I’m in Geneva, for the ITU’s workshop on Internet Governance. It’s a post-WSIS affair. Here’s the summary/background for my 15 minute presentation on Friday. It’s based on a working paper I’ve co-written with a wonderful duo, David Johnson and Susan Crawford, called “The Accountable Net” (inspired by the last Aspen Institute Internet Policy Project meeting and the work of Esther Dyson, who wrote about it here, and others). The full text of my submission, including the current working draft of the Accountable Net paper, is here.
* * *
This submission introduces the notion of peer production of governance, to which sovereigns ought to defer in instances where it can work. Only as a second option should we turn to a more centralized mode of governance, in which event meaningful public participation and the enfranchisement of individuals in developing countries ought to be first-order priorities.
We have been called, by the World Summit on the Information Society (WSIS) Action Plan, “to investigate and make proposals for action, as appropriate, on the governance of Internet by 2005.” This submission addresses some of the public policy questions raised by this hard and, to date, intractable question.
* Scope: Determine when (not if) Net governance is needed.
We have moved past the debate of the late 1990s about whether the Net can or should be governed. We acknowledge at the outset that traditional sovereigns can and should play an important role in regulating many actions and actors that affect the internet. There are collective action problems that arise on the global, unitary, public network of networks that we call the Internet that require some form of governance to resolve. Some problems that plainly call for some form of governance include, but are not necessarily limited to, spam, identity theft, network security, and certain aspects of technical coordination of the network. Our job, a hard one, to be sure, is to determine what the most pressing of problems are, and not to insist upon global governance schemes where they are not needed. We should not default to a single global Internet governance scheme. This scoping question is critical to get right as a threshold matter. (See Don MacLean’s submission for this conference for a helpful series of starting points).
* Framework: Use the most wise, fair and effective form of governance for any given issue.
For those problems on the Net where some form of governance is required, we ought to test a series of possible schemes of governance to determine which scheme is the most wise, fair and effective. We are unlikely to conclude that a single mode of governance is the most wise, fair and effective means of resolving every difficult problem on the Net. (Among others, Karl Auerbach has directed our attention to this issue in the second of his submissions to this conference, in which he highlights the need for “Tailoring the Mode of Governance To The Matter Needing To Be Governed.”) We ought to be open to disaggregation of Net governance: the notion that certain problems will call for different forms of governance. We ought also to consider ways in which peers, or groups of peers, may be able to produce effective forms of governance for some problems that crop up on the Net.
* Principle: Individual choice, participation, and diversity are paramount.
We should start with the premise that the individuals ought to have a reliable means of participation in any scheme of governance of the Internet that we develop or that otherwise emerges. If a governance scheme can drive choice to the individual level and if peers can produce their own system of governance, sovereigns ought to defer to this peer production of governance. Where peer production of governance is not the most wise, fair and effective mode of governance, any sovereign empowered to make and enforce rules must be accountable to those who defer to and grant that sovereign its authority.
* Principle: Any governance scheme must enfranchise individuals in developing countries.
Our consideration of Internet governance schemes must build in from the start, not as a tack-on at the end of a long process, a conscious effort to give adequate voice (and, only if we’ve chosen a representative system, then votes) to individuals in the developing world. We should learn from the abject failures of many of our early efforts to enfranchise developing countries in the Internet decision-making context. The empowerment of individuals in developing countries in a manner that corresponds to the empowerment of individuals in more developed countries should be a core tenet of any Net governance scheme.
* Conclusion: Sovereigns should defer to peer production of governance where it can work.
In the context of an accountable internet, in which individuals connect only with those who have shown they are worthy of trust, the peer production of governance can address some of the toughest internet governance problems. This accountable internet is emerging through authentication and accreditation technologies and other phenomena of life on the Net. We should consciously build into the accountable internet the safeguards necessary to preserve the values that we hold dear.
* Conclusion: Where a centralized group must be given the right to tell others what to do, meaningful public participation, in the form of voice or, where needed, votes, and recourse against the decision-maker must be assured.
The accountable net means not only that peers are accountable to one another, but that any sovereign is accountable to those by whom it been granted power to make and enforce rules. Despite significant experimentation in this regard, no effective means of ensuring such accountability, and, correspondingly, legitimacy in the sovereign entity, has yet emerged in the Internet governance space.
John Palfrey 