This paper examines the emergence of diversity as a guiding principle of Internet governance. This paper compares how diversity is being interpreted and applied in Internet governance discourse and related research with its interpretation and application in traditional communications policy–making and policy research. As this paper illustrates, a somewhat more narrow conceptualization of diversity has developed within the Internet governance context than has been the case in traditional communications policy contexts. This pattern is also reflected in Internet governance–related diversity research to date. To some extent, the Internet governance conceptualization of the diversity principle is becoming diffused into other related policy principles (e.g., access, localism), which may undermine its significance as a guiding principle of Internet governance.
The diversity principle in communications policy
The diversity principle in Internet governance
Principles are central to the policy–making process. As Anderson (1992) states, “In order to make a policy decision, one must invoke some criteria of evaluation. We cannot decide whether a proposal for public action is desirable or undesirable, whether the results of a public program are to be adjudged a success or a failure, except in light of a standard” . This standard is usually defined in terms of “a finite and bounded set of classic principles” . For these principles to be useful to policy–makers and policy analysts, they must have clear, agreed–upon interpretations, so that they contain within them substantive and reasonably stable evaluative standards. Otherwise, “political argument can fasten arbitrarily on one or a few of these concepts and ... they can be arranged in different patterns in ideological thinking, invested with a variety of meanings and given different degrees of emphasis” .
This pitfall has been particularly acute in communications policy–making, where central guiding principles have suffered from ambiguity, inconsistency, and manipulation (Karppinen, 2013; Napoli, 2001). Concepts such as diversity, pluralism, the public interest, and universal service long have been prominent buzzwords in communications policy–making, but often these concepts have not been infused with the specific, concrete meaning necessary for them to serve as meaningful and effective tools for designing, implementing, and analyzing policies (Hitchens, 2006; Karppinen, 2013; Napoli, 2001).
Efforts to develop such guiding principles for the Internet have proliferated, as evidenced by the congestion of declarations of Internet principles by governments and international organizations, such as G8, the Council of Europe and OECD (see Kleinwächter, 2011). In addition, more sustained debates around the principles of global Internet governance continue to take place within the U.N.–sponsored Internet Governance Forum (IGF), a global multi–stakeholder forum that has convened annually since 2006.
To some extent it is inevitable that the broad articulations of established communications policy priorities are going to be applied to new communications platforms and find their way into these articulations of the guiding principles for new media policymaking. The well–documented interpretive flexibility surrounding communications policy principles (see, e.g., Napoli, 2001) suggests a degree of malleability that would facilitate established communications policy principles gaining a foothold in new policy contexts. At the same time, the structural, technological, and institutional dynamics may be significantly different within the context of the new platform, raising questions of “fit” between the established policy principle and the new policy context.
And so, perhaps not surprisingly, some of the emergent guiding principles for Internet governance have migrated from the realm of traditional communications policy–making. Such a process raises a variety of potential challenges. Consider, for instance, the somewhat disconnected relationship that has emerged thus far between policy–making for traditional media and new media contexts such as Internet governance. Because of media convergence, distinct normative traditions associated with print, broadcasting and telecommunications have clashed, and the Internet, as a combination of many media platforms, has inherited a “baggage of discordant normative traditions” . For Duff, these developments have yielded a “normative crisis of the information society” — and a need for new and rigorous normative principles on which public policies should be based.
Little if any research has, at this point, traced the migration of communications policy principles from traditional media to new media contexts. This paper begins to address this gap by mapping various ways in which the diversity principle has received attention in the context of Internet governance and compares these to the uses of the diversity principle in traditional communications policy. The principle of diversity, which has a long and well–researched history as a guiding policy principle in the realm of traditional media, has emerged as a central guiding principle of Internet governance.
This paper identifies the points of intersection and disconnect that have emerged thus far in terms of how the diversity principle has been conceptualized and applied as an analytical tool in traditional media and Internet governance contexts. The goal of this analysis is to provide at least an initial step towards a much–needed bridge between established articulations and applications of the diversity principle in communications policy–making and policy research and its emerging articulations and research applications in the realm of Internet governance.
The focal point for this analysis is the annual Internet Governance Forum (IGF). The IGF has essentially served as the laboratory in which guiding principles of Internet governance are developed and debated. The IGF itself has no formal decision–making capacity, as it is mainly seen as “a discursive exercise aimed at creating common ground for discussing Internet–related policy issues” (Epstein, 2011). This has also been raised as a criticism of studies that focus on the few formal global institutions of internet governance. Van Eeten and Mueller , for instance, call for researchers to focus more on the “real–world activities that actually shape and regulate the way Internet works” instead of discursive arenas that have little impact on actual decision–making. It is important to note, however, that the aim of this paper is not to explain policy formation or to evaluate the effectiveness of different measures that aim to promote diversity. Instead, the paper looks at the IGF simply to map some of the different conceptualizations and uses of the diversity principle that are beginning to emerge in the context of Internet governance.
As interpretative or discursive approaches to policy analysis emphasize (e.g., Fischer, 2003), the shaping of policy discourse and the definition and framing of issues and principles can often have far–reaching consequences in their own right. This is especially true in contexts of policy change and the emergence of new policy problems, when existing paradigms are challenged and it is necessary to conceptualize new goals and find criteria for assessing them. This certainly applies to the IGF, where central struggles have focused on the politics of agenda setting and the shaping of the language used in framing discussions (Dutton, et al., 2007; Epstein, 2011; Mueller, 2010).
It is for these reasons that this paper uses the IGF as the point of entry for analyzing how the diversity principle is taking form as a guiding principle of Internet governance, and how its development and application map against the history of the diversity principle in traditional communications policy. This analysis is derived from transcripts of the 2006 through 2012 IGF proceedings and preparatory meetings, participant observation by the lead author at the 2007 IGF in Rio de Janeiro, official documents prepared by the U.N. and the U.N.–created Internet Governance Working Group, as well as position papers and scholarly papers prepared by various stakeholder groups, including civil society organizations, government agencies, industry associations, and academics in connection with the IGF proceedings.
The first section of this paper provides an overview of how the diversity principle has been conceptualized and applied in traditional communications policy contexts. The second section focuses on a comparative analysis of how the diversity principle has thus far been conceptualized and applied in the realm of Internet governance. The final section offers some concluding observations about the process of translating an established policy principle such as diversity into new policy contexts, as well as suggestions for future research.
The diversity principle in communications policy
The place of the diversity principle in communications policy–making is an expansive topic (see, e.g., Hitchens, 2006; Karppinen, 2013; Napoli, 2001). The purpose here is not to provide a detailed account of the topic, or of the substantial literature that has developed around it (see Roessler, 2007). The goal here, rather, is to establish the basic contours of the principle and how it has been used in communications policymaking and policy research, in an effort to establish a vantage point from which to assess the principle’s transition to the realm of Internet governance.
Amongst all of the overarching themes reflected in the IGF’s agenda, the theme of diversity perhaps has the deepest roots in other areas of communications policy–making. Within the context of traditional communications policy–making, the diversity principle has been conceptualized primarily in terms of the promotion and preservation of a diverse array of sources of information, as well as a diverse array of ideas, viewpoints, and content options (Hitchens, 2006; Horwitz, 2005).
Figure 1: The diversity principle in communications policy: Key components and presumed relationships.
Note: Adapted from Napoli (2001).
Figure 1 presents Napoli’s (2001) model of the diversity principle that will serve as the primary point of reference for this analysis. The goal of this model was to identify the key conceptual elements that have characterized the application of the diversity principle in communications policy–making, policy analysis, and policy advocacy, and to identify the relationships that stakeholders frequently have presumed to exist between these various conceptual elements. It is important to note that, while this model was constructed via an analysis of communications policy in the United States, the model has achieved fairly extensive international applications , making it a potentially useful baseline from which to assess the transition of the diversity principle to the international context of Internet governance.
At the most basic level, Figure 1 illustrates that the principle of diversity can be broken down into three inter–related components: source, content, and exposure diversity. Source diversity refers to the extent to which the media system is populated by a diverse array of content providers. This focus on content providers can emphasize the ownership of either the media outlets or the underlying content, with the specific diversity criteria taking a variety of forms, ranging from ownership race/ethnicity or gender, to various dimensions of organizational or economic structure (e.g., public, private, for–profit, non–profit, independent, group–owned). Source diversity also has, at times, been conceptualized in terms of the diversity (in terms of gender, ethnicity, etc.) of the individuals working within the media outlets.
A key point reflected in this model is that such source diversity often has been presumed to lead to diversity of content. Content diversity has been operationalized in a variety of ways, including in terms of the diversity of program types or genres available, the diversity of ideas or viewpoints expressed, or in terms of the demographic diversity of those depicted in the content (see Roessler, 2007). This causal relationship between source and content diversity has, at various times, been questioned, and research addressing this relationship has not provided definitive evidence of a systematic relationship (Horwitz, 2005). This issue becomes particularly important within the context of policy debates about whether the promotion of a diversity of sources is an important policy objective in its own right, absent any clear indication that such source diversity enhances diversity of content (see, e.g., Baker, 2007).
The final element of Figure 1 is exposure diversity. This term refers to the extent to which audiences consume a diverse array of content. There traditionally has been a presumption that increasing diversity of content promotes diversity of exposure, as audiences have a greater array of sources and content options to choose from (Napoli, 2001). Here again, however, legitimate questions arise as to whether this causal relationship holds, as some studies suggest that many media consumers utilize greater diversity of available content in ways that narrow the range of content they consume (Hindman, 2009; Webster, 2007; Webster and Ksiazek, 2012).
From a policy standpoint, this issue becomes particularly important if increasing audiences’ exposure to diverse sources and content is the ultimate goal of any diversity–enhancing policies. At the very least, however, understanding how changes in source and content diversity impact exposure diversity is fundamental to policy–makers’ understanding of the production and consumption dynamics of any communications system. Perhaps not surprisingly, exposure diversity has resided at the fringes of contemporary communications policy discourse (Helberger, 2011; Napoli, 2011a; Webster, 2007). If and how policy–makers should concern themselves with increasing the extent to which audiences consume a diverse array of sources, in keeping with the underlying logic of the effective functioning of a robust marketplace of ideas, remains a difficult question.
Diversity concerns have been central to a wide range of contemporary communications policy issues, ranging from media consolidation, to the privatization and commercialization of media ownership, to minority ownership of media outlets, to content regulations and programming carriage requirements, to, most recently, Internet governance. Consequently, recent years have seen substantial growth in various forms of diversity research, not only within academia, but also within the policy–making sector. Efforts ranging from the U.S. Federal Communications Commission’s controversial Diversity Index to the German Commission on Concentration in Media’s (KEK) weighting approach that accounts for possible influences of various media on the diversity of opinion, to the Ofcom (U.K.) public interest or plurality test, all have sought to empirically assess one or more of the components of the diversity principle discussed above (see Just, 2009; Karppinen, 2013). The most comprehensive effort so far, the Media Pluralism Monitor (MPM) introduced by the European Commission in 2009, was able to develop no less than 166 different indicators for measuring media pluralism and diversity (Interdisciplinary Centre for Law and ICT (ICRI) — Katholieke Universiteit Leuven, et al., 2009). Generally, though, these efforts have focused most heavily on the process of measuring the diversity of available sources.
The diversity principle in Internet governance
Obviously, the Internet represents a very different communications environment in which the diversity principle is being conceptualized and applied. The issues of spectrum scarcity, license allocations, and high barriers to entry that historically have characterized the traditional media are more or less non–issues within the context of the Internet. At least superficially, this would seem to allow the Internet to provide the kind of choice and multiplicity of sources that extend far beyond what could be achieved via traditional media, particularly given the Internet’s more global orientation. From this standpoint, one could assume (as some policy–makers have) that the Internet essentially “solves” all diversity policy concerns (see Baker, 2007).
Such an assumption would, however, be misguided (Baker, 2007). Although the Internet may alleviate many diversity policy concerns, it also raises new ones (see Napoli, 2011b). The visions of unlimited diversity have recently been questioned by many scholars, who have pointed out new types of control and warned of threats to the open and distributed nature of the internet. Enduring asymmetries of power together with the emergence of new dominant players, such as Google, thus continue to guide people’s attention to a limited number of sources (e.g., Curran, et al., 2012; Hindman, 2009). 2012 IGF participant Eli Noam went so far as to proclaim that, “when it comes to media pluralism, the Internet is not the solution, but it is actually becoming the problem,” due to the “fundamental economic characteristics of the Internet” (such as scale economies, capital intensivity, etc.) .
These worries are discernible in a variety of ways in the debates on Internet governance. Understood broadly as the widest possible distribution of voices, the principle of diversity seems to underlie some of the basic architectural principles of the Internet. The decentralized structure and the so–called “end–to–end design” of the Internet, for example, mean that the network serves as a relatively neutral and transparent platform, which in turn has been seen to promote the widest possible variety of content .
The open and decentralized nature of the Internet is one of the key values supported by a broad consensus of Internet governance stakeholders. Similar concerns also surface in debates on a wide variety of issues, from net neutrality to intermediary liability, filtering and blocking of Web sites, not to mention the myriad of questions associated with copyright. Understood broadly enough, concerns associated with diversity are thus at least implicit across a wide range of issues on the Internet Governance Forum agenda.
Most of the time, however, these issues are not discussed explicitly under the theme of “diversity” but under other Internet governance terms such as “free flow of information” or “openness.” This has also meant that there are very few explicit connections between these debates and the ways in which diversity has been used in other areas of media and communications policy.
One reason for the failure to conceptualize these issues in terms of diversity may also be that instead of the established goals of source and content diversity, concerns around new intermediaries, such as search engines, tend to pertain more to the aspect of exposure diversity, which still remains less established in policy–making (see below). Yet, many of these issues that have so far not been articulated in terms of diversity would seem to be directly relevant to what Helberger (2011) has recently called “diversity by design,” or “creating an architecture or service that helps people to make diverse choices” .
One can ask, of course, if it really matters whether these debates take place under the label of diversity or not. The debates on digital divide, network neutrality and universal access, for instance, may get to the heart of the same type of questions and policy concerns without the need to invoke “the diversity principle” as an overarching label.
The aim of this paper is not to argue for diversity as a better term for these concerns. However, it remains important to identify, for the sake of conceptual clarity alone, how one concept can have different meaning in different policy contexts and how the same type of issues may be discussed under various labels, particularly given that these different labels can have different types of political resonance. The following section focuses on explicit references to diversity in the context of the IGF, and thus provides a basis for comparing them with traditional communications policy.
Cultural and linguistic diversity as key concerns
The diversity concerns that have arisen explicitly within the context of Internet governance have so far revolved around the issue of language. Today, questions of cultural and linguistic diversity of the Internet are routinely mentioned as key principles in various declarations on Internet principles and they have also become a particular area of focus for UNESCO (see, e.g., Council of Europe, 2011; UNESCO, 2003; World Summit on the Information Society (WSIS), 2003, 2005). The specific focus has been on the linguistic diversity of the content available online. For many Internet users, the potential benefits of the tremendous variety of content options available online from a vast array of sources essentially run aground against the fact that much of this information may not be available in their native language. As was noted in the IGF 2010 panel on linguistic diversity, there are more than 6,000 languages in the world, though only about 350 of them are represented online (see also UNESCO, 2009) . And, not surprisingly, there has been an overwhelming proportion of English–language content online, relative to English speakers’ representation in the global population and the online population (UNESCO, 2009).
It is important to emphasize that the issue of linguistic diversity online reflects traditional concerns about preserving and promoting cultural diversity (UNESCO, 2002). A common focal point of contemporary Internet governance discourse involves the importance of the world’s cultural diversity being accurately reflected in the online realm . One participant in the 2007 IGF suggested that “Linguistic diversity is for human society what biodiversity is for nature.”  On the other hand, as Tim Berners–Lee, the inventor of the World Wide Web, emphasized in his keynote speech to the 2009 IGF, the universal reach of the Internet across people from different cultures, with different languages, and with different abilities is also central to the unifying ideal of “the humanity connected by technology.”  In these regards, linguistic diversity can be seen as both facilitating the maintenance of cultural differences as well as forging larger global communities.
One IGF participant went so far as to declare that a “multilingual Internet should be viewed as a citizen right and a government obligation.”  As one UNESCO representative speaking at the 2007 IGF noted, “The ability to use one’s language on the Internet will determine one’s ability to participate in the Information Society.”  This same representative suggested that the unavailability of native–language content online may even represent a more significant component of the digital divide than infrastructure imbalances.
The definition of diversity in this context, however, has also created some confusion. Uncertainty about the meaning and scope of diversity as one of the main themes of IGF has thus been a recurring subject at many IGF sessions. One participant at the 2009 IGF, for example, objected to the notion of diversity as condescending: “We do not want to be seen in the Arab world as a diversity. We are not. We are just as equal as anybody on the Internet world.”  The discourses about diversity in Internet governance thus continue to be surrounded by ambiguity, as different actors associate diversity with different issues, each with their own political undercurrents. As noted above, such problems are not foreign in the use of guiding principles in traditional communications policy either. However, efforts to articulate and implement appropriate guiding principles are particularly challenging when the context is global in scope (Dutton, et al., 2007; Mueller, 2010; Mueller, et al., 2007). Such is the case in the realm of Internet governance, where the definition of guiding principles confronts not only diverse political priorities but also the added challenge of cultural differences and different national needs.
The intersection of diversity and other communications policy principles
This linking of linguistic diversity with the notion of the digital divide discussed above represents the primary way by which the principle of diversity has, within the context of Internet governance, become tightly intertwined with the principle of access (see, e.g., UNESCO, 2003). This intersection of policy principles became especially evident in the context of the IGF when previously separate sessions on diversity and access were merged starting in 2010 into a common theme of “access and diversity.”  Within this theme, discussions of linguistic diversity have typically been discussed in parallel with other accessibility issues faced, for example, by disabled people.
The key point in this linkage between diversity and the digital divide is that only by ensuring that a greater diversity of language populations has access to the Internet can greater linguistic diversity be achieved online. Reflecting this perspective, many participants in ongoing Internet governance discussions have emphasized the need for the local production of online content . For instance, Jerry Yang, co–founder of Yahoo!, noted at the 2009 IGF that: “The Internet isn’t just about getting as many people online as possible, but making sure that once they get online, they have something productive to do, something to gain, and something meaningful to experience. We must provide relevant, local content. ... I encourage Internet users to continue to create local language content. I encourage governments to create rules that allow users to create that content to flourish.” 
It has been argued that the significance of locally produced content even increases as the Internet diffuses into less economically developed regions of the world. As Mike Silber, a panelist in the 2010 IGF session on access and diversity emphasized, when it comes to going online, “for many people, the ability to engage with a global community is a driving factor, but in a rural village, as a subsistence farmer, the ability to connect with people in another continent, speaking a different language, may not actually be that appealing. ... it’s the ability to access relevant, local content, key information that affects the lives of those people that is the major driver.”  The intersection between linguistic diversity and access becomes even tighter in light of research indicating that there is a significant inverse relationship between the amount of local language content available online and the cost of Internet access (Bruegge, et al., 2011).
The centrality of local content to the notion of linguistic diversity also creates a significant point of overlap between the principle of diversity as it relates to Internet governance and the principle of localism — which also has been a prominent and long–standing principle in communications policy. In the communications realm, localism has been reflected in efforts to structure media markets in ways that promote local ownership of media outlets and that foster (and in some cases, mandate) the production of locally oriented content (see Napoli, 2001). In the case of the Internet, in which the scope is expanded to the global level, thus making the issue of language differences of paramount importance, the achievement of linguistic diversity is most likely only achievable via mechanisms that promote the local production of content. Of course, the term “local” is being adjusted to accommodate the scope of the policy space at issue. Within the context of national–level communications policy, localism typically is operationalized in terms of individual cities or communities. Within the context of global Internet policy, localism is adjusted to units of analysis related to nations, or language communities within or across nations.
In terms of the practicalities associated with the linguistic diversity issue, challenges arise not only in terms of the production of native–language content, but also in terms of the underlying system of domain name and number registration. The Internet Domain Name System (DNS) has been based on the American Standard Code for Information Exchange (ASCII), which is limited to Latin letters, digits, and the hyphen. Therefore, it has not traditionally been able to deal with languages consisting of non–Latin characters or even European languages (such as French and German) containing letters with diacritics.
Continuing a series of efforts to address this issue, ICANN started in 2009 a new fast–track process to create internationalized country code top–level domains (IDN ccTLDs), making entirely multilingual domains available. The first of these was launched in 2010, and by 2012 over 30 languages have been approved by ICANN.
While being widely recognized as an important technical step, the internationalized domain names alone do not solve all technical problems. A recent report on the deployment of IDNs highlights several remaining problems, including lack of e–mail functionality and limited support for IDNs by Internet service providers, browsers, and many other popular applications and Web sites (EURid–UNESCO, 2012). These and other technical problems in building a genuinely multilingual Internet ecosystem continue to be recurring themes in the IGF debates on linguistic diversity.
Clearly then, within the context of Internet governance, the principle of diversity contains significant points of interaction with other established communications policy principles such as access and localism. This point is reflected in Figure 2, in which the principles of diversity, access, and localism all intersect around the concept of linguistic diversity, given the necessity of locally produced content to the achievement of linguistic diversity and the importance of access to facilitating the production and distribution of the locally produced content that contributes to linguistic diversity.
Figure 2: The intersections of diversity, localism, and access principles in Internet governance.
Comparing the uses of diversity in traditional and new media
Clearly, diversity’s incarnation as a principle of Internet governance possesses some fundamental differences from its incarnation as a guiding principle of traditional communications policy. There is, for instance, a greater degree of intersection between the predominant conceptualizations of the diversity principle and other communications policy principles in Internet governance than has historically been the case in terms of diversity’s conceptualization in traditional media contexts. This has occurred despite the fact that the predominant conceptualization of diversity in Internet governance contexts (linguistic diversity) is far more limited and uni–dimensional than has been the case in traditional communications policy contexts. This extent to which the predominant interpretation of the diversity principle is simultaneously relatively narrow and overlapping with other guiding principles of Internet governance suggests a significantly diminished role in the Internet governance context than has been the case in traditional communications policy contexts.
This latter point reflects a certain degree of disconnect at the conceptual level between the articulation of the diversity principle in traditional communications policy and Internet governance contexts. This disconnect is reflected in the conceptualization of diversity in the IGF agenda. Some individual workshops, such as a workshop on media pluralism and freedom of expression at the 2012 IGF, have explicitly discussed the impact of the Internet on the established notions of diversity and pluralism in communications policy, but so far these attempts to join together the agendas of Internet governance and traditional communications policy have remained on the fringes of the debates; while issues, such as multilingualism, that have a more entrenched presence on the IGF agenda have received the majority of attention.
In so far as diversity online has been conceptualized primarily in terms of language, the debates at the IGF have remained particularly isolated from broader academic or policy debates on diversity in communications policy. For instance, the persistent discussion and debate about the relationship between source and content diversity that has characterized the diversity principle’s place in the regulation of traditional media has been largely absent, to this point, in the discussion of the diversity principle in the realm of Internet governance. That is, while various stakeholder groups, ranging from scholars to the courts, to regulators, have questioned whether enhancing the diversity of sources was a necessary — or even viable — method of enhancing the diversity of content available to the citizenry (see, e.g., Horwitz, 2005), this has not been the case (at least not yet) within the realm of Internet governance. Within the context of Internet governance, there seems to be something approaching a consensus that diversifying the range of individuals and organizations with an opportunity to communicate online is an essential mechanism for achieving the kind of content diversity that has been deemed lacking in the online environment.
Finally, although policy scholars in both the traditional and online media realms have advocated a greater focus on the issue of diversity of exposure (i.e., the extent to which audiences access a diverse array of sources and content options; see Helberger, 2011; Hindman, 2007; Webster, 2007), this concern has failed, at this point, to gain significant traction in the realm of Internet governance (though the issue did finally receive explicit discussion in a 2012 IGF workshop on media pluralism) , just as it has largely failed gain significant traction in the realm of traditional media regulation and policy–making (Napoli, 2011a). This situation persists despite a growing body of research documenting the extent to which audience attention online is tightly clustered around relatively few content options (see Hindman, 2007, 2009, 2011).
Diversity research for Internet governance
Just as empirical approaches to the diversity principle have become increasingly prominent in traditional areas of communications policy–making, so too are we now seeing efforts to develop empirical tools for assessing diversity in ways that can potentially inform and guide Internet governance. Consequently, it is important to also comparatively assess how the diversity principle has been translated into empirical tools to inform policy–making in both traditional media and Internet governance contexts. As this section illustrates, the efforts to date to empirically assess diversity in ways that reflect and inform the priorities that have emerged within the context of Internet governance map fairly closely against established patterns of diversity assessment in the traditional media.
The need for reliable indicators and figures on which informed evaluation or policy decisions could be based is a common theme that has been emphasized in a variety of workshops and debates on diversity within the IGF. For example, it is now commonly acknowledged that the problems of market dominance and concentration of media power have not disappeared in the Internet environment. But as was noted in the 2010 workshop on how to measure communication and media in the digital converged era, the degree of concentration is increasingly difficult to measure in the online environment when there are no commonly accepted means to define relevant markets or assign market shares to different types of sources .
Similar concerns about the lack of reliable indicators to assess the development of content in different languages are also raised in discussions on linguistic diversity. As a 2009 UNESCO report notes: “In order to create a meaningful linguistic policy, the first step is to obtain relevant figures quantifying the situation, so as to be able to assess and follow up the effects of the policy, based on reliable indicators. ... The need for reliable data on the presence of languages in the Internet naturally follows.” Unfortunately, the report notes that existing linguistic measurement has been characterized by a lack of scientific rigor and even misinformation. Instead of publically available data, much of the Internet demographic data is being created and held by the commercial sector, which has also led to problems in transparency of research methodologies .
As was noted previously, within the traditional media context, source and content diversity have received the bulk of both the policy and empirical attention, while notions of exposure diversity have tended to reside at the margins of both policy discourse and empirical diversity research. We seem to be seeing something similar happening within the context of online linguistic diversity assessment. For instance, a UNESCO (2005)–sponsored assessment of the measurement of linguistic diversity on the Internet that was prepared in conjunction with the World Summit on the Information Society outlined three primary methodologies for assessing linguistic diversity.
The first approach involves the analysis of the user profiles (including native language) of the online population. Via this approach, the number or proportion of active Internet users in each language group is the focal point of analysis. Data for this approach are derived from various commercial organizations (SIL International, Global Reach) that publish aggregate data on user profiles across all nations.
The second approach involves analysis of the languages employed by users online. This approach has been described as “an absolute or relative measure of the actual use of a language on the Internet” . This method involves the analysis of a specific online communications context such as e–mail or discussion group postings in order to analyze the linguistic behavior of an identified user community (see, e.g., Climent, et al., 2003; Durham, 2003). Recent applications of this approach have focused on platforms such as Wikipedia and related user–generated encyclopedias (see, e.g., Liao, 2009).
The third approach involves the analysis of the languages employed by individual Web sites. This approach focuses on the number or proportion of Web pages written in each language. Representative global (or national, depending upon the focus of the analysis) samples of Web pages are drawn and assessed in terms of the languages they employ in an effort to determine the “Web presence”  of different languages (see, e.g., Lavoie and O’Neill, 2001; O’Neill, et al., 2003) .
If we consider these three methodological approaches in light of the model of the diversity principle discussed earlier, we see that the first approach essentially involves the analysis of source diversity. That is, in the online environment, in which the content provider–audience dichotomy has been blurred to the point of near indistinguishability, it seems appropriate to consider any assessment of the number or proportion of Internet users who speak a particular language as a (certainly imperfect) indicator of the linguistic diversity of the sources of online communication. This is not to say that such an approach necessarily represents the best, or only, approach to assessing source diversity online, but it is an approach that reflects the distribution of online speakers in relation to their linguistic affiliation.
And, as has long been the case in the traditional media realm, such an assessment of source diversity can be argued to represent a meaningful proxy for content diversity, given the likelihood that sources affiliated with a particular linguistic group will communicate using that language. But here, as in the traditional media realm, this is an assumption that’s certainly open to question, as a speaker’s primary language affiliation is not necessarily the language that the speaker uses when communicating online.
The second two approaches tap at the concept of content diversity. Both the analysis of online user communities and the analysis of individual Web pages represent efforts to assess the linguistic diversity of the content available online. The only real difference is in their unit of analysis, with the former approach generally focusing on the analysis of a fairly narrow, discretely defined online space (e.g., online encyclopedias, emails of a particular language population) and the latter approach explicitly identifying individual Web pages as the unit of analysis.
As should be clear, missing in this UNESCO overview of diversity assessment methodologies are efforts to consider the extent to which online attention is distributed across a diverse array of linguistic groups (i.e., exposure diversity). This is an omission that has not gone unnoticed, even if it has not yet resonated strongly within the Internet governance discourse or research. As the UNESCO (2005) report on linguistic diversity on the Internet noted, “We can easily produce a random count of Internet pages by using any number of commercial search engines, but we cannot judge how often Web pages are read ...” . A paper by Daniel Pimienta (2005) contained within this same report recognizes the value of an analytical focus on exposure diversity, noting that:
Our experience in the field has made us think that a promising approach that does not yet seem to be used would be a method similar to that used by Alexa to paint a portrait of the most visited sites and to provide other invaluable information. Alexa compiles data on the behaviour of a large number of users who have accepted to download spyware to their computers; this then provides extremely detailed statistics. Following the same method, we can imagine a programme that would be capable of measuring the languages used in a variety of contexts which would be relevant to indicators such as the composing and reading languages of emails, languages of sites accessed, etc. 
In recent years, some scholars have attempted related analyses, though these analyses generally have not been grounded in Internet governance concerns and thus have typically been more narrowly focused on exposure diversity within a single language population and within a specific type of online content (see, e.g., Hindman, 2009, 2011). Indeed, it is worth noting that in the most recent overview of the state of linguistic diversity research, this analytical approach receives no discussion (UNESCO, 2009). Most recently, at the 2012 IGF, policy researcher Peggy Valcke noted that there remains “no well–developed methods” to measure diversity of exposure online . It remains to be seen whether such analytical approaches become integrated into the policy discourse on the diversity principle and its role in Internet governance. Figure 3 maps these methodological approaches to online diversity assessment against the model of the diversity principle presented in Figure 1.
Figure 3: The diversity principle online: Analytical approaches.
This paper has provided an examination of how the diversity principle is migrating into the realm of Internet governance, in terms of how it is being conceptualized in policy discourse and applied in policy research. As should be clear, the diversity principle is developing in ways in the Internet governance context that both reflect and diverge from its conceptualization and application in traditional communications policy–making and policy research.
One specific source of disconnect in the conceptualization of diversity as a principle of Internet governance arises from the discrepancy between a broad understanding of diversity, concerned with the distribution of power in the online realm more generally, and a more narrow conceptualization, in which diversity is associated specifically with language issues. So far, these discussions have been largely unconnected both with each other and thus also with existing conceptualizations of media diversity in traditional communications policy–making.
This pattern may be a reflection of the broader tension that has in many ways become a defining characteristic of Internet governance discourse, and of the IGF in particular — the tension between whether Internet governance should be considered as primarily a technical enterprise, concerned exclusively with technical issues related to the operation of the Internet; or whether Internet governance should be conceptualized more broadly, to address a wider range of social and cultural issues that are embedded in how the relevant technologies operate and are accessed and used (see, e.g., Mueller, 2010; Wilson, 2005). As the discussion above indicated, the issue of linguistic diversity is tightly intertwined with one of the fundamental technical issues associated with Internet governance (the assignment of Internet domain names), which can perhaps explain why this dimension of diversity has resonated quite strongly; whereas other dimensions of diversity that are less explicitly linked to fundamental technical dimensions of Internet governance have gained less traction. Linguistic diversity is essentially an element of the broader diversity principle that has the capacity to resonate within both perspectives on the appropriate scope of Internet governance.
From a research standpoint, as the various stakeholders involved in the Internet Governance Forum, and in the broader process of establishing a coherent Internet governance regime, continue to move forward in their efforts to meaningfully conceptualize and apply the diversity principle, it is essential that policy researchers work to infuse these discussions with useful research. As this paper has made clear, there is, at this point, a lack of relevant policy–oriented research. Little of the research that has been conducted draws upon or extends the substantial body of research (and critiques of this body of research) that has examined diversity issues within the context of traditional media. This paper represents an initial effort to identify connections between these two bodies of literature. Continued efforts at drawing such connections may prove useful in terms enhancing the extent to which Internet governance issues resonate with a broader spectrum of policy advocates and policy scholars, and in terms of assuring that all stakeholders recognize the full range of relevant political and cultural implications that can underlie the diversity principle’s interpretation and application within the context of Internet governance.
About the authors
Philip M. Napoli is a Professor of Journalism & Media Studies in the School of Communication and Information at Rutgers University.
E–mail: philip [dot] m [dot] napoli1970 [at] gmail [dot] com
Kari Karppinen is a Postdoctoral Researcher in Media and Communication Studies at the University of Helsinki. His research interests include media and communication policy, media and democracy, and political philosophy. He currently works on the project Facing the Coordination Challenge: Problems, Policies and Politics in Media and Communication Regulation, funded by the Academy of Finland.
E–mail: kari [dot] karppinen [at] helsinki [dot] fi
1. Anderson, 1992, p. 387.
2. Anderson, 1992, p. 390.
3. Anderson, 1992, p. 395.
4. Duff, 2012, p. 13.
5. Van Eeten and Mueller, 2013, p. 721.
6. Specifically, this model of the diversity principle and its meaning in communications policy–making has been utilized by public broadcasters in Finland in connection with their strategic planning (Jääsaari, et al., 2004), and has been used as a framework for analyzing Mexican (Rendón, 2004), Korean and Australian media policy (Herd, 2006). Assessments of the diversity and pluralism in European media policy also have incorporated the core elements of this model (see, e.g., Interdisciplinary Centre for Law and ICT (ICRI) — Katholieke Universiteit Leuven, et al., 2009; Valcke, 2011; van Cuilenburg, 2002).
7. See transcript of IGF 2012 session on Media Pluralism and Freedom of Expression in the Internet Age (available at http://www.intgovforum.org/cms/transcripts).
8. Mueller, et al., 2007, p. 247.
9. Helberger, 2011, p. 442.
10. This estimate comes from the opening statement of Viola Krebs at the IGF 2010 session of the Dynamic Coalition for Linguistic Diversity (transcript available at http://www.intgovforum.org/cms/dynamic-coalitions/691-dc1-icann).
11. As one IGF 2007 panelist stated, “To reduce cultural diversity is to jeopardize the possibility for our species to evolve and adapt” (from participant–observation notes taken at the 2007 Internet Governance Forum.
12. This quote comes from participant–observation notes taken at the 2007 Internet Governance Forum.
13. Opening Ceremony of IGF 2009, p. 16 (transcript available at http://www.intgovforum.org/cms/2009-igf-sharm-el-sheikh).
14. Access and Diversity Theme Session p. 5. Statement of Manal Ismail (transcript available at http://www.intgovforum.org/cms/component/content/article/102-transcripts2010/644-access-and-diversity).
15. Quote from participant–observation notes taken at the 2007 IGF.
16. Diversity Theme Session, p. 3, Statement of Talal Abu–Ghazaleh (transcript available at http://www.intgovforum.org/cms/2009-igf-sharm-el-sheikh).
17. See the opening statement of Antanas Zabulis for the 2010 IGF session on Access and Diversity (transcript available at http://www.intgovforum.org/cms/component/content/article/102-transcripts2010/644-access-and-diversity).
18. See, e.g., the IGF 2012 session on Local Content Production and Dissemination as a Driver of Access (transcript available at http://www.intgovforum.org/cms/transcripts).
19. Opening Ceremony of IGF 2009, pp. 20–21 (transcript available at http://www.intgovforum.org/cms/2009-igf-sharm-el-sheikh).
20. Access and Diversity Theme Session, p. 4 (transcript available at http://www.intgovforum.org/cms/component/content/article/102-transcripts2010/644-access-and-diversity).
21. See the statement of Peggy Valcke in the Workshop on Media Pluralism and Freedom of Expression in the Internet Age (transcript available at http://www.intgovforum.org/cms/transcripts).
22. Workshop on How to Measure Communication and Media in Digital Converged Era (transcript available at http://www.intgovforum.org/cms/2010-igf-vilnius/transcripts). See also, the 2012 Workshop on Media Pluralism and Freedom of Expression in the Internet Age (transcript available at http://www.intgovforum.org/cms/transcripts).
23. UNESCO, 2009, pp. 7–8.
24. Gerrand, 2007, p. 1,300.
25. Gerrand, 2007, p. 1,301.
26. It’s worth noting, however, that this methodological approach has grown increasingly unreliable due to the way that the operation of search engines has evolved. Specific factors affecting the viability of search engine–based linguistic diversity research include the fact that search engines now represent an estimated 30 percent of the content online (compared to over 80 percent in the past, and that search returns are increasingly responsive to commercial criteria that lead to an English language bias; see UNESCO, 2009).
27. UNESCO, 2005, p. 6.
28. Pimienta, 2005, p. 33, emphasis added.
29. See the statement of Peggy Valcke in the Workshop on Media Pluralism and Freedom of Expression in the Internet Age, p. 8 (transcript available at http://www.intgovforum.org/cms/transcripts).
Charles W. Anderson, 1992. “The place of principles in policy analysis,” In: John Martin Gillroy and Maurice Wade (editors). The moral dimensions of public policy choice: Beyond the market paradigm. Pittsburgh, Pa.: University of Pittsburgh Press, pp. 387–409.
C. Edwin Baker, 2007. Media concentration and democracy: Why ownership matters. New York: Cambridge University Press.
Chris Bruegge, Kayoko Ido, Taylor Reynolds, Cristina Serra-Vallejo, Piotr Stryszowski, and Rudolf Van Der Berg, 2011. “The relationship between local content, Internet development, and access prices,” report prepared for the Internet Society, UNESCO, and the OECD, at http://www.unesco.org/new/fileadmin/MULTIMEDIA/HQ/CI/CI/pdf/local_content_study.pdf, accessed 19 November 2012.
Salvador Climent, Joaquim Moré, Antoni Oliver, Míriam Salvatierra, Imma Sànchez, Mariona Taulé, and Lluïsa Vallmanya, 2003. “Bilingual newsgroups in Catalonia: A challenge for machine translation,” Journal of Computer–Mediated Communication, volume 9, number 1, at http://onlinelibrary.wiley.com/doi/10.1111/j.1083-6101.2003.tb00360.x/full, accessed 15 June 2012.
http://dx.doi.org/10.1111/j.1083-6101.2003.tb00360.x, accessed 5 December 2013.
Council of Europe. Committee of Ministers, 2011. “Declaration by the Committee of Ministers on Internet governance principles,” adopted by the Committee of Ministers on 21 September at the 1121st meeting of the Ministers’ Deputies, at https://wcd.coe.int/ViewDoc.jsp?id=1835773, accessed 5 December 2013.
James Curran, Natalie Fenton, and Des Freedman, 2012. Misunderstanding the Internet. London: Routledge.
Alistair S. Duff, 2012. A normative theory of the information society. New York: Routledge.
Mercedes Durham, 2003. “Language choice on a Swiss mailing list,” Journal of Computer–Mediated Communication, volume 9, number 1, at http://onlinelibrary.wiley.com/doi/10.1111/j.1083-6101.2003.tb00359.x/full, accessed 22 November 2012.
http://dx.doi.org/10.1111/j.1083-6101.2003.tb00359.x, accessed 5 December 2013.
William H. Dutton, John Palfrey, and Malcolm Peltu, 2007. “Deciphering the codes of Internet governance: Understanding the hard issues at stake,” Oxford Internet Institute and e–Horizons Institute, Forum Discussion Paper, number 8, at http://www.e-horizons.ox.ac.uk/pdfs/FD8.pdf, accessed 5 December 2013.
Michel J.G. van Eeten and Milton Mueller, 2013. “Where is the governance in Internet governance?” New Media & Society, volume 15, number 5, pp. 720–736.
http://dx.doi.org/10.1177/1461444812462850, accessed 5 December 2013.
Dmitry Epstein, 2011. “Manufacturing Internet policy language: The inner workings of the discourse construction at the Internet Governance Forum,” TPRC 2011, at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1989674, accessed 1 December 2012.
EURid–UNESCO, 2012. “World report on internationalised domain names deployment 2012,” at http://www.eurid.eu/files/publ/insights_2012_idnreport.pdf, accessed 1 December 2012.
Frank Fischer, 2003. Reframing public policy: Discursive politics and deliberative practices. Oxford: Oxford University Press.
Peter Gerrand, 2007. “Estimating linguistic diversity on the Internet: A taxonomy to avoid pitfalls and paradoxes,” Journal of Computer–Mediated Communication, volume 12, number 4, pp. 1,298–1,321, at http://onlinelibrary.wiley.com/doi/10.1111/j.1083-6101.2007.00374.x/full, accessed 22 November 2012.
http://dx.doi.org/10.1111/j.1083-6101.2007.00374.x, accessed 5 December 2013.
Natali Helberger, 2011. “Diversity by design,” Journal of Information Policy, volume 1, pp. 441–469, at http://jip.vmhost.psu.edu/ojs/index.php/jip/article/view/59, accessed 1 December 2012.
Nick Herd, 2006. “Communications policy and the diversity question: Does regulation matter?” paper presented at the Communications Policy and Research Forum, September 2006, Sydney, Australia, at http://www.networkinsight.org/verve/_resources/Herd_N_paper.pdf, accessed 8 June 2012.
Matthew Hindman, 2011. “Less of the same: The lack of local news on the Internet,” report prepared for the Federal Communications Commission, at http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-307476A1.pdf, accessed 15 November 2012.
Matthew Hindman, 2009. The myth of digital democracy. Princeton, N.J.: Princeton University Press.
Matthew Hindman, 2007. “A mile wide and an inch deep: Measuring media diversity online and offline,” In: Philip M. Napoli (editor). Media diversity and localism: Meaning and metrics. Mahwah, N.J.: Lawrence Erlbaum Associates, pp. 327–348.
Lesley Hitchens, 2006. Broadcasting pluralism and diversity: A comparative study of policy and regulation. Portland, Ore.: Hart Publishing.
Robert B. Horwitz, 2005. “On media concentration and the diversity question,” Information Society, volume 21, number 3, pp. 181–204.
http://dx.doi.org/10.1080/01972240490951908, accessed 5 December 2013.
Interdisciplinary Centre for Law and ICT (ICRI) — Katholieke Universiteit Leuven, Jönköping International Business School, Central European University, and Ernst & Young, 2009. “Indicators for media pluralism in the member states — Towards a risk–based approach,” final report prepared for the European Commission, at http://ec.europa.eu/information_society/media_taskforce/doc/pluralism/pfr_report.pdf, accessed 5 December 2013.
Johanna Jä&suml;saari, Juha Kytomaki, and Erja Ruohomaa, 2004. “Audience report ’04: Diversity as a quality factor for YLE,” Helsinki, Finland: YLE, at http://yle.fi/yleisradio/sites/default/files/yle_yleisokertomus_2004.pdf, accessed 22 November 2012.
Natascha Just, 2009. “Measuring media concentration and diversity: New approaches and instruments in Europe and the US,” Media, Culture & Society, volume 39, number 1, pp. 97–117.
http://dx.doi.org/10.1177/0163443708098248, accessed 5 December 2013.
Kari Karppinen, 2013. Rethinking media pluralism. New York: Fordham University Press.
Wolfgang Kleinwächter, 2011. “Internet principle hype: How softlaw is used to regulate the Internet,” .Nxt Newsletter, at http://news.dot-nxt.com/2011/07/27/internet-principle-hype-anon, accessed 4 December 2012.
Brian F. Lavoie and Edward T. O’Neill, 2001. “How ‘world wide’ is the Web? Trends in the internationalization of Web sites,” Journal of Library Administration, volume 34, numbers 3–4, pp. 407–419.
http://dx.doi.org/10.1300/J111v34n03_21, accessed 5 December 2013.
Han–Teng Liao, 2009. “The plausibility of estimating language diversity on the Internet through online encyclopedias,” paper presented at the WebSci ’09 Conference, Athens.
Milton Mueller, 2010. Networks and states: The global politics of Internet governance. Cambridge, Mass.: MIT Press.
Milton Mueller, John Mathiason, and Hans Klein, 2007. “The Internet and global governance: Principles and norms for a new regime,” Global Governance, volume 13, number 2, pp. 237–254.
Philip M. Napoli, 2011a. “Exposure diversity reconsidered,” Journal of Information Policy, volume 1, pp. 246–259, and at http://jip.vmhost.psu.edu/ojs/index.php/jip/article/view/40/30, accessed 5 December 2013.
Philip M. Napoli, 2011b. “Diminished, enduring, and emergent diversity policy concerns in an evolving media environment,” International Journal of Communication, volume 5, pp. 1,182–1,196, and at http://ijoc.org/index.php/ijoc/article/viewFile/1074/595, accessed 5 December 2013.
Philip M. Napoli, 2001. Foundations of communications policy: Principles and process in the regulation of electronic media. Cresskill, N.J.: Hampton Press.
Edward T. O’Neill, Brian F. Lavoie, and Rick Bennett, 2003. “Trends in the evolution of the public Web, 1998– 2002,” D–Lib Magazine, volume 9, number 4, at http://www.dlib.org/dlib/april03/lavoie/04lavoie.html, accessed 22 November 2012.
Daniel Pimienta, 2005. “Linguistic diversity in cyberspace: Models for development and measurement,” In: Measuring linguistic diversity on the Internet. Paris: UNESCO, pp. 13-42, and at http://unesdoc.unesco.org/images/0014/001421/142186e.pdf, accessed 5 December 2013.
José Carlos Lozano Rendón, 2004. “Public policies and research on cultural diversity in Mexican television,” paper presented at the Seminar on Media and Globalization in North America, Monterrey, Mexico.
Patrick Roessler, 2007. “Media content diversity: Conceptual issues and future directions for communication research,” Communication Yearbook, volume 31, pp. 464–520.
UNESCO, 2009. Twelve years of measuring linguistic diversity on the Internet: Balance and perspectives. Paris: UNESCO, and at http://www.unesco.org/new/en/communication-and-information/resources/publications-and-communication-materials/publications/full-list/twelve-years-of-measuring-linguistic-diversity-in-the-internet-balance-and-perspectives/, accessed 5 December 2013.
UNESCO, 2005. Measuring linguistic diversity on the Internet. Paris: UNESCO, and at http://www.unesco.org/new/en/communication-and-information/resources/publications-and-communication-materials/publications/full-list/measuring-linguistic-diversity-on-the-internet/, accessed 5 December 2013.
UNESCO, 2003. Recommendation concerning the promotion and use of multilingualism and universal access to cyberspace. Paris: UNESCO, and at http://www.unesco.org/new/en/communication-and-information/about-us/how-we-work/strategy-and-programme/promotion-and-use-of-multilingualism-and-universal-access-to-cyberspace/, accessed 5 December 2013.
UNESCO, 2002. Universal declaration on cultural diversity. Paris: UNESCO, and at http://unesdoc.unesco.org/images/0012/001271/127160m.pdf, accessed 5 December 2013.
Peggy Valcke, 2011. “Looking for the user in media pluralism: Unraveling the traditional diversity chain and recent trends of user empowerment in European media regulation,” Journal of Information Policy, volume 1, pp. 287–320, and at http://jip.vmhost.psu.edu/ojs/index.php/jip/article/viewArticle/31, accessed 5 December 2013.
Jan van Cuilenburg, 2002. “The media diversity concept and European perspectives,” paper presented at the Media Economics, Content and Diversity Seminar, Finnish Academy of Sciences, Helsinki.
James G. Webster, 2007. “Diversity of exposure,” In: Philip M. Napoli (editor). Media diversity and localism: Meaning and metrics. Mahwah, N.J.: Lawrence Erlbaum Associates, pp. 309–326.
James G. Webster and Thomas B. Ksiazek, 2012. “The dynamics of audience fragmentation: Public attention in an era of digital media,” Journal of Communication, volume 62, number 1, pp. 39–56.
http://dx.doi.org/10.1111/j.1460-2466.2011.01616.x, accessed 5 December 2013.
Ernest J. Wilson, III, 2005, “What is Internet governance and where does it come from?” Journal of Public Policy, volume 25, number 1, pp. 29–50.
http://dx.doi.org/10.1017/S0143814X0500019X, accessed 5 December 2013.
World Summit on the Information Society (WSIS), 2005. “Tunis agenda for the information society” (18 November), at http://www.itu.int/wsis/docs2/tunis/off/6rev1.html, accessed 5 December 2013.
World Summit on the Information Society (WSIS), 2003. “Geneva declaration of principles” (12 December), at http://www.itu.int/wsis/docs/geneva/official/dop.html, accessed 5 December 2013.
Received 6 December 2012; revised 3 October 2013; accepted 10 October 2013.
This paper is licensed under a Creative Commons Attribution–NonCommercial 4.0 International License.
Translating diversity to Internet governance
by Philip M. Napoli and Kari Karppinen.
First Monday, Volume 18, Number 12 - 2 December 2013