One of the fundamental challenges facing ICANN, and regularly articulated at its recent 49th meeting in Singapore, is how to reach consensus amongst the many different stakeholders with interests in the future of the Internet. Having been doing research over the last 15 years on how to ensure success in multi-stakeholder partnerships (see for example my recent 2013 post, and an older 2012 post on partnerships in education) as well as working with a range of groups on consensus decision making, I find these discussions fascinating, not least for their theatrical quality but also for the apparent lack of knowledge exhibited on the very extensive research that has already been done on managing multi-stakeholder partnerships.
Two intersecting themes seem relevant, not only to ICANN, but also more widely to the many other ongoing international debates on global governance, particularly with reference to ICTs. These are hugely complex issues, far too challenging to resolve in a simple blog post, but what I want to do here is summarise what I see as being the main issues that require resolution:
- Multi-stakeholderism representation. I have to admit hugely to disliking the term multistakeholderism, despite the fact that I frequently plead for people to use the term “multi-stakeholder” rather than “public-private” to refer to the kinds of partnership that are necessary to deliver effective ICT for development initiatives. “Multi-stakeholder” is preferable because it emphasises that such initiatives require a more diverse set of stakeholders than just the private and public sectors, and that they particularly need to involve civil society. Most research on multi-stakeholder partnerships has focused on how to bring partners together to deliver particular initiatives at a national or local scale, and far less in the context of reaching international agreements (although see Jens Martens’ important work on the latter). The use of the term “multi-stakeholder” has nevertheless been clearly recognised by ICANN (albeit defining it in a very particular way, as treating “the public sector, the private sector, and technical experts as peers”), but a fundamental challenge is to identify the means through which each group can, or should, be represented in international discussions on critical ICT issues. Four issues seem particularly problematic and pertinent:
- Defining multi-stakeholders groupings. Most work on multi-stakeholder partnerships recognises a triadic typology of “states”, the “private sector” and “civil society”. However, there are additional types of entity over and beyond these that might be involved under these headings, including international organisations, foundations, and indeed user groups. These are sometimes treated as sub-sets of civil society, but on other occasions as distinct entities in their own right that could be grouped into additional categories.
- Numbers and scale. In global bodies concerned with international treaties, such as UN bodies including the ITU, governments usually have the dominant say, albeit that this say is increasingly being challenged. It is relatively easy to choose the entities that represent governments – they are, after all, finite in number – but for the private sector and especially civil society it becomes much more problematic. UNDESA’s integrated Civil Society Organizations (iSCO) System thus currently maintains a database of more than 24,000 entries (see also the UN Global Compact’s list). How can representation from this diversity of stakeholders be included, especially when it is often unclear who exactly these civil society organisations represent?
- Representative democracy. Invariably it is only the larger and richer companies and civil society organisations that are able to participate in major international gatherings – often quite simply because of the cost of so doing – although many UN bodies do indeed welcome civil society participation once they have been recognised in some way as members. In crafting such partnerships, and in line with the notion of representative democracy, there can be value in seeking to involve some kind of representative mechanism, whereby stakeholders elect from their membership people or institutions to speak on their behalf. This prevents the decision making process becoming too unwieldy, but those not elected onto the “Board” can feel aggrieved and not-represented.
- Governance structures. The mechanisms for selecting representatives also depend heavily on the kinds of governance structure that are deemed to be appropriate for the purpose in hand. Even here there are difficulties because someone has to determine these criteria in the first place. At a simplistic level, it would be possible to imagine a multi-stakeholder decision making body made up of a set number of members from each of the three key sectors of “governments”, “companies” and “civil society”. Within this, there would then need to be mechanisms for determining how the elections would take place, and what the constituencies should be. In the ITU, for example, members of the Council and the Radio Regulation Board are elected based upon regional groupings.
- Consensus decision making and democratic representation. One of the most fascinating aspects of seeking to reach global agreement on particular issues is the choice of the process that is used to seek consensus. When combined with representative mechanisms, most consensus building models use an aggregative process, whereby agreement is sought at one level (for example the “local”), and then representatives from that level meet at a higher level (such as the “regional”) to seek wider consensus. This can be a very effective mechanism for reaching consensus, but the ways in which the governance of such structures operate can lead to very different outcomes. This is highly pertinent to discussions about governance of the Internet and ICTs. Six main principles and issues seem particularly pertinent here:
- Consensus building requires good will on behalf of all of those involved. Put simply, if there is not a desire to reach agreement on the part of some of those involved, then no amount of skilled negotiation will reach a successful outcome. The first stage of any consensus building process must therefore be the need to convict all participants of the benefits of reaching a consensus. Ultimately, those not willing to commit to this need to be excluded in the interests of reaching agreement among those who are willing to engage in the process.
- Generally speaking, it often makes sense to try to reach agreement on the most contentious issues at the lowest/local scale, because most time can usually be devoted to reaching consensus here. For example, if it is expected that different ethnic groups have very different views on a subject, then it makes sense for the difficult issues to be resolved at the lowest scale that can combine these multiple different ethnic perspectives. However, this does not always work, since unexpected disagreements can emerge later in the process, which can prevent the final reaching of a consensus.
- Moderation of the consensus building process requires great skill and patience. All too often, inexperienced chairs or moderators are charged with seeking to reach agreement among a particular constituency, and this can rapidly lead to dissatisfaction and disenfranchisement with the entire process.
- The choice of representatives to carry forward the discussion at a higher level is critical. Such people need to combine excellent negotiation skills with empathy for the different perspectives that they need to represent. They also need to be trusted by their constituencies.
- Despite a tendency to wish to return to the lowest level to get final agreement on the principles agreed at a higher level, this often leads to the unraveling of the process. This is largely because consensus decision making requires skillful bargaining, and not everyone involved at the earliest stages of a process may be aware of the issues that emerge later in the process that require resolution. It is, though, particularly useful if the higher level discussions are open to participation from anyone who wishes to be an observer from the lower levels in the process, since this can serve as a useful check on the probity of the representatives and negotiators.
- Ultimately, those involved in building consensus need to adhere to the fundamental negotiating principle that they should focus particularly on “What can’t you live with; what can’t you live without“!
If, and it is a big if, the global Interent governance agenda is seen as being concerned with reaching agreement amongst “governments”, “private sector companies” and “civil society”, then drawing on the above two main alternative model structures can be conceptualised:
- Model A – initial consensus building at a national level
- The lowest level discussions take place in national forums that bring together representatives of governments, the private sector and civil society
- National representatives (not necessarily drawn from governments) then meet to reach regional consensuses, such as for East, North, Southern and West Africa.
- Finally, representatives from these global regions meet to thrash out global agreements.
- Model B – initial consensus at a sectoral level
- The lowest level discussions take place in regional sector-specific global forums one in each region (such as East, North, Southern and West Africa) for representatives of governments, another for the private sector and a third for civil society.
- Representatives from each of these regional sector meetings (or indeed subdivisions within them) then meet to reach a global consensus. For example, there would be a global private sector meeting bringing together regional private sector representatives, and similar fora for governments and for civil society.
- Finally, representative of each of the three main groupings meet at a global meeting to bring together the three broad swathes of governments, the private sector and civil society.
To date, it would seem that Model B has often been the preferred modality of consensus building in discussions about Internet and ICT governance. The ITU, for example, holds regional meetings in advance of its major conferences, where it seeks to reach agreement on key issues.
Significantly, most of the major international bodies working in the field of ICTs and the Internet claim in some way to be multi-stakeholder. However, the driving force for each entity usually tends to be from one or the other sectors, be they governments, the private sector or civil society. Against this context, broadly speaking, ICANN (a private sector, non-profit corporation) has tended to focus on the interests of the private sector, the IGF as a multi-stakeholder policy dialogue (purportedly supporting the UN Secretary General) is widely seen as being the main vehicle for civil society participation, and the ITU is the UN agency generally accepted as being a predominantly governmental body (although defining itself as a “public-private partnership”). A real challenge is how to bring these together – or whether indeed there is actually real interest in so doing. Attempts to create a truly global forum, including the ill-fated Global Alliance for ICTs and Development (GAID) have largely failed, although the WSIS+10 process led by the ITU and involving other UN agencies continues to strive to bring a wide range of participants together.
This post is already too long, and barely scratches the surface of these complex issues! However, we have to find a way to stop holding the same conversations in different circles, and actually create structures and consensuses that serve the interests of the poorest and most marginalised!