6.4.1. Representation

If formal decisions of the IGF are not to be made directly by its plenary body but through the proposed multi-stakeholder bureau, then the IGF becomes closer to a representative than a direct democracy. However, the bureau does not represent the plenary in the same sense that the parliament of a liberal democratic state represents the demos. Specifically, there is no need for it to act as a filter for the plenary’s views, since the plenary is itself to adopt a deliberative process. Rather, the fact that the bureau’s recommendations are to reflect the plenary’s deliberative consensus means that it acts strictly as a mirror of the plenary.[1] In fact if anything, the image of the plenary that is reflected in the bureau may be slightly biased towards the interests of the disadvantaged, by reason of the criteria employed by the nominating committee to ensure that the bureau’s composition is substantively democratic and diverse.

Exactly what processes should such a multi-stakeholder bureau then employ in making its formal decisions? There is no close functioning model that conveniently answers this question. Francis Muguet describes the IGF as a UN multi-stakeholder process that

is completely new at the UN and in International Public Law anywhere. There are no UN rules concerning a full, equal footing, multi-stakeholder process, only concerning intergovernmental processes, it is a simple as that ... For example, the very question concerning “observers” is meaningless because there are no “observers” since all stakeholders are on an equal footing. ... There is no need to bend existing UN rules, but to find new ones, when none exist. UN multi-stakeholder rules of procedures and working methods are to be invented.[2] The UN multi-stakeholder process

This does overstate the point slightly, in that the development of a multi-stakeholder process for the United Nations has been a concern of the organisation since at least the release of the Cardoso Report on UN–Civil Society relations that was released in 2004, which recommended that “[s]ince networked governance is clearly emerging as an important aspect of policy-making, the United Nations must embrace and support it more overtly if it is to remain at the forefront of global policy-making.”[3] Moreover, there are precedents for the UN multi-stakeholder process in two of the other organisations addressed in this chapter: UNICTTF, and its successor GAID.[4]

Taking UNICTTF first, this was an example of a multi-stakeholder organisation, formed by intergovernmental agreement, with equal decision-making power distributed among all its members. Like the IGF, it was directed not “to take over or supersede other important processes in this area,” being “not envisaged as an operational or executing agency,” yet at the same time it was required to “submit an annual report to the Secretary-General which will focus on the major emerging issues, including recommendations thereon.”[5]

The main difference between UNICTTF and the IGF in structural terms is that the former had a defined membership of 55 who were appointed in representative capacities, along with a Chair whom the body would elect itself.[6] In this respect therefore UNICTTF can be viewed as something of a hybrid between the IGF in plenary session and the multi-stakeholder bureau proposed here.[7]

Although the detail of its working processes was not the subject of UN resolution, in practice decisions of UNICTTF were made by “broad consensus” or “general agreement,” as assessed by its Chair.[8]

GAID illustrates the evolution of the UN multi-stakeholder structure and process towards a post-WSIS model that is closer to that of its sister body, the IGF. Thus GAID, like the IGF and unlike UNICTTF, does not possess a fixed membership; being structured as a decentralised governance network that is open to all interested stakeholders.

With this looser structure has come a restriction of its policy-making capacity. GAID is described explicitly as a “channel for multi-stakeholder input to policy debate to be conducted in intergovernmental organs,”[9] which “will not have an operational, policy-making or negotiating function” in its own right.[10] Thus unlike the IGF and UNICTTF, the making of recommendations does not form part of GAID’s formal mandate.

GAID’s decision-making organs are a multi-stakeholder Strategy Council and a Steering Committee, both of which are appointed by the Secretary-General. The Strategy Council of sixty is divided into equal numbers of governmental and non-governmental members, and pursuant to its terms of reference “[p]rovides overall strategic guidance and vision to the Alliance.”[11]What few decisions the GAID Strategy Council has made pursuant to these very general terms of reference have been made by consensus, as assessed by the Chair of the Steering Committee and as reduced to writing by its Secretariat.[12]

The Steering Committee is a smaller group of twelve led by a Chair, who are appointed by the Secretary-General to act in their personal capacities. The Steering Committee has a more detailed terms of reference which includes the approval of applications for the establishment of Communities of Expertise (GAID’s equivalent of working groups) and stakeholder networks, the review of progress reports from these bodies, the approval of recommendations from the Secretariat, and the delivery of “inputs” (not “recommendations”) on ICT for development issues to the Secretary-General.[13] The Steering Committee’s decisions are also made by consensus, as assessed by the Chairman and reported by the Secretariat.[14]

The evolution of the UN multi-stakeholder process from UNICTTF through to GAID and the IGF illustrates that as the openness and inclusiveness of the plenary body has been increased, so it has been disempowered, with decision-making authority instead being concentrated at the highest executive level; or in practice, in the organisations’ UN-appointed Chairs and Secretariats. The UNICTTF’s defence is that “stakeholders involved in the ICT TF sometimes have competing or even conflicting aims. Organizations, including multilateral, bilateral and civil society groups are often competing for influence, leadership, support, and attention.”[15]

Whilst this is undeniable, to accept it as a justification for the relegation of those stakeholders to a lowly advisory role flies in the face of UN rhetoric about the need for multi-stakeholder public policy governance, for which the proffered alternative of participatory democratic consultation is a poor substitute. The democratic principle, which specifies that governments must operate with the consent of the governed, and the principle of subsidiarity which requires that governance should be exercised at the lowest practical level, require more. So too, for that matter, does the Tunis Agenda. A consociational multi-stakeholder process

How then, should the bureau produce its recommendations or other formal output, given that in doing so it is to act as a mirror of the plenary body, for which the United Nations multi-stakeholder process offers an inadequate model?

Referring back to Section 6.3.2, which listed seven substantive functions that it had been found such a bureau would need to perform, the process of making formal recommendations is made up of the last two functions: assessing the consensus of the plenary body as to the policy objectives to be achieved and the appropriate means of achieving them, and preparing any appropriate formal output for the IGF based on that consensus, such as recommendations for input into other organisations.

Since it has now been determined that the bureau should be consociational in form, it is also necessary that once draft recommendations or the like have been prepared, they must be separately approved by each stakeholder group within the bureau, by giving that group the opportunity to exercise its right of mutual veto.

These three phases, then—determining the consensus of the plenary body on a particular issue, drafting any applicable recommendation or statement that may be required in response, and then seeking the assent of each stakeholder group—form the core of the consociational multi-stakeholder process by which the bureau represents the views of the plenary body. The phases may also be iterative. That is, if a stakeholder group determines to exercise its veto, this may indicate that:

The processes involved in the first phase outlined above, assessing the consensus of the plenary forum, will be considered under the following heading of consent, along with the analogous processes of assessing the consensus of workshops and dynamic coalitions.

As for the second phase, in which the bureau develops a formal expression of the public policy upon which the plenary forum has reached a consensus, the processes involved are simply those required to deliberatively develop an agreed text within a relatively small, multi-stakeholder group, much as WGIG did when producing its report. These processes will also be dealt with under the next heading, when considering the use of democratic deliberation within each of the IGF’s organs. However two distinguishing features of these processes within the context of the multi-stakeholder bureau should be noted here:

The third phase in the bureau’s decision-making process is that which provides each stakeholder group with the opportunity to veto any decision of the bureau at large that may have been reached during the second phase. Unlike the second phase which involves deliberation amongst the bureau’s members as a multi-stakeholder entity, in the third phase of the process, the stakeholder groups decide whether to exercise their collective powers of veto separately.

Such separation of the stakeholder groups during the third phase carries two advantages, other than those already noted during the discussion of consociation and in considering the structural advantages of a consociational bureau.[17]

The first is that since the form of the consociational bureau is not prescriptive about how a stakeholder group should decide to exercise its power of veto, this phase allows for the different characteristic working methods of each stakeholder group to be accommodated.

So for example, governments may require a face-to-face meeting between diplomats to decide upon their exercise of the power of veto, while civil society may be accustomed to making such decisions by rough consensus using online tools, and the private sector representatives might even prefer to delegate the decision to a representative association such as the ICC. The segmentation between stakeholder groups in a consociation brings all of these modes of decision-making together within the bureau.[18]

The second benefit of each stakeholder group within the bureau sharing the power of mutual veto during the third phase of decision-making is that this will influence the shape of deliberations during the second phase; not only by reinforcing the equality of each group as previously noted, but also by streamlining the internal politics of the bureau. Avoiding the unwieldy need for a particular form of recommendation to be negotiated between fluid networks formed from amongst (say) forty individuals, instead it will be possible to move forward so long as the three stakeholder groups, who hold ultimate decision-taking power, are able to reach a rough accord between themselves. Collaboration and compromise between stakeholder groups then takes priority over power relations between the individuals who happen to constitute the bureau at any given time. As well as better meeting the deliberative democratic conditions of freedom and equality,[19] this also assists the bureau to overcome the concern noted by Kenneth Cukier that

[t]he process of Internet governance fails because so far the notion of collaborative policy-making is completely missing—there are no ideological camps, no political parties or coalitions in which groups are forced to sublimate their ideal self-interest for a suitably acceptable compromise, in order to attain the benefits of the workable system as a whole.[20]

One possible criticism of a divided bureau is that the separation between stakeholder groups could create conflict and induce participants, cloistered in their fixed constituencies, to adopt entrenched positions. This in turn could limit the range of options and the perspectives surrounding a given issue from being fully considered, which works against the subsequent formation of consensus. This problem is an instance of the “silo effect” which produces conflict between separated organisational units,[21] and which in turn results from the phenomenon of groupthink that exists within such homogeneous units.[22]

Without necessarily accepting that the stakeholder groups in the present instance are particularly homogeneous, this potential weakness is overcome here by the fact that the third phase of decision-making, in which stakeholders deliberate separately upon the exercise of their right of mutual veto, is preceded by the second phase of full multi-stakeholder deliberation, which enables all perspectives to be aired at an early stage (not to mention that democratic deliberation has also already taken place within the plenary body when substantive consensus was reached).

A related criticism is that not only might the third stage of decision-making within a particular stakeholder group not be deliberative, it might not even be democratically accountable and transparent. Indeed if ICANN’s GAC, or even the IGF’s Advisory Group for that matter, is taken as a likely model upon which for the governmental stakeholder group to structure itself, this can be taken as a given. For now the same answer can be given to this as to the preceding criticism—that it is sufficient that the multi-stakeholder phase of deliberation within the bureau is open and transparent—though this criticism will also be further addressed below.[23]

A third possible criticism is that, in common with other consociational structures, the bureau will require a strong set of shared norms that inhibit the use of the power of veto except where necessary to protect a stakeholder’s core interests,[24] in order to mitigate the risk of consensus being blocked by a small minority during the third phase.[25]

In response to this, it should be noted that any recommendation that has passed the second phase is one which already carries the support of the majority of the bureau, as determined either by voting or rough consensus. Except where support for the recommendation falls strictly along stakeholder group lines (rather than, for example, between Forum hawks and Forum doves), this support is also likely to extend to some degree into each stakeholder group. It is therefore more likely that a proposal will fail at an earlier phase than that it will be vetoed out of hand once it reaches the third phase.

More importantly however, the process of democratic deliberation that resulted in a recommendation receiving majority support within the bureau will itself tend to induce the development of the group norms such as trust, cooperation and equality that are necessary to prevent the power of veto from being misused.[26] These norms will also be reinforced over time as the IGF’s social capital is developed through its successful operation as a governance network.[27]



See Section


Muguet, Francis, A Legal Analysis of the Internet Governance Forum Process (2007), 19


Cardoso, Fernando H, Cardoso Report on Civil Society (2004), 33


Beyond those, see Martens, Jens, Multistakeholder Partnerships: Future Models of Multilateralism? (2007).


ECOSOC, Information and Communications Technology (ICT) Task Force (2002), 4–6


ECOSOC, Information and Communications Technology (ICT) Task Force (2002), 6; Martens, Jens, Multistakeholder Partnerships: Future Models of Multilateralism? (2007) , 28. Originally there were 37, then 40 members.


Though UNICTTF additionally had its own bureau of six, as well as a Panel of Advisors of thirty; which was also known as its Advisory Group and was a precursor to that of the IGF.


See UNICTTF, Summary of Conclusions and Decisions (2003).


UNICTTF, Principles and elements of a Global Alliance for ICT and Development (Multi-stakeholder Forum) (2006), 2


UNICTTF, Principles and elements of a Global Alliance for ICT and Development (Multi-stakeholder Forum) (2006), 4


GAID, Terms of Reference—Strategy Council (2006)


GAID, Strategy Council Meeting: Informal Summary (2006)


GAID, Terms of Reference—Steering Committee (2006)


GAID, Meeting of the Steering Committee of the GAID: Informal Summary (2007)


UNICTTF, Strategic Plan (2002), 7


See Section 6.2.


See Section and Section


Compare Muguet, Francis, A Legal Analysis of the Internet Governance Forum Process (2007), 20.


See Section


Cukier, Kenneth N, Slouching Towards Geneva: Ten Unappreciated Axioms of Internet Governance (2005), 5


Burke, Wyatt W, Organization Change: Theory and Practice (2002), 163


See Section


See Section


Skelcher, Chris, Jurisdictional Integrity, Polycentrism, and the Design of Democratic Governance (2005), 105


See Section


Uphoff, N, Understanding Social Capital: Learning from the Analysis and Experience of Participation (1999)


See Section 4.3.2.