Skip to main content.

Red background with group of raised fists in the center, one holding a smart phone. Central image surrounded by four round icons containing the numbers 99%, the Anonymous logo, the Wikileaks logo, and the SOPA logo.

Networked Activism, Hybrid Structures, and Networked Power

John Jones

Contemporary activism is increasingly dominated by networks and networked forms of organization. This is not a completely new phenomenon. As both Manuel Castells (Rise xviii) and Armand Mattelart (7-8) note, networks have existed as a form of organizing human action for as long as humans have lived together in societies. Consequently, the use of network structures for activism predates the current form of the Internet. For example, in his influential 1991 book, The Temporary Autonomous Zone, Ontological Anarchy, Poetic Terrorism, Hakim Bey suggests activists can evade the influence of state power by enacting “Temporary Autonomous Zones” (TAZs), spaces independent of state control that allow for free expression unconstrained by state power. Bey argued a TAZ could be a physical location, such as a commune or retreat, yet he acknowledged it could also be the product of communication technologies that allow for temporary, localized protest that is able to disperse and reassemble in different spaces. With the advent of “digital networking technologies,” which have enabled network structures to “overcom[e] the traditional limitations of networking” (Castells Rise xviii)—specifically through the ability of these technologies to manage the complexity of large network structures—such networked activism has become increasingly common.

However, the rise of networked activism has not eliminated previous forms of power and resistance. As networked activism becomes increasingly ubiquitous, it has been accompanied by questions relating to how this activism functions relative to other means of political resistance and the ultimate efficacy of networked tactics. One such question deals with the relationship between networks and hierarchies. Many analyses of networked power either take the form of assertions that networks pervade all levels of society and are the primary mechanism of social power (Castells Rise) or, in contrast, argue that networks, while increasingly pervasive, tend to fail when compared to hierarchies (cf. Gladwell; Eilstrup-Sangiovanni and Jones). Such positions frequently fail to address the ways in which networks and hierarchies influence each other. In this discussion, by networks I refer to organizational structures that are characterized by multiple connections, both vertical and horizontal, between individual nodes lacking strong, top-down control. Hierarchies, in contrast, are organizational structures characterized by primarily vertical connections between individuals and clear, top-down governance. In this article, I wish to explore the interrelatedness of these structures, which I call hybrid structures, or structures that depend on a combination of networked and hierarchical forms of organization. To do so, I examine a case study of networked activism, a series of leaks of U.S. government materials in 2010 by the whistleblower organization Wikileaks.

Assange on Network Activism

In an essay and related blog post, Wikileaks founder Julian Assange (Assange “Conspiracy”; Assange “Sun 31 Dec”) outlines the theory of networked activism that animates the site, arguing an effective means of countering, and ultimately defeating, “bad governance” (“Conspiracy” 1) is to simply expose the existence of secret government communication networks. While Assange does not mention any particular government or even Wikileaks as a means for implementing this strategy for activism, he posted these writings soon after the launch of the leak-exposing site (Benkler 315), and Wikileaks’s subsequent actions appear to follow the strategies they outline.

Assange defines networks as “conspiracies” (“Conspiracy” 2), and as Aaron Bady puts it, these conspiracies aren’t necessarily cloak-and-dagger affairs, but rather are “fairly banal” attempts at secrecy characterized by “any network of associates who act in concert by hiding their concerted association from outsiders” (n.p.). According to Assange, conspiracies hide their associations because they know that if their actions were public, those actions, and ultimately the entire conspiratorial network supporting those actions, would encounter resistance. Because the potential for such resistance depends on exposure, an effective response to conspiratorial networks, Assange writes, is to reveal their workings—the connections within the network and its communications. Once their secrets are exposed, conspirators will become fearful of communicating with other nodes in the network, crippling the network’s ability to function. As Assange puts it: 

We can deceive or blind a conspiracy by distorting or restricting information available to it. We an reduce total conspiratorial power via unstructured attacks on links or through throttling and separating. A conspiracy sufficiently engaged in this manner is no longer able to comprehend its environment and plan robust action. ("Conspiracy" 5)

As individual nodes and sub-networks limit communications with each other, the network becomes less efficient and, therefore, less effective. By “distorting or restricting” information flows between nodes and “throttling” communication and “separating” the connections between conspirators in the network, activists can challenge these conspiracies and their contributions to “bad governance.” As Bady explains, “the more opaque [the conspiracy] becomes to itself (as a defense against the outside gaze), the less able it will be to ‘think’ as a system, to communicate with itself” (n.p.). In the context of network theory, Assange’s arguments can be seen as a means of limiting the advantages digital communication offers to networks, rendering them inefficient and unwieldy.

However, there are reasons to question the effectiveness of this strategy. While Assange’s theory takes into account the growing ubiquity of networks enabled by digital technologies and explicitly attempts to disrupt them, his theory relies on the assumption that government power is inherently—perhaps exclusively—networked. After all, rendering conspiratorial networks inert would be ineffective if true power in governance is enacted via hierarchical structures rather than networked ones. Yet, while governments clearly make use of networked structures, those networks remain closely bound to hierarchical forms of power. Indeed, a key feature of the U.S. response to Wikileaks in 2010 was the use of both hierarchical and networked tools to attack the organization. Further, it is not clear whether Assange’s theory of challenging networks is enough to deter the conspiratorial networks he describes. In a network analysis of Assange’s tactics, Kay Hamacher argues these tactics do not take into account redundancies in conspiratorial networks, particularly the ability of networks to cooperate with each other via “mutual support” to resist attacks (4). While Hamacher’s model is simply that—a mathematical model of network behavior—other studies of Wikileaks seem to support her claims. Yochai Benkler’s case study of Wikileaks demonstrates how, in the aftermath of the organization’s 2010 release of U.S. diplomatic cables, this mutual support—via open, rather than conspiratorial, channels—was used to attack the organization (330). Similarly, in his analysis of the aftermath of Wikileaks’s release of “Collateral Murder” (“Collateral Murder”), a video depicting the death of a Reuters reporter by U.S. forces, Marouf A. Hasian describes the means by which multiple individuals and groups collaborated via open channels to “domesticate,” or justify, the use of deadly force depicted in the video, despite the exposure of outright fabrications on the part of the military and U.S. government (Hasian 191). While Wikileaks’s goal in each of these releases was to expose conspiratorial networks, the result of its actions was not the failure of those networks due to restrictions brought on by that exposure. Instead, they demonstrated their resiliency by altering their communication channels to counter the attack. The exposure of the conspiratorial forces in the network mainly served to reveal the wider network of support that enabled the conspirators’ actions.

These analyses suggest Assange’s theory of networked activism is, at best, incomplete. Simply exposing a conspiratorial network does not render that network inert. Rather, networks are able to draw on support both from hierarchical structures and outside networks that help them achieve their goals through reorganization. What, then, is the relationship between networks and hierarchies in governance and activism? And how can we understand Assange’s theory of networked activism in the context of both networks and hierarchies? To address the first question, I will briefly review two critiques of the efficacy of networked activism. To address the second, I will provide a short case study of Wikileaks’s 2010 release of U.S. government materials and the response to those leaks, showing how that response and the actions of Wikileaks, and its allies, suggests a more complex picture of how networked activism should be seen. Wikileaks’s actions—and the subsequent attacks on the organization by governments and a loose network of government supporters—illustrate the complex connections between networks and non-networked organizational structures in both resistance to state power and official challenges to that resistance. While Assange’s theory of networked activism may not be sufficient to overthrow conspiratorial power, I argue it is effective at challenging hybrid power structures by isolating and attacking the networks embedded in them. That is, rather than providing a means for “throttling” network conspiracies, Assange’s strategies represent a means for identifying conspiratorial networks and separating them from their hierarchical support, a move that is necessary to make those networks vulnerable to other forms of resistance.

Networks Versus Hierarchies

In their analysis of the threat of networked terror organizations, or “illicit networks,” Mette Eilstrup-Sangiovanni and Calvert Jones counter what they identify as the accepted wisdom in international relations, that “it takes a network to fight a network” (8), arguing instead that networks exhibit some inherent disadvantages when compared to hierarchies. Whereas hierarchies, with their strong central organization and rule-based interactions, are able coordinate and communicate with general effectiveness, networks can lack the means for doing so. The authors identify six potential problems with illicit networks: “information limitations and communication failure” (19), “poor decision-making and excessive risk-taking” (21), “restrictions on scope and structural adaptability” (22), “collective-action problems due to coordination” (27), “security breaches” (29), and “learning disabilities” (31). These problems ultimately stem from a lack of communication; the authors note the illicit networks they study tend not to be “all-channel network[s] where everybody is connected to everybody else,” but rather rely on autonomous units communicating through a series of relays (13). This particular structure leads to choke points that prevent effective communication, limiting the organizational benefits of networks and making effective collaboration difficult or impossible.

Relying in part on Eilstrup-Sangiovanni and Jones’s study, in his 2010 New Yorker essay entitled “Small Change,” Malcolm Gladwell argues networked activism is fundamentally disadvantaged when combating hierarchical power structures and, as a result, networks are ultimately ineffective against hierarchical powers like the state. He writes that in order for activists to challenge such “a powerful and organized establishment” they must rely on the strategic and organizational benefits that come with hierarchical structures. As he puts it: 

Because networks don’t have a centralized leadership structure and clear lines of authority, they have real difficulty reaching consensus and setting goals. They can’t think strategically; they are chronically prone to conflict and error. How do you make difficult choices about tactics or strategy or philosophical direction when everyone has an equal say? (Gladwell)

According to Gladwell, drawbacks like weak leadership and inability to set goals mean networked activism tends toward low-risk activities, such as liking a cause on Facebook, instead of high-risk activities, like placing oneself in bodily danger at a protest. Interestingly, his opening example of high-risk activism, activism that was ultimately successful at defeating “a powerful and organized establishment”—the 1960 lunch counter sit-ins in North Carolina—were not exclusively the result of hierarchical decision making, but had characteristics of viral, networked activities. As he notes, these sit-ins spread virally across the south, and network analysis of the movement has shown individuals who participated in sit-ins were more likely to do so if they had “close friends” within the movement. Certainly the civil rights movement in 1960 was not networked in the way Wikileaks is, and Gladwell provides clear examples of the hierarchical discipline imposed on the group. Despite these caveats, however, individual participation at the grassroots level of this movement was fundamentally a networked phenomenon, not characterized by the “impersonal, rule-guided relations” of hierarchies but the “direct personal contacts” of networks (Eilstrup-Sangiovanni and Jones 12). A small group made a decision, their actions were copied by other, loosely connected small groups without direct central planning, and these actions, in aggregate, constituted a movement with a hierarchical leadership. That these actions were subsumed under the broader hierarchy of the movement does not undermine their character as a viral and essentially networked activity.

The lunch counter sit-ins of the 1960s Gladwell describes suggest the complicated relationship between hierarchies and networks in both traditional power structures and activism. Gladwell’s analysis (with Eilstrup-Sangiovanni and Jones’s helping hand) sheds light on the nature of Assange’s networked activism strategies, as well as the apparent disconnect between his description of this strategy and its ultimate success. Eilstrup-Sangiovanni and Jones provide some support for Assange’s claims, showing how limiting communication within a network can make that network inefficient and its actions increasingly ineffective, while Gladwell’s analysis inadvertently provides evidence that even supposedly hierarchical movements contain networked elements. As already noted, Hamacher argues that as a description of network power, Assange’s strategy isn’t effective. But if networks are entangled with hierarchies—as Eilstrup-Sangiovanni and Jones put it, if “the boundaries between networks and hierarchies are not always clear-cut”—then many networked organizations might struggle to challenge these hybrid entities (12). Pitting networks against hybrid structures allows us to examine Assange’s strategy in a different light. Namely, as a means of disconnecting a network from the hierarchy that supports it, thus enabling challenges to that network via other sources of networked power. Where Assange’s original analysis seems to be wrong is in its assumption that this process is enough to cripple the network in question. It is not. Rather, it simply enables additional forms of networked power to be brought to bear on such networks. Such a move does not immediately disable the network, for it can respond using other forms of networked power. In the following case study of Wikileaks’s release of U.S. diplomatic cables in 2010, I will explore how Assange’s exposure strategy helped to identify a conspiratorial network, as well as how the group used other forms of networked power to challenge this network.

Wikileaks and Networked Activism

Castells has argued the two primary forms of networked power are programming and switching, where “programming” represents the ability to set the goals for a network and “switching” represents the ability to make connections between networks to achieve those goals (Castells Communication Power 45). The networked activism of Bey and Assange can be seen as a means of attempting to disrupt the programs of networks, or to re-program them “[t]o alter the outcomes of the network” as Castells puts it (Communication Power 20). Here I wish to focus on switching, arguing that the way in which Wikileaks released materials in 2010, the subsequent attacks on the organization, and its response all suggest applications of switching, or utilizing connections between networks to increase the overall power and effectiveness of a network and achieve its goals.

In 2010, Wikileaks released a series of formerly secret U.S. government materials, including the “Collateral Murder” video (“Collateral Murder”), the “Afghan War Logs” (“The War Logs”), and the State Department cables. However, their tactics did not simply follow the strategy for destroying networked conspiracies outlined in Assange’s early essays. Rather, they made use of traditional media networks to garner attention for the releases. Instead of simply releasing these materials in bulk on its site, the organization provided select members of the news media—the New York Times and the Guardian for English speakers, Der Spiegel, Le Monde, and El País for German, French, and Spanish speakers, respectively—with exclusive access to the documents prior to their general release, using the authority of these publications to market the information and ensure wide dissemination. As Assange explains, “it’s about supply and demand. Zero supply equals high demand, it has value. As soon as we release the material [en masse on the Web], the supply goes to infinity, so the perceived value goes to zero” (Nystedt). What Wikileaks had discovered in previous releases was that providing materials directly to the Web did not ensure their coverage in the media, and as a result the group altered their tactics to increase media demand for their leaks. This change in tactics appeared to be effective, as the cables were widely reported in the news media (“United States Diplomatic Cables Leak”). In her analysis of Assange’s theory of networked activism, Hamacher found the most effective means of disrupting a networked conspiracy was through the combination of many networks “compet[ing] . . . for the extinction of” that network (4). Wikileaks’s change in tactics represents a realization on the part of the group that it needed to involve multiple networks in its activities, and, as Hamacher argues, this tactic had a better chance of ultimately defeating the conspiratorial networks it targeted.

The acquisition of the materials by Wikileaks allowed for the exposure of this conspiratorial network. However, as Assange points out (Nystedt), simple exposure on the Wikileaks site did not cripple these networks, as simply making materials available was not enough to bring them to the attention of the governed, thereby causing the reaction that would hinder communication in these networks. To acquire this attention, Wikileaks had to leverage other media networks through switching, in this case by providing exclusive access that would allow these networks to publish the materials as breaking news. This, in turn, created the potential for the situation outlined by Hamacher—many networks working together to challenge the target network. Yet, while this strategy brought great attention to these materials, it ultimately did not lead to the extermination of the networks involved. As then Secretary of Defense Robert M. Gates stated at the time, while the release was “embarrassing” and “awkward,” its “[c]onsequences for U.S. foreign policy” would be, in his opinion, “fairly modest” (“DOD News Briefing”). This result was partly due to the resilience of the networks in question, and partly due to the ultimate disinterest in Wikileaks’s media partners in the “extinction” (Hamacher 4) of those networks. As Jay Rosen put it, Wikileaks, as a “stateless news organization” concerned itself with “the release of information without regard for national interest” (Rosen). However, its media partners, particularly the New York Times, were deeply invested in U.S. government interests. As Glenn Greenwald bluntly states, “there are few countries in the world with citizenries and especially media outlets more devoted to serving, protecting and venerating government authorities than the U.S.” (n.p.). More diplomatically, Rosen notes the Times was placed—or placed itself—in the position of mediating between Wikileaks and the U.S. government, rather than acting as an adversarial network intent on destroying the networked conspiracy represented by the released materials.

While the cable release’s exposure of conspiratorial networks made them vulnerable to forms of networked power like switching, Wikileaks was not the only group that engaged in switching practices. After the release of the cables, Wikileaks faced a series of attacks, both from governments and private organizations. While the group was challenged with the threat of legal penalties, these threats were difficult to prosecute because of the statelessness of the organization and the lack of clear legal jurisdiction. Consequently, the most effective and prominent attacks came after U.S. Senator Joe Lieberman made an unofficial request for the entities doing business with Wikileaks to end their relationships with the group. In response, the private companies that hosted Wikileaks’s webservers, as well as the banks and credit card processors that handled its finances, abruptly cut off all service (Benkler 313-14). While Benkler notes any official demand for these actions to occur was unlikely to have been achieved for the same reason other legal challenges had failed—”the high barriers erected by First Amendment doctrine in cases of prior restraint” (Benkler 340), barriers that constitute the hierarchical control of U.S. law—this action nevertheless came about through the individual decisions of private companies responding to Lieberman’s request (Benkler 340). Similarly, Assange, as the public face of the organization, was subject to a series of criminal proceedings that, Benkler argues, heavily suggest they were strictly “part of a campaign to undermine Wikileaks” (Benkler 146). In short, the response to the networked activism of Wikileaks and its founder—not to mention the alleged leakers of documents to the organization (Benkler 345)—suggests an unofficial, networked response unhindered by the exposure of any conspiratorial networks, and, as Hamacher’s analysis predicts, driven by supporting networks, in this case those of private companies.

This networked response, while causing Wikileaks to go offline for a few days, did not destroy the group. Like the government, Wikileaks and its allies responded in networked fashion. This response consisted of a combination of legal, discursive, and technological tactics, from shifting its operations to countries with strong protections for whistleblowers, to making the case for Wikileaks’s status as a journalistic organization, to hosting redundant copies of the site to prevent its complete removal. And just as the U.S. government was supported by its allies in their attacks on Wikileaks, supporters of Assange’s group launched their own networked-based attacks in the form of distributed denial of service attacks on organizations who cooperated with the government’s efforts and the quick formation of a number of copycat leaks sites that replicated the same activist model (Benkler 349). As Benkler concludes, this networked skirmish between Wikileaks and its attackers ended in a stalemate, insofar as the goal of the government’s response to was to “tame the beast of distributed online systems providing avenues for leaking documents outside of the traditional responsible media system” (Benkler 350).

What, then, can this stalemate tell us about networked activism? Benkler’s analysis of the distributed response to Wikileaks by the U.S., other world governments, and private corporations illustrates the extent to which networks abet hierarchies, allowing for new forms of government control and coercion. Benkler underscores the following point. While hierarchical in many respects, Wikileaks’s targets were capable of responding in networked fashion to counter the networked threat created by the group. Where hierarchical response—as represented by the U.S. legal system and the checks and balances that weigh “disclosure’s beneficial effects against its harmful ones” (Fenster 753)—was limited, networked responses filled in the gaps. Benkler describes the situation this way:

The integrated, cross-system attack on Wikileaks, led by the U.S. government with support from other governments, private companies, and online vigilantes, provides an unusually crisp window into the multi-system structure of freedom and constraint in the networked environment and helps us to map the emerging networked fourth estate. The attack’s failure provides us with insight into how freedom of action is preserved primarily by bobbing and weaving between systems to avoid the constraints of those subsystems under attack and harness the affordances of those that are out of reach of the attacker. (330)

In other words, Wikileaks’s status as a networked entity, rather than an easily recognizable, traditional journalistic organization, prompted the activation of the similarly networked “multi-system structure” that allowed for the group to be punished in extra-legal ways. While Benkler focuses on the ways in which such a system limits press freedom, I argue it was the initial actions of Wikileaks in exposing the networked conspiracy that distanced that network from its hierarchical support systems, ultimately necessitating the ensuing networked response. The ultimate “failure” of this response—it was “dramatic, extensive, overwrought, and ineffective,” Benkler argues (351)—does not take away from its networked nature or, as Benkler notes, the “insight” it provides into networked activism and power. While Assange’s theory of networked activism cannot destroy conspiratorial networks on its own, it can identify and isolate those networks, thus making them vulnerable to other forms of networked attack. However, this strategy also frees these networks to defend themselves using the same tools of networked power. As a theory of networked activism, Assange’s strategy is only an opening gambit, one that must be accompanied by an attempt to address the networked nature of its targets by bringing forms of networked power like switching and programming to bear against it. The case of Wikileaks represents not simply the features of networked activism, but the networked capacity of established powers to counter that activism and target activists through what Benkler has called “extralegal partnership[s]” (312) with non-government entities, as well as entrenched hierarchies. Acknowledging the continuing hybrid nature of both state power and activism suggests a need to continue exploring how power and counterpower affect such hybrids.

Works Cited

“Collateral Murder”. Last Update n.d. Web. 1 Apr 2013.


“DOD News Briefing with Secretary Gates and Adm. Mullen from the Pentagon”.
      U.S. Department of Defense. Last Update 30 Nov 2010. Web. 28 Mar 2013.

“United States Diplomatic Cables Leak”. Wikipedia: The Free Encyclopedia. Last
      Update n.d. Web. 28 Mar 2013.

“The War Logs”. The Guardian, Last Update n.d. Web. 28 Mar
      2013. <>.

Assange, Julian. “Conspiracy as Governance”. Last Update 3 Dec 2006. Web. 28
      Jan 2013.

---. “Sun 31 Dec 2006 : The Non Linear Effects of Leaks on Unjust Systems of
      Governance”. 31 Dec 2006. Web. 30 Jan 2013.

Bady, Aaron. “Julian Assange and the Computer Conspiracy; ‘to Destroy This
      Invisible Government’“. Zunguzungu. 11 Nov 2010. Web. 28 Jan 2013.

Benkler, Yochai. “A Free Irresponsible Press: Wikileaks and the Battle over the
      Soul of the Networked Fourth Estate.” Harvard Civil Rights-Civil Liberties Law
      Review 42.2 (2011): 311–97. Print.

Bey, Hakim. The Temporary Autonomous Zone, Ontological Anarchy, Poetic
      Terrorism. 1985. Brooklyn, NY: Autonomedia, 1991. Web. 19 Mar 2012

Castells, Manuel. Communication Power. Oxford: Oxford UP, 2009. Print.

---. The Rise of the Network Society. 2000. Information Age. Vol. 1. 2nd ed.
      Oxford: Blackwell, 2000. Print.

Eilstrup-Sangiovanni, Mette, and Calvert. Jones. “Assessing the Dangers of Illicit
      Networks: Why Al-Qaida May Be Less Threatening Than Many Think.”
      International Security 33.2 (2008): 7-44. Print.

Fenster, Mark. “Disclosure’s Effects: Wikileaks and Transparency.” Iowa Law
      Review 97.3 (2012): 753-807. Print.

Gladwell, Malcolm. “Small Change: Why the Revolution Will Not Be Tweeted.”
      The New Yorker 4 Oct 2010: n.p. Web. 30 Jan 2013

Greenwald, Glenn. “Wikileaks Reveals More Than Just Government Secrets.” 30 Nov 2010. 28 Mar 2013.

Hamacher, Kay. “Resilience to Leaking: Dynamic Systems Modeling of
      Information Security.” PLoS ONE 7.12 (2012): 1-7. Print.

Hasian, Marouf A. “Watching the Domestication of the Wikileaks Helicopter
      Controversy.” Communication Quarterly 60.2 (2012): 190-209. Web.

Mattelart, Armand. Networking the World, 1794-2000. Trans. Liz Carey-Libbrecht
      and James A. Cohen. Minneapolis: U of Minnesota Press, 2000.

Nystedt, Dan. “Wikileaks Plans to Make the Web a Leakier Place”. 9 Oct. Web. 28 Mar 2013.  

Rosen, Jay. “The Afghanistan War Logs Released by Wikileaks, the World’s First
      Stateless News Organization”. PressThink. 26 July 2010. Web. 28 Mar 2013.

John Jones is an assistant professor in the Professional Writing and Editing program at West Virginia University. He is currently researching the effects of rhetoric and digital technologies on online writing and community formation.

Share this