Introduction
In the September issue of the Swiss journal for judges («Justice - Justiz - Giustizia»), a practitioner expressed criticism regarding the proportional distribution («Parteienproporz») of judges according to party affiliation and related mandate taxes.[1] When current developments in such a forum are being discussed, their relevance for the law is undisputed. At the same time, only a small group of people can access these comments. The password-protected area is only available to subscribers. Should they not be openly accessible? Are there reasons to take action, so that scholarly publications can be made accessible for everyone? In particular the adoption of the Swiss National Strategy on Open Access (OA) and the following action plan («Aktionsplan»), approved in February 2018 by the plenary assembly of Swissuniversities, puts this issue on the agenda in Switzerland. The strategy and action plan place great emphasis on flexibility for the universities, especially in the process of adapting the framework.[2]
Nevertheless, many aspects still remain unclear. While there are many factors, which play a role in the discussion, this paper will focus on making an inventory of the legal regimes which govern the question of OA in different jurisdictions. One of the said unresolved issues, related to the relationship between researchers, their home institutions, publishers and society, with a doctrinal focus on identifying and interpreting the rules applicable to OA, is the subject of this publication: Can and should university employees be obligated to provide Open Access to their publications?
I. Current Situation
In the various academic disciplines, different OA approaches, models and attitudes can be found.[3] While OA journals have gained a strong footing in the natural sciences in particular,[4] conventional journal publishing is still dominating the market in the legal sphere.[5] This prominent position of publishers in the field of print and digital journal subscriptions has led to increasing costs for libraries in recent years.[6] Only very few people are aware of the sums actually paid by the libraries.[7] Declining library budgets and the sheer increase in information volume, mean that a significant portion of scholarly literature is no longer accessible. Hence, the interest in OA is growing. Theoretically, and from the perspective of a publishing scholar, two paths that can be taken. Firstly, the publication in a journal, which provides content freely accessible online - the «gold road».[8] Secondly, the «green road», which takes a publication that has previously been published elsewhere and makes it available in an online repository.[9]
Another dimension is the institutionalization of the idea of OA. Recent developments, such as the aforementioned action plan under the national OA strategy,[10] have made the topic appear political.[11] According to Vision 2024, the aim was to have 100 percent of the publications from Swiss universities published OA by the year 2024.[12] Meanwhile, the University Council of the Swiss University Conference has approved the action plan.[13] These institutionalization steps stand between the poles of national and international copyright law on the one hand, and the idea of OA as global scholarly communication on the other.[14] In practice, different forms of institutionalizing OA have been developed, with specific features determined by their geographic perimeters. Some of them, by practical relevance, will be presented in the following.
II. International Approaches
1. Harvard University
In 2008, the Faculty of Arts and Sciences at Harvard University unanimously voted in favor of transferring the non-exclusive rights of its researchers' writings, on all forms of non-commercial use, to the university.[15] Other institutions followed.[16] From the perspective of copyright, the mentioned model is nothing more than a non-exclusive license.[17] The university will gain the rights to all works by their own employees.[18] Once the university has received the final works, they will be saved in the DASH repository,[19] accessible to anyone without registration. A not-for-profit and non-derivative use by third parties is possible, but must conform to the DASH conditions of use.[20] As a rule, the universities are trying to get written grant licenses.[21] Storage in the DASH repository, if stored as a first publication, thus corresponds to a gold road publication in another medium.[22]
Still, exemptions from this obligation to OA, known as «waivers», remain possible. Provisions are also made in respect to possible embargo periods on the part of the repository.[23] But how can a person employed at the university as a researcher benefit from the aforementioned exemption («opt out»)? To do so, they must apply to the dean of the faculty - sometimes with the requirement of giving reasons. Most of the time, the exemption is approved.[24] However, such requests are only made in about five percent of cases.[25] A second exception to the OA obligation is referred to as the «dark deposition». Here, contractual provisions between the researcher (author) and the publisher preclude the work from simultaneously being made accessible in the repository. After the expiration of an approximately six-month embargo, the «lift date» is reached and the publication is made public for everyone.[26]
Despite the continued success of this model, discussions about the infringement on academic freedoms will continue. The researcher is no longer able to decide where, when and what to publish.[27] Another, sometimes technical, challenge is the classification of the different versions. In the databases, a distinction is made between the author's manuscript, the published version and other versions.[28] The distinction is crucial, with some publishers prohibiting the distribution of the «version of record», the final version of the publisher.[29] What remains unclear are the criteria for rejecting or accepting the waiver. While some faculties require extensive justifications, others grant approval without further inquiry.[30] In relation to the specific situation in the United States, it is also questionable whether the statutory «work made for hire» scheme[31] could result in the university, as the employer, being regarded as the author, rather than the researcher herself or himself.[32]
Nevertheless, the Harvard model appears as a useful complement to the US legal system concerning copyrights. One reason is, that the system alone - even with its limitations and exceptions for the benefit of education, science and archiving - does not meet the full potential of the new technologies for educational institutions.[33]
2. University of Suffolk: UK Scholarly Communication License (UK-SCL)
In the United Kingdom, a variety of OA approaches and institutional regulations has evolved.[34] Here, some universities already have a budget for gold road publications. In this context, OA approaches are being discussed[35] and increasingly supported.[36] At the University of Sussex, for example, an approach[37] based on the Harvard model is being pursued. [38] In addition to making writings available in a database, a focus is placed on the reuse of the stored works.[39] This is done by means of standardized Creative Commons (CC) licenses, which enable third parties to use the particular work.[40] The author himself or herself can decide about the rights enjoyed by other users.[41] Proponents of this approach stress that CC licenses better meet the needs of the copyright owner, compared to the traditional reward-based system.[42] This model differs from the Harvard model due to its use of CC licenses. However, according to the DASH conditions, a subsequent use is possible there too.
At the University of Suffolk, under the «UK Scholarly Communications License», the «CC‑BY‑NC» basic form of the CC license is applied.[43] This is comprised of the «attribution» («BY») and the «non-commercial use» («NC») components. Possible additions can include the «share alike» («SA») and «no derivatives» («ND») components.[44] The necessary precondition for any type of use is the attribution of the work to the author.[45]
Furthermore, the concept of the «moving wall» license is in use. The storage of the writing in the repository allows for metadata to be made visible.[46] The UK‑SCL also provides for special exemptions. However, unlike the Harvard approach, more room is given for the institutions, whose conditions for refusing the waiver can be set forth individually.[47] This model, too, follows the assumption that the copyright of a work will be attributed to the author and not the university.[48]
3. Baden-Wuerttemberg
In Germany, the idea of OA was promoted by the Alliance of Scientific Organizations («Allianz der Wissenschaftsorganisationen»)[49] and, to the astonishment of many, by the legal committee of the Bundestag.[50] Section 38 para. 4 of the German Copyright Act (UrhG), which came into force on 1 January 2014, grants researchers, who finance their work predominantly with state funding, a nationwide self-activating republication[51] right - after a contractually mandatory waiting period of one year.[52] In the explanatory memorandum[53] of the German legislators, central justifications of the Open Access paradigm are being taken into consideration for the first time.[54] Furthermore, in 2017 the federal government voted to accept the «Draft Law on the Alignment of Copyright with the Current Requirements of the Knowledge Society» (UrhWissG).[55]
The obligation to OA has been particularly notable since the third University Law Amendment Act of April 1, 2014 (3. HRÄG), and the adoption of § 44 para. 6 of the State Higher Education Act (LHG) in Baden-Wuerttemberg. By statute, researchers are obliged to exercise their rights from the aforementioned § 38 para. 4 of the Copyright Act.[56] Thus, a disciplinary-juridical protection is in force. The right becomes a duty.[57] Still, the question of the constitutionality of such an obligation arises.[58] The Federal Constitutional Court (Bundesverfassungsgericht) places a researcher's decision about where to publish under the fundamental right of freedom of research (Art. 5 para. 3 Constitution - Grundgesetz [GG]).[59] In addition, the secondary publication requirement (Zweitveröffentlichungsrecht) contradicts § 3 para. 2 LHG, in which the freedom of research is anchored.[60] To these substantive aspects comes the question as to whether or not the federal states have legislative authority. The answer must be no if, in the process of determination, a barrier can be seen to copyright and thus substantive copyright law.[61] Do individual rights for researchers consequently emerge from the LHG? The law directs the statute obligation to universities, not directly to individuals.[62] Thus, the successful introduction of an OA obligation, as well as any actions against such a requirement, ultimately depend on the statutes of the respective universities and their implementation.[63]
III. Implementation in Switzerland
The question of whether university employees may or should be obligated to OA cannot be answered solely from the perspective of university laws. Rather, recent legal developments, implementation approaches and future challenges must be taken into consideration.[64] This notably includes cantonal law, due to the universities' funding[65] as outlined in the federal law. The Federal Act on Funding and Coordination of the Swiss Higher Education Sector (Higher Education Act, HEdA),[66] in particular, must be taken into account.[67]
1. Affected Areas of Law
Open Access and any related obligation thus concerns various legal bases. In the following, aspects of constitutional, copyright and competition law will be discussed in particular. A distinction will be made between the perspective of the university, and that of its research staff.
a) Constitutional Law
From a constitutional point of view, the first question is whether an obligation to OA could be supported by the scholarly freedom provided for in Art. 20 of the Federal Constitution of the Swiss Confederation (BV).[68] Science,[69] as a part of society's opinion-making process, requires a high degree of institutional embedding.[70] For universities, as a prerequisite for the independence of researchers, this means organizational autonomy.[71] This results in a directive to the state to promote scientific research and innovation (Art. 64 para. 1 BV).[72] Still, it needs to be clarified to what extent academic freedom is compatible with the claims of OA.
In general, academic freedom does appear to correspond with OA. An image of science whose main function is based on «cognition by method»[73] benefits from the aim of free accessibility of scientific information.[74] A part of science is the distribution of such information in an open, rational discourse.[75] In addition to determining the respective publication for distribution, free publishing also involves having the option to choose the date and channel of publication.[76] This aspect shows that an obligation to OA is in tension with academic freedom. Accordingly, such an obligation must take considerations of proportionality into account from the very start.[77]
Not only the researchers as natural, but also as legal persons may have fundamental rights under Art. 20 BV. Still, it remains unclear if and how universities should be involved.[78] Academic freedom, as part of the third-party applicability of fundamental rights («Drittwirkung der Grundrechte»), not only applies in relation to the state, but must also be considered in the interpretation between private individuals.[79] In light of an OA obligation, the interpretation of the Federal Act on Copyright and Related Rights (Copyright Act, CopA)[80] in particular appears primarily relevant. If and to what extent an interpretation of fundamental rights is relevant in practice, remains a matter of dispute.[81] Yet the aim of the norm restricts interpretations from being too extensive.[82] Furthermore, an interpretation is not automatically compliant with the Federal Constitution of the Swiss Confederation if it corresponds to academic freedom; rather, other constitutional provisions must be taken into consideration.[83] One such example involves property rights.[84] Intellectual property rights are considered a part of the property rights under Art. 26 BV.[85] Interestingly enough, fundamental liberties and copyright law reflect diametrically opposed interests. As a result of debates related to the digital age, this fact is moving into the public focus.[86]
b) Copyright and Associated Issues
Compared to other jurisdictions,[87] the Swiss CopA offers flexible responses to the development of information technology.[88] This is facilitated by the design of the usage rights as a general clause,[89] with corresponding samples («Regelbeispiele»).[90] But do they also allow for an obligation to OA?[91] From a copyright perspective, it seems clear that a university does not automatically hold the rights to works that their paid researchers produce.[92] Therefore, a publication in a university repository requires an explicit grant of rights.[93] This stems in part from the creator principle («Schöpferprinzip») ofArt. 6 CopA,[94] and of Art. 9 para. 2 CopA,[95] whereby the author can decide what, where and when to publish. If deciding to negotiate directly with the publisher, within the scope of the author's contractual autonomy, the authors typically find themselves in a weaker negotiating position.[96] As a result, the author often - sometimes unwittingly - transfers the corresponding «online rights».[97] Nevertheless, simple licenses[98] could facilitate the author's retention of rights without any problem.[99] Templates for such licenses can be found online, for example on the SPARC website.[100] They can then be incorporated into a corresponding contract.[101]
The default rules of the law shall apply, if parties did not derogate from these rules by agreement.[102] Where Swiss law applies, Art. 381 para. 1 of the Federal Act on the Amendment of the Swiss Civil Code (Part Five: The Code of Obligations, OR)[103] is of particular relevance. According to this article, only those rights shall be transferred that are actually needed for the implementation of the contract.[104] If the author concludes a publishing contract with an OA publisher or repository, the reproduction[105] or access rights[106] are no longer in the foreground, their place being taken by the technical and sales-related services of the provider.[107] The interest of the author in granting a non-exclusive license corresponds with that of the university as operator of the repository. Ideally, the latter has no interest in an extensive or even commercial exploitation of the rights.[108] For this reason, problems arise especially when the author is bound by a publishing contract.[109] In such cases, negotiations between the university and the publisher seems to be the most promising for all involved.[110] As in the United States,[111] cooperating publishers can then benefit from their appearance on the university website.[112]
Altogether, it can be stated that the existing Swiss CopA enables, but does not promote, OA publications. What changes in favor of OA would be possible on the part of the legislature or universities? One approach would be the combination of public and copyright law rules.[113] Through mandatory rules, the right to make publicly accessible would remain with the copyright holder. Immediately, or after a specified period, the author would have the option or obligation to deposit the work in the repository.[114] A further proposal to introduce compulsory licenses appears as a relic of an old-fashioned copyright system and is thus only partially effective.[115] Barrier schemes («Schrankenregelungen»)[116] that benefit the operator of the repository must respect the requirements of the three-step test[117] - and seem difficult to enforce politically, because they would deprive the author of a part of his or her copyright from the outset.[118] It should be noted that most of these proposals do not include OA obligations, but try to promote the aims by other means. Suggestions for OA obligations generally relate only to higher education law, as well as to regulations of funding organizations, such as the Swiss National Science Foundation (SNF).
c) Competition Law
Some publishers of reputable scholarly journals hold a quasi-monopolistic position. This is reinforced by several trends. Aspects of competition law are also noted in online platforms with a solid customer base. While offering customers added value, they complicate the entrance into the market for future competitors.[119] On an individual basis, this situation can be avoided in part by self-archiving. Collectively and sustainably, this can happen through the creation and preservation of OA business models.[120]
In the promotion of Open Access, there is regular mention of antitrust price control over the large publishing companies, and thus of measures for guaranteeing competition.[121] If, in the national OA agenda, an obligation to publishing Open Access in institutional repositories is pursued, these repositories become competitors, because they effectively constitute a part of the publishing industry.[122]
2. The Example of the University of St.Gallen and the National Science Foundation
Self-regulation is clearly on the agenda in the field of research. It determines the forms for publication, communication and reception.[123] As a result, many institutions in Switzerland already have OA servers or archives.[124] The various OA strategies at the universities have to be considered separately from the technical implementation. They differ in terms of form, content and institutional embedding.[125] The University of St. Gallen, for example, has an OA regulation that governs various aspects. The duties of the researchers include providing bibliographic data, full texts as a pre- or post-print version, and ensuring non-exclusive distribution rights in publishing contracts. A higher level of liability arises with the Implementation Provisions for Doctoral Studies of May 2, 2017. Pursuant to Art. 31 para. 1, the doctoral candidates must declare that they transfer to the university the rights to save the dissertation electronically, to make the dissertation publicly available in data networks and to provide further copies to the professors of the University of St. Gallen.
In Switzerland, the Swiss National Science Foundation (SNF) attracted particular attention with its obligation requirements. The aforementioned Vision 2024 aims at introducing and coordinating OA policies, negotiations with publishers (known as offset agreements, or the «offsetting road»), the use of alternative forms of publication,[126] the coordination and pooling of resources,[127] the formation of a supportive regulatory framework, the establishment of a national monitoring system, and raising of awareness in general. Different scenarios are feasible.[128] As co-signatory of the Berlin Declaration, the SNF sees itself as a promoter of OA efforts and is trying to develop a corresponding political framework. It is therefore not surprising to contribution recipients that Open Access publication is required.[129]
3. Challenges for the Establishment of OA Obligations
Different interests of the parties involved - the institutions, individual researchers and representatives of the publishing industry[130] - as well as the international context of OA[131] make a possible introduction of OA obligations into a complex project. This circumstance becomes particularly evident when a functioning and reliable OA obligation rule must be transferred to another institution.[132] For Switzerland, with a decentralized education and research system,[133] this is not much different.[134] The legal rules and institutional basis of OA obligations would be of importance for the implementation, but they are no guarantee of success. Instead, individual researchers need to be convinced of the idea.[135] Such conviction may arise, in particular, if a similar boost of reputation can be achieved through OA publications. Furthermore, personal advocacy is needed on the part of universities.[136] While a wide access scheme is in demand for foreign research results, the university's own publications are largely still being held under wraps.[137]
IV. Conclusion
When addressing Open Access, the specific implementation must be distinguished from the vision.[138] In answering the question of whether an OA obligation can be introduced for researchers in Switzerland, a more concrete framework would be desirable. What must be clarified, in particular, is whether the Swiss legislator in the ongoing revision of the Copyright Act is willing to introduce a secondary publication right («Zweitver-öffentlichungsrecht») for researchers.[139] Furthermore, in relation to Art. 20 BV, concretization is needed with regard to the question of whether an OA obligation would be consistent with fundamental rights.[140] At the moment, only sparse case law exists.
OA strategies and rules, as bases of an OA obligation, not only concern the legal sphere, but also social and political issues. Especially when it comes to promoting innovation, a discourse between representatives of the scientific, social, economic and technological spheres would be fruitful.[141] While supporters see the OA obligation as an effective way to achieve the OA vision and view it as a future direction for research,[142] opponents see it as a threat to the existing legal understanding of the publishing industry, and to the culture of books.[143]
Keeping this is mind, the question arises as to how to legislate a possible OA obligation: as compulsory or as a «mandatory option» with the possibility of opting out?[144] Three proposals are being considered as plausible - which ideally would be combined altogether They are: An OA obligation in higher education law with an opt-out option, based on the English model,[145] the introduction of a secondary publication law, similar to the one in the Federal Republic of Germany,[146] and an antitrust[147] price control in view of quasi-monopolistic publishing groups.[148]