Beyond the democratic state: anti-authoritarian interventions in democratic theory

took her daughter
Ashora platoons now moving from valiasr toward National Tv staion. mousavi's 
supporters are already there. my father is out there!
Iranian activists could communicate information about police movements throughout Tehran 
immediately, enabling others to quickly respond and change plans – knowing where exactly to 
be in order to avoid, or engage, the police. As the use of Twitter in Iran demonstrates, the 

---

177 
decentralized nature of this type of communication is precisely what makes them such useful 
tools for coordination.
Drawing on Chisholm (1989, 186-87) and the above example, the main advantages of 
decentralized coordinative mechanisms are: 1) they increase the accuracy of information insofar 
as information can be produced and processed by a wider range of actors, each with their own 
social position; 2) they tend to be much quicker; 3) they are better able to respond to change and 
unanticipated problems, whereas formal mechanisms tend to take time to alter after unpredicted 
issues arise; and 4) informal systems of coordination tend to promote the values of 
“differentiation, representation and reliability…more effectively than a formal centralized 
system.”
Networked Organization
With this backdrop in mind – that is, with a basic theory for the possibility of 
“coordination without hierarchy” – we are now in a position to think about the possibility for 
networked organization to function as a kind of self-governance. It is important to note, that the 
concept of networks can be used to describe a range or modes of coordination and organization.
Hardt and Negri (2004, 56-57) outline three distinct types of networks (in a military context): the 
hub network, the polycentric network and the distributed network:
The traditional military structure can be described as a hub…network in which all 
lines of communication and command radiate from a central point along fixed 
lines. The guerrilla structure suggests a polycentric network, with numerous, 
relatively autonomous centered clusters, like solar systems, in which each hub 
commands its peripheral nodes and communicates with other hubs. The final 
model in the series is the distributed, or full-matrix, network in which there is no 
center and all nodes can communicate directly with all others” (Hardt and Negri, 
56-57). 

---

178 
Thus, it is not simply that all networks are desirable from a democratic or anti-authoritarian 
perspective.
38
Rather, it is this last network model, the “distributed network,” that overlaps with 
the above discussion of non-hierarchical coordination and that I use for theorizing a networked 
form of self-governance. However, even distributed networks are not 
necessarily
democratic.
However, much like direct action, they have the potential, when mobilized in certain ways, to be 
radically
democratic. Or, as Hardt and Negri (2004, 93) put it: “We have to look not only at the 
form but also the content of what they do. The fact that a movement is organized as network or 
swarm does not guarantee that it is…democratic.” One can think about direct action as the 
“content” and network as “the form” of many contemporary social movements. In the previous 
chapter, I outlined criteria to help distinguish democratic from non- (or anti-) democratic action.
Hardt and Negri (
ibid
., 54-55) focus on two core features of a distributed network.
One essential characteristic of the distributed network form is that is has no 
center. Its power cannot be understood accurately as flowing from a central 
source or even as polycentric, but rather as distributed variably, unevenly, and 
indefinitely. The other essential characteristic of the distributed network form is 
that the network constantly undermines the stable boundaries between inside and 
outside. This is not to say that a network is always present everywhere; it means 
rather than its presence and absence tend to be indeterminate…Networks are in 
this sense essentially elusive, ephemeral, perpetually in flight. Networks can thus 
at one moment appear to universal and at another vanish into thin air. 
In contrast to the hierarchical model of organization – characteristic of both the hub and, to a 
lesser extent, the polycentric network, a distributed network: 1) has many nodes but no center 
and most, if not all, of those nodes can communicate with each other; and 2) has unstable 
boundaries such that its “members” – who is, and who is not a part of the network – are always 
38
The Foucauldian conception of power as something that circulates, for example, could be well be seen as kind of 
network that is to be resisted by democrats and anti-authoritarians. Indeed, Hardt and Negri (2004, xii), working in 
a Foucauldian framework, raise concerns that “‘network power,’ a new form of sovereignty, is now emerging, and it 
includes as its primary elements, or nodes, the dominant nation-states along with supranational institutions, major 
capitalist corporations, and other powers.” Thus, even though the sovereignty of the nation-state and the classic 
model of imperialism have fallen somewhat by the wayside, this is being replaced by a newer and more diffuse 
network that they identify as “empire.” 

---

179 
shifting. “[A] distributed network such as the Internet is a good initial image or 
model…because, first, the various nodes remains different but are all connected in the Web, and, 
second, the external boundaries of the network are open such that new nodes and new 
relationships can always be added” (
ibid
. xv). 
That said, even these networks should not be considered perfectly non-hierarchical or 
egalitarian. Research into a diverse array of networks – from the internet, to citation patterns to 
food webs in ecosystems – has shown that networks are built around their own internal 
hierarchies, in which some parts are more important than others. 
[I]n most real networks the majority of nodes have only a 
few
links and…these 
numerous tiny nodes coexist with a few big hubs, nodes with an anomalously high 
number of links. The few links connecting the smaller nodes to each other are not 
sufficient to ensure that the network is fully connected. This function is secured 
by the relatively rare hubs that keep real networks from falling apart…We see a 
continuous hierarchy of nodes, spanning from the rare hubs to the numerous tiny 
nodes (Barabasi 2003, 70). 
Thus, while networks are decentralized and encompass many different nodes, not all of those 
nodes are as important, influential or powerful as others. For example, on the internet, websites 
that already have many existing links attract newcomers to link to them, as well. “Nodes that 
have been around for awhile…have distinct advantages over newcomers” (Dean 2009, 30).
Therefore, it is important not to idealize networks as being perfectly egalitarian or spontaneous.
Networks have both hierarchies and histories, and both of these (at least potentially) reflect 
unequal distributions of power. Just as internet networks have hierarchies, networks of 
organizations and activists have less important nodes and more important hubs. The hubs are 
those individuals or groups that have better connections and greater influence. Agents could 
hold those privileged positions simply because they are older and more established, or because 
they have greater access to resources, or because they have done the hard work of establishing 

---

180 
relationships with many others. Whatever the reason, an important point is simply that such 
privileged positions exist in networks.
This does not mean that networks should be scrapped as a potential mode of coordination 
and self-governance. Though they too suffer from problems of hierarchy, the contention 
underlying support for networked organization is that the level of hierarchy present in networks 
is frequently much less severe, and much more flexible, than in centralized models of 
organization. Critically, in a distributed network no single node or hub dominates the network 
insofar as no single node can issue commands that others must follow. Moreover, 
communication is possible between many (though perhaps not all) nodes in the network, 
enabling multiple nodes to be sources of innovation and action. Beyond that, networks have 
proven to be successful at coordinating social movements and mobilizations, even as mainstream 
observers have struggled to understand how these movements work – grasping for a way to 
describe the so-called “leaderless” nature of these movements, groping to make sense of a model 
of coordination that looks so different from the hierarchical models that dominate our politics 
and economics. For instance, in describing the successful shutdown of the WTO in Seattle in 
1999, activist and author Starhawk (2002, 404) speculates: “My suspicion is that our model of 
organization and decision making was so foreign to their [the police and city officials’] picture of 
what constitutes leadership that they literally could not see what was going on in front of them.”
It is this model of action and decision-making that merits further development and consideration 
within democratic theory. In the following section, I illustrate how social movement networks 
can function as a form of decentralized coordination on both global and local scales. 

---

181