Part III - The Scope and Mechanisms of International Intellectual Property Treaties

8 Universalism in International Copyright Law as Seen through the Lens of Marrakesh

Graeme B. Dinwoodie

Table of Contents

1. A. Introduction 236

2. B. Universalism in International Copyright Law 237

3. C. Mechanisms of Universalism 241

4. D. Marrakesh as Inflection Point … (but on Universality)? 248

   1. I. Exceptions and Users’ Rights 249

   2. II. Mandatory Exceptions 250

   3. III. Human Rights 254

   4. IV. Scope of Application: Domestic Copyright Law 255

   5. V. Cross-border Exchange 257

5. E. Conclusion 259

A. Introduction

The Marrakesh Treaty to Facilitate Access to Published Works for Persons Who Are Blind, Visually Impaired or Otherwise Print Disabled (hereinafter “Marrakesh Treaty”) was adopted in 2013 and came into effect in 2016.¹ It has already been ratified by more than a hundred countries.² It mandates exceptions in national copyright laws to ensure access to special-format copies of books, including audio books and digital files, for people who are blind and visually impaired, and it establishes mechanisms to facilitate cross-border access to those formats.³ The treaty had an important effect even in countries that already had exceptions for the blind on the statute book. For example, the United States legislation had – for twenty years – contained limitations and exceptions that allowed certain authorized entities to provide published works in accessible formats (e.g., Braille, audio, and large print) to those who have print disabilities. However, in 2018, Congress revised that provision and inserted a new section (Section 121A) allowing authorized entities (typically non-profits operating to assist visually impaired people) to both export and import works in accessible formats between the United States and other countries that have signed the Marrakesh Treaty.⁴

The Marrakesh Treaty was heralded by a wide range of commentators as a step in a new direction in international copyright law. Justin Hughes, who headed the US delegation to the diplomatic conference at which the treaty was adopted, described it as a “new, quite extraordinary, multilateral” instrument.⁵ Marketa Trimble, who has written about various complexities within the treaty, noted that “the Marrakesh Treaty is a different species of international IP treaty.”⁶ Mihály Ficsor reported its frequent description as “exceptional, unique and even historical.”⁷ And numerous other participants and commentators have described the conclusion of the treaty as “the Marrakesh Miracle.”⁸

This chapter assesses the Marrakesh Treaty’s significance through the lens of a consideration that has been an implicit – and sometimes explicit – part of international copyright debates since before the adoption of the Berne Convention in 1886, namely, the push for a universal copyright law. I explore whether and in what ways the Marrakesh Treaty might alter our understanding of the notion (and mechanisms) of universalism in international copyright law. To facilitate that assessment, Part B of the chapter considers the universalist aspiration in the development of international copyright law. Part C sets out the range of mechanisms by which universalism might now be being pursued in copyright law. And Part D explains the conceptual features of the Marrakesh Treaty, touching on the universalism debate that might cause it to signal a change in the international copyright landscape. The treaty was a milestone in international copyright law, as many scholars have argued. But its importance might lie in several aspects that are not often highlighted.

B. Universalism in International Copyright Law

In his groundbreaking study of the first hundred years of the Berne Convention, Sam Ricketson suggested that the structure and content of the Berne Convention in 1886, and the evolution of Berne ever since, have been shaped by a battle between what he called the “universalists” and the “pragmatists.”⁹ At the time of the intergovernmental meeting in 1883 to form the Berne Union, attempts were made – particularly by the German delegation – to institute an internationally uniform copyright code that would apply in each member state.¹⁰ The argument was that these universal norms would guarantee uniformity and predictability, which would result in the improved international circulation of works of authorship.¹¹

This “universalist” vision, as Ricketson termed it, did not prevail in 1886. Instead, those adhering to what he called the “pragmatist” vision carried the day, with the adoption of a system grounded on the twin pillars of national treatment and low-level substantive minima. That is, signatory states undertook to provide authors from, or works first published in, other signatory states with protection as generous as that afforded to domestic authors and works. And they agreed that their national laws would adhere to a cluster of minimum substantive standards. For example, the Convention listed the types of works that a signatory state must protect. Signatory states could offer greater protection to authors but were obliged only to satisfy the minimum levels.¹²

Under this approach, the creation of a work results in a bundle of independent copyrights in all copyright-respecting nations.¹³ The principle of territoriality that this approach endorses has a solid conceptual foundation.¹⁴ As a matter of instrumental analysis, the optimal balance between incentive and access that determines the content of copyright law will necessarily vary between states of different economic, social and cultural contexts. And insofar as copyright law shapes the cultural landscape of a particular territory, sovereign states have the right to determine how that landscape should look. But this approach raises the prospect of variation in protection from one country to another.¹⁵

The battle between the universalists and the pragmatists has been a feature of the development of the Berne Convention ever since, just as much as the contest between adherents of the notion of droit d’auteur and advocates of a more utilitarian system of copyright.¹⁶ The universalists succeeded in tweaking this model throughout the twentieth century, serially revising upwards of the minimum standards, although the pragmatists pressed to maintain as much room within the system as possible to accommodate a diverse range of participating sovereign states.

But in some sense, the pragmatists were in fact universalists. Or, at least, to state it less dramatically, the so-called pragmatists pursued a vision based upon a slightly different sense of universalism. They argued that the less demanding and less prescriptive the obligations, the more countries were likely to become members of the Union and to enforce basic notions of copyright. The group of copyright-respecting nations, in which authors’ rights were protected, would thus become enlarged, and core copyright protection would be made more universal.¹⁷ That is to say, the vision of “universalism” driving the pragmatists was focused not so much on the uniformity of the content of copyright law but on the geographic scope of the international system. And the fact that Berne now has 178 contracting parties surely suggests there is some basis to that strategy.¹⁸

Jane Ginsburg, reflecting on Ricketson’s account, appeared partially to agree with this recharacterization when she differentiated between “true universality” and the “minimal universality” that the pragmatists sought.¹⁹ But there is the sense from reading Ginsburg’s account that the adoption of a more pragmatic approach is seen as an unfortunate but inevitable concession to reality.²⁰ Likewise, Silke von Lewinski in her text on international copyright law talks of the “idealistic” but “hardly practical” philosophy of universalism.²¹ Universal standards is a goal that is tempered by political reality.

In fact, a different vision of “universality” – which stresses the global reach of the basic system over the uniformity of content – might be both more idealistic and more practical. Whether this is true will depend on the “ideal” that we have in mind in constructing the international system. This is a function as much of our relative normative commitments to difference and uniformity as it is to a particular vision of the form of copyright protection. Moreover, the balance may change over time and by context. The assessment of the question of practicality, in contrast, depends on what one thinks are the best institutional and political arrangements by which to achieve the ideal – which might also be informed by our experience in pursuing convergent norms over the last century, as well as a prediction of possible new arrangements in the future. That is to say, my (arguably too semantic) assessment of Ricketson’s labels and von Lewinski’s description is premised upon a more multi-dimensional understanding of universalism, and arguably on a different sense of the normative force that universalism might possess in international copyright law.

Consideration of the extent to which we should move from the national to the universal has been an ongoing assessment for international copyright policymakers and thinkers at different staging posts on the road from Berne to Marrakesh.²² To give one notable and important scholarly assessment, in 1999, Jane Ginsburg asked at the ALAI Congress whether and to what extent we now had a supranational copyright law that had displaced national laws.²³ She started her talk by commenting that when invited to address the congress on the role of national copyright in an era of international copyright norms – which was one of the themes of the 1999 Congress – she inferred a question mark at the end of what appeared to be a declaratory statement. Ginsburg noted that her inference might be surprising, because it was a core tenet of international copyright law that we have a system of “interlocking national copyrights, woven together by the principle of national treatment.”²⁴ Yet, after an insightful tour of both TRIPS and the WIPO Copyright Treaty, as well as developments in the European Union, she concluded that “‘[i]nternational copyright’ can no longer accurately be described as a ‘bundle’ consisting of many separate sticks, each representing a distinct national law, tied together by a thin ribbon of Berne Convention supranational norms. Today’s international copyright more closely resembles a giant squid, whose many national law tentacles emanate from but depend on a large common body of international norms.’²⁵

The dynamics that Professor Ginsburg detected two decades ago have only intensified. International norms continue to grow, although not with the same pace or through the same type of broad public international instruments – TRIPS and the WIPO Copyright Treaty – that characterized the five years preceding the 1999 ALAI Congress.²⁶ And we continue on the long and perhaps never-ending march toward the universality seen as too idealistic in 1886. But perhaps a different flavor of universalism is now being offered.

C. Mechanisms of Universalism

In the concluding Part D of this chapter, I consider whether and in what ways the Marrakesh Treaty speaks to the question of universalism. But that inquiry raises some preliminary questions about what we mean by that term. Thus, in this part of the Chapter, I discuss the different mechanisms by which universalism occurs in international copyright law.

As mentioned above, the foundational choice that confronted negotiating countries in 1886 is often presented as a choice between a pluralist territorial model and a more comprehensive universal copyright law that establishes uniform standards to be applied in all adherent countries. We have tended to label the latter approach as “universalist,” but its distinguishing feature, which sets it apart from the model ultimately adopted, was its desire for substantive uniformity.²⁷

A pragmatist’s vision might have been equally universal, but their focus was on the scope of the system. Their universalist philosophy prioritized that there be an international copyright system that was as broad as possible rather than that there be an international copyright law. And this preference might be grounded as much in a concern for appropriately tailored copyright law as resignation to the second-best reality of international relations. Focusing on extending the geographic reach of the system and allowing for variation among countries might be both a differently idealistic and a more practical course of action.

But disagreement about the concept of “universal copyright law” might be even more complicated. That is to say, universalism can be gauged – and pursued – in a number of intersecting ways, in addition to uniformity of content or geographic reach of the system. For example, in terms of universality of scope, we might wish to consider not only the geographic reach of norms but the scope of application more broadly speaking.

The Berne system was designed to apply to the treatment of foreign works in international settings. Under Article 5(3) of the Berne Convention, “protection in the country of origin is governed by domestic law.”²⁸ The rights guaranteed by the Convention only apply “in respect of works for which [authors] are protected under this Convention, in countries of the Union other than the country of origin.”²⁹ If one accepts the notion of universality that drives much thinking about human rights, that might be an inadequate implementation of universal aspirations. Although it is a contested understanding, the universality of human rights is such that those principles should apply in internal settings and modulate the normal sovereignty of states in domestic matters. And Article 27(2) of the Universal Declaration of Human Rights (1948) provides that “everyone has the right to the protection of the moral and material interests resulting from any scientific, literary or artistic production of which he is the author.”³⁰

Universality might alternatively be pursued in respect of particular issues central to the international flow of copyrighted works. The territorial model results not only in variable laws but also potentially inconsistent allocation of rights.³¹ The international copyright system devotes less time to this territorially induced dilemma than international trademark or patent law. This is because the prohibition on formalities now found in Article 5(2) of the Berne Convention effectuates an immediate vesting of rights internationally, in a far more comprehensive fashion than the Paris Convention does for industrial property. However, the Berne Convention provides little guidance as to the identity of an author under public international law.³² Thus, national laws have substantial latitude as to the ways in which they approach the question of authorship. Thus, as to initial authorship, US law reflects its predominantly instrumentalist orientation by recognizing employers as authors of works prepared by employees within the scope of their employment,³³ whereas French law links ownership to the personality of the individual author by treating the employee as the author in the same circumstance.³⁴ Many countries that treat the individual employee as the author also include presumed transfers of rights in their law, although this is not the same as an allocation of authorship.³⁵

On this issue of initial ownership, universality (of a slightly different flavor) would be enhanced by ensuring a single global author or owner of a copyright work.³⁶ This could be achieved by a lex originis rule as the choice-of-law rule for initial authorship.³⁷ And it might be a nudge toward universalism in a more effective way than a single rule on ownership in an international treaty. Of course, one might have a genuine debate about whether diverse labor structures around the world should make us hesitate before instantiating a lex originis rule to initial authorship, but that point could perhaps be handled by a more flexible policy-based approach to choice of law, with a default of lex originis.

Indeed, as this example shows, any mechanism that elevates a single applicable rule to any issue of copyright law will operate to advance the cause of universalism. Indeed, the Montevideo Convention, which endorsed a country-of-origin rule over the territorial lex protectionis found in the Berne Convention system,³⁸ has been described by Silke von Lewinski as grounded in “universality” – I assume for that very reason.³⁹

Other instruments that achieve the same effect through an effective country-of-origin rule include the EU Cable and Satellite Directive, which designates the copyright law of the country of the uplink as applicable law in the event of a dispute.⁴⁰ Designating a single right owner from whom a user has to acquire rights in order to operate globally would advance universality.⁴¹ Of course, the harmonized substantive rules in the European Union make possible what might be impossible globally. Without harmonization of substantive law, the designation of the place of uplink to govern globally might invite the prospect of copyright havens. But rules can be constructed to prevent the development of safe havens if this approach were extended internationally, as was proposed by Jane Ginsburg in her 2000 WIPO paper, in which she suggested a cascade rule for online choice of law that was also accompanied by a backstop minimum of international copyright law.⁴²

Universality might be enhanced by ensuring that a global dispute about the use or exploitation of a copyright reaches a single resolution facilitating global distribution. Even substantially uniform laws can be interpreted quite differently by national courts, and the open-ended nature of many copyright principles elevates the role of the courts in law formation. A single global court applying uniform norms is an unrealistic (and perhaps undesirable) pipe-dream.⁴³ But procedural mechanisms that allow disputes that cross borders to be resolved in a single national forum, perhaps with appropriate recognition of national variation at the edges, might result in greater universality – without formal uniformity of norms.⁴⁴

Yet, international policymakers have been reluctant to create what might be called “public private international copyright law,”⁴⁵ that is, instruments dictating or confining the choices that countries can make as to their rules on jurisdiction, choice of law, and recognition of judgments in copyright cases.⁴⁶ In 1991, the Hague Conference on Private International Law, at the request of the United States, embarked on a quest to negotiate a jurisdiction and judgments convention of general applicability in civil and commercial matters.⁴⁷ Those efforts floundered in 2000–2001, in large part because of disagreement over how to handle intellectual property cases, forcing the conference to scale back its efforts and concentrate on a 2005 convention that validated exclusive choice-of-court clauses in business-to-business (B2B) contracts.⁴⁸ Intellectual property (including copyright) is excluded from the scope of the later Hague Convention on the Recognition and Enforcement of Foreign Judgments in Civil or Commercial Matters.⁴⁹

Other mechanisms can also contribute. Universality might be enhanced by knowing that nationally acquired rights can travel with a user abroad, extending the model of the EU Portability Regulation. That regulation enables consumers to access their portable online content services when they travel elsewhere in the European Union in the same way they access them at home.⁵⁰ Indeed, the European Union offers a number of exemplars in this space. For example, the EU Orphan Works Directive permits certain organizations to reproduce and make available copies of so-called “orphan works,” that is, works in relation to which all or some of the right holders cannot be identified or located despite a “diligent search” having been carried out.⁵¹ Although the diligent search condition is typically satisfied by being performed in the “Member State of first publication,”⁵² if a work is regarded as orphan in one member state, then that status should be recognized in other member states.⁵³

Likewise, the recent Digital Single Market Directive mandates extended collective licensing⁵⁴ to allow cultural heritage institutions to copy and provide access to out-of-commerce works (i.e, works not available to the public through customary channels of commerce, after a reasonable effort has been made to determine this) that are permanently in the collection of the institution. But the directive also requires that such licenses granted must allow the use of out-of-commerce works or other subject matter by cultural heritage institutions in other member states.⁵⁵

Stated more generally, these mechanisms from Europe show that if conduct in one country that is recognized as meeting a defined standard of legality were to be recognized elsewhere, under a system of mutual recognition, that would be a move toward greater universality. To some extent, this dynamic might underlie a strong doctrine of exhaustion. Even in countries adhering to a principle of international exhaustion, ordinarily a copy of a work lawfully made in one country by virtue of an exception or limitation in that source country cannot automatically be lawfully imported into another country. A copy made by virtue of a national exception is not placed on the market “with the consent of the copyright owner,” even if this stance interferes with the free movement of goods. This would appear to be clear in the European Union, notwithstanding a system of regional exhaustion.⁵⁶

The language of the first sale provision is more ambiguous in the United States. Section 109(a) makes the exhaustion turn on the copy in question being “lawfully made.”⁵⁷ Although the US Supreme Court has endorsed the concept of international exhaustion,⁵⁸ the goods in the case embracing that principle were lawfully put on the market because the copyright owner had itself consented to the foreign sale. It is not settled whether goods lawfully placed on a foreign market because of an exception in that country should be regarded as “lawfully made,” and litigants have recently sought (unsuccessfully) to bring the issue before the US Supreme Court.⁵⁹ It might be a radical attenuation of prevailing understandings of territoriality if the US courts found that the law of the place of foreign manufacture applied, as this would effectively extend the exceptions of other countries into the United States.⁶⁰ However, this is another example of the mechanism being deployed in Europe: if conduct in one country recognized as meeting a defined standard of legality were to be recognized elsewhere, that would be a move toward greater universality.

Finally, universality might also be enhanced by ensuring that an actor can operate online, and thus improve global exchange, by complying with a single set of regulations. Or universality might even be enhanced by recognizing increasing interdependence and exploiting cross-border capacity to enable the fruits of creativity induced by the copyright system to be distributed globally. I come back to these two options below in Part D of this chapter. But importantly, this discussion shows that furthering universality does not always involve the articulation of a defined international norm of public international law in the way that the universalists of 1883 thought to be the case.

D. Marrakesh as Inflection Point ... (but on Universality)?

For present purposes, it is sufficient to focus on the three main provisions of the Marrakesh Treaty without going into greater detail. Article 4(1) provides as follows: “Contracting Parties shall provide in their national copyright laws for a limitation or exception to the right of reproduction, the right of distribution, and the right of making available to the public … to facilitate the availability of works in accessible format copies for beneficiary persons [meaning mostly persons who are blind or have visual impairment].” The exception can be confined to works that are not available commercially in accessible format on reasonable terms, and exercise of the exception can be made subject to remuneration.

Article 4(2) offers an exemplary means of providing such an exemption, most notably including the involvement of “authorized entities,” who are non-profit organizations established to provide assistance to persons who are blind or have visual impairment and who will assist in making and distributing the copies of works in accessible formats. But contracting parties can formulate an exception in their own terms to comply with their Article 4(1) obligation, as long as⁶¹ it complies with the three-step test from Berne and TRIPS (and the WIPO Copyright Treaty).⁶²

Articles 5–6 of the Marrakesh Treaty facilitate the cross-border exchange of accessible-format copies. Article 5 requires that contracting parties allow an accessible-format copy made under a limitation or exception or pursuant to operation of law to be exported by an authorized entity “to a beneficiary person or an authorized entity in another Contracting Party.” Again, the treaty provides an exemplar of how a contracting party might comply with this obligation, but leaves it to contracting parties to implement the obligation in other ways too. Article 6 complements Article 5 by providing that, to the extent that the national law of a contracting party would permit the making of an accessible-format copy of a work, that law shall also permit the import of such copies by a beneficiary person or authorized entity.

Why has all of the above been said to be significant, and how real are those claims? And do any of the claims bear on the question of universality?

I. Exceptions and Users’ Rights

The Marrakesh Treaty has been described as “the world’s only IP treaty dedicated to harmonizing exceptions and limitations.”⁶³ Such a development might be seen as the international manifestation of the users’ rights rhetoric that has been a prominent part of the domestic landscape over the last two decades. Certainly, this is the first multilateral copyright treaty focused exclusively on exceptions and limitations to (apart from augmentation of) authors’ rights.

Of course, the significance of this point can be overstated. Many copyright treaties have contained provisions regarding exceptions and limitations. To be sure, most of these earlier provisions simply permit exceptions and limitations, which is conventionally thought to be a relatively soft tool of universal harmonization.⁶⁴ For example, Article 10bis of the Berne Convention explicitly allows exceptions for the use of works for reporting of current economic, political, or religious events, provided the source is clearly indicated.⁶⁵ And the most transcendental provision on exceptions in earlier treaties – the three-step test – has served primarily to constrain exceptions and limitations, and thus (consistent with the objectives of Berne) to protect authors’ rights against substantial erosion.⁶⁶ On its face, the Marrakesh Treaty requires any of the exceptions adopted under Articles 4–6 to comply with the three-step test, which is consistent with the dictates of this prior international copyright law. Therefore, the claim needs some refinement.

II. Mandatory Exceptions

The Marrakesh Treaty not only addresses or permits exceptions; it mandates them. Optional exceptions and limitations were less effective in achieving universality because some countries did not act on their authority to create exceptions, and even when they did so, there was little uniformity of approach. Mandatory exceptions ameliorate the first part of that problem by requiring contracting parties to enact exceptions.⁶⁷

But again, this feature is not entirely new. The Berne Convention, according to almost all commentators, contains mandatory exceptions – such as Article 10(1), which requires that “[i]t shall be permissible to make quotations from a work which has already been lawfully made available to the public, provided that their making is compatible with fair practice, and their extent does not exceed that justified by the purpose, including quotations from newspaper articles and periodicals in the form of press summaries.”⁶⁸ And both TRIPS and the WIPO Copyright Treaty⁶⁹ have been read by courts to mandate non-protection for ideas or facts, which can be understood as subject matter exclusions or an exception.⁷⁰ The Comprehensive Economic and Trade Agreement (the CETA) concluded between the European Union and Canada contains mandatory exceptions or immunities.⁷¹ And, at the regional level, we find mandatory exceptions in the recent EU Digital Single Market Directive,⁷² adding to the couple we already found in the Information Society Directive and the Software Directive.⁷³

However, mandatory exceptions, at least at the international level, might not always fulfill their universalizing objective. As Tanya Aplin and Lionel Bently recount in their analysis of the mandatory exception in Article 10(1) of Berne, a provision that they contend should have required “global mandatory fair use” has resulted instead in what they call “dysfunctional pluralism.”⁷⁴ Current judicial understandings of Article 10(1) may be unlikely to sustain the strong form of Aplin and Bently’s argument.⁷⁵ But, whether one accepts the details of how Aplin and Bently read Article 10(1), there is certainly no universal understanding of what the article means or how it is applied – which we might, in theory, have expected to be the benefits of mandatory exceptions.

The Marrakesh Treaty may obviate some of these problems by including relatively detailed exemplars of how to implement Article 4 and Article 5. If experience of implementation of EU directives is anything to go by, the availability of turnkey provisions will be an attraction to legislators.⁷⁶ Not all countries possess the domestic legislative capacity to transpose general international principles into workable domestic mechanisms. The turnkey provisions might encourage such countries to make those provisions self-executing, or to copy them verbatim. It might have been better in encouraging such adoptions had the treaty drafters felt able to declare that the exemplars passed the three-step test and thus operated as what Larry Helfer, Molly Land, Ruth Okediji, and Jerry Reichman have termed a “safe harbor,”⁷⁷ or (although complicated in terms of inter-treaty relations and arguably Articles 19–20⁷⁸ of the Berne Convention) to provide that the three-step test was inapplicable to these exceptions.⁷⁹ But given the approach of WTO panels to the harmonious interpretation of copyright instruments,⁸⁰ it is highly unlikely that a country enacting the exemplar will be successfully challenged under TRIPS.⁸¹

Of course, national courts may well yet interpret their common provision in disparate ways, interfering with the aspirations of the parties. And the treaty itself recognizes that different countries might adopt different conditions on commercial availability and remuneration.⁸² Finally, the treaty leaves hanging a number of important choice-of-law questions that could affect the cross-border availability of works.⁸³

All that speaks to, however, is the looser institutional structure of international law; no level of mandatory exception or even WTO enforcement can absolutely achieve universality. Indeed, the “autonomous reading” that the Court of Justice of the European Union has given to supposedly optional exceptions in Article 5 of the Information Society Directive may ensure a more uniform, and perhaps under one metric, more universal position than would the adoption of mandatory provisions drafted for more heterogenous international application, but lacking direct enforcement power.⁸⁴

III. Human Rights

A third way in which the Marrakesh Treaty is said to be unique is its endorsement of (universal) human rights, the right to read and the rights of the disabled.⁸⁵ Again, this point is overstated, even without referencing Article 27(2) of the Universal Declaration of Human Rights, which guarantees that “everyone has the right to the protection of the moral and material interests resulting from any scientific, literary or artistic production of which he is the author.”⁸⁶ Clearly, the right to read was part of any number of debates in Berne Conferences surrounding exceptions for developing countries, most notably in the Berne Appendix.⁸⁷ And the 1996 WIPO Copyright Treaty recognized “the need to maintain a balance between the rights of authors and the larger public interest, particularly education, research and access to information, as reflected in the Berne Convention.”⁸⁸

Perhaps the Marrakesh Treaty is the first copyright treaty expressly to reference in its preamble a particular human rights instrument conferring rights on a group other than authors, namely, the UN Convention on the Rights of Persons with Disabilities.⁸⁹ More important than these textual niceties, however, the core human rights dimension might speak to the universality of the underlying norms.⁹⁰ As noted above, the universality of human rights is understood to mean that they should apply even in internal settings.

IV. Scope of Application: Domestic Copyright Law

This last point connects well with another new feature of the Marrakesh Treaty that sets the treaty apart from existing international copyright law. The obligation of Article 5(1) of the Berne Convention applied only to “in respect of works for which [authors] are protected under this Convention, in countries of the Union other than the country of origin.”⁹¹ As Article 5(3) of the Berne Convention lays out even more explicitly, “protection in the country of origin is governed by domestic law.” Thus, member states are not required to grant treaty rights to domestic works; Berne only applies when there is a difference between the country for which the protection is sought and the country of origin. For example, under Section 411 of its Copyright Act, the United States imposes on US works the formality of registration prior to an action for infringement, which Article 5(2) of the Convention bars being required of foreign works.⁹²

This feature of the Berne system derogated from its universality, measured in scope of application. But it was an understandable feature given the roots of the Berne system in the desire to ensure that foreign works were granted protection. And as a practical matter, notwithstanding the US rule in Section 411 noted above, national political realities meant that most protections required by treaty to be conferred on foreign works would be extended to local works.

This political tempering of theoretical legislative latitude works well for the rights of authors. It might work less well for exceptions. Thus, the French Supreme Court has held that Berne-derived rights to make a brief quotation was not applicable to an alleged infringement in France of a painting of French origin, where an auctioneer had reproduced a painting in its auction catalog.⁹³ Thus, under the Berne model, contracting parties could deny the application of a mandatory exception to French works, while being required to allow such exceptions to be exercised as regards foreign works. The political realities do not work the same way here. This would rarely be a problem when the international copyright instruments were largely concerned with the rights of authors. But as the debate shifts to encompass mandatory exceptions – which might be called the rights of users – this issue requires the treaties to have a broader and more universal scope of application.⁹⁴

Article 4(1) of the Marrakesh Treaty requires that “Contracting Parties shall provide in their national copyright laws.” Sam Ricketson reads this as applying to domestic and foreign works and has asked whether this “Marrakesh deviation” will become the model in other areas where international agreements on exceptions are being considered, such as libraries and educational institutions.⁹⁵ Although there is no clear indication in the records of the conference that this language reflected an awareness of this dynamic, the desire for universality does indeed support this approach being replicated. The political winds that prevent this derogation from the scope of the Berne system being serious might blow differently with exceptions.⁹⁶

Indeed, more broadly, this development might make us begin to question more broadly the concept of “country of origin” in the Berne system. Article 5(4) provides a very complex definition of “country of origin.”⁹⁷ That concept has been severely tested by online publication,⁹⁸ which may – if read wrongly (as it may well have been by some national courts) – unintentionally evict works from the protection of Berne.⁹⁹

V. Cross-border Exchange

Even with international mandatory exceptions, which require that national laws permit particular uses, the principle of territoriality can interfere with the effectiveness of those exceptions in ensuring the universal application of the norm. Capacity to take advantage of those exceptions may vary widely from one country to another, limiting the overall distribution of socially valuable works (and in the case of Marrakesh, of accessible-format copies of works).

The Marrakesh Treaty (mirroring to some extent, if only implicitly, the treatment of patent compulsory licenses and access to medicines under the Doha Declaration and Article 31bis of TRIPS)¹⁰⁰ requires member states to allow for the cross-border exchange of special-format copies of books, including audio books and digital files, and other print material between those countries that are parties to the treaty (and who comply with the standard three-step test for exceptions in international copyright law).¹⁰¹ This should reduce the high costs of converting books into accessible-format copies by eliminating the duplication of efforts and allow many copies to be made in one country and supplied to beneficiary persons in several countries. The scheme set up by Articles 5–6 may encourage countries to take particular positions on the exhaustion question (despite the Marrakesh text bracketing the exhaustion question).¹⁰²

Rochelle Dreyfuss and I have previously suggested that an international IP acquis should include principles that reflect the increased interdependence of nation-states.¹⁰³ On the one hand, this requires greater attention to enforcement for authors. But it also suggests exploiting the potential for pooling of resources and thus for securing the social gains for which the copyright system exists. This is a mechanism by which to make more real both the universal aspiration of the “right to read” and the universal remit of the mandatory exception in the Marrakesh Treaty.

As noted above, in the regional context, the European Union has recently adopted a number of cross-border mechanisms that allow determinations or rights in one country to have effect in another.¹⁰⁴ This approach is likely less controversial in the more limited European context, where much of substantive copyright law has been harmonized. Even within the European Union, those mechanisms make important contributions to expanding the geographic area in which works might with certainty be exploited, because even with common legal rules, the facts and circumstances on which those rules operate may vary from country to country. Likewise, rights granted under common substantive rules might have been allocated to different grantees or granted under different conditions among different countries.

But the background commonality on substantive norms is likely to make the erasure of formal territoriality less troubling to states now required to treat a particular exploitation on their territory as lawful, without regard to domestic facts that would ordinarily inform that assessment. The realignment of outcome that is effected is likely to be minor (both because of the commonality of legal rules, but also because the more homogenous social and economic conditions make it more likely that the factual context would – if separately assessed – be similar). This seems a small price to pay for the reciprocal gains that are likely to come from these “mutual recognition” systems, especially when they contribute positively to the wider exploitation of the work, consistent with the objectives of international copyright law.

The implementation of a similar system at the global level is more radical because the global context removes the factual and legal commonality that is present within the European Union. But the Marrakesh mechanisms suggest variables that will inform the extent to which global replication is likely to ruffle feathers and the devices that might be the basis for rendering any such system more acceptable. Making the exceptions mandatory was important, as was the focus of the treaty on a topic on which there was broad agreement; most countries favored this type of exception. But so too was the provision of turnkey exemplars that is likely to effect soft harmonization as the treaty is implemented. Likewise, provisions in the treaty delineating the ways in which countries might vary their implementation (such as imposing a “commercial unavailability” condition) will, in practice, serve to cabin the extent to which the latitude to tailor implementation allows for deviation from a single norm.¹⁰⁵ Indeed, while the focus on the three-step test might have been problematic in some respects, it is the type of provision that helps to ensure that the cross-border mechanism cannot be exploited by a rogue country. Only certain countries can be admitted to the system of mutual recognition. Finally, the involvement of intermediaries whose affinities and sources of expertise often reach across borders is also likely to exert soft harmonizing influence. All this should help to make this mechanism seem less disruptive to actors who are accustomed to operating a national model.

E. Conclusion

Universalism has been a goal of the international copyright system from the outset. But once the concept is understood as something more than uniformity of content, it becomes apparent that all the participants in the development of the system over the last century can be understood as universalists. Less semantically, it also calls our attention to an expanded range of mechanisms that can fulfill that aspiration. In a number of features, the Marrakesh Treaty reflects that expanded, multi-dimensional view of universalism, and as such is a significant milestone (and a potential lodestar) in the further development of international copyright law.

9 Measuring the Scope of Obligations under International Treaties (To What Extent) Are IP Conventions Binding on Paris or TRIPS-Plus Legislation?

Annette Kur

Table of Contents

1. A. Introduction 261

2. B. Qualification of Subject Matter as Intellectual Property Rights 263

   1. I. Supplementary Protection Certificates 263

   2. II. Non-Original Databases and Press Publishers’ Rights 266

      1. 1. Background 266

      2. 2. Are the Database Makers’ and Press Publishers’ Rights a Form of Intellectual Property? 268

         1. a. What About Copyright? 268

         2. b. A (New) Kind of Industrial Property? 270

      3. 3. Carte Blanche for Discrimination Based on Nationality? 273

3. C. Limits for TRIPS-Plus Content in Intellectual Property Rights: Does the Greater Include the Lesser? 276

   1. I. Background 276

   2. II. What – If Any – Impact of Art. 1(1) 2nd Half-sentence TRIPS? 279

   3. III. Inherent Qualifications of ‘The Greater Includes the Lesser’ 281

   4. IV. The Need for Further Research 285

4. D. Conclusions and Summary 286

A. Introduction

The issue addressed in this chapter first emerged in the framework of the SPC Study that was conducted and coordinated by Roberto Romandini at the Max Planck Institute for Innovation and Competition in Munich (MPI).¹ ‘SPC’ stands for Supplementary Protection Certificate, a legal title in the European Union (EU) by which exclusive rights for the marketing of pharmaceuticals or agrochemical products are granted, for a limited period, after the lapse of patents pertaining to the underlying invention.² Among other issues,³ the study concerned the legal option of introducing a so-called manufacturing waiver, that is, the possibility to produce medical products for export or stockpiling while they were still protected by an SPC.⁴ In that framework, it had to be assessed inter alia whether such a rule complies with international law, in particular TRIPS.

Concerning stockpiling, a negative answer had been given to that question by a WTO Panel;⁵ however, the underlying dispute had been about patents alone. So, what about SPCs? Does the answer have to be the same? Or are they, by contrast, not even among those industrial or intellectual property (IP) rights that fall within the ambits of the Paris Convention (PC)⁶ and TRIPS?⁷ Would that mean that such rights are beyond any – even the most basic – obligation under the two treaty systems, such as granting national treatment? On the other hand, if it is assumed that SPCs are IP rights covered by the Paris Convention and TRIPS, does that automatically mean that all obligations under those agreements apply to the full extent? In other words, what exactly are the legal consequences of the fact that SPCs are TRIPS-Plus elements which do not form part of the mandatory minimum standard established under international IP conventions?

Having come so far in our deliberations, we realised that while SPCs present a complex and timely scenario for a case study on the implications of TRIPS-Plus legislation in the light of international law, that scenario is certainly not the only one. For instance, what about other sui generis rights, such as the right of database makers and press publishers?⁸ What about (potential) national legislation which makes registration an obligatory prerequisite for asserting copyright protection beyond fifty years after the death of the author, or for claiming unitary copyright on a regional (EU) level – in addition to national law?⁹ What about extending patent protection to diagnostic and/or therapeutic methods while denying access to civil remedies for infringement?¹⁰ Does the right to reproduce design-protected parts of complex products for repair purposes have to meet the criteria of the three-step test in Art. 26(2) TRIPS, even though Art. 25 does not make the design protection of parts mandatory?¹¹

Those issues, specific and diverse as they may appear, concern the same theoretical theme, namely, whether and to what extent international conventions produce legal effects beyond their core scope.¹² The situations giving rise to such questions can be roughly distinguished as follows. First, the issue may arise with regard to subject matter which is closely similar, yet not identical, to that covered by international IP conventions. Second, the problem may show up when national IP laws grant more generous protection than what is mandatory, but on the other hand impose certain qualifications that are generally objectionable under the relevant conventions.

The paper considers some of the issues identified above, being conscious that only a fragmentary and tentative approach can be offered here. The presentation proceeds as follows. First, the question is posed as to how and why subjective entitlements are qualified as industrial or IP rights, and what consequences ensue when they fail that mark (section B). As examples, I consider SPCs on the one hand (subsection I) and the rights of database makers and press publishers on the other (subsection II). Second, I address the impact of mandatory requirements of international law on regulations of national (or regional) IP law falling within the ambit of international conventions but granting more generous protection than the bare minimum (section C). More specifically, this section includes a discussion of whether and to what extent TRIPS-Plus legislation enjoys leeway under the principle that the broader option – not to protect at all – includes the narrower one, that is, granting more limited rights than what is otherwise mandatory under TRIPS (‘the greater includes the lesser’). The chapter ends with a brief summary and conclusions (section D).

B. Qualification of Subject Matter as Intellectual Property Rights

I. Supplementary Protection Certificates

Supplementary protection certificates¹³ were created as a sui generis right on the basis of EU regulations enacted in 1992 (for medicinal products)¹⁴ and 1996 (for agrochemical products) respectively. The legislative aim was to make sure that holders of patents pertaining to medical or agrochemical substances or procedures were able to make up for the time lost by the need to obtain market authorisation for products covered by the patent. The legislation was set off by similar schemes previously enacted in the USA¹⁵ and in Japan.¹⁶ In those countries, however, legislation took the form of an extension, or restoration, of the patent term with regard to the products concerned. The right granted was thus integrated into national patent legislation instead of constituting a sui generis right. In the EU, that route was barred at the relevant time, as the European Patent Convention (EPC)¹⁷ in its original version restricted the option of patent prolongation to certain emergency situations (Art. 63 EPC 1973). By establishing a sui generis system, the EU legislature tried to escape the necessity to initiate and promote a revision of the EPC,¹⁸ which – it was feared, not without reason – would be too cumbersome to accomplish in due time.¹⁹

In the context considered here, this scenario raises the question of whether the solution chosen produces any effects on the international level, or whether SPCs, as sui generis rights, are exempt from obligations under the Paris Convention and TRIPS. While it seems clear that a regulation within patent law such as in the USA and Japan (and others) would naturally fall within the ambit of industrial or IP law, the issue is no matter of course for SPCs. A negative answer could derive from the fact that SPCs are neither listed in the definition of industrial property as set forth in Art. 1(2) PC nor appear in Part II sections 1 to 5 TRIPS as referred to in Art. 1(2) of the TRIPS Agreement. However, an interpretation of those articles in good faith, as commanded by Art. 31 of the Vienna Convention on the Law of Treaties (VCLT), must take account not only of the ordinary meaning of the terms as such but must also consider the context as well as the object and purpose of the treaty concerned.

For the Paris Convention, this means in particular that account must be taken of Art. 1(4) PC, which specifies that ‘[p]atents shall include the various kinds of industrial patents recognized by the laws of the countries of the Union, such as patents of importation, patents of improvement, patents and certificates of addition, etc.’ Furthermore, Art. 1(3) PC stipulates that ‘[i]ndustrial property shall be understood in the broadest sense…’. Under the broad interpretation thus warranted, the term ‘patent’ must be understood as encompassing any kind of right by which national legislation rewards technical achievements.²⁰ Importantly, pursuant to Art. 1(4) PC, the term is not reserved for new inventions but applies equally to the mere importation of such achievements (‘patents of importation’).²¹ The focus thereby lies on the fact that the holder of the right has brought new and useful subject matter to the (national) market for which the right is granted.

The same is true for SPCs: there too, the right pertains not to the invention as such; the protection granted is rather meant to reward and incentivise the effort and investments needed to develop a pharmaceutical or agrochemical invention into a marketable product. Furthermore, EU legislation on SPCs was clearly intended to enable a kind of patent term extension resembling foreign legislation in most of its crucial features, though consciously avoiding, for the reasons set forth above, any reference to the term ‘patent’. But nomenclature cannot be decisive; what counts are the contents and objectives of the legislation. This point invites the conclusion that SPCs are ‘patents’ in the broad definition of Art. 1(4) PC and thereby form part of industrial property as defined in Art. 1(2) PC.

Indeed, the European legislature never disputed the contention that SPC legislation was fully encompassed by international IP law, including the obligation to grant national treatment.²² Thus, different from other fields of EU legislation,²³ it was never intended to make the grant of such rights subject to reciprocity. On the contrary, because the EU wanted to achieve a level playing field with international market actors who had already introduced similar protection clearly falling into the scope of the Paris Convention, the same approach was readily accepted for SPCs.

Accepting that SPCs are patents in the meaning of Art. 1(2) PC also settles the matter for TRIPS, at least insofar as amenability to dispute settlement proceedings as well as application of the general provisions in Part I of the agreement are concerned. This follows from the Appellate Body’s findings with regard to trade names in the US – Sec. 211 Omnibus Expropriation Act (Havana Club).²⁴ Pursuant to the report, rights protected under the Paris Convention are included within the ambit of TRIPS by virtue of Art. 2(1) TRIPS, irrespective of whether they are explicitly listed in Part II.²⁵

II. Non-original Databases and Press Publishers’ Rights

1. Background

The question as to whether new types of sui generis rights form part of the system established by the international IP conventions was a matter of debate, inter alia, in connection with the introduction of the Database Directive (DBD) in the EU in 1996.²⁶ Art. 7 DBD establishes a right protecting database makers against non-authorised users extracting or re-utilising substantial (or, in the case of repeated and systematic use, also insubstantial) parts of a database which has been the result of substantial qualitative and/or quantitative investment in either the obtaining, verification or presentation of the contents. Furthermore, pursuant to Art. 8 DBD, lawful users of the database may not perform acts that clash with normal exploitation of the database or unreasonably prejudice the legitimate interests of the database maker. As emphasised in the preamble, the right is not meant to create a new right in the works, data or materials themselves,²⁷ and it shall in no way extend copyright to mere facts and data.²⁸

Having thus distinguished the new right from copyright in terms of its objectives and scope, the EU legislature felt entitled to forgo the principle of national treatment by restricting application of the right to database makers or right-holders who are nationals of a Member State or who have their habitual residence in the territory of the Community (now the EU; Art. 11(1) DBD). For all others, the grant of protection was made subject to reciprocity²⁹ or to conditions established in bilateral agreements with the EU.³⁰

To some extent this move may have been a reaction to previous legislation in the USA protecting the circuit design of semiconductor chips.³¹ Protection for non-US nationals or companies was made dependent on the respective countries undertaking ‘good faith and reasonable efforts’ to provide similar protection. Restricting protection to nationals in that case had triggered similar legislation inter alia³² in the EU,³³ and even resulted in the adoption of a specific international treaty.³⁴ However, any hopes possibly nourished by the EU that a similar effect would arise with regard to protection of database makers were in vain; the USA and several other countries have, to date, refrained from such legislation.³⁵ However, and in spite of doubts articulated in the meantime about the aptitude of the legislative model to trigger investment in new and useful databases,³⁶ the EU model remains in operation and has been adopted in a number of other jurisdictions.³⁷ Whether and to what extent the reciprocity clause helped in that process can only be guessed.

The press publishers’ right in Art. 15 of the Digital Single Market (DSM) Directive³⁸ presents another more recent example of legislation reserving legal entitlements to nationals³⁹ only. Different from the database makers’ right with its particular prerequisites and scope, the press publishers’ right to prevent online uses is tailored after copyright protection against reproduction and making available to the public under Art. 2 and 3 of the Infosoc Directive⁴⁰ (with the exception of hyperlinking, and not relating to single words or ‘very brief extracts’), and it is subject to the same limitations and qualifications as set forth in Art. 5 to 8 of that directive. Nevertheless, the right is conceived as an instrument which rewards press publishers’ financial and organisational contributions to maintain a free and pluralist press and ensure quality journalism and the information of the public rather than the intellectual creation of individual authors.⁴¹ Accordingly, the right is claimed to form a legal title sui generis, which is separate and independent from copyright pertaining to the works contained in the publication.⁴² The EU legislature therefore felt free to grant the right solely to publishers of press publications established in an EU Member State (Art. 15(1) DSM Directive).

2. Are the Database Makers’ and Press Publishers’ Rights a Form of IP?

a. What About Copyright?

In both cases, databases as well as press publications, the contention that the right granted to an investor does not trigger the obligation to apply national treatment under the international IP conventions is open to challenge.⁴³ Starting with the press publishers’ rights, the distinction made vis-à-vis copyright might sound hollow in view of the fact that the scope of the right follows exactly the relevant stipulations in the Infosoc Directive.⁴⁴ On the other hand, the legislative intent to create an aliud meant to protect financial and organisational input instead of incentivising creative activities seems to be genuine; there are no indications to the contrary.⁴⁵ Nevertheless, the question remains whether it is possible and legitimate to complement an existing IP right by another legal title that emulates the contours of the former, yet differs in its subject matter to an extent that takes it beyond the exigencies of international law to which the ‘twin right’ is subject.⁴⁶

For the answer, one must again, in good faith, revert to the interpretation of the Berne Convention.⁴⁷ On the one hand, the press publishers’ rights, just as copyright of journalists, are set off when (part of) texts are reproduced or made available to the public, and there is no doubt that texts are ‘literary works’ for which copyright protection is mandatory under Art. 2(1) Berne Convention. On the other hand, the new right does not detract in any manner from the availability of copyright pertaining to the text; it only constructs an additional right with a different ‘DNA’ in spite of its mirror-image appearance. The Berne Convention is satisfied with the copyright granted; neither the wording of Art. 2(1), nor its context, purpose or objectives appear to prevent the establishment of the second ‘same outlook, yet different substance’ type of right.⁴⁸

For the database makers’ right, the distinction vis-à-vis copyright is likewise clear. Apart from legal systems accepting a ‘sweat of the brow’ doctrine,⁴⁹ copyright protection only vests in the structure or arrangement of the contents compiled in the database (see Art. 10(2) TRIPS).⁵⁰ Mere efforts and investments remain outside of that realm, and, if a demand for such protection exists, it can only be covered by a sui generis right.⁵¹ It is true that the database makers’ right still remains related to copyright, and the same can be said about the press publishers’ right addressed above. However, in the context considered here, that qualification does not help much, as there is no all-encompassing international convention, similar to the Berne Convention that concerns related rights in general.⁵² The existing conventions are specifically targeted to individual groups of right-holders – performing artists,⁵³ phonogram producers⁵⁴ and broadcasting companies⁵⁵ – and do not apply beyond their specific scope.

b. A (New) Kind of Industrial Property?

If the sui generis rights established by the database and DSM directives cannot be squeezed into the frames of the Berne Convention or other copyright-related instruments, one could explore instead whether there is a place for them in the somewhat broader realm of the Paris Convention for the protection of industrial property.⁵⁶ Considering that the right of database makers and press publishers is not about protecting the creative spark of an author’s mind, but rather encourages and rewards entrepreneurial spirit, organisational skills and financial investment, it is certainly no misnomer to allocate this form of protection to the ‘industrial’ side of the IP spectrum.⁵⁷ It also helps that the Paris Convention takes an open, rather sweeping approach to the subject matter included. Notably, the definition in Art. 1(2) also includes geographical indications⁵⁸ and ‘the repression of unfair competition’. The latter term might be broad enough to encompass at least the database makers’ right,⁵⁹ which, prior to harmonisation, had been a matter for unfair competition law in a number of EU Member States.⁶⁰ Indeed, the EU legislature motivated the grant of the right in part by the need to prevent parasitic competition.⁶¹ Whether similar considerations could apply to the press publishers’ rights is less clear. While some kind of free-riding certainly did play a role, the chief motivation seems to have been the wish to preserve the economic basis for activities highly valued in a democratic society, rather than fighting some kind of parasitic behaviour.

For a more informed assessment of what is meant by ‘repression of unfair competition’, account must be taken of the relevant provision in the Paris Convention, Art. 10bis PC. Pursuant to Art. 10bis(2) PC the obligation of Member States to provide effective protection against unfair competition concerns any ‘act of competition contrary to honest practices in industrial or commercial matters’. The meaning of that principle is further exemplified in Art. 10bis(3) PC, which refers to acts of commercial communication such as creating confusion or making discrediting or misleading statements, but does not mention parasitic behaviour in the sense of using input procured by others to one’s own commercial advantage.⁶² That omission is easily explained by the difficulties of defining the tort so that it does not become excessive; after all, using others’ input for one’s own purposes can, and often will be, a normal feature of sound competition.

On the other hand, there is nothing in Art. 10bis PC signalling that such acts, if they are considered as unfair for certain reasons, are not meant to be captured by the provision.⁶³ Accordingly, it seems to be understood that if national law qualifies certain behaviour as parasitic under its unfair competition law, national treatment will apply. The same follows from the argument that due to the specific character of the provision, it is not necessary anyhow to rely on national treatment. If conduct qualifies as an act of unfair competition under national law, the provision itself makes it mandatory for the national legal system to prevent such acts within its own jurisdiction, irrespective of the nationality of the person who is adversely affected by the impugned behaviour.⁶⁴

That clarity, however, does not put an end to the question considered here. The two sui generis rights are distinct from unfair competition in their legal structure and in their contents. Both rights grant an exclusive position in relation to an object, whereas unfair competition rules are conduct-related.⁶⁵ Furthermore, while preventing parasitic competition may account for part of the motivation underpinning the two rights, this is not their only – or even the most prominent – objective. As emphasised in Recital 42 of the DBD, ‘the right to prohibit extraction and/or re-utilisation of all or a substantial part of the contents relates not only to the manufacture of a parasitical competing product but also to any user who, through his acts, causes significant detriment, evaluated qualitatively or quantitatively, to the investment’. The acts of users addressed in the second part of the sentence, though harming the economic position of the database maker, hardly qualify as ‘acts of competition’ in a narrow sense; users take advantage, but they do not compete.⁶⁶ The same is true for the press publishers’ right: news aggregators and press publishers are not competitors, or at least are not direct competitors; they act on different markets. Furthermore, while news aggregators may cause economic problems to publishers, they render a genuine service instead of simply engaging in usurpation of third-party efforts.⁶⁷ The EU legislature therefore did not even allege that the legislation was about fighting parasitism, but relied nearly entirely on the need to preserve the economic basis for quality journalism.⁶⁸

Unfair competition protection in the meaning of Art. 10bis PC can therefore hardly provide the ultimate key for including the two sui generis rights into the Paris Convention. Nevertheless, Art. 1 PC deserves a second look. As noted above (B.1.), the article provides additional clues regarding its breadth, for instance in Art. 1(4) PC with regard to patents. On a more general note, Art. 1(3) PC stipulates that ‘[i]ndustrial property shall be understood in the broadest sense’. Considering that industrial property under the convention is even found to subsist in the rather loose body of rules governing unfair competition, it could make even more sense to apply the term where rules governing conduct with regard to intangible assets have solidified into a system of rights and obligations. Under this analysis, the case for qualifying sui generis rights such as the two species considered here as ‘industrial property’ in the meaning of Art. 1 PC appears to be tenable in principle.

However, there are a number of counterarguments. First, the primary concern underlying of Art. 1(3) PC was to ensure that Member States do not exempt certain economic sectors from the effects of the convention;⁶⁹ the idea of eventually encompassing newly emerging rights was not on the horizon. To confirm this point, nothing in the wording of Art. 1(2) PC indicates that the list spelled out therein is meant to be only of an exemplary, non-exhaustive character. Second, if the interpretation suggested above were correct, it would also apply to other kinds of rights, including those related to copyright. Indeed, the rights of phonogram producers and broadcasters⁷⁰ are no less ‘industrial’ in their substance than those of database makers and press publishers. The fact that in spite of this, related rights were never regarded as subject matter to which the Paris Convention might apply indicates that pursuant to a general understanding, the term ‘industrial property’ is restricted to subject matter expressly listed in Art. 1(2) PC. To corroborate the position one can further observe that plant breeders’ rights as well as those of designers of topographies of semiconductor chips – which both would fit under the broad definition of ‘industrial property’ suggested above – were not considered to fall within the realm covered by the Paris Convention. Instead, both rights – just as the rights of phonogram producers and broadcasters⁷¹ – became the subject of separate treaty systems,⁷² restricting national treatment to other Members of the same agreement.

Thus, even though the wording of the Paris Convention does not completely exclude a broad interpretation of the subject matter covered, subsequent practice in the meaning of Art. 31(3)(b) VCLT clearly points in a different direction, with the result being confirmed by the negotiation history (Art. 32 VCLT).

3. Carte Blanche for Discrimination Based on Nationality?

The most obvious effect of intangible subject matter not qualifying as ‘industrial’ or, in a broader sense, ‘intellectual’ property is that such rights are not subject to the principle of national treatment set forth in Art. 2(1) PC and Art. 3 TRIPS. Persons and entities may therefore suffer different treatment than that enjoyed by the beneficiaries of the respective legislation, depending on their nationality or place of business. While such legislation – as in the examples of EU law concerning database makers and press publishers – is obviously legitimate in the sense that it complies with current international obligations, this does not mean that the wisdom and appropriateness of that result could not be challenged. Would granting equal treatment to foreign-based entities not be more conducive to general principles of equity and fairness than strictly clinging to the letter of international IP conventions?

In the field of human rights it is an established principle anyhow that discrimination violates the basic rules of customary law. Thus, Art. 7 of the Universal Declaration of Human Rights states that ‘All are equal before the law and are entitled without any discrimination to equal protection of the law…’. Furthermore, Art. 2(2) of the International Covenant on Economic, Social and Cultural Rights (ICESCR) stipulates that ‘[t]he States Parties to the present Covenant undertake to guarantee that the rights enunciated in the present Covenant will be exercised without discrimination of any kind as to race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status’. While this wording does not exclude differential treatment, especially that which is aimed at counteracting indirect discrimination,⁷³ it is generally accepted that legislation clashing with one of the prohibited grounds listed in Art. 2(2) ICESCR violates fundamental human rights norms.⁷⁴

The situation is obviously different insofar as economic entities are concerned. National treatment, and discrimination based on national origin as its necessary counterpart, are a fully accepted element of international commercial law. Indeed, discrimination of foreign business actors is the rule rather than the exemption. National treatment needs to be treaty-based in order to become effective;⁷⁵ this implies an option for negotiating parties to craft exceptions as they see fit. Much effort is therefore spent on the issue in international trade and investment treaties, where equal access to markets and/or resources presents a central bargaining point.⁷⁶

Likewise, the international IP conventions reserve national treatment to a clearly defined circle of beneficiaries connected by nationality, residence or activities to one of the Member States. Moreover, in certain instances, international IP law itself allows setting aside national rules vis-à-vis ‘foreign’ subject matter. Well-known examples for that are Art. 2(7) and Art. 7(8) of the Berne Convention. If works of applied art are only eligible for design protection in their country of origin, or if the term of protection in the country of origin is shorter than in the country where protection is sought (though still meeting the minimum standard set forth in the convention), Berne Member States are only obliged to grant the same term, or the same kind of protection, as applies in the origin country.⁷⁷ In that light, it appears only consequent to completely withhold from non-nationals and non-residents any rights not falling within the conventions.

However, it may be time after all to reconsider the attitude reflected therein, as its foundations may have become outdated and obsolete in the area covered.⁷⁸ Restricting national treatment to Members was an important or even indispensable element of international IP conventions at the time when commonly accepted minimum standards did not exist and foreigners were frequently precluded from obtaining protection. It was crucial to reserve equal treatment to beneficiaries of the treaties so as to provide an incentive for others to join the new instruments. Now that TRIPS has turned IP protection into a standard feature of legislation practically world-wide, the old coercion mechanism is no longer needed for achieving its basic goal. Its perpetuation as a lever to enhance membership in special agreements or make others adopt sui generis legislation paralleling one’s own⁷⁹ seems somewhat pointless in comparison to the original motivation. Using the lever is also redundant in view of the fact that nowadays trade negotiations have become the primary tool for promoting treaty membership and mandating specific IP legislation – probably doing a more thorough and effective job than withholding national treatment ever could.⁸⁰

A counterargument could derive from the consideration that if and insofar as protection is based on investment, states may have a justified interest in rewarding only those whose financial input benefits the national economy. That is certainly true insofar as the bargain between foreign direct investment and free access to markets or subsidies is concerned. Regarding sui generis rights adjacent to IP, however, the argument is not that straightforward. In the examples of databases and press publications considered here, it is rather so that activities springing from the relevant investments are regularly available to a global, digitised market, benefiting not only a regional clientele. Furthermore, it hardly does justice to IP and adjacent rights to equate the legal position conferred – the exclusive right with its specific conditions and qualifications – to plain economic transactions. This is not simply about granting financial benefits of some kind. The crafting of IP rights as well as of sui generis rights adjacent to them regularly concerns a complex cluster of interests involving beneficiaries, users and the public at large, in consideration of which it has been decided by the national or regional legislature that granting the right would be in the best interest of public welfare. Those considerations do not become meaningless simply because someone has a different nationality or domicile.

This does not mean that any differentiation, including different treatment based on nationality or place of establishment, is pernicious per se. There can be valid reasons for such treatment. The argument proffered here only posits that the discriminatory effect resulting from denying national treatment for sui generis rights not falling within the canon explicitly listed in the international IP conventions should be taken seriously. This is more than a mere technical detail following, as a matter of course, from the form of regulation chosen. It consciously withholds a form of protection that is otherwise considered to be beneficial and appropriate vis-à-vis everyone.

In other words, current international IP law should not be understood as giving carte blanche for discrimination of foreigners or foreign-based companies with regard to sui generis rights. Although it does not apply on a mandatory basis, the broad spectrum of subject matter protected under the Paris Convention, from exclusive rights to mere regulation of market conduct, provides a basically appropriate measurement of what should ideally be covered by the non-discrimination principle of national treatment. Exceptions are acceptable and may be needed in certain situations, but only to the extent that they are based on sound and fair policy objectives – other than the simple wish to ‘keep the others out’.⁸¹

C. Limits for TRIPS-Plus Content in IP Rights: Does the Greater Include the Lesser?⁸²

I. Background

Whereas in Part B the focus was on sui generis rights which do not fall within the ambit of international IP conventions (or in relation to which doubts, although ultimately unfounded, might be raised – as in the case of SPCs), we now turn to ‘genuine’ IP rights that are unequivocally covered by the Paris or Berne Conventions as well as by TRIPS. In view of such rights, the question arises to what extent the obligations set forth in the conventions are valid even where national legislation extends beyond the mandatory requirements prescribed therein. For instance, submitting that SPCs are to be treated as patents in the meaning of Art. 27 TRIPS,⁸³ the fact that they extend the lifetime of a patent well beyond the minimum term of twenty years might liberate the legislature from the obligation to fully observe the exigencies of the three-step test in Art. 30 TRIPS.⁸⁴ The same argument could be of relevance when copyright extending beyond the minimum term is subjected to formalities, or when patents are granted for therapeutic methods with the owners not being entitled to claim civil remedies,⁸⁵ or in a variety of other TRIPS-Plus situations when the borderlines of what is permitted under the mandatory terms of the relevant rights are transgressed.

It is not possible in the framework of this chapter to embark on an in-depth investigation of all instances in which the issue could be of relevance. Instead, the following remarks only concern the general argument that derogations from the obligations usually applying under TRIPS Part II can be based on the principle that conduct is not illegitimate if it yields more than what is owed (‘the greater includes the lesser’).

The maxim is derived from Roman law⁸⁶ and to date still applies in a number of national jurisdictions.⁸⁷ However, the caveat must be made that there is no clear basis for assuming that this principle also forms a valid interpretative tool for identifying duties or fallacies under international law. In particular, the principle is not mentioned in the VCLT, nor does Roman law figure among the sources of law listed in Art. 38 of the statutes of the International Court of Justice. Nevertheless, the principle merits consideration due to its inherent logic. It makes sense to assume that sovereign parties negotiating and concluding an international agreement do not act with the intention to bind themselves beyond the specific minimum standards set forth therein, and that they renounce to flexibility – including the freedom to balance rights and obligations in a manner best suiting their own socio-economic prerogatives – in ultra-obligatory legislation. The hypothesis seems consequent therefore that TRIPS-Plus legislation enjoys larger freedom to prescribe restrictions than where only the minimum threshold is met.

For the sake of clarity, it must be added that this is not to be confused with the principle ‘in dubio mitius’. The latter principle, which restricts international commitments to the least obliging understanding of relevant provisions, is incommensurate with the maxim of treaty interpretation in good faith as set forth in Art. 31(1) VCLT; it is therefore largely regarded as obsolete.⁸⁸ Here, we do not talk about interpretation of individual treaty provisions, but rather about the internal structure of the treaty and the dynamics of its elements in relation to each other. The question is whether over-fulfilling the obligations in one aspect means that other elements in the treaty become moot insofar as the add-on is concerned (alternative one), or whether the binding effect of such other elements remains intact insofar as subject matter is concerned which in its substance – irrespective of certain excess features – continues to be captured by the treaty (alternative two). Under the principle that the greater includes the lesser, alternative one would prevail.

There are indeed indications signalling basic acceptance of that precept. For instance, it was found acceptable – and declared to be compatible with Art. 27 TRIPS – in the USA that patents for therapeutic methods were granted, while civil remedies for infringement of such rights are denied under Sec. 287(c) of the Patent Act.⁸⁹Furthermore, the fact that under EU design legislation the term of the Unregistered Community Design Right (UCDR) is limited to three years after publication never gave rise to discussions, in spite of the fact that Art. 26(4) TRIPS requires a minimum term of ten years. It is obviously taken for granted that, because the UCDR is an add-on to the registered community design with its 25-year maximum duration, curtailing the protection term does not present an issue under TRIPS.⁹⁰ On the other hand, such examples are still too few to be considered as subsequent practice in the meaning of Art. 31(3)(b) VCLT.

There are also indications to the contrary. For example, concerning formalities in copyright, Stef van Gompel argues that establishing registration as a prerequisite for profiting from an extended protection term would clash with Art. 5(2) Berne Convention⁹¹ if it applies to works of foreign origin that are entitled to protection for the full term of the right. No objections would arise from the Berne Convention only if, on the basis of the so-called comparison of terms (Art. 7(8) Berne Convention), the duration of protection granted to foreign works can be limited to the (shorter) term granted in their country of origin.⁹² With regard to industrial designs, Joseph Straus claims that introducing a repairs clause allowing independent manufacturers to reproduce design-protected spare parts for repair purposes clashes with the three-step test in Art. 26(2) TRIPS, without considering that under Art. 25 TRIPS parts of complex products could be excluded from protection for good.⁹³ The matter therefore seems to be contentious and must be examined further.

II. What – If Any – Impact of Art. 1(1) 2nd Half-sentence TRIPS?

In his book on the TRIPS regime of patent rights, Nuno Pires de Carvalho denies legislative flexibility with regard to TRIPS-Plus rights.⁹⁴ His position is based on the argument that according to Art. 1(1) 2nd half-sentence, TRIPS Member States may implement more extensive protection than what is required by the agreement, but only under the condition that ‘such protection does not contravene the provisions of this Agreement’. This is claimed by Carvalho to mean that any decision to extend protection must be consistent with the provisions pertaining to the respective rights. For patents, this is said to mean for instance that prohibition of discrimination under Art. 27 TRIPS as well as the conditions and requirements laid down in Art. 28, 30 and 31 TRIPS prevail even during periods of time extension. In consequence of that approach, Carvalho concludes that the EU SPC regime violates Art. 27 TRIPS because it is limited to medicinal and plant protection products. From the same approach it would follow that introducing a stockpiling and export⁹⁵ waiver after lapse of the mandatory protection term is irreconcilable with Art. 30 TRIPS, as the same result (concerning stockpiling) was endorsed with regard to patents by a WTO Panel scrutinizing Canadian patent legislation.⁹⁶

Carvalho’s argument runs parallel to the second of the two alternatives presented above (C.1.), namely that all individual treaty obligations remain binding, independently of whether the features of the right that are concerned by the qualifications exceed the minimum standard. However, Art. 1(1) 2nd half-sentence TRIPS can hardly be quoted in support of that position. An interpretation in light of the first recital in the preamble, where the need ‘to ensure that measures and procedures to enforce IP rights do not themselves become barriers to legitimate trade’ is emphasised, rather suggests that it is the purpose of Art. 1(1) 2nd sentence TRIPS to fend off too far-reaching protection that could stifle competition, free trade and the exchange of ideas and information.⁹⁷ For instance, if patent protection were granted for subject matter that is not inventive, such ‘generosity’ might, in the meaning of Art. 1(1) 2nd sentence, contravene⁹⁸ Art. 27 of the TRIPS Agreement.

Apart from establishing a barrier against clearly dysfunctional developments of that kind, the TRIPS Agreement does not contain specific rules limiting legislatures’ freedom in case of extension of rights. The situation is different for instance from EU harmonisation directives which spell out certain options for those who want to ‘do more’ than implement the bare minimum. In the EU context, those options become binding insofar as national legislatures are precluded from choosing a middle path between the minimum rule and the option granted;⁹⁹ in that regard, the principle of ‘the greater includes the lesser’ is set aside. This approach is owed to the goal of harmonisation that underlies EU directives, which might be severely compromised if the leeway given for national solutions were too broad. In the framework of TRIPS, however, that scenario does not apply. Neither does TRIPS set forth specific TRIPS-Plus options, nor is the goal to reduce distortions in international trade¹⁰⁰ comparable to establishing the conditions for the functioning of a single internal market, as in the EU.

Carvalho’s reference to Art. 1(1) 2nd half-sentence TRIPS is therefore inconclusive. However, this does not necessarily settle the matter for good. The position holding that legislatures are completely free to derogate from international obligations if and to the extent that they have added certain non-mandatory features to the protection granted under national law is not free from doubt either. It is not inconceivable that a general public policy concern exists in safeguarding the transparency and consistency of legal systems, beyond mere transposition of minimum rules. Some form of self-binding effect of legislative decisions might arguably arise, meaning that if legislatures voluntarily adopt TRIPS-Plus protection, they must stick to the rules of the game at least to some extent.¹⁰¹ That proposition is considered more closely in the following section.

III. Inherent Qualifications of ‘The Greater Includes the Lesser’

In an analysis based on US jurisprudence, Michael Herz calls the argument that the greater includes the lesser ‘tremendously attractive to lawyers and judges’, but points out that it is also ‘a trap’, ‘because it is only sometimes true’.¹⁰² In particular, certain qualifications must be heeded which, if ignored, would produce skewed results.¹⁰³

Three different types of potential flaws are distinguished. The first and second derive from the inner logic of the argument itself. First, the argument is obviously only valid if in fact the greater power exists.¹⁰⁴ Second, the argument does not work properly if one of the propositions is not in fact ‘the lesser’ of the other one.¹⁰⁵ The third and arguably most important qualification results from the legal framework within which the rule operates. In the context considered by Herz, this means that the argument is flawed where ‘exercise of the “lesser” power implicates constitutional considerations not present in the exercise of the “greater” power’.¹⁰⁶

Although the applicability of the principle in the context of international law remains doubtful,¹⁰⁷ I suggest that the parameter set forth above can be used, mutatis mutandis, as an analytical tool for a further probe into the issue discussed here. In particular, the scheme could provide valuable insights insofar as the inner logic of the argument is concerned (the first and second condition identified above). Regarding the first point – whether the greater power exists – one might indeed wonder whether the concept of ‘greater power’ in the meaning of the argument can be equated with the ‘freedom not to act’ (beyond the prescribed minimum) which is at stake here. However, freedom from binding obligations is not so unlike greater power to act, as both enlarge the number of legal options available, including, in the cases considered here, the option to grant a limited right (instead of no right at all).

Concerning the second point – whether one proposition is indeed ‘lesser’ compared to the other one – the decisive question would be whether allocation of a right which is restricted in its contents or encumbered by formalities etcetera is, in all its aspects, comprised in the entitlement not to grant the right in the first place. Typically, that condition would not be met in the case of ‘package deals’, when the grant of TRIPS-Plus advantages is compensated by TRIPS-Minus restrictions,¹⁰⁸ even if the latter are small compared to the former. Thus, even a long-term extension of copyright would not include the freedom to introduce a registration requirement throughout the full lifetime of the right – that is, also during the first fifty years for which Berne requires protection as a minimum. The corollary is that if registration only applies as a condition for enjoying the time extension, this would arguably be ‘lesser’ in comparison to not prolonging the term at all. On the other hand, the answer is not always straightforward. For instance, it can be asked whether granting a right which is curtailed in its substance by specific limitations is rather something different from not granting the right at all, meaning that the alternatives cannot be conceptualised in the simple categories of ‘more’ or ‘less’.¹⁰⁹ The same problem arises if an additional right is granted on a higher institutional level – such as a unitary EU copyright complementing the national copyright systems – but is made dependent on an act of registration which is prohibited by the Berne Convention.¹¹⁰ Is that ‘lesser’ than not granting an EU right at all, or are we again talking about two different things that do not fit into a simple scaling scheme? And what about the UCDR with its short term, compared to not granting an unregistered design right at all?

The third element in the analysis is even more problematic, as it refers to ‘constitutional considerations’, which are difficult if not impossible to identify in the legal framework of international conventions.¹¹¹ Nevertheless it is suggested here that the notion can be used in the present context at least insofar as very basic defining elements of the system are concerned. This applies in particular to national treatment,¹¹² which corresponds to what in Herz’s scheme is labelled ‘equality concerns’.¹¹³ There is no doubt that irrespective of legislatures’ freedom to act where no minimum rights apply, the principle of national treatment must nevertheless be respected. For instance, assuming that in a TRIPS Member State the duration of patents were prolonged from twenty to twenty-five years, a clear and unequivocal obligation exists under the Paris Convention as well as TRIPS that this must apply to all patent holders, irrespective of nationality, seat or establishment. A deviation from that rule is only permitted where an express dispense applies under the relevant convention, such as – in the case of copyright – Art. 7(8) Berne Convention.¹¹⁴

As clear as this seems to be, compliance with the principle is not always guaranteed. This tendency is shown by the following example. Pursuant to Art. 11 in conjunction with Art. 110a(5) of the Community Design Regulation (CDR), protection for UCDRs is triggered by publication of the design ‘in the territory of the EU’, meaning that no such protection is granted if the design was first published (‘made available to the public’) outside the EU. The provision is flawed for many reasons,¹¹⁵ but most critically, it is incompatible with international obligations. There is no doubt that UCDRs are industrial designs as defined in Art. 1(2) PC and in TRIPS¹¹⁶ and are thus encompassed by the obligation to grant national treatment under Art. 2(1) PC and Art. 3(1) TRIPS. The fact that industrial designs are also available through registration does not change that clear result. It is true that Art. 11 CDR does not refer to nationality but only to the place of publication. However, in case of copyright, the CJEU pointed out inter alia in Tod’s v Hérault¹¹⁷ that withholding legal privileges on the basis of the place of first publication is but another covert method of discriminating against foreign nationals.¹¹⁸ Likewise, in EU – GIs,¹¹⁹ the WTO Panel found that Art. 3(1) TRIPS is violated where the difference in treatment affects the ‘effective equality of opportunities’ between the nationals of other WTO Members and the European Communities.¹²⁰ This point clearly bans legal rules that are formally (‘de jure’) non-discriminatory, but which lead – and are meant to lead – to actual disadvantages for non-residents.¹²¹

That consideration marks a difference between the ‘constitutional’ concerns raised by the discriminating effect of Art. 11 CDR on the one hand and by the limited duration of UCDRs (three years instead of a minimum of ten as prescribed in Art. 26(4) TRIPS) on the other. While the principle ‘the greater includes the lesser’ is clearly and fatally compromised where concerns of the first type¹²² are triggered, it may be able to prevail in situations of the second type, unless its operation is precluded by other reasons (for instance, because granting a short-term unregistered design right in addition to the full-term registered right is not considered as ‘lesser’, but rather as an aliud vis-à-vis not granting such a right at all).¹²³

The question remains whether, apart from national treatment, there are other ‘quasi-constitutional’ concerns irrevocably tainting TRIPS-Plus IP legislation in spite of the fact that it yields more than the bare minimum. For an informed response, one would first have to define what ‘unconstitutional’ means in the context of international IP law. Does the concept of ‘constitution’ – if it is accepted at all for the sake of argument – relate to rules having attained the status of customary law? If so, the restriction would be basically meaningless and redundant, as those rules will prevail in any case. Alternatively, in line with what was said about the equality and national treatment principle, one could conceptualise as ‘constitutional’ those elements within international conventions that establish the essential groundwork for their operation.¹²⁴ For instance, it could be argued that protection without registration is a constitutional element of the Berne Convention. In that light, requiring registration as a prerequisite for enjoying longer protection than what is prescribed in Art. 7(1) would be incompatible with international law, even if it only applies during the additional protection period. Or what about granting TRIPS-Plus rights in the form of mere liability rules? Is exclusivity (notwithstanding the admissibility of limitations) necessarily a defining – and therefore constitutional – feature of IP rights? If so, the grant of patents for therapeutic methods without any access to civil remedies would be incompatible with Art. 28 TRIPS, irrespective of the fact that providing for such patents is not mandatory under Art. 27.

IV. The Need for Further Research

The quest for ‘constitutionalisation’ of certain elements of international IP conventions, or the global IP system in its entirety, is nothing new. To be highlighted is in particular the endeavour by Rochelle Dreyfuss and Graeme Dinwoodie in their book on a neo-federalist vision of TRIPS to identify an ‘acquis’ of basic principles forming the background norms animating the IP system.¹²⁵ For such efforts to gain further impact, it needs to be clarified whether there is a place in the hierarchy of international norms for a set of core principles which have not risen to the level of customary law, but which, by virtue of their ranking in the framework of international norms and based on a common or at least widespread understanding, cannot be discarded by national lawmakers or in international negotiations. While there does not seem to be a clear precedent on the international level for such constitutionalisation of basic tenets,¹²⁶ the concept might arguably work in densely regulated areas such as IP, where defining elements can be extracted from a plethora of existing norms and a trove of documented practice.

The larger implications of such a move – if it were successful – are, however, not quite clear. On the one hand, constitutionalisation of core IP principles might offer a resilient defence warding off imbalanced and intrusive measures pushed through in international negotiations by powerful actors.¹²⁷ On the other hand, as shown by the considerations above, elevating certain elements of IP conventions to the status of (quasi-) constitutional norms could impose substantial limits on legislatures’ freedom to frame TRIPS-Plus legislation.¹²⁸

Regarding more specifically the principle that ‘the greater includes the lesser’, the caveat must be repeated here that transposing a scheme developed to operate within a clear-cut legal framework to the loosely structured area of international IP law is somewhat questionable. Nevertheless, the thought experiment explained above has shown two things. First, the issue needs further exploration and discussion. Second, whatever scheme is adopted at the end, this is definitely a matter for differentiation depending on the kind of right granted and the gravity of the violation of international law which is allegedly caused. Generally speaking, the principle of ‘the greater includes the lesser’ – where it applies – might lose its convincing force only where limitations or conditions appear to be incompatible with the basic tenets of the international IP system and the very essence of the relevant right as enshrined in TRIPS and the earlier conventions. As an example, one might think of legislation accepting smell marks for registration, which is not mandatory under Art. 15 TRIPS, but limiting the duration of the right to ten years without a possibility for renewal. That would be both illogical in view of what trademark protection is about and ignorant of the fundamental policy considerations underlying Art. 18 TRIPS.

However, such extreme examples are rare. Furthermore, the misgivings would not preclude legislation pursuing equal goals in a different form. Thus, no international rule would hinder sui generis legislation granting a time-limited exclusive right to creators of smells that are novel and distinct. Similarly, even assuming that it would be incompatible with the very DNA of copyright to stipulate that after lapse of the minimum term the right is transformed into a mere liability rule, international law would not prohibit that copyright protection terminates at the prescribed term and a domaine public payant¹²⁹ is established thereafter. In that situation, the more relevant question would be the one considered in Part 2 of this chapter, namely to what extent such rights must – or should – nevertheless respect the principle of national treatment.

D. Conclusions and Summary

This chapter has developed two strands of arguments concerning TRIPS-Plus rights. The first one concerns novel forms of rights to intangible assets that neither belong in the universe of copyright covered by the Berne Convention nor, under an interpretation in the light of Art. 31 VCLT, constitute industrial property rights in the meaning of the Paris Convention, although being close enough at least to the latter. The question is posed, with regard to those rights, whether they should and can remain exempted from application of the fundamental rule of national treatment. It is posited here that the reasons for limiting national treatment to the closed catalogue of rights and conventions expressly demanding its application are dwindling and that discrimination on the basis of nationality should be discouraged more generally, except where barring others from protection is justified by sound policy reasons.

The second strand of arguments deals with the question of under which circumstances TRIPS-Plus legislation in the field of IP is liberated from constraints that otherwise follow from the agreement and the conventions it incorporates. The contention that any derogation from the obligations enshrined in Part II would be in violation of Art. 1(1) 2nd half-sentence TRIPS is rejected, but it is also doubted that legislatures are entirely free to introduce conditions and restrictions with regard to TRIPS-Plus elements. It is further suggested that the principle ‘the greater includes the lesser’ with its inherent qualifications may provide some guidance. However, because the application of that principle poses its own conundrums, more research is needed to arrive at a clearer picture. Such investigations link to the question of constitutionalising certain elements of international IP law, including whether there is a place in the hierarchy of international norms for a core set of legal principles which have not risen to the level of customary law, but which nevertheless command compliance beyond that owed to ‘simple’ conventional norms.

10 Floors and Ceilings in International Copyright Treaties Berne, TRIPS, WCT Minima and Maxima

Jane C. Ginsburg

Table of Contents

1. A. The Two Pillars of International Copyright Treaties 289

   1. I. Meaning of “Minima” and “Maxima” 290

   2. II. Policy Underlying Berne “Maxima” and Its Preclusive Effect 295

2. B. Berne/TRIPS/WCT Maxima Applied: The Case of the DSM Directive Art. 15 Press Publishers’ Right 296

   1. I. Why Create a Press Publishers’ Right? 297

   2. II. Does Art. 15 DSM Bestow Copyright Protection on Berne-Excluded Subject Matter? 298

   3. III. May the EU Protect Berne-Adjacent Subject Matter through Sui Generis Systems? 301

   4. IV. Even Were a Sui Generis Right in Berne-Excluded Subject Matter Permissible, Must Exceptions to That Right Be Interpreted Co-extensively with the Berne Art. 10(1) Quotation Right? 304

3. C. Conclusion 307

A. The Two Pillars of International Copyright Treaties

The Berne Convention and subsequent multilateral copyright accords rest on two pillars: national treatment and supranational substantive obligations. National treatment is a rule of non-discrimination: a Member State may not accord foreign authors less protection than it grants its own. But a second principle buttresses the first: whatever level of protection national law provides, a treaty Member State must grant foreign authors protection commensurate with the treaty’s substantive standards. Most often that obligation means that Member States whose domestic laws fall below the treaty minima must accord more protection to foreign authors. Berne’s drafters anticipated that the political precariousness of such an outcome would result in a general raising of the level of domestic protection as well.¹ In the case of Berne maxima, in theory a Member State could deny foreign Berne works the protections that it extends to local authors, if that coverage concerns subject matter that the treaties exclude – or rights that a mandatory exception mitigates. But, as the drafters may also have anticipated, most national laws are likely to incorporate Berne’s mandatory exclusions and exceptions, so that a downward discrepancy between local law and the Berne norms seems improbable – or did until the passage of art. 15 of the Digital Single Market Directive (hereafter DSM), on press publishers’ rights, called that assumption into question.

This paper addresses the “floors” (minimum substantive international protections) and “ceilings” (maximum substantive international protections) set out in the Berne Convention and subsequent multilateral copyright accords. While much scholarship has addressed Berne minima,² the maxima have generally received less attention.³ I first address the general structure of the Berne Convention, TRIPS and WCT regarding these contours, and then analyze their application to the recent “press publishers’ right” promulgated in the 2019 EU DSM.⁴

I. Meaning of “Minima” and “Maxima”

The Berne Convention contains many mandatory obligations regarding subject matter and rights. These are the provisions denoted by “shall,” for example, regarding protected subject matter, art. 2(1): “The expression ‘literary and artistic works’ shall include every production in the literary, scientific and artistic domain, whatever may be the mode or form of its expression ….” Another example is art. 2(3): “Translations, adaptations, arrangements of music and other alterations of a literary or artistic work shall be protected as original works without prejudice to the copyright in the original work.”⁵

By contrast, some subject matter provisions clearly signal their optional character. For example, art. 2(4), “It shall be a matter for legislation in the countries of the Union to determine the protection to be granted to official texts of a legislative, administrative and legal nature, and to official translations of such texts” (emphasis added here and subsequent citations). The formulation “It shall be a matter for legislation in the countries of the Union” tells us that protection for the object is permitted, not required – nor prohibited.⁶

With respect to rights, the same expressions identify the right or exception as mandatory or left to local legislation. Hence, for example, art. 8 proclaims: “Authors of literary and artistic works protected by this Convention shall enjoy the exclusive right of making and of authorizing the translation of their works throughout the term of protection of their rights in the original works.”⁷ But art. 11bis(2) states: “It shall be a matter for legislation in the countries of the Union to determine the conditions under which the rights mentioned in the preceding paragraph [various forms of communication to the public] may be exercised ….”⁸

Now consider ceilings. Berne and subsequent treaties allow Member States to create exceptions and limitations to exclusive rights, generally subject to a variety of conditions. With one exception, Berne does not impose any restrictions on the scope of rights.⁹ Because these derogations from exclusive rights are optional, they are not maxima. We return to the one rights restriction that Berne prefaces with “shall” after considering maximum subject matter.

The TRIPS and the WCT expressly incorporate the “idea–expression dichotomy,” that is, the exclusion of ideas, methods and processes from the subject matter of copyright.¹⁰ The Berne Convention does not explicitly adopt this rule, although it may be implicit in the overall concept of “literary and artistic works” or through state practice, given that most or all Member States are likely – by text and/or by caselaw – to exclude these elements from the scope of protection.¹¹ The Berne Convention goes further than the later accords in also removing facts from protection (though this exclusion may also be implicit in those agreements). Art. 2(8) states: “The protection of this Convention shall not apply to news of the day or to miscellaneous facts having the character of mere items of press information.” As a result, a Member State may not grant copyright protection to the ideas or facts (as opposed to their expression) contained within the works of foreign authors. Again, the Berne minima and maxima apply only to works of foreign Berne origin: “Protection in the country of origin is governed by domestic law.”¹²

Turning to maximum rights under Berne, the art. 10(1) quotation provision is a “shall” clause, qualified by a variety of conditions. However, on its face, it is a direction to Member States to permit the making of “quotations from a work which has already been lawfully made available to the public, provided that their making is compatible with fair practice, and their extent does not exceed that justified by the purpose, including quotations from newspaper articles and periodicals in the form of press summaries.” This Comment does not explore art. 10(1) in depth,¹³ but examines the premise that it establishes a mandatory quotation “right” within its purview. Absent a mandatory character, art. 10(1) would not be a true maximum, and any ceiling it imposes would in fact be retractable.

Aplin and Bentley contend that there are several reasons to believe that this provision imposes a mandatory requirement for Member States to provide for a quotation exception.¹⁴ The first is the text: the language “shall be permissible” indicates that the quotation provision is obligatory.¹⁵ That interpretation is bolstered by the contrasting language used in other provisions that are optional. With the exception of art. 10(1), Berne allows Member States to institute copyright limitations and exceptions, but does not impose them. For example, the very next provision of art. 10 specifies that limitations related to certain educational uses “shall be a matter for legislation in the countries of the Union.”¹⁶

Second, the records of the Stockholm Conference of 1967, where the present language of art. 10(1) was adopted, also support the notion that art. 10(1) is mandatory. The language of art. 10(1) was initially proposed by the 1963 Study Group, which repeatedly referenced the “right of quotation” and the “right to make quotations” (emphasis added), again suggesting that the exception is required.¹⁷ Finally, Aplin and Bently point to existing commentary interpreting art. 10(1) as mandatory.¹⁸ Among these, some commentators have suggested that the exceptional mandatory status of art. 10(1) reflects its dual operation: it is a limitation that curbs one author’s right in order to benefit not only the general public but also other authors, who in many fields rely upon the ability to quote other works.¹⁹ This rationale, however, does not explain why permitted exceptions, many of which also further downstream authorship, should not also be mandatory.²⁰

Nonetheless, not all commentators agree that art. 10(1) is mandatory; some²¹ contend that the provision merely permits rather than requires a quotation right. Ficsor, for example, has argued that because Berne expressly provides that Member States can enter into agreements providing higher levels of protection,²² art. 10(1) is not obligatory, at least in principle.²³ Ficsor also notes that the practice of Member States, specifically the European Union, has been to interpret art. 10(1) as optional.²⁴ In particular, the InfoSoc Directive expressly provides that “Member States may provide for exceptions and limitations” (emphasis added) regarding

quotations for purposes such as criticism or review, provided that they relate to a work or other subject-matter which has already been lawfully made available to the public, that, unless this turns out to be impossible, the source, including the author’s name, is indicated, and that their use is in accordance with fair practice, and to the extent required by the specific purpose.²⁵

Aplin and Bentley also acknowledge that the practice of EU members has not been to treat art. 10(1) as obligatory, with Sweden being one of the very few countries – or the only one – that has enacted domestic legislation that fully implements the requirements of art. 10(1).²⁶ They contend, however, that the EU’s seemingly optional implementation of a quotation exception in the InfoSoc Directive is not necessarily in conflict with the art. 10(1) requirement, as the Directive covers both Berne and non-Berne works.²⁷

In any event, while the scholarship is somewhat divided, the weight of authorities seems to favor the interpretation that art. 10(1) is mandatory.²⁸ Even among the commentators who agree that quotation right is obligatory, however, there remains some disagreement about the implementation of the right. Professors Goldstein and Hugenholtz, for example, argue that “[a]lthough Article 10(1) is mandatory rather than permissive, national legislatures presumably are free to prescribe the conditions on which quotation is permitted,” and thus see no conflict in principle or practice with the InfoSoc Directive.²⁹ For purposes of this Comment, we grant the premise that art. 10(1) is mandatory and therefore consider its application to the new EU press publishers’ right.

II. Policy Underlying Berne “Maxima” and Its Preclusive Effect

A concern to maintain the free international flow of basic elements of information appears to animate and unite the Berne maxima. These provisions offer the Convention’s strongest expression of solicitude for the broader public interest, notwithstanding the Convention’s overall goal to protect the rights of authors. The Convention cannot prevent a Member State from locally privatizing information its own authors generate – that is the consequence of art. 5(3) – but it can require that Member States preserve the freedom of these excluded elements when the works that contain them traverse borders. Thus, if national legislation purports to grant protection to Union authors in such cases, this must be contrary to the Convention.

Nor would Berne art. 19 change that conclusion. That provision declares that “[t]he provisions of this Convention shall not preclude the making of a claim to the benefit of any greater protection which may be granted by legislation in a country of the Union.” It addresses protection for works of authorship and therefore still comes within the general Berne framework. Under art. 5(1), authors enjoy rights “in respect of works for which they are protected under this Convention ….” Works, or elements of works, omitted or excluded from Berne subject matter thus fall outside the ambit of art. 19, and Union authors therefore have no treaty entitlement to protection for such subject matter. But the concept of Berne maxima goes further, in that it would deny Member States the option of according foreign Union authors copyright protection to certain subject matter (the news of the day). By the same token, while art. 19 clearly extends to rights in protected subject matter that are not specified among the mandatory minimum rights of the Convention, it should be understood as entitling Union authors to claim “greater protection” in Member States so long as their domestic law is not inconsistent with Berne norms. Member States may supplement Berne minimum rights but may not undermine the policies underlying the principle of maximum protection. Whether as a matter of national treatment under art. 5(1), or of claim to greater rights under art. 19 (which, in this respect, reinforces the rule of national treatment to make clear that the rule extends beyond Convention’s minima), if domestic protection is “greater” because, for example, the Member State does not provide for quotation rights, that State may not insulate foreign Berne works from acts coming within the scope of art. 10(1), because the Member State would thus be rendering impermissible that which Berne declares “shall be permissible.”

This reading of art. 19 can draw on support from Berne art. 20. This provision permits Berne Union members to enter into “special agreements among themselves, in so far as such agreements grant to authors more extensive rights than those granted by the Convention, or contain other provisions not contrary to this Convention.” If those agreements exceed the Berne maxima, then they contravene art. 20. One might imply a similar limitation in art. 19. Admittedly, one might instead contend, by way of negative inference, that the absence of a similar proviso in art. 19 suggests that Union authors may claim greater protection in a Member State even if that state’s domestic protection contravenes Berne. Such a textually permissible reading, however, seems inconsistent with the overall structure and goals of Berne.

On the other hand, the “special agreements” that art. 20 references concern authors’ rights; they are copyright agreements. If Berne, TRIPS and the WCT prohibit copyright coverage of ideas and facts, does it follow that Member States may not protect those elements by other means – such as a sui generis neighboring right (which would, in effect, remove the malodor by applying any other name to the same stinkweed) or by resort to another international norm, such as the Paris Convention art. 10bis guarantee of protection against unfair competition?³⁰ Can one derive a preclusive effect of Berne, TRIPS and WCT from those exclusions, or does the path remain open to Member States to pursue protection by other means? DSM Directive art. 15 casts those questions into sharp relief, as we see in the next part.

B. Berne/TRIPS/WCT Maxima Applied: The Case of the DSM Directive Art. 15 Press Publishers’ Right

First, an overview of the provision and its rationale, as set out in the accompanying Recitals. Art. 15 provides, in relevant part:

Protection of press publications concerning online uses

1. 1. Member States shall provide publishers of press publications established in a Member State with the rights provided for in Article 2 and Article 3(2) of Direcve 2001/29/EC [reproduction and communication to the public] for the online use of their press publications by information society service providers.

   …

The rights provided for in the first subparagraph [of art. 15(1)] shall not apply in respect of the use of individual words or very short extracts of a press publication.

Art. 2(4) defines “press publications” as follows:

• ‘[P]ress publication’ means a collection composed mainly of literary works of a journalistic nature, but which can also include other works or other subject matter, and which:

  1. (a) constitutes an individual item within a periodical or regularly updated publication under a single title, such as a newspaper or a general or special interest magazine;

  2. (b) has the purpose of providing the general public with information related to news or other topics; and

  3. (c) is published in any media under the initiative, editorial responsibility and control of a service provider.

Periodicals that are published for scientific or academic purposes, such as scientific journals, are not press publications for the purposes of this Directive.

I. Why Create a Press Publishers’ Right?

The EU Commission perceived that the practices among third-party online services of news aggregation and other copying from the websites of newspapers and periodicals threatened those publications’ continued existence.³¹ The Commission therefore provided a two-year³² neighboring right³³ of the “the same scope as the rights of reproduction and making available to the public provided for in [the Information Society] Directive” and subject to “the same provisions on exceptions and limitations as those applicable to the rights provided for in [that] Directive, including the exception in the case of quotations for purposes such as criticism or review provided for in art. 5(3)(d) of that Directive.”³⁴ The objectives are clear, namely to insulate press publishers from online services’ predatory practices and to require remuneration for the services’ copying and communication to the public.³⁵ But art. 15’s subject matter coverage is unclear.³⁶ On the one hand, Recital 57 states: “The rights granted to publishers of press publications should not … extend to mere facts reported in press publications.” Recital 58 reinforces that exclusion. While extending the neighboring right to “parts of press publications,” it cautions,

such uses of parts of press publications have also gained economic relevance. At the same time, the use of individual words or very short extracts of press publications by information society service providers may not undermine the investments made by publishers of press publications in the production of content. Therefore, it is appropriate to provide that the use of individual words or very short extracts of press publications should not fall within the scope of the rights provided for in this Directive. Taking into account the massive aggregation and use of press publications by information society service providers, it is important that the exclusion of very short extracts be interpreted in such a way as not to affect the effectiveness of the rights provided for in this Directive.

(Emphasis supplied.)

The highlighted phrase suggests that the meaning of “very short extracts” may depend on the significance of the economic impact of the appropriation of those extracts. In some instances, “the use of individual words or very short extracts of press publications by information society service providers may [might] not undermine the investments made by publishers of press publications in the production of content,” but in other cases, the practice of “massive aggregation” of small amounts of content by service providers could cumulatively cause economic harm. An effective remedy therefore might need to apply granularly. But would such relief run afoul of Berne’s subject-matter limitations?

II. Does Art. 15 DSM Bestow Copyright Protection on Berne-Excluded Subject Matter?

To the extent that art. 15 DSM provides extra-national copyright protection to the “news of the day” or “mere items of press information,” it would violate art. 2(8) Berne. Whether the rights conferred qualify as copyright, that is, “the protections of [the Berne] Convention,” or are more accurately characterized as a sui generis system of protection, is discussed in the section below. Here, the question is whether “press publications” include the subject matter expressly excluded under art. 2(8) Berne.

To begin, it is necessary to determine the scope of the exclusions in art. 2(8). What qualifies as “news of the day” or “items of press information”? The provision excluding the news of the day and items of press information from protection was moved from art. 9 to art. 2 during the 1967 Stockholm Conference revisions.³⁷ As the Records of the Conference indicate, “[t]he precise meaning of the provision is far from clear.”³⁸ The question of whether the provision could be improved or clarified was first raised by the Permanent Committee at its 1958 session in Geneva and subsequently discussed by the Study Group in its 1963 Report.³⁹ In its report, the Study Group ultimately adopted the following understanding of the provision:

The correct meaning of this provision is that it excludes from protection articles containing news of the day or miscellaneous information, provided that such articles have the character of mere items of news, since news of this kind does not fulfill the conditions required for the admission to the category of literary or artistic works.⁴⁰

Thus, the role of the provision was merely “to recall the general principle whereby the title to protection of articles of this kind, as in the case of other intellectual works, pre-supposes the quality of literary or artistic works within the meaning of the Convention.”⁴¹

Note that the Study Group perceived the exclusion to apply to entire articles and not merely to the information they contained. It appears that the Study Group assumed that the articles would be so devoid of authorship as to fail to qualify as a “literary or artistic work.” As such, the Study Group considered the “news of the day” exclusion to be a “superfluous element,” but retained the provision nonetheless. Moreover, although there had been some discussion of modifying the provision to improve its clarity, the Study Group concluded that no modification was necessary as “it would be sufficient to discuss the question of interpretation in the documents of the Conference.”⁴² That position was reaffirmed in the Study Group’s 1964 Report.⁴³

The report of the Main Committee on the Programme of the Conference reiterates this view. The report concluded that “the provision only seeks to establish that the Convention does not protect mere items concerning the news of the day or miscellaneous facts (and, a fortiori, the news or the facts themselves).”⁴⁴ The provision was not intended, however, to exclude “articles” or “other journalistic works reporting the news … if they can be considered as works within the meaning of the Convention.”⁴⁵ On this point, the Committee believed, it could “hardly be claimed that there [was] any obvious need to clarify the text of the Convention….”⁴⁶ Thus, art. 2(8) appears to function less as a provision of exclusion so much as a reiteration that recitations of facts that do not themselves qualify as intellectual creations, and which therefore are not literary or artistic works, are not included.

The commentary on art. 2(8) is in accord with that conclusion. The 1978 Guide to the Berne Convention interprets the provision to exclude not only news and facts but also “the simple telling of them, since matters of this kind lack the necessary conditions to be considered as falling into the category of literary and artistic works.”⁴⁷ Masouyé, the author of the 1978 WIPO Guide, viewed the provision as merely confirming “the general principle that for a work to be protected, it must contain a sufficient element of intellectual creation.”⁴⁸ Thus, while stories “related with a measure of originality” are protected under art. 2(1), “simple account[s], arid and impersonal, of news and miscellaneous facts” are not.⁴⁹

Given the above understanding, art. 15 DSM would violate art. 2(8) Berne if its protection of press publications extends either to facts themselves or to mere recountings of facts that lack sufficient original expression. As defined in the DSM Directive, press publications are certain collections “mainly composed of literary works of a journalistic nature, but which may include other works or other subject matter.”⁵⁰ While “literary works,” and “works” generally, are properly the subject of copyright protection under Berne⁵¹, the possibility of inclusion of “other subject matter” within the scope of protection raises a potential conflict with art. 2(8) Berne. Specifically, would the “news of the day” and “items of press information” be included within this “other subject matter” and consequently protected?

Recital 56 of the DSM Directive provides some elaboration on the scope of protection. In particular, Recital 56 clarifies that “press publications contain mostly literary works, but increasingly include other types of works and other subject matter, in particular photographs and videos.”⁵² Though presumably not exhaustive, the illustrative examples of photographs and videos as other types of work and subject matter suggest the Directive is not intended to cover the otherwise unprotectable “news of the day” or “items of press information,” since photographs and videos generally qualify as artistic works. Recital 57 is more explicit: the rights granted to publishers of press publications “should also not extend to mere facts reported in press publications.” Still, as explained above, art. 2(8) appears to extend slightly beyond the facts themselves and also excludes sterile accounts of facts, regardless of length. Thus, while art. 15 may not protect facts or “individual words or very short extracts of a press publication,”⁵³ to the extent that it protects press publications that include factual accounts too lacking in originality to support a copyright,⁵⁴ the Directive may be covering subject matter excluded under art. 2(8) Berne. Moreover, as discussed above, the potential for coverage of economically valuable “very short extracts” might create tension with Berne art. 2(8).

III. May the EU Protect Berne-Adjacent Subject Matter through Sui Generis Systems?

Art. 2(8) Berne excludes certain subject matter from copyright protection, but it generally does not prevent Union members from protecting that subject matter under different regimes, including sui generis forms of protection (see discussion below). An initial question then is whether art. 15 DSM vests publishers with copyrights in press publications or instead establishes a sui generis system. Although art. 15(1) DSM nominally provides the same copyright protections as conferred in arts. 2 and 3 of the InfoSoc Directive, it limits those rights in important ways not consistent with other copyright protection. Perhaps most importantly, the primary beneficiary and holder of the right is not necessarily the author(s) but the publisher.⁵⁵ Second, the term of protection is limited to just two years, beginning with publication – in contrast to Berne’s minimum fifty years post mortem auctoris).⁵⁶ Additionally, the scope of the art. 15 right is limited specifically to “online use[s]” by information service providers and does not apply “to the acts of hyperlinking.”⁵⁷ Recital 55 also makes clear that the rights granted are not copyrights per se but “rights related to copyright.” Finally, the granting of rights is not expressly predicated on the presence of original expression but rather the “organisational and financial contribution of publishers in producing press publications.”⁵⁸

Given these significant differences from the traditional copyright regime, there is a strong argument that the rights granted in press publications are not just copyright by another name, but instead are genuinely sui generis. One then must ask whether the protection of this Berne-adjacent subject matter through a sui generis regime is permissible. As Henning Grosse Ruse-Khan and Annette Kur observe, the ability to protect Berne-excluded subject matter through different means is problematic.⁵⁹ Nonetheless, both the records of the Stockholm Conference and the commentary on the 1971 Paris text of the Berne Convention agree that such protection is permissible. As described in the Conference Records, one of the utilities of art. 2(8), despite its otherwise superfluous nature, was to “permit the conclusion that if the articles concerned are protected by other legal provisions – for example, by legislation against unfair competition – such protection is outside the field of the Convention.”⁶⁰ Similarly, the provision helped to fix “the line of demarcation between copyright and other means of protection.”⁶¹ Thus, the possibility of other means of protection was expressly contemplated and was not accompanied by any signs of disapproval.

Commentary on art. 2(8) also endorses the view that sui generis protection is permissible. Professors Goldstein and Hugenholtz state that “[l]ike ideas, news of the day and data compilations may be protected outside copyright under unfair competition law, neighboring rights, or sui generis regimes.”⁶² Similarly, in the 2003 WIPO Guide, Ficsor notes that the subject matter of art. 2(8) can be protected “on the basis of some legal institutions other than copyright – such as a sui generis system for the protection of databases and their contents, or unfair competition….”⁶³ Other commentators agree.⁶⁴ Indeed, although they acknowledge that the results may be troublesome, Kur and Grosse Ruse-Khan emphasize that “the relevance of Art. 1(1) TRIPS [and art. 2(8) Berne] is limited to mandatory exclusion of subject matter from copyright, whereas it does not appear as a tenable position to argue that it also applies if information or data are under a sui generis regime deliberately established for the purpose of granting such protection.”⁶⁵

The European Union’s adoption of the European Database Directive⁶⁶ occasioned concrete application of the principle that a sui generis right might supply protection withheld by the Berne Convention. Similarly to art. 15 of the DSM Directive, the Database Directive provides sui generis protection with respect to the substantial investment in the compilation of otherwise unprotectable data. While the Database Directive has incurred both practical and theoretical objections,⁶⁷ these criticisms have not evoked an underlying incompatibility with Berne art. 2(8). Similarly, while a draft treaty proposing international protection for databases was not adopted at the Diplomatic Conference on Certain Copyright and Neighboring Rights Questions in 1996 (or anytime thereafter), there is no record of objections premised on the exclusion of such subject matter from copyright.⁶⁸

Rather, the debate has centered around which form of legal protection – a sui generis intellectual property protection or a misappropriation right sounding in unfair competition – was best suited to the task of protecting the investment in compiling databases. Advocates of more expansive and definite protection preferred a sui generis right with more precise details, well-defined term of protection, and greater facility for licensing.⁶⁹ Skeptics of the economic benefits or necessity of database protection favored the more limited protection of misappropriation claims.⁷⁰ In any event, the ability of Berne members to establish other forms of protection, including sui generis intellectual property rights, seems to have gone unquestioned.

IV. Even Were a Sui Generis Right in Berne-Excluded Subject Matter Permissible, Must Exceptions to That Right Be Interpreted Co-extensively with the Berne Art. 10(1) Quotation Right?

In the absence of a full examination of what constitutes a “quotation” under Berne art. 10(1),⁷¹ one may nonetheless question whether the press publisher’s right is compatible with Berne art. 10(1). Art. 15(3) of the Berne Convention directs that the exceptions set out in the 2001 InfoSoc Directive “shall apply mutatis mutandis in respect of the rights provided for in paragraph 1 of this Article.”⁷² The incorporation in the InfoSoc Directive of an optional quotation privilege in terms nearly identical to Berne art. 10(1)⁷³ suggests that one may avoid discrepancies between the two instruments by interpreting art. 15 DSM co-extensively with InfoSoc Directive art. 5(d), which, in turn should track Berne art. 10(1). In that event, the copyright-adjacent nature of art. 15, while potentially problematic with respect to covered subject matter, will not immunize the press publishers’ right from third parties’ quotation rights.

There are two caveats to the above. First, the quotation right in Berne, according to most commentators, is mandatory: Member States must allow quotations (within the contours of the right). By contrast, the InfoSoc Directive leaves the list of permitted exceptions and limitations in art. 5(3) up to national adoption (or not). On the other hand, DSM Directive art. 17(7) makes the quotation exception (among others) mandatory with respect to content posted by users to Online Content Sharing Service Providers.⁷⁴ While art. 17, on the liability of Online Content Sharing Service Providers for infringing content posed by their users, addresses a different problem from the one that occasioned art. 15, there may be some overlap between the entities that are Online Content Sharing Service Providers under art. 17 and the “information society service providers” that are subject to the press publishers’ right. Arguably, the European Union may be creeping toward substantive equivalence with the Berne norm with respect to its mandatory character.⁷⁵

Second, even assuming third parties will enjoy quotation rights in press publications, the scope of the quotation exception may differ between InfoSoc Directive art. 5(3)(d) and DSM Directive art. 15(3). The same words may mean different things in different contexts, and the requirement that the quotation be “in accordance with fair practice” may impose different constraints on the exercise of the quotation right. Given that the practice of news aggregation spurred the enactment of art. 15, the size and amount of the “quotations” and their economic impact may bear more heavily on the assessment of incompatibility with fair practice for press publications than courts might tolerate for works of authorship. Thus, interpretation of the two instruments may not be fully coextensive: the principles may be the same, but their application may not yield identical results. (However, the same might be said of the assessment of “fair practice” across different kinds of works of authorship, or regarding different purposes for the quotations.)

The adoption in art. 15 of the InfoSoc Directive exceptions avoids a confrontation between Berne norms and an unbounded sui generis right over subject matter that includes works of authorship as well as Berne-excluded content. Recall that DSM Directive art. 2(4) defines press publications to cover “an individual item within a periodical”; that item generally will be a whole article or a substantial extract (Recital 58 generally excludes “very short extracts,” though the meaning of the term may vary with economic impact). Acknowledging that a neighboring rights regime over Berne-excluded subject matter may coexist with copyright,⁷⁶ one may still inquire whether Berne maxima should exert a preclusive effect when the subject matter of the sui generis right includes works of authorship. If, for example, art. 15 covered both copyrightable and non-copyrightable content, but did not also incorporate copyright exceptions, so that a quotation exception would not limit the scope of the press publishers’ right, then publishers could invoke the sui generis right to prevent quotations from the same copyrightable content to which their rights under copyright must yield.⁷⁷ The argument for a Berne-preclusive effect seems strongest when the sui generis right covers both copyrightable and non-copyrightable content. It should not be permissible to end-run the Berne quotation right by resort to sui generis protection against copying the same subject matter.

Chapter III, “Sui Generis Right,” of the 1996 Database Directive, however, may belie that proposition. As we have seen, the Database Directive covers both original and non-original databases and provides a sui generis right against extraction and reutilization of substantial parts (whether or not copyright-infringing) of databases that are the fruit of substantial investment. While Chapter II, “Copyright,” of the Directive permits Member States to provide for copyright exceptions “traditionally authorized under national law,”⁷⁸ Chapter III sets out three specific exceptions and limitations (which do not include a quotation provision), without Chapter II’s open-ended catch-all.⁷⁹ The restriction in Chapter III of the extraction right to “insubstantial parts” of the database⁸⁰ will place some quotations outside the ambit of the database holder’s exclusivity. But to the extent that a copyright-permissible “quotation” may be qualitatively or quantitatively substantial,⁸¹ Chapter III of the Database Directive would appear to grant the rightholder a remedy, whereas Chapter II would allow an exception. If non-copyright material entirely comprises the “quotation,” then once one has admitted the premise that Berne Member States may establish sui generis rights in copyright-excluded content, perhaps copyright limitations need not constrain the scope of rights in that subject matter.⁸² But if the quotation comprehends a substantial extract of copyrightable expression, then Chapters II and III appear to be in tension.⁸³

C. Conclusion

Within the universe of multilateral copyright obligations, the Berne maxima, buttressed by the TRIPS and WCT exclusion of protection for ideas, methods and processes, should promote the free cross-border availability of facts and ideas, as well as of exercise of the quotation right. Individual Berne countries of origin may protect excluded subject matter in their own works of authorship, but not in foreign Berne works. Conversely, those countries must apply the quotation right to foreign Berne works, but need not do so to their own. Nonetheless, there exist at least two challenges to this equilibrium. The first, as we have seen, concerns the potential for Berne members to protect excluded subject matter, or to avoid the quotation right, by resort to sui generis regimes. The second concerns the EU principle of non-discrimination:⁸⁴ Berne may limit protection in excess of its maxima to the country of origin, but EU norms require Member States to accord full national treatment, thus granting to works by other EU nationals the same scope of protection as the EU country of origin provides its own authors.

This cornerstone of EU law potentially places EU Member States in conflict with their international obligations: on the one hand, they may not – by copyright – protect Berne-excluded subject matter in foreign works, including works by EU nationals; on the other hand, EU norms oblige Member States to extend to other EU nationals the protections Berne would deny them. The EU regional public law obligation of national treatment thus may clash with international obligations enshrined in Berne, TRIPS and the WIPO Copyright Treaties. As we have seen, denominating the protections as sui generis might not suffice to shield them from the international treaties’ preemptive effect. If not, then, as the editors of this volume observe in their preface, the international IP treaties produce “a level of global harmonization of IP protection whose cumulative effect leaves little space for different national policy choices.”