I have just uploaded the first complete draft of my PhD thesis, titled Multi-Stakeholder Public Policy Governance and the IGF. From that page you can read it online, download a copy in PDF or various other formats, or even annotate or suggest edits to the thesis (click "About" on that page to read how).

Of course, it will be undergoing considerable revision before it is submitted (including the approximate halving of its length!), but I thought I would put together a summary of the thesis in its present form, in a few paragraphs, for those who are disinclined to read all 215 000 words of the current draft. So here goes:

If the IGF did not exist, it would be necessary to invent it. Governance of the Internet (either by rules, or by more decentralised mechanisms such as norms, markets or architecture) inherently engages issues of public policy with transnational scope. None of the above mechanisms alone are able to deal with such issues with sufficient legitimacy (for example, governments have no legitimate authority over transnational civil society), or effectiveness (for example, norms are insufficient to combat cybercrime). Instead, a hybrid mechanism, governance by network, is required; the hallmark of which is that it brings all the other mechanisms together and draws upon the composite legitimacy of all the stakeholders who employ them.

That Internet governance should be the province of a network of all stakeholders was recognised at WSIS (albeit that formal policy authority was reserved to states), and the IGF was intended as the fulfillment of that principle. But more is required to ensure the legitimacy of such a network than simply the involvement of all stakeholders. It must also be democratic, in the sense that it operates with the consent of those it governs. This cannot mean, in the context of the Internet, that members of the network are elected. Instead, the network can be democratised by designing it to bring in all affected viewpoints and subject them to reasoned deliberation. The process by which Internet standards are developed by "rough consensus' is a close analogy.

The IGF cannot yet claim to be such a network. In fact, it does not even accept that its mandate requires it to develop the capacity to make policy recommendations. This is because those in power in the old Internet governance regime, such as governments and the technical community, fear the inclusion of other stakeholders in governance. To address this, and also to legitimise the IGF's structure which is currently beholden to the intergovernmental authority of the UN, a new multi-stakeholder bureau of the IGF is required, which would make formal decisions only by the consensus of all stakeholder groups. It in turn would draw from the ultimate authority of a reformed plenary body which would be empowered to deliberate democratically, both in person and online.