. Some authors, such as Jonathan Zittrain, who phrases the utilitarian maxim of “maximum good for the greatest number” in the context of Internet content to maximisation of the overall capacity [of the Internet grid] to produce generativity (Zittrain, 2006a, p. 1980) (essentially putting the generation of information and creativity at the forefront of protectable values) cautions against liability imposed on intermediaries for the actions of their users, for the fear of over-blocking of content that would follow suit because of intermediaries trying to escape liability and instead removing any kind of information that would have the slightest hint of illegality (Zittrain, 2006b, p. 262). Other authors, such as Doug Lichtman and Eric Posner, who also emphasise the need of cost and effectiveness in regulatory approaches (at least in the sphere of copyright) (Lichtman, Posner, 2006, p. 256–259), suggest expanding the limits of Internet intermediary liability while protecting the interests of rightsholders that could more easily defend their rights while addressing their claims at intermediaries, rather than individual infringers, who might be too difficult to track (Ibidem). Even other authors suggest at least considering forsaking liability rules altogether and instead establishing some sort of alternative compensation mechanisms for rightsholders, this way trying to balance both the free flow of information and fulfilment of rightsholders interests through compensation for the use of the intellectual property as well (Mizaras, 2019). Even though all of these authors consider different values as priorities in intermediary regulation (as well as different ways to achieve them), they also hold the effects or rather - consequences - that specific regulatory suggestions could bring about, as the main guideline when discussing regulation, which helps to separate it from a more deontological approach which will be discussed further.

It should stand to mention separately that, out of all these different utilitarian approaches to Internet intermediary liability, none, however, are as influential as the concept developed by Professor Reiner Kraakman almost four decades ago, marking the beginning of what is now widely referred to in academic literature as the “Gatekeeper” theory. In his article “Gatekeepers: The Anatomy of a Third-Party Enforcement Strategy”, Professor Kraakman provided an analysis of the application of liability for third-party misconduct by private entities that are able to disrupt misconduct by withholding their cooperation from direct wrongdoers (thus, becoming “gatekeepers”) and compared this liability to other enforcement strategies. Of particular importance in this theory is the cost-benefit analysis, in light of which Prof. Kraakman identifies four criteria that can help to assess when gatekeeper liability could be applied: firstly, serious misconduct that practicable penalties cannot deter; secondly, missing or inadequate private gatekeeping incentives; thirdly, gatekeepers who can and will prevent misconduct reliably, regardless of the preferences and market alternatives of wrongdoers; and, fourthly, gatekeepers whom legal rules can induce to detect misconduct at reasonable cost (Kraakman, 1986, p. 61). Although Prof. Kraakman’s theory was not directed at Internet intermediaries, but primarily spoke of the liability of legal and financial service providers, its internal logic can be perfectly applied in the Internet context as well, with legislators trying to strike the right balance between applying liability in order to improve the position of rightsholders, and in certain cases excluding intermediaries from liability, in this way giving priority to other values, such as the freedom of economic activity of intermediaries as business subjects (which would undoubtedly be constrained to some extent by setting limits on the extent to which Internet intermediaries should be liable for the actions of the users of their services too broadly).

Good examples of this attempt to balance different values through legal norms are the E-Commerce Directive and Regulation (EU) 2022/2065 of the European Parliament and of the Council of 19 October 2022 on a single market for digital services and amending Directive 2000/31/EC (“the Digital Services Act”), which are the central pieces of legislature in the field of Internet intermediary liability in the EU. Articles 12–15 and 4–8 of these legal acts respectively provide for “safe harbours”, which establish the conditions under which Internet intermediaries providing “simple conduit”, “caching” and information hosting services may be exempted from liability for online infringements by third parties. This “safe harbour” regulation takes into account the fact that the imposition of an absolute control regime, where intermediaries would be liable for any infringing information transmitted or stored, would effectively mean a factual obligation to monitor all information traffic, which would impose an extremely heavy financial burden on intermediaries, as well as a potential risk of removing legitimate content, where intermediaries would remove even slightly suspicious content that might actually turn out to be legitimate, in order to evade liability. Maximum control is inefficient simply because it is too expensive (Kleve, Mulder, 2005), both in terms of the financial burden on intermediaries (potentially by shutting down business altogether) and the price paid by Internet users in terms of reduced potential to share information. Accordingly, Internet intermediaries of this nature are not currently required to monitor all information transmitted (Article 15 of the E-Commerce Directive, Article 8 of the Digital Services Act), just as hosting service providers are not required to check the legality of the information stored or transmitted on their own initiative, having to remove it only after they become aware (are informed) of the illegal information (see Article 14 of the E-Commerce Directive, Article 6 of the Digital Services Act). Failure to comply with this obligation would exclude the application of the safe harbours and could potentially render the intermediary concerned liable for the acts of third parties, if the national law provides that the conditions for liability are met. It could be argued that this fulfils the fourth condition of the above-mentioned model set by Prof. Kraakman, that liability could be imposed, but only where the requirements do not impose a disproportionate burden on intermediaries. The other three conditions are dictated by the reality of the conduct of infringements and market forces: firstly, they are carried out despite the possibility of direct liability for primary infringers (which is inefficient); secondly, by being able to avoid the obligation to play the role of Internet “policeman”, intermediaries presumably would not undertake this role without the intervention of the legislator; and, thirdly, the centralisation of the Internet means that intermediaries are able to prevent (or to address post factum) the vast majority of infringements4.

However, even without linking the model of the E-Commerce Directive and the Digital Services Act to Prof. Kraakman’s theory, it is possible to discern strong utilitarian motives underlying this type of regulation in broader terms. For example, it could be claimed that the indicated legal acts display a strong expression of a need to balance different societal values, the internal logic of the “safe harbours” clearly reflects the importance of the cost-benefit assessment in the structure of such a model, and the main motivational weight of the regulation is given to the possible consequences that would occur in the absence of liability rules. In this way, the “safe harbour” legislation could be classified as a more utilitarian legislation. These legal acts, being the central and the most important sources of legislation regulating the liability of intermediaries at the EU level, accordingly also shift the whole system of liability application to a more utilitarian model. Furthermore, as the Digital Services Act continues the ideas set out in the E-Commerce Directive (albeit with some specificities), utilitarian approach could be seen to continue to be dominant. Nevertheless, this does not mean that certain deviations do not occur or are not possible. As discussed below, examples of legislation that can be described as more “deontological” are increasingly relevant to the overall development of the legal liability (and responsibility) model for Internet intermediaries. And the beginning of their prominence is just starting.

3. Deontological approach to liability

Legislation providing for “safe harbours” is not the only example of regulation based on utilitarian considerations. Indeed, almost every piece of legislation that passes through the normal legislative process is assessed on the basis of the real economic, social, criminological and similar effects that it might have on social relations and is designed to achieve a certain beneficial outcome. Where regulation touches on a number of separate values that cannot be protected without restricting others, most legislators also try to reflect this in one form or another in the legislative process. However, as mentioned earlier, assessing which of the competing values should be given priority is difficult, especially given that not all socially recognised goods can be easily converted into economic parameters. It is therefore often the case that societal acceptance, that certain goods should be protected because they are intrinsically valuable, would be considered a sufficient motive to impose regulation. For example, it is often considered immoral and unjust to allow others to commit violations of the law. Accordingly, the cases where Internet intermediaries do not stop online infringements, especially when they profit by doing so through advertising revenues, would also be considered immoral and wrong (Spinello, 2014, p. 312), which would potentially suggest a possible need to regulate their liability (should we consider that (in)action of intermediaries is sufficiently vicious to merit regulation not only in terms of morality, but also in terms of law). This approach, which emphasises the immorality or injustice of certain actions in themselves (with less weight given to the possible consequences), could, in a sense, be called “deontological”5, i.e. view that actions should be evaluated as such regardless of the consequences they may have.

Although the academic literature is somewhat less rich in works based on this approach than utilitarian theories, a number of authors have also devoted research to potential deontology-based liability systems for Internet intermediaries. As an example, Marcelo Thompson’s comprehensive paper “Beyond Gatekeeping: The Normative Responsibility of Internet Intermediaries” proposes a consideration of a liability framework based not on the assessment of the consequences of Internet intermediaries’ actions taken in order to control the availability of illegal information (e.g. by looking at whether or not they have removed the infringing information after notification of potential infringement by rightsholders), but on the reasonable care given to the very thought processes by which intermediaries choose their reasons for action (Thompson, 2016, p. 848), in other words, by assessing whether intermediaries have made sufficient and reasonable efforts to remove the infringing material (irrespective of whether or not the infringement is removed). As Marcelo Thompson himself argues, this system would not be utilitarian, as it would give weight to the analysis of the actions of the intermediaries, rather than their effects (Ibidem). Other authors, while not proposing new systems of liability, also often hint that actions by which conditions for the commission of wrongdoing are created should be assessed as they are in themselves (rather than through the lenses of consequences) (see, for example, Spinello, 2005, p. 127, where it is referred to the wrongfulness of acts as such). In fact, it could be argued that many commentators on issues relating to Internet intermediaries, while not necessarily expressing support for a deontological position, would support the view that the actions of intermediaries in helping third parties to infringe the rights of others are wrong or immoral as such. As Folkert Wilman observes, the feeling that intermediaries should act more responsibly is broadly shared (Wilman, 2020, p. 379), but almost each and every attempt to translate that feeling into a legally binding text proves highly controversial (Ibidem).

The Directive (EU) 2019/790 of the European Parliament and of the Council of 17 April 2019 on copyright and related rights in the Digital Single Market amending Directives 96/9/EC and 2001/29/EC (hereafter – Digital Single Market Directive) could arguably be considered as one of the examples of these controversial attempts to implement more deontological ideas into intermediary liability regulation (or, at the very least, it could be considered a conceptually mixed example). Specifically, this Directive establishes a model of intermediary liability that goes beyond the limits of the “safe harbour” system (see Article 17 (3), stating that “safe harbours” are not applied for online content sharing-service providers6 that enable their users to share copyrighted works online), and de facto requires such intermediaries to implement online filtering mechanisms by requesting that online content sharing-service providers would ensure that illegal content would be removed not only after the first upload, but after all the additional uploads as well (thus requiring the use of content identification and removal technologies) (see Article 17(4), which requires online content sharing-service providers to act expeditiously, upon receiving a sufficiently substantiated notice from the rightsholders, to disable access to, or to remove from their websites, the notified works or other subject matter, and make best efforts to prevent their future uploads). Such regulation puts aside the fear of over-blocking that automatic filtering mechanisms might invoke (Hilty, Bauer, 2017) and arguably concentrates much more on the role of intermediaries to combat illegal activities online and makes them take on much bigger responsibility for the actions of their users, reflecting the deontological notion that intermediaries should stop the illegal, and thus, immoral, actions of their users no matter the consequences that this might bring about.

Nevertheless, the Digital Single Market also states that it seeks a balance between the interests of rightsholders and the users of their works (Recital 6) and, as way to implement this goal, establishes such exceptions and limitations to copyright as permitted use in the case of caricature, parody, pastiche, quotation, criticism and review (Article 17 (7)). It is also limited in its application in regards to the platforms that need to adhere to the indicated rules (making certain exceptions for new online content-sharing service providers) and has some other safeguards for limiting its overextension, making it a more conceptually mixed legislation. Furthermore, it could be argued that the more rigorous defence of rightsholders’ interests does serve an overall more beneficial purpose of fostering original creation even if it can result in some cases of over blocking, in this way justifying the Digital Single Market Directive through the lenses of utilitarianism as well. However, this acceptance of the possible risks of over blocking is in itself and indicator of a certain shift from the purely utilitarianism based status quo and only time will tell if these risks will crystallise once the new model has had enough time to be tested.

Other signs of a shift from a utilitarian system to a more deontological thought model could also be seen in tools for increasing responsibility beyond the normal liability rules (typically through self-regulation), such as codes of conduct, as well as soft law standardisation, application of self-imposed filtering mechanisms, three-strike schemes, voluntary online search manipulation, follow-the-money strategies and private denial of service content regulation7 (Frosio, Husovec, 2020, p. 615). All these measures are united by the idea of the “enhanced responsibility”, whereby Internet intermediaries are held responsible for the distribution of illegal information and the actions of their users even beyond the normal scope of the safe harbour system. Many of these measures are the result of the soft law regulatory incentives to make the intermediaries take on a higher burden of responsibility for online content (for example, see the Communication from the EU Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of the Regions on Online Platforms and the Digital Single Market, COM/2016/0288 final), where the regulator states that intermediaries should act in a way to prevent infringements (and paying less regard to balancing different interests). Some measures are imposed, however, as a direct intervention by the legislator, thinking of online filtering tools as the ultimate response to combating illegal activities online. Nevertheless, whatever the form they take, these tools of regulation could be seen as a departure from a utilitarian framework, as they give particular importance to the proactivity of Internet intermediaries, who are encouraged, and, at times, required, to take voluntary actions to control the flow of information online despite the possible negative outcomes of their intervention without any requests of rightsholders. This departure marks a philosophical shift away from the idea of safe harbours system, which attempts to balance the protection of the freedom of information and rightsholder interests (and also expresses the idea that Internet intermediary passivity and reduced burden of responsibility may have positive effects as well).

The Digital Single Market Directive and other tools marking this shift are often criticised because of the threat of infringements of the freedom of speech and self-expression (for example, see the arguments raised by Poland in Court of Justice of the European Union 26 April 2023 Decision, case C-401/19, Republic of Poland v European Parliament and Council of the European Union), fear of the increased significance of private ordering, whereby most of the decisions related to moderation of content would be taken by private companies without public oversight, the risks posed by algorithms that would inevitably have to be used in order to better control online information (at times failing to recognise and legitimate information) and the possible violation of the prohibition of the general monitoring duty (Kuczerawy, 2020, p. 540). Nevertheless, it should not be immediately accepted that the measures referred to above should be completely rejected and that the turn from more utilitarian regulation should be reversed. The argument, that facilitating infringement should be considered unethical, works as a good incentive to empower Internet intermediaries to help protect the rightsholders’ interests, which should be considered just as legitimate as the interests of the users of the intermediaries’ services and the interests of the intermediaries themselves. After all, the vast majority of information on the Internet passes through a handful of giant technology companies, which, by the virtue of their effective position of power on the Internet, should also assume a certain amount of authority in the fight against these infringements. These are independent ethical agents who are not isolated in the information exchange system from rightsholders and the general public, and their decisions to help or not to help in the fight against infringements also carry a moral weight that can lead to them being judged through the prism of legal responsibility and liability (once we agree that the actions of the intermediaries are ethically important enough to be regulated). And while the risks associated with the above-mentioned restrictions on information dissemination, the right of expression remain, it may be a little premature to proclaim the death of freedom of expression on the Internet. Should these risks crystallise, the utilitarian approach, based in particular on rational calculation, still has a strong basis in the EU regulation on the liability of Internet intermediaries and can always be turned back to. And it is likely to remain there, at least in the short term.

Conclusions

1. The centralisation of the Internet leaves little room for “online anarchism” and its related ideas, although some movement calling for a return to the days of Internet decentralisation can be detected.

2. Legislation regulating safe harbours, such as the E-Commerce Directive or the Digital Services Act, to this extent is most consistent in its conceptual principles with the utilitarian approach to the liability of Internet intermediaries. Such regulation seeks to balance the protection of different types of values and emphasises the consequences of imposing liability, or granting immunity from liability, not only in regards to defending rightsholder interests, but also for enjoyment of freedom of information and incentivising Internet intermediaries to conduct further business.

3. Recent (self)regulatory initiatives, such as content filtering mechanisms, mark a conceptual shift away from a more utilitarian regulation towards deontological approach to the liability of Internet intermediaries. These regulatory examples raise some concerns about limitation of the dissemination of information, freedom of expression and speech. Nevertheless, an immediate return to a purely utilitarian system could be not as desirable as it first appears because of the lack of enforcement of the interests of rightsholders under such system, which is, presumably, what led to the spread of a deontological approach in the first place. The complexity of these issues would suggest a cautious approach to further legislation, having utilitarian regulation as the central column for the whole system, in case “deontological” initiatives fail.

Bibliography

Legal acts

European Union legal acts

Directive (EU) 2000/31/EC of the European Parliament and of the Council of 8 June 2000 on certain legal aspects of information society services, in particular electronic commerce, in the internal market.

Directive (EU) 2019/790 of the European Parliament and of the Council of 17 April 2019 on copyright and related rights in the Digital Single Market and amending Directives 96/9/EC and 2001/29/EC.

Regulation (EU) 2022/2065 of the European Parliament and of the Council of 19 October 2022 on a Single Market For Digital Services and amending Directive 2000/31/EC (Digital Services Act).

Case law

Court of Justice of the European Union 26 April 2023 Decision Republic of Poland v European Parliament and Council of the European Union, C-401/19, EU:C:2022:297.

Specialised literature

Fonseca, G. Z. G. (2016). Intermediaries Liability for Online Copyright Infringements: The Duty to Cooperate Under E.U. Law (February 28, 2014). WIPO Academy, University of Turin and ITC-ILO, Research Papers Collection, 2013–2014.

Frosio, G. (2016). Internet Intermediary Liability: WILMap, Theory and Trends. Indian Journal of Law and Technology, 13(1), Article 2. https://doi.org/10.31235/osf.io/3n5h4

Frosio, G., Husovec, M. (2020). Accountability and Responsibility of Online Intermediaries. The Oxford Handbook of Online Intermediary Liability. Oxford University Press.

Hilty, R., Bauer, A. (2017). Use of Protected Content on Online Platforms. Modernisation of the EU Copyright Rules Position Statement of the Max Planck Institute for Innovation and Competition. Max Planck Institute for Innovation & Competition Research Paper, No. 17–12.

Johnson, D. R., Post, D. (1996). Law and Borders: The Rise of Law in Cyberspace. Stanford Law Review, 48(5), p. 37.

Kovacs, A. Z. (2001). Quieting the Virtual Prison Riot: Why the Internet’s Spirit of ‘Sharing’ Must Be Broken. Duke Law Journal, 51(2), 753-785

Kraakman, R. H. (1986). Gatekeepers: The Anatomy of a Third-Party Enforcement Strategy. Journal of Law, Economics, & Organization, Spring, 2(1), 53–104. https://doi.org/10.1093/oxfordjournals.jleo.a036906

Kreimer, S. F. (2006). Censorship by Proxy: The First Amendment, Internet Intermediaries, and the Problem of the Weakest Link. University of Pennsylvania Law Review, 155(1), 11-101. http://dx.doi.org/10.2307/40041302

Kuczerawy, A. (2020). From ‘Notice and Take Down’ to ‘Notice and Stay Down’: Risks and Safeguards for Freedom of Expression. Oxford Handbook of Online Intermediary Liability.

Lessig, L. (2006). Code and Other Laws of Cyberspace. 2nd Ed. Basic Books.

Lichtman, D., Posner, E. (2006). Holding Internet Service Providers Accountable. Supreme Court Economic Review, 14, 221-259.

Mizaras, V. (2009). Autorių teisė: monografija. T. 2. Vilnius: Justitia.

Riordan, J. (2016). The Liability of Internet Intermediaries. Oxford University Press.

Spinello, R. A. (2005). Secondary liability in the post Napster era: Ethical observations on MGM v. Grokster. Journal of Information, Communication and Ethics in Society, 3(3), 121-130. https://doi.org/10.1108/14779960580000266

Spinello, R. A. (2014). Intellectual Property: Legal and Moral Challenges of Online File Sharing. Ethics and Emerging Technologies (Ronald L. Sandler (ed.), Palgrave Macmillan).

Stjernfelt, F., Lauritzen, A. M. (2020). Trust Busting the Tech Giants? Your Post has been Removed. Springer Cham.

Thompson, M. (2016). Beyond Gatekeeping: The Normative Responsibility of Internet Intermediaries. Vanderbilt Journal of Entertainment & Technology Law, 18(4).

Wilman, F. (2020). The Responsibility of Online Intermediaries for Illegal User Content in the EU and the US. Edward Elgar Publishing Limited.

Zittrain, J. L. (2006a). The Generative Internet. Harvard Law Review, 119(7), 1974-2040.

Zittrain, J. L. (2006b). A History of Online Gatekeeping. Jonathan Zittrain, A History of Online Gatekeeping, 19 Harvard Journal of Law and Technology 253.

Other sources

Anderson, J., Rainie L. Concerns about democracy in the digital age [online]. Available at: https://www.pewresearch.org/internet/2020/02/21/concerns-about-democracy-in-the-digital-age/ [accessed at 23rd August 2023].

Barlow, J. A Declaration of the Independence of Cyberspace [online]. Available at: https://www.eff.org/cyberspace-independence [accessed at 23rd August 2023].

Chohan, U. W. Web 3.0: The Future Architecture of the Internet? [online]. Available at: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4037693 [accessed at 23rd August 2023].

Farmer, C. (2018). Five projects that are decentralizing the web in slightly different ways. [online]. Available at: https://medium.com/textileio/five-projects-that-are-decentralizing-the-web-in-slightly-different-ways-debf0fda286a [accessed at 23rd August 2023].

Kleve, P., Mulder, R. V. D. (2008). Code is Murphy’s Law [online]. Available at: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1138659 [accessed at 23rd August 2023].

Mizaras, V. ES autorių teisių bendroje skaitmeninėje rinkoje direktyva: politikų bei lobistų kompromisas, galimai tapsiantis standartiniu eksperimentu visuomenei [online]. Available at: https://www.teise.pro/index.php/2019/03/31/v-mizaras-es-autoriu-teisiu-bendroje-skaitmenineje-rinkoje-direktyva-politiku-bei-lobistu-kompromisas-galimai-tapsiantis-standartiniu-eksperimentu-visuomenei/ [accessed at 23rd August 2023].

Ormandy, R. The Death of Centralized Web [online]. Available at: https://romanor.medium.com/the-death-of-centralized-web-d26d41a3afb [accessed at 23rd August 2023].

Post, D. (2001). Should the Internet be Regulated? Contra: Why the Internet Should not be Regulated [online]. Available at: https://www.cesifo.org/en/publications/2001/article-journal/should-internet-be-regulated-contra-why-internet-should-not-be [accessed at 23rd August 2023].

Riley, M. C. (2013). Anarchy, State, or Utopia? Checks and Balances of Power in Internet Governance [online]. Available at: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2262055 [accessed at 23rd August 2023].