The study of political behavior and collective action in connection with digital media is in a rich period of theoretical and empirical advance. After nearly two decades of development, the literature is flourishing with theoretical accounts of the changing structure of collective action and participation as well as nuanced empirical analyses testing specific causal mechanisms at the individual level. The flood of new forms of large-scale data coming from social media is fueling research, as are dramatic events in the real world of politics, especially where protest and challenge to traditional political authority are concerned.
Several challenges lie ahead for future research. The first involves moving beyond the treatment of digital media use as an individual-level variable. The second is finding common ground among the variety of post-bureaucratic theories of the structure of collective action that are now available. The last is finding better connections between structural-level theories and individual-level models of behavior, such that studies about digital media and collective action at the structural and individual levels can better complement one another. This essay addresses these tasks as a theoretical undertaking.
Conceptualizing Technology and Collective Action
The Traditional Formulation: Technology Use As Variable
Whether use of digital media is associated with participation in politics, engagement in civic affairs, or contributions to collective action and public-goods production has been under discussion for several decades. By the 1990s, the question was part of an active debate about community and democratic practice (e.g., Etzioni, 1993; Margolis & Resnick, 2000; Sandel, 1996), and the problem of whether changes to politics would serve to reduce or reinforce existing inequalities was important.
Pursuing answers to these questions has typically meant comparing people online and off, or those who use digital media more with those using it less. Treating digital media use as an individual-level variable has been less conclusive than researchers wanted. Many associations between technology use and political behaviors have been reported, chiefly in the United States, based primarily on single cross-sectional surveys (e.g., Bakker & de Vreese, 2011; Bimber, 2001, 2003; Boulianne, 2009, 2011; Cho et al., 2009; Gibson, Lusoli, & Ward, 2005; Mossberger, Tolbert, & McNeal, 2008; Prior, 2007; Shah et al., 2007; Xenos & Moy, 2007). Boulianneâs (2009) seminal meta-analysis put these findings in perspective, showing many null results and a very small average effect size. Other work has shown that relationships between simple measures of Internet and various political actions are not stable over time; moreover, political interest can moderate these relationships in either the positive or negative direction (Copeland & Bimber, 2015; Bimber, Cantijoch, Copeland, & Gibson, 2015).
Ganesh and Stohl (2014) illustrate vividly the problem of trying to measure individual-level media use as a variable. In the field, they find that Occupy protesters in New Zealand were often unable to explain which media tools provided them information about the protest. In trying to describe how they were prompted to participate, protestors conflated different media, sometimes remembering the sender but forgetting whether the medium was YouTube, the Web, Facebook, or Twitter; some protestors heard about the event âeverywhere.â In the case of collective action in Arab Spring, Tufecki and Wilson (2012) show that social media mattered for many people who did not use it at all themselves, because people offline were connected through face-to-face networks to those who were. This lack of boundaries in media are an element of what Chadwick (2013) refers to as media hybridity and what Vaccari (2010) calls the commodification of digital media. Media are a seamless part of many peopleâs life experience rather than a discrete tool or set of tools whose use can meaningfully be isolated, quantified, and correlated with other aspects of life (Bimber, Flanagin, & Stohl, 2012). At this point, the upshot of dozens of studies is that the old rubric of treating digital media use as a variable is confronting limits.
A Better Formulation: Technology As Context
A way around these limits is to approach media as part of the context of politics. Treating media as context rather than a variable accommodates the fact that people sometimes perceive messages to be âeverywhere,â that boundaries within the media system are blurred, and that digital media are widely used commodities whose precise distribution is less important than the fact of their presence in the human environment.
The main implication is that more opportunities for collective action exist than did in the mass media context. The expanded set of action possibilities arises because collective action (as well as many other classes of behavior) requires information and communication. In the context of abundant, widely distributed information and ubiquitous communication possibilities, barriers to collective action are lower.
The idea of expanded action possibilities should not be interpreted to mean more than it does. Nothing about lowered barriers to action compels behavior or establishes a motive force for action. In the context of expanded opportunities, people may just as well sit still at home doing nothing as take to the streets, as they may prefer to be entertained or attracted by what is most popular and highly filtered in their environment (Hindman, 2009; Prior, 2007). Expanded opportunities for action do not necessarily imply liberty as an end result, either. New opportunities for action lead to contests over power, and it may be that people, groups, or whole states employ expanded possibilities for action to exploit or infringe on the liberties of others. Thinking in terms of reduced limits on behavior places technology out of the way, so to speak, and returns agency to people. It directs attention to asking why people act as they do rather than some other possible way, given an expanding range of ways to act in the changing media context.
Post-Bureaucratic Theories of Digital Media and Collective Action
Framing collective behavior in terms of expanded opportunities for action is helpful to the problem of organization. Traditionally, organizations of various kinds have occupied a central position in accounts of collective action and political participation. Generations of graduate students have been taught that collective action happens because organizations accumulate and spend resources, frame messages, create collective identity, coordinate with one another, set action agendas, and so on (e.g., Benford & Snow, 2000; Jenkins, 1983; McAdam, McCarthy, & Zald, 1996; McCarthy & Zald, 1977; Tarrow, 1998, 2006; Walker, 1991). Scholars have moved well beyond the Olsonian rational actor framework, but as theories of social movements and political action have grown over the decades, organizations have not moved far from the central place that Olson assigned them.
We can now understand these theories of collective action to be rooted in the assumption of informational and communicative limits on individual human action. These limits are Weberian, in the sense that bureaucratic structures were long thought necessary for efficient production of action. A paradigmatic statement of Weberian limits on behavior from the resource-mobilization perspective is that of Walker (1991, p. 94), who wrote that âmobilization is seldom spontaneous,â and âorganizations must be formed, advocates must be trained, and the material resources needed to gain the attention of national policy-makers must be gatheredâ before collective action can occur. A parallel claim from the social movement literature is that of Snow, Rochford, Worden, and Benford (1986), who observe that participation in social movements is largely limited to involvement in the activities of social movement organizations.
To be sure, there have been debates about the need for bureaucratic forms. Piven and Cloward (1977) argued that building organizations can actually undermine the effectiveness of âpoor peopleâsâ movements. Snow and colleagues (1986) conceded that mobilization can occur in the absence of social movement organizations (SMOs), but called this a theoretical puzzle. The ânew social movementsâ literature, which is no longer new, observed a reduced dependence on formal organization and bureaucratic control as limits on behavior (e.g., Melucci, 1980; Offe, 1985), but critics pointed out exceptions (Pichardo, 1997). Today, a good deal of literature still adopts the assumption that collective behavior is subject to substantial limits due to the need for bureaucratic structures (Bennett, 2014).
The recent body of literature on the changing structure of collective action has moved away from this assumption, seeing action as much less constrained by the need for the services of bureaucratic structures. This post-bureaucratic direction in theory highlights expanded opportunities to act and the interactions among individual activists, networks, and organizations. Its most important tenets are that formal organization need not precede collective action, and that more opportunities for initiating action exist than in the context of mass media. For instance, the literature on Arab Spring shows that people who initiated the protests were not limited by the need for bureaucratic solutions to planning, framing, or the call to action. In Egypt, traditional organizations had been crucial in contributing to the deeper contexts for revolution (Wolfsfeld, Segev, & Sheafer, 2013), but collective action started through citizens communicating directly with one another, and then the organizations subsequently joining in (Howard & Hussain, 2013; Tufecki & Wilson, 2012). Another illustration from the realm of routine politics comes from research on citizen petitions to government. As Margetts, John, Escher, and Reissfelder (2011) show, petitions for policy change are not constrained by the strategies of organizations in a context where citizens can act publicly and thereby signal to one another directly, without intermediaries brokering messages. Literature focusing on online political tactics (Earl & Kimport, 2011) has also shown that people can overcome collective action obstacles in the absence of organization entirely.
In some cases, attention by citizens creates opportunities for organizations to act, rather than the other way around. Karpf (2012) argues that a new style of organization has adapted to the digital media environment by taking its agenda from rising and falling cycles of public attention to issues. These organizations then provide citizens opportunities for action, but in a way that is opportunistic and reflective of expanded citizen agency. Another way that more traditional bureaucratic structures exploit and contribute to expanded opportunities for action is by presenting multiple faces to the public, offering opportunities for different citizens to relate to the same organization in personalized ways (Bimber et al., 2012).
Bennett and Segerbergâs (2013) theory of connective action provides one of the best synthetic statements of the new direction in collective action theory, because it accounts for a spectrum of structural arrangements from traditional brokering by formal organizations to ad hoc, viral networks. A key theme in their account, to which I return later, ties individual and structural theories together. They argue that organizations make choices about whether to frame and structure action on their own terms, or to step back and facilitate the personalized styles of networked action and expression that some people seek today. Organizational choices interact with citizenship styles to shape the structure of collective action. Chadwickâs (2013) theory of media hybridity is consistent with this, providing an even broader account centering on the collapse of boundaries among media types and the processes of adaptation and mutual adjustment among actors engaged in struggles for power through political communication.
It would be greatly premature to suggest that post-bureaucratic theories have replaced classical ways of thinking about collective action. The new theories are now in somewhat controversial tension with traditional theories. Absent a core theoretical claim against which to push, critics of contemporary collective action theory have argued that in the end not much has really changed, or that new literature has simply selected on the dependent variable to tell the stories of a few outlying cases (see Bennett, 2014; Tarrow, 2014).
What might a synthetic account look like, framed in the most general terms? Most of the contemporary theories concur that expanded opportunities exist for organizing collective action and that these are due to the changed media context. The reason is that information and communication underlie identity building, framing, coordination, persuasion, and most other aspects of collective action. Indeed, to make a classical assertion that collective action entails framing by organizations, or the building of identity through SMOs, or the logistical power of interest groups to organize action, is to assume a set of arrangements for political communicationâwho has what information and who can communicate with whom. In the mass media context, these arrangements reflected relatively high costs of communication and information, which are best born by organizations. Lowered costs of information and communication in the digital media context permit a variety of structures for accomplishing that set of functions: formal organizations; networks comprising individuals and organizations; informal, adaptive organizational structures; and individual action (Bennett & Segerberg, 2013). Of course, not all structural arrangements are equally successful at all the functions and goals of collective action. Some are likely better at producing sustained pressure on political institutions, some at working across national borders or drawing together disparate interests under a single banner, some at rapid influence on the political agenda, and so on. Which structures are most successful for which goals has not been established and is a ripe topic for future study.
Rather than asking the old question of how organizations accomplish collective action and whom they are able to mobilize, it is better for researchers to begin with the range of structural pathways now available for collective action. They can then ask why, in any particular instance, one pathway is taken rather than another, and why some citizens may prefer one pathway to another.