CGCS Post-Doctoral Research Fellow Ben Wagner reflects on the April 2014 NETmundial Global Multistakeholder Meeting on the Future of Internet Governance, discussing the decision making procedures and structure of the meeting.
Despite coming to a close, NETmundial 2014 in São Paulo remains a strange political beast. The organization and decision making-processes involved during the conference itself swung wildly between haphazardly obscure, open and participative, and mere political grandstanding. In a sense, it was great political theater to watch, but it was also deeply frustrating for anyone trying to get anything done.
There has been enough historicizing about what NETmundial was meant to be doing but little conversation on what it was actually doing. While Stefania Milan has already penned an excellent piece on the sociology of NETmundial, a sociology which included a wide array of techno-activist groups previously not present at the table, this post will instead focus on the decision making procedures and structures of the meeting.
Who is Assigned to Seats of Power?
At the outset, it should be noted that there is a substantial legitimacy problem in how the NETmundial Executive Multistakeholder Committee (EMC), the High-Level Multistakeholder Committee (HLMC), and the members of the drafting committee were selected. Some committee members were only informed that they would be on these committees a week in advance. Others participated in a process in which names for various positions were proposed to the organizers. However the process by which individuals were eventually appointed to these committees is entirely unclear.
Many of the stakeholders who might have had an interest in this conference were likely completely unaware of email lists such as [governance], [BestBits], or [GigaNet] through which much of the coordination and voting for appointed members took place. The people on key NETmundial committees therefore overwhelming represented established elites, individuals who have been working on internet governance for decades and share many of the same norms and values. Their assumptions about what process is appropriate, and what ‘the internet’ is and how it should be shaped are therefore based on decade long internal conversations of the ‘in crowd.’
The terms multistakeholder and rough consensus have been papered over all manner of cracks in the internet governance ecosystem. Even when these terms were put to the test in practice, the resulting process was decidedly brittle. After the long list of introductory speeches that lasted most of the first day, NETmundial basically devolved into a participant statement collection mechanism.
Participants lined up in queues that were arranged around stakeholder groups and gave opinions to the chairs of their groups. These interventions lasted between thirty seconds and two minutes depending on how much time was left. While a shortening of speaking time was indeed welcome, the format tended to lead to a drone of publicly outputted statements that could equally have taken place online. The alternative to this is taking more time and structuring the statement collection process differently. While it was designed to prevent diplomatic speeches, its ad hoc time shifts and stakeholder quotas were wholly inadequate to deal with the challenges at hand.
Where Sausages are Made
At the same time, most of the actual bartering was taking place in the drafting room. Here, drafting committee members were supposed to be making decisions and allowing ‘participants’ to watch them but not interact with them. The reality was of course different. From asking the ‘participants’ watching the draft committee questions, to requesting clarifications regarding individual input, to sending away ‘multistakeholder groups’ with the request: ‘come back when you’ve found a consensus,’ the whole process was incredibly ad hoc and lacked any kind of structure.
Most decisions were made in this drafting room, thus those who had the most agency in making changes to the outcome document were those who happened to be in the drafting room at the right time and were prepared to engage. Organizations and groups with high organizational capacity and effective coordination were therefore most likely to be there, and, in most cases, these were the groups from the business community.
This poses significant legitimacy challenges as “the people who happened to be in the room” at any given time is not a good proxy for legitimate representation. This is particularly the case as such participation opportunities were not announced and thus not even available to every participant in equal measure, let alone to those who were not present. While the drafting-room bartering did serve to make many of the more ugly sides of multistakeholderism more transparent, it also served to show that nothing much has changed. In this sense, multistakeholderism was politics as usual with large business groups dominating the debate and government acquiescence. Such machinations are a far cry from the participatory and integrative claims of multistakeholder governance. This can only be remedied by going back to the multistakeholder drawing board and ensuring democratic participation at the beginning of the negotiating process. It also involves significant reconsiderations of how stakeholders are selected to participate and how voting is actually meant to take place.
Fragmented Civil Societies
In contrast to the highly organized business sector, civil society groups, as well as the technical and academic communities, were all over the place. While neither technical nor academic communities were anticipated to organize around much, NETmundial made it painfully obvious that – at least among those present – there was, and remains, little or no consensus among civil society on these topics.
At a cursory level, groups working on human rights, privacy, and surveillance almost uniformly wanted much stronger language in the principles document, legitimately assuming that these issues were part of the conference’s purpose. The perfunctory consensus that seemed to exist between civil society groups over the tone of the outcome document quickly deteriorated. At the same time, the fragmentation of opinions and positions within civil society makes it very difficult for civil society to be effective at these kinds of meetings. Lack of clear positioning and internal coordination are obvious stumbling blocks to achieving any kind of effective impact on outcomes of NETmundial.
The deeper problem of course is that various civil society groups have very different interests and arenas to address these interests. Perpetuating multistakeholder dialogue is what keeps a lot of internet governance folks at the table. Only by keeping the multistakeholder model afloat are they able to travel from ICANN meetings to the IGF, to the other multistakeholder fora and be consistently labelled as experts. Their expertise rests primarily on their relevance in understanding the model and their seniority in having been part of the process for a long time. As noted in a previous blog post by Stefania Milan, these internet governance experts have very different positions from the techno-activists or many other parts of the NGO community. The discordant parts of the civil society community thus find it difficult to find a clear common ground, or even a minimal common understanding, of what their goals and strategies are.
Double Veto Power
At the end of the day it was two states that had ultimate veto power on the outcomes of the final NETmundial statement document: the United States and Brazil. Thus for all of the conversations about multistakeholderism and equal footing for stakeholders, two states still defacto ran the show.
This was powerfully evident during the final High Level Multistakeholder Committee (HLMC) meeting at the end of the conference. Brazil and the United States battled over the outcome while other stakeholder groups barely had a say. It is not that Brazil had more seats on the committee, but there was an implicit assumption evident in HLMC and EMC discussions that the opinions of these two states were not to be challenged. Thus, even if in theory the U.S. or Brazilian representatives could have been outvoted, no challenges of their positions were made by other actors during the key final deliberations. This meeting is when most of the stronger language on surveillance, intermediary liability, and net neutrality were removed from the document. Although the HLMC were not quite sure whether or not they were even allowed to vote, they went ahead and did so anyway.
Absurdist Theater or Innovation in Governance?
It is in a sense remarkable that the final document includes significant reference to access and the right to development, concepts that key influencers, including many business stakeholders and the U.S., were initially completely opposed to. There are many other phrases in the statement that provide room for hope. For all of the many criticisms of the multistakeholder model and its illegitimacy, the conference did produce something. Whether many organizations agree or disagree with this document is beside the point—it exists and has normative power through its very existence.
The final NETmundial outcome document is likely to be around for a while as the ad hoc follow-up to the WSIS process. As Milton Mueller and others have suggested, we do indeed now live in a new post-WSIS internet governance world. NETmundial responded less to Snowden and more to the inability of IGF, CSTD Working Group, and ICANN to develop any substantive progress on multistakeholder governance. This progress was now rammed through at the expense of strong normative positions on Snowden or surveillance.
All theater is politics, all politics is theater. While multistakeholder internet governance may seem particularly like a Dadaist practical joke, it is precisely these humoristic qualities that obscure its actual politics. In an ad-hoc internet governance world in which decisions are made on the fly and consensus is manufactured so rapidly, you fail to take multistakeholder internet governance seriously at your own peril.