4.1.3. Criticisms

On the other hand, there are also other factors to be considered in assessing the appropriateness of anarchistic organisation for a governance network. At root, these factors are the network’s legitimacy and its effectiveness.[1] It has already been determined that a network involving all stakeholders is the most legitimate mechanism of public policy governance for the Internet, and there is no reason to reopen that question here.[2] However, where that network is organised along anarchistic lines, its involvement of all stakeholders to the requisite degree is likely to be more difficult to demonstrate.

The two main indicia of an organisation’s legitimacy are its accountability and transparency. These are important to whichever organisational structure is adopted by a governance network, and hence although they will be considered here to illustrate the main shortcomings of anarchistic ordering for a governance network, they are to be discussed in more depth at Section 4.3.3.

Briefly then, considering accountability in respect of the anarchistic structure in particular, the fluidity with which the arrangements between stakeholders in such a structure may form and reform make it difficult to ensure that at any given point, those arrangements are adequately inclusive of all stakeholders. Governments may, for example (and do), enter into arrangements between themselves without consultation with the private sector and civil society, or vice versa. Even where all stakeholder groups are represented, their power within the network may be unequal, replicating similar inequalities in the larger international system.[3]

This is not important in other contexts—Wikipedia, for example, publishes an encyclopædia; it does not deliberate upon issues of public policy that might affect stakeholders other than its own users, and therefore is not required to be accountable to them. But for a public policy governance network, stakeholders’ uncertainty of the state of power relations in a governance network at any given time, and (by definition) their lack of control over the same, is a serious shortcoming of the anarchistic structure.

As for transparency, the London Action Plan, for example, has a “Members Only” section of its Web site that conceals a number of the details of its activities. Less formal governance networks of anarchistic structure are likely to be even less transparent.

As far as effectiveness is concerned, certainly, the efficiency of anarchistic ordering, and its cultural appropriateness (which increases the likely effectiveness of its internal governance by norms), both of which have already been noted, are positive indicators. On the other hand, detracting from the effectiveness of an anarchistically ordered governance network is its very voluntariness. Without denying that norms and architecture can effectively channel cooperative behaviour, there are also cases in which if stakeholders are given the choice to cooperate or to act strategically on their own, they will take the latter course.

In fact, rational choice theorists claim that strategic behaviour will be the rule rather than the exception in such cases. The prisoner’s dilemma famously and simply illustrates the problem. If two accused prisoners in adjoining cells could cooperate and minimise each other’s sentence by remaining silent, or could betray the other in order to go free themselves, it would always be in their rational self-interest to betray each other (or to “defect” in the language of game theory). This is so even though cooperation would have resulted in the least collective deprivation of liberty for them both.[4]

Whilst the anarchist’s response is that cooperation does begin to become a rational strategy over the long term (which it does), even in such cases it is rarely the only rational strategy.[5] Thus for liberals, the institution of the state is necessary to protect individual rights that might be infringed by the strategic behaviour of others that game theory predicts.[6] In other contexts, it may be necessary for some other hierarchical authority to take the state’s place in performing this function.

It may also be properly objected by anarchists that rational choice theory tends to leave aside the forces of norms and institutions which can result in cooperative behaviour emerging sooner than rational choice theory might predict;[7] however in the present context the point remains that there is little empirical evidence of stable governance networks that are truly inclusive of governments, the private sector, civil society and international organisations forming spontaneously.[8]

But for the supranational authoritative force of the United Nations, for example (admittedly exercised through a soft law process), it is doubtful that the IGF would have come into being. There are many more examples (and the IGF may become one of them) in which governments have strategically gained the dominant position in what should have been a multi-stakeholder network—in the ITU, for example—and other cases in which non-governmental stakeholders have done so, as in the case of the ill-fated IAHC and its gTLD-MoU.

What is perhaps needed to overcome the problem of short-term strategic behaviour, and the other problems of anarchistic ordering that have been noted, is a hybrid which preserves the efficiency and cultural fit of decentralised collective action, with the greater (or at least more certain) accountability, transparency and effectiveness made possible by more structured organisational forms that utilise governance by rules. One such hybrid is called co-regulation. But it is hardly anarchistic; which takes us to the second of the four organisational structures to be considered in this chapter.



See Section 3.4.1.


See Section 2.4.


See Section


Poundstone, William, Prisoner’s Dilemma (1992), 120


Rutten, Andrew, Can Anarchy Save Us From Leviathan? (1999), 587–591; Hardin, Russell, Political Obligation (1989) , 115


Sugden, Robert, Maximizing Social Welfare: Is it the Government’s Business? (1989), 83


Green, Donald & Shapiro, Ian, Pathologies of Rational Choice Theory: A Critique of Applications in Political Science (1994), 121–123


Vedel, Thierry, Four Models for Internet Governance (2005), 65