Abstract
The article analyzes one of the earliest intergovernmental initiatives regarding Internet constitutionalism. Based on an analytical framework that combines argumentative discourse analysis with elements from actor–network theory, it assesses the actors and discourses involved in the preparation of an international recommendation on universal access to the Internet, adopted by the United Nations Educational, Scientific, and Cultural Organization (UNESCO) in 2003. During the contentious negotiations, UNESCO had to find a balance between the divergent positions of its member states and to account for the different levels of technological development amongst them. By retracing the performative and discursive struggles that UNESCO had to face to reach consensus, the article contributes to the general understanding of ideas and practices behind Internet policy making in international settings.
Keywords
Introduction
UNESCO, as the ‘intellectual’ Specialized Agency of the United Nations, is a natural forum for building consensus on issues related to cyberspace. The Organization will […] take the lead in drawing up a body of principles applicable to cyberspace, with a view to reaching universal agreement at the threshold of the twenty-first century. (UNESCO, 1997a: 3)
Among UNESCO's many efforts in this regard, there is one episode in particular that lends itself to studying early efforts of digital constitutionalism, understood here as ‘initiatives that have sought to articulate a set of political rights, governance norms, and limitations on the exercise of power on the Internet’ (Gill et al., 2015: 2). In 2003, just before the first WSIS phase, UNESCO adopted an international instrument on Internet-related issues ranging from universal access and cultural diversity to problems concerning the public domain and intellectual property rights (IPR). This ‘Recommendation concerning the promotion and use of multilingualism and universal access to cyberspace’ was the result of a 6-year drafting process during which UNESCO had to overcome many contentions among its member states and the involved experts to reach a global consensus on policy problems that emerged due to the pervasive yet unequal spread of digital networks. The instrument corresponded to the four dimensions that, according to Gill et al., characterize attempts at digital constitutionalism: it addressed ‘broad and fundamental political questions’ related to the Internet; it was directed at a ‘particular and defined political community’, namely UNESCO's member states; as an intergovernmental recommendation, it aimed at ‘formalized political recognition’; and, finally, by covering a large set of issues, it showed a high ‘degree of comprehensiveness’ (2015: 3). Hence, the instrument's preparation can be considered one of the first intergovernmental attempts to adopt a global framework of Internet norms and principles, offering a unique opportunity to study the discourses and power struggles in early international debates about the Internet's impact on society.
The article scrutinizes the difficult history of UNESCO's 2003 recommendation by studying the actors and discourses involved in its preparation and how they interacted to shape UNESCO's first official policy response to the Internet. To this end, the article first briefly introduces the conceptual framework. It then retraces the recommendation's drafting process with a focus on the power struggles among its protagonists. Lastly, it assesses the competing discourses influencing the recommendation and provides a short interpretation of the discursive interactions and their role for Internet policy making.
Analyzing policy discourse ‘in the making’
The analysis of the recommendation's preparation process is guided by the aim to understand the dynamics of UNESCO’ policy-making practices. Moving the assessment away from the established teleological perspective of policy analysis, the empirical assessment does not focus on the input and output of policy processes and their causal relations. Instead, it is interested in the creation of policy discourse through the meaning-making capacities of involved actors, including the deliberations, interactions, and power relations between actors. To account for this particular interest, the article is based on a conceptual framework that links the analysis of discursive struggles to the concrete processes in which they occur. Instead of scrutinizing the content of the final recommendation text itself, the assessment follows the idea that, more than anything else, it is the arguments used in policy debates that define the conceptualization and solution of policy problems. Therefore, it draws on Marteen Hajer's transformation of this idea into the approach of argumentative discourse analysis (ADA) (Fischer and Forester, 1993; Hajer, 1993; Hajer and Wagenaar, 2003), which allows us to assess the discursive dimension of policy making. This means it is used to analyze the argumentative structures and discursive exchanges during UNESCO's policy debates and to identify ‘discourse coalitions’ (Hajer, 1993: 45) seeking to institutionalize their own ideas in the recommendation (Hajer, 2005: 303).
The focus on discourse theory is combined with selected concepts of actor–network theory (ANT). This second approach is used to analyze the performative dimension of policy making, meaning the social creation of meaning and objects in heterogeneous policy-making environments. The article draws particularly on the ideas of ANT authors related to production processes and social ordering (Akrich, 1994; Latour, 1987, 1999). Most importantly, the assessment is based on an ANT-inspired relational and nondeterministic concept of power. Believing that power is a result of social interactions rather than their cause (Latour, 1986: 264ff), the analysis aims to retrace the concrete way in which actors shape their power relations while seeking to create—together, yet in constant struggle—a convincing policy discourse that could be inscribed into UNESCO's cyberspace recommendation (Pohle, 2016a: 174ff). Consequently, the combination of these ANT-inspired observations with an ADA-inspired discourse analysis makes it possible to observe meaning ‘in the making’ (Latour and Crawford, 1993: 265). 1
The conceptual approach's translation into empirical research builds on the study of the official documents and verbatim records of UNESCO's governing bodies between 1996 and 2003. In addition, it is based on the extensive analysis of all relevant archival records, including draft versions of the recommendation, comments submitted by member states and experts, handwritten minutes, and summarized proceedings of meetings as well as UNESCO Secretariat correspondence. To clarify findings with the help of subjective in-depth accounts, the archival research was complemented with nine semi-structured interviews with former UNESCO staff and experts, including four Assistant Director-Generals, three of the staff members in charge of the recommendation's drafting process and two observers. 2 More than 140 documents were screened for valuable arguments, of which 59 were eventually coded within Dedoose, an application for qualitative data analysis. Lastly, to detect how the recommendation's text changed over time, the different draft versions were analyzed and compared manually.
Seeking a global consensus on Internet-related matters
The idea of preparing a legal instrument on Internet-related matters was first discussed by UNESCO's governing bodies in 1996. It was supposed to serve as UNESCO's first official position statement regarding the changes triggered by the spread of the Internet as a communication and information technology and its progressive commercialization during the 1990s. Yet it took UNESCO more than six years until the ‘Recommendation concerning the promotion and use of multilingualism and universal access to cyberspace’ was eventually adopted in 2003. Due to many seemingly insurmountable differences in interests and ideas, the organization could only reach a consensus after refining of the most contested paragraphs.
Performative dimension: Negotiating a UNESCO-sponsored framework for the Internet
The formulation of a normative framework applicable to cyberspace was one of the most contested actions that UNESCO's member states community considered to ‘place UNESCO at the forefront of the international debate on this important subject’ (UNESCO, 1996: 1). For some countries, such an initiative evoked memories of the organization's last involvement in international information issues, the NWICO debate, in which developing countries’ demands for media independence and a more balanced information flow had been fiercely opposed by Western countries (UNESCO, 1997b: 225). UNESCO's involvement in the NWICO movement had culminated in the withdrawal of the United States, the United Kingdom, and Singapore, causing an institutional crisis from which the organization recovered only very slowly (Carlsson, 2003: 52ff). Yet, merely 15 years later, in 1997, UNESCO's general conference decided to take another chance and prepare an international instrument concerning the new medium of the Internet and access to online information (UNESCO, 1997c).
Initially, many tensions regarding this decision evolved around the question of whether the new instrument should represent a legal framework regulating issues concerning the new field of ‘cyberspace law’, a binding declaration or rather a nonbinding recommendation addressing ethical and social principles that would merely impose a moral obligation on UNESCO's member states to respect them (UNESCO, 1998, 1999). It was soon recognized by UNESCO's Sector for Communication and Information (CI Sector)—the sector in charge of preparing the instrument—that only the latter would be acceptable to the majority of member states and compatible with the organization's intellectual, cultural, and scientific mandate (Montviloff, 2013; Yushkiavitshus, 2015). Thus, over the following years, the text of the planned recommendation was slowly drafted and refined during several rounds of consultation with member states, the international council of UNESCO's Information For All Programme, the national UNESCO Commissions, and a few selected experts. During the written consultation processes, involved actors could submit comments to the most recent drafts to support or oppose parts of the recommendation, suggest new elements, or propose a different wording. In addition, during each meeting of UNESCO's governing bodies, member states’ representatives discussed the progress of the recommendation and their concerns with the text. While the recommendation provoked fierce controversies in almost all of those meetings, the conflicts between member states about key elements of the recommendation and its potential adoption escalated during the General Conference in 2001 (Koven, 2015; Montviloff, 2013). To calm the situation and avoid any further political tensions that could harm the organization as a whole, UNESCO's Director-General personally intervened to withdraw the topic from the agenda (Montviloff, 2013; Rivière, 2015). Through this highly unusual interference, the decision regarding the recommendation was postponed for another 2 years and its content put up for discussion once more (UNESCO, 2002: 1). 3
It would be too simplistic to reduce the disputes encountered during these first years of the preparation process to content-related differences alone. Rather, they are also an expression of the power relations between the different groups of actors involved in the process and their slow shift over time. In the beginning, it was the UNESCO Secretariat, more precisely its CI Sector, that held a particularly influential position, allowing it to take programmatic decisions about the draft recommendation's content and general tone. First of all, it proposed the four topics the recommendation should cover, namely (1) universal access to Internet networks and services, (2) the promotion of multilingualism online, (3) the development of public domain content, and (4) the application of copyright exemptions (UNESCO, 2003). By doing so, the responsible staff members within the CI Sector were able to introduce ideas into the draft that they had been trying to promote for a number of years, such as extending the fair use principle to online content and defining information as a common good. Although some of these ideas were simultaneously discussed in other forums, they were new to UNESCO, where they were immediately highly contested and perceived by some as overly critical of economic players and their interests. To support these controversial ideas through scientific research, the CI Sector also selected the experts who provided the input reports informing the recommendation's first drafts and who shared their very critical view of the increasing commercial usage and provision of online information (Quéau, 2000, 2013). Although the Sector's dominant position was challenged later, throughout the whole process the secretariat maintained what in ANT tradition is called the role of an Obligatory Passage Point (Callon, 1986): Since it was the secretariat who collected and reviewed all comments received during the consultations and integrated them into the draft text, all other actors necessarily had to interact with it and respond to its proposals and ideas to achieve their own goals.
The biggest challengers of the UNESCO Secretariat and, more precisely, the CI Sector's prominent position, were UNESCO's member states. This antagonistic constellation is surprising, because, according to UNESCO's constitution, the secretariat is supposed to simply act as a facilitator for the member state community and not as an independent actor with its own decision-making capacities. But whereas the draft recommendation was generally supported by the Group 77 and other countries closely involved in the drafting process (e.g., France, Germany, Russia, Korea, and Iceland), it encountered strong objections from some Western governments, such as the Dutch, Japanese, and Finish delegations. 4 Over the entire process, their criticism led to the repeated reformulation of key elements and a general attenuation of the initial propositions.
Interestingly, the fiercest opposition, however, was not voiced by a UNESCO member state but by two different actors. The first one was a group of nongovernmental organizations (NGOs), including the US-based Motion Pictures Association of America (MPAA), the International Association of Scientific, Technical, and Medical Publishers (STM) and the World Press Freedom Committee, as well as the International Publishers Association (IPA). These NGOs, some of them among the drivers behind the coalition against the NWICO movement in the 1970s, initiated a vigorous attack against the recommendation that in many regards brought back memories of this former politicized controversy. Claiming that the positions in the proposed draft recommendation would once more represent a threat to the free flow of information, they warned that the planned instrument could be misinterpreted as an invitation for censorship and closed markets for information and communication (MPAA, 2001). Through their actions, they successfully mobilized the opposition of the United States, which was the only non-member state involved in the deliberation process as it had not yet re-joined UNESCO since its withdrawal in 1984. As a result, the US government warned UNESCO of ‘getting embroiled in another international fracas which might further delay the re-joining of the USA’ (UNESCO CI/INF, 2001a).
This threat provoked immediate reactions by UNESCO, thereby leading to a power shift in favor of the United States, which, as a non-member state, officially had no voice in the organization's internal policy debates. The motive was quite simple: Since the appointment of Koïchiro Matsuura as Director-General in 1999, the strategic goal of convincing the United States to rejoin UNESCO had become the organization's main priority (Rivière, 2015; Yushkiavitshus, 2015). Thus, the US government was given access to all draft versions of the recommendation and was included as an observer in all member states' consultations and official meetings (UNESCO ADG/CI, 2002). Moreover, the UNESCO Secretariat arranged for meetings between the US observer, UNESCO's Director-General and the director of its cabinet, which eventually led to the reformulation of the draft recommendation's most contested paragraphs and helped to partially smooth the objections (Rivière, 2015). But the criticism of the USA and its allied NGOs was not only targeted at the recommendation's content. It also concerned a particular UNESCO staff member whom they viewed as the source of the controversial discourse related to free access, copyright exemptions, and information commons (Quéau, 2013). In 2003, after the Director-General had released the contested person from his director position with the CI Sector and appointed him head of UNESCO's field office in Moscow instead, the USA rejoined the organization, and the contentious Internet recommendation was eventually adopted by the General Conference (UNESCO, 2003). Despite remaining reservations, the member states' community agreed that any further deferment of the adoption would not only make any consensus within UNESCO on the topic unlikely; it would also mean that UNESCO would not be able to present an official contribution on Internet matters to the first phase of the WSIS, which took place later the same year. Yet, due to the many hurdles UNESCO had to overcome to reach consensus, the legal instrument remained at the stage of a nonbinding recommendation; it was not, in a second step, followed up by a binding convention, ratified by governments, as it had been initially envisaged by the UNESCO Secretariat.
In sum, on the performative level, the 6-year preparation process of UNESCO's Internet recommendation was marked by different actors, who jointly produced the final text while constantly competing for influence on the processes of social production. Although the secretariat initially held a powerful position that allowed it to promote certain ideas and filter those proposed by others, it was soon outpaced by the only non-member state involved in the process, namely the USA. In contrast, most member states and the expert community, which, according to UNESCO's organizational settings, should both have the strongest voices in policy debates, primarily played a supporting role. But through their support, they equally contributed to shifts in power relations and, thus, ultimately to the processes of social ordering within UNESCO.
Discursive dimension: Universal access to the Internet as a global public good
The shifting power relations among the actors involved in the preparation process were also reflected on the discursive level. Although a nonbinding UNESCO instrument does not have great impact on national or international policy agendas, all involved actors were fully aware that every postulation in the text would be part of the organization's official policy discourse; consequently, every subsequent text could draw on the recommendation's wording and policy makers could potentially justify their actions by referring to its consensus-based adoption. For exactly these reasons, actors vigorously competed over ‘inscriptions’, which in ANT-tradition represent the process through which protagonists of production processes embed their ideas and visions in the material structures of objects and artefacts (Akrich, 1994). Hence, by trying to inscribe their preferred discourses and arguments in the recommendation's text, actors sought to institutionalize them within UNESCO's official policy response to the Internet.
As a result, the various drafts versions underwent an important development, with their general tone and the scope of the proposed measures changing quite fundamentally. But not all parts of the draft recommendation caused the same kind of contentions. The four sections of the recommendation related to different dimensions of universal access to the Internet and proposed different ways of reaching this overarching goal. By including four aspects of universal access, the drafters combined two different, and in many ways contradictory, objectives in a single instrument: On the one hand, the recommendation's first two sections concerned physical access to networks and services and the issue of multilingual information. Thus, they related to the problem of universal access from a development perspective: By framing the gap between information-rich and information-poor countries as the main obstacle to universal access, the respective measures were of particular relevance to developing countries that lagged behind regarding their inclusion in global information networks. On the other hand, the last two sections proposed to enhance universal access through public domain content and copyright exemptions. Hence, they regarded the problem from a more socioeconomic perspective: Viewing universal access as a question of balance between commercial and public interests on the Internet, the proposed measures mainly concerned highly developed countries in a more advanced state of digitalization. As a consequence, the recommendation was caught in a discrepancy between an initiative aiming to bridge the global digital divide and an initiative tackling broader societal problems arising from the increasing commercialization of the Internet and online content. Then again, the question of physical access was in line with UNESCO's long-standing tradition of technical assistance in the field of communication; the issue of linguistic diversity of online information reflected UNESCO's ongoing and very prominent debate on cultural diversity. By contrast, both the issue of the public domain and the question of IPR exemptions were introduced to UNESCO's policy debates only some years earlier and had immediately encountered criticism as they were seen to interfere with member states' economic interests and political priorities. Consequently, the recommendation's first two sections were significantly less controversial than its last two, also because member states could agree much more easily on concrete measures supporting developing countries than on complex socioeconomic problems (Montviloff, 2013; Yushkiavitshus, 2015).
It would go beyond the scope of this article to assess in detail how all four sections of the recommendation were drafted and shaped through the exchange of competing arguments. Instead, the analysis focuses on identifying the major discourse coalitions that vied with each other over the contested paragraphs on public domain information and copyright, representing the socioeconomic and thus the more controversial of the two objectives. According to Hajer, actors build discourse coalitions when they share the usage of a particular discourse over a longer period of time, without necessarily agreeing on every detail or having the same fundamental values (2006: 70). In the context of UNESCO's 2003 recommendation, we can identify at least three of these coalitions, although there are certainly many more nuances that would need to be accounted for (see also Pohle, 2016a: 402ff).
The first discourse coalition was marked by a very critical, almost alarmist perspective on the changes triggered by the Internet. This coalition was represented by influential members of the UNESCO Secretariat and the experts chosen by the CI Sector to be involved in the drafting process. They were united by the belief that access to information, particularly in science and research, was increasingly threatened by the economic structures defining its availability in the digital sphere. The coalition emphasized the importance of online content belonging to the public domain, insisting that it would be ‘fundamental to look for strategies and actions to broaden the proportion of global knowledge that can be categorized as public domain information’ (UNESCO CI, 2000: 19). Access to free information was considered not only a counterweight to commercially owned and distributed information but also as an important instrument for fighting social and global inequality. Accordingly, the coalition viewed any kind of access restriction as a threat to this goal. It challenged particularly existing IPR legislation and claimed that fair use exceptions for scientific and education use of protected information needed to be preserved and expanded on the Internet—a claim that was aptly summed up by one of the experts supporting the position of the CI Sector: The relation between the principle of intellectual property and the principle of free access to information are [sic] not defined by natural law and is not immutable; […] It is counterproductive to insist on existing rights, laws, rules and regulations when there is agreement on the need to overcome the global digital divide and this has not been met in the past despite the promises of the information age. (Kuhlen, 2001)
Unsurprisingly, this critical perspective was not shared by the majority of UNESCO's highly industrialized member states, most prominently Denmark, Sweden, Finland, the Netherlands, and Japan. Together with the United States and the publisher associations among the NGOs involved in the preparation process, they built the second discourse coalition. In direct opposition to the first coalition, they advocated for the preservation and reinforcement of existing IPR legislation, warning about the negative impact the recommendation could have on the differentiation between public domain content and proprietary information: The Draft Recommendation gives the misleading impression that exemption from copyright infringement and limitations on the level of copyright protection lies in Member States' best interest, when in reality it is the publishers' investment in creativity—which copyright protects—which truly supports the information society. (IPA and STM, 2001) No new balance needs to be found. […] In fact, the balance is contained in existing conventions and treaties on the law of copyright, such as the Berne/TRIPS and WCT three-step test. It is this existing balance which was developed in the analogue world, that needs to be applied or transposed to the electronic environment. (IPA, 2001) An appropriate balance should be sought between the legitimate rights of the creators/producers of information and of those who use it. Balances must also be found between the interests of the public and private sectors of society, especially in connection with public access to information. (Canada, 2001)
Conclusion
This article aimed to understand the conflicts around UNESCO's first instrument concerning Internet-related matters by analyzing the performative and discursive dimension of its preparation process. It thereby adds to the general understanding of the problems international institutions need to face when preparing legal instruments that are equally valid for all parts of an increasing globalized world. In the case of UNESCO's cyberspace recommendation, the confrontations between competing discourse coalitions resulted in a final text that meant a step backwards compared to the very critical viewpoints dominating the early drafts. Although the difficult questions of access to public domain content and copyright exemptions were still mentioned, the recommendation did not propose any innovative solution that could allow for counterbalancing the economic forces at play. At the same time, the unfruitful struggles about the contested paragraphs steered the attention away from the urgent issues addressed by the other two sections, which focussed more closely on the situation of the developing world. As during the NWICO episode, the discursive confrontations and power struggles amongst its member states once more in UNESCO's history hampered joint reflection about the adequate balancing of interests and about the diverging needs and necessities of different countries with regard to information and communication.
Then again, the different positions and arguments of the actors involved in preparing the recommendation show that, even in the early times of digitalization, it appeared almost impossible to agree internationally on a policy discourse that would not be contrary to some perspectives on digital information and technology. Back then, just as today, the Internet resulted for some parts of the world in an increasingly hyperconnected environment, which becomes more and more synchronized and delocalized. Yet some parts of the world's population have remained largely excluded from this development, despite being irrevocably affected by the changing nature of information and communication. Consequently, a comprehensive policy approach that seeks to address the increasing pervasiveness of digital information infrastructures and their impact on society necessarily has to account for the inherently global nature of digital networks and has to address problems concerning all countries. At the same time, such a policy approach also needs to acknowledge inequalities in a globally networked society, with the result that not all countries face the same challenges and societal consequences; thus, it has to account for the different needs and interests of countries affected by the globalized networks.
The difficulty to find a balance between the different needs and interests of all parties involved is also reflected in many attempts at digital constitutionalism that could be witnessed on the international level in recent decades (Gill et al., 2015; Santaniello et al., 2016: 328). While, the increasing number of national parliamentary initiatives toward Internet constitutionalism, such as recent initiatives in Brazil or Italy, struggle to find compromises on regulatory norms and the limitations of power on the Internet, international initiatives are under additional pressure to accommodate different national priorities. This is even more the case for an intergovernmental organization such as UNESCO, which regroups a large number of member states with different cultural, social and political backgrounds and diverging normative traditions.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research was supported by a PhD fellowship of the Research Foundation Flanders (2010–2015).
