An extended role for the IGF in filling the gaps in the Internet governance ecosystem

  • Entitled by: Jeremy Malcolm
  • Region: Malaysia
  • Organization: Personal submission
  • Sector: Civil Society
  • Keywords: IGF, multi-stakeholderism, stakeholders, roles, consensus


This submission builds on the joint civil society submission developed through the Best Bits platform, taking as a starting point the distributed model of Internet governance put forward in that submission. But it adds two notes of caution that the Best Bits submission does not cover, and proposes possible solutions to these, based around an extended role for the Internet Governance Forum.


This submission takes as its starting point the submission from participants of the Best Bits network on “Roadmap for the Further Evolution of the Internet Governance Ecosystem – institutional mechanisms” (, which I participated in drafting and generally agree with (“the Best Bits submission”).

In particular, I take for granted the following points that are either drawn from, or at least are fully consistent with the Best Bits submission, and I will not explain them further here:

There are certain policy issues that cannot be addressed within the current Internet governance ecosystem in a way that adequately fulfils basic procedural criteria, due to underlying institutional deficits.

  1. In addressing these deficits certain criteria must be fulfilled, including a rights-based approach, preservation of the global nature of the Internet, and the use of multi-stakeholder democratic processes.

  2. To redress these deficits requires a new coordinating mechanism to direct stakeholders towards the appropriate existing fora to deal with policy issues, as well as one or more new multi-stakeholder working groups to address issues that have no existing home.

  3. The new multi-stakeholder working groups referred to would in most cases develop soft law only, rather than binding rules such as treaties, though they could in appropriate cases make recommendations to external treaty bodies or to the UN General Assembly.

  4. It would be appropriate for the Internet Governance Forum (IGF) to take a key role in the operationalisation of these required changes, which would require the most significant reforms to the IGF since its establishment.

There are however two problems that the Best Bits submission does not explicitly acknowledge, nor suggest a way to overcome, which are the subject of this submission:

(1) There remains a substantial disagreement between (and even within) stakeholder groups, over whether governments should have a preeminent role and responsibility over international public policy issues pertaining to the Internet, or whether the roles of all stakeholder groups in Internet policy development processes should be equal. Practically speaking, there is no likelihood that NETmundial will see any resolution of this fundamental disagreement.

In the narrower context of ICANN, the GAC is an unsuccessful attempt to compromise between these two demands, by giving governments certain rights in the policy development process that no other stakeholder group has (which upsets those who demand equality between the stakeholder groups), but by making its recommendations purely advisory (which upsets those who demand government sovereignty).

It may seem that this problem is irreconcilable, in that any governance institution must choose either one or the other – either making all stakeholders equal, or giving governments sovereign rights. But counter-intuitively, it may be possible to satisfy both demands, and to do so without creating any new body or process, other than those referred to in the Best Bits submission. The solution involves the following two elements:

(1a) First, there are some issues and contexts in which, as a factual matter, governments do not claim a primary role in policy development, nor ever have (and the same is true in other governance domains also). As these typically involve technical issues, the drafters of the Tunis Agenda accounted for this fact by drawing a (simplistic) division between “international public policy issues pertaining to the Internet” and “day-to-day technical and operational matters, that do not impact on international public policy issues”.

A less simplistic approach requires a case-by-case analysis: whenever the coordinating function as described in the Best Bits submission is exercised, it should consider not only whether and where policy is being developed, but also what the appropriate roles of stakeholders are for that issue. Factors relevant to that determination include agreed high-level global Internet governance principles (if any), roles and responsibilities historically exercised in respect of that issue, andthe governance mechanisms that are likely to be effective in dealing with the issue.

To give some possible examples, it may be found that governments are entitled to take a leading role in developing global rules on money laundering via the Internet, but that they should take a back seat in the development of spam filtering standards, and should participate as equals with other stakeholders in the development of policy for transnational intellectual property enforcement policies.

By engaging in this analysis (which would itself be conducted by a multi-stakeholder deliberative body), a more reasoned determination can be made of the appropriate roles of governments and other stakeholders in a given issue area, than can be had by resorting to fixed descriptions of stakeholder roles, or an arbitrary demarcation between “public policy” and “technical and operational” issues. The result will be that the appropriate role of governments in determining any given policy issue can be defined with more particularity than simply making a blanket claim of sovereignty.

(1b) Second, for areas in which the coordinating function assesses that governments do have an appropriate role to play along with other stakeholders, the only way in which to resolve the disagreement mentioned above is if the multi-stakeholder policy development process is so designed that the stakeholder groups are formally equal – but allows governments to veto decisions. Logically the only way in which both those conditions can both be true is if all stakeholder groups can veto decisions.

This describes a consociational multi-stakeholder process, where a consociation is a power-sharing arrangement comprising a coalition of stakeholder groups who cooperate in making decisions together, but who share a right of mutual veto allowing any stakeholder group to block a joint decision,if there is consensus to do so within that group. Note that governments, having a legal monopoly on the use of force, have a de facto right of veto in any multi-stakeholder policy development process anyway; the consociational structure simply formalises this and grants the other stakeholder groups the same right (Malcolm 2008, pp.293-294, 466-467, 478-482).

Does this mean that the power of veto would always be exercised, so that nothing would ever be agreed? Of course it is possible, as it is in any organisation that depends upon consensus. But by adopting deliberative democratic processes, the likelihood of a veto power being used strategically become less likely,especially as the social capital invested by the stakeholders into the IGF compounds. A stakeholder group will not use its power of veto lightly if this may damage its own longer-term prospects of using the IGF to advance its own interests later on.

(2) Whilst I support the decentralised model of Internet governance put forward in the Best Bits submission, there is a second problem that the submission does not acknowledge. The problem lies in the fact that it allows for the fact that there are a multiplicity of existing fora where global Internet public policies are developed, but also demands that all of them adopt (afresh, any many cases) a multi-stakeholder policy development process. Although absolutely sound in theory, there are two practical faults with this:

(2a) It is very optimistic, to say the least, to call for these existing multiple fora to be transformed into multi-stakeholder formats. In many cases – the ITU, for example – it will only take one member state to object to multi-stakeholder reforms in order to close down such transformation. Therefore in practical terms, this ideal is likely to be impossible to achieve.

(2b) Even if the previous objection were somehow overcome and there were a multiplicity of fully accessible and multi-stakeholder fora dealing with Internet policy issues, the task of addressing all these fora would massive exceed the scope of civil society’s limited financial and human capacity. As the least well resourced stakeholder group, this would put civil society at a further disadvantage when compared to governments and the private sector.

To correct for these faults requires a slight gloss upon the recommendations in the Best Bits submission, that I outline here. Essentially, it involves allowing for a broader role for the Internet Governance Forum, at least in the short term, whereby the IGF – as the most accessible, central policy forum for Internet policy discussion – would be tasked with developing recommendations for other institutions in the Internet governance ecosystem, if those institutions do not themselves yet have a similar multi-stakeholder process.

This conforms with the IGF’s existing mandate to “promote and assess, on an ongoing basis, the embodiment of WSIS principles in Internet governance processes” (Tunis Agenda paragraph 72(i)), and to “identify emerging issues, bring them to the attention of the relevant bodies and the general public, and, where appropriate, make recommendations” (paragraph 72(g). It is also consistent with its mandate to be non-duplicative (paragraphs 72(b) and 77), since a multi-stakeholder recommendation to an external institution is not to be considered duplicative of that institution’s less inclusive decision-making process, even if they both concern the same subject matter (Malcolm 2008, p.439).

Thus notwithstanding that there may be a plethora of other Internet governance bodies, the development of high-level principlesto guide the work of bodies that are not multi-stakeholder in their own right remains an appropriate activity for a reformed IGF.

Whilst the Best Bits submission does contend that “it is a mistake to think that one body, and one set of experts, could possibly be responsible for effective policy making on all Internet-related matters”, this is not what I propose. The IGF would continue to have a very broad mandate, yes. But so too does the General Assembly of the United Nations, and we do not hear complaints that it lacks expertise to cover all subject areas. This is because the principal contribution of the General Assembly is not its expertise, but its legitimacy – and, so it is with the IGF, only extending to other stakeholder groups besides governments.

Naturally the operational decision making that takes place by reference to the high-level principles that the IGF develops will remain the responsibility of more specialised bodies, just as the implementation of resolutions of the General Assembly falls to national parliaments and specialised agencies, in accordance with the principle of subsidiarity.

In summary and conclusion, I fully accept the merits of a distributed model of Internet governance as set out in the Best Bits submission. But I also caution as to two limitations that the submission does not cover.

The first is that it does not address the fundamental and irreconcilable disagreement over whether governments should or should not have greater policy authority than other stakeholders.I suggest a way of overcoming that disagreement by firstly engaging in a more thorough analysis of the claims of each stakeholder stakeholder group to participate in policy development in a given issue area. For issue areas in which governments are found to have a role, I suggest that the concerns of governments that their sovereignty is threatened could be addressed by structuring the policy development process according to a consociational model, that would give governments (and other stakeholder groups) a right of mutual veto over recommendations that they cannot support.

My second caution is that trying to retro-fit multi-stakeholder democratic legitimacy to a multitude of existing bodies, in the short term, is likely to prove a quixotic endeavour. It is also unnecessary when there is already a multi-stakeholder Internet governance institution – the IGF – thatpossesses a UN mandate to make recommendations across a broad range of Internet public policy issues, already has the trust and support of all non-governmental stakeholders, and is much more accessible to civil society than that plethora of other bodies could ever be.

In this way the IGF, with its proposed coordinating function, can act as a proxy to receive multi-stakeholder input into the development of high-level principles for the guidance of other, less multi-stakeholder institutions. At the same time, the IGF can host multi-stakeholder working groups to develop policy on issues for which there is no existing body to deal with them. In both cases, significant reforms to the IGF would be required – but these are much more achievable than the similar reforms required to transform the entire ecosystem of Internet governance institutions.


Malcolm, Jeremy. Multi-stakeholder Governance and the Internet Governance Forum. Terminus Press, 2008, available at

Martin Networth

Internet Researcher at
I am in constant support of internet privacy, and actively campaign against government intervention.

Latest posts by Martin Networth (see all)