Introduction
For over 30 years researchers from a range of disciplines have been concerned with exploring the possibilities computer-mediated communication (CMC) offers for facilitating deliberative processes and effective information exchange, thereby strengthening democratic processes. The relation between information and communication technologies (ICTs), social movements, and democracy has been a focus of work from a wealth of disciplines (Della Porta, Reference Della Porta2011).
Critiques of the state of the art note a number of key problems in current research on the relation between ICTs and social movements. First, much research studies online participation exclusively and/or separately from offline participation (Ducheneaut and Watts, Reference Ducheneaut and Watts2005; Tufekci, Reference Tufekci2014), often focusing on interactions within online communities or between social movements and their online publics (Ayres, Reference Ayres1999; Myers, Reference Myers2001; Gillan et al., Reference Gillan, Pickerill and Webster2008; Atkinson, Reference Atkinson2010; Coleman, Reference Coleman2012; Castells, Reference Castells2013). This is problematic because activists communicate within a hybrid media ecology in which on and offline communication interacts in myriad ways (Etzioni and Etzioni, Reference Etzioni and Etzioni1999; Cammaerts et al., Reference Cammaerts, Mattoni and McCurdy2013; Flesher Fominaya, Reference Flesher Fominaya2014a; Tufekci, Reference Tufekci2014). In addition, methodologically, studying online participation exclusively cannot tell us anything about non-participation, it only captures the behaviour of those who are already participating (Tufekci, Reference Tufekci2014). This makes it harder to explore factors that inhibit online participation, a key issue for social movements seeking to maximize the democratic potential of ICTs. Second, there is a tendency to select on successful cases (or the dependent variable), neglecting the reality that many social movement uses of ICTs are unsuccessful or do not lead to positive outcomes (Earl and Kimport, Reference Earl and Kimport2011; Tufekci, Reference Tufekci2014). Third, the literature on ICT use in social movements is strongly focused on its use in mobilization or ‘cyberactivism’, rather than on the day-to-day communication that sustains social movement groups over time (Haug, Reference Haug2013). This conflation of social movement activity with mobilization is accompanied by an emphasis on the strategic and reflexive use of ICTs, neglecting the more routine or quotidian aspects of their use. An emphasis on mobilization also directs attention to the use of ICTs in external communication (with publics or outsiders), but neglects attention to the reciprocal effects of on and offline communication within social movement groups or communities (Haug, Reference Haug2013; Jordan, Reference Jordan2013). Finally, there is a tendency to focus on the impact of the technological aspects of ICTs (i.e. costs, affordances, and leveraging) rather than the emotional and subjective aspects of their use, and their impact on social relations (Bimber, Reference Bimber2001; Gillan, Reference Gillan2008; Jordan, Reference Jordan2008, Reference Jordan2013; Ganesh and Stohl, Reference Ganesh and Stohl2010). While the technical and design attributes of specific ICTs do have specific effects, it is also true that the social context in which they are used renders their effects more variable and contingent than a primarily technological emphasis suggests, as I will show.
A key concern in the discussion of ICTs and activism is the degree to which ICT use can sustain geographically dispersed networks (Lovink, Reference Lovink2011). Much less attention has been paid to the relationship between ICT use and sustainability in local networks or groups. Despite the increase in ICT use, face-to-face meetings are still crucial for many social movement groups (Mosca, Reference Mosca2008). However, with important exceptions (Cronaeur, Reference Cronauer2004; Horton, Reference Horton2004; Pickerill, Reference Pickerill2004; Kavada, Reference Kavada2007, Reference Kavada2009, Reference Kavada2010) much less attention has been paid to the impact of ICTs on the internal communication and cohesion of face-to-face social movement groups, leaving the relationship between ICT use and face-to-face group dynamics largely unexplored. Despite a rapidly evolving media environment for social movement groups (Juris, Reference Juris2012; Mattoni, Reference Mattoni2012), one of the central ICT tools still used by social movements for internal communication is e-mail (Cronaeur, Reference Cronauer2004; Kavada, Reference Kavada2007; Ganesh and Stohl, Reference Ganesh and Stohl2010), yet studies exploring the internal use and impact of e-mail are rare.
In this article I use ethnographic data to analyse the use and impact of e-mail on the internal dynamics of two face-to-face groups in the Global Justice Movement network in Madrid, Spain. I focus specifically on the unintended or negative consequences of e-mail use on the processes of group cohesion and collective identity formation that sustain groups over time. This ethnographic approach allows me to analyse the subjective experience of e-mail use and non-use, as well as its impact on social relations between group members. It enables an analysis of the reciprocal interaction between on and offline ICT use, as well as an exploration of the challenges ICTs pose for social movements, a necessary corrective to a tendency to take for granted their benefits for collaboration and to overlook their routinized rather than strategic use. This article, therefore, contributes to the literature by addressing all four central critiques of the state of the art noted above.
The two groups analysed shared an explicit political ideological commitment to practices of deliberation and ‘horizontality’ as central features of their practice and identity, consistent with the ‘autonomous’ assembly (asamblearia) tradition (Flesher Fominaya, Reference Flesher Fominaya2007, Reference Flesher Fominaya2014a, Reference Flesher Fominayab). Autonomous actors distinguish themselves from the practices of the institutional left, rejecting representative democracy and majority rule and instead defending more participatory models, based on leaderless ‘horizontal’ (non-hierarchical) structures, consensus decision making (if possible and necessary) in an open assembly, and rarely with fixed roles for individual members. In contrast to formally organized social movement organizations (SMOs), autonomous groups lack formal institutional structures, memberships, and resources. Historically, autonomous movements have been closely intertwined with a range of digital democratic imaginaries, meaning that ICTs and the internet are seen as facilitating greater and more democratic forms of participation (Juris, Reference Juris2005; Flesher Fominaya, Reference Flesher Fominaya2014a). This belief was widespread though unevenly shared in the Madrid network at the time of this study. Despite these ideological commitments to ‘horizontality’, I discovered an intriguing puzzle: e-mail use reinforced and magnified existing hierarchies even when activists were committed to their elimination, and hindered rather than facilitated the reaching of group ideals.
Overall, I found that the use of e-mail in these two groups increased existing hierarchies, hindered consensus, decreased participation, and worked towards marginalization of group members. I found that the power relations established within the group and the communication patterns in face-to-face assemblies were reproduced and even heightened in the virtual sphere, rather than diminishing or neutralizing them. In addition, the negative and unintended consequences of e-mail use affected both groups, independently of the evaluation of activists with their experience in their face-to-face assemblies (one of which was overwhelmingly perceived as positive and one of which was perceived as negative).
While this finding is important for all social movement groups using e-mail, it is particularly important for autonomous or non-institutionalized groups, because in contrast to formally organized SMOs where institutional frameworks and formal resources play key roles in sustainability, for autonomous movement groups it is affective ties and positive perceptions of group interaction and experiences that nourish a strong group collective identity and become crucial means of maintaining group cohesion over time (Flesher Fominaya, Reference Flesher Fominaya2010a, Reference Flesher Fominaya2014b). The analysis here is also particularly important for any group – avowedly political or not – who explicitly set out to work collectively based on democratic, horizontal, inclusive, and participatory methods and ideals, by showing that a lack of reflexivity about e-mail can work against those ideals. To my knowledge, this research is unique in tracing the interaction effects between face-to-face and online interaction as they affect feelings of group collective identification over time.
My emphasis is on what Horton (Reference Horton2004: 743) refers to as ‘reinforcement e-mails’, those e-mail lists that are internal to social movement groups, and that serve to ‘maintain and build on the strong ties of activists who already meet face-to-face’, although I do include analysis of the use of e-mail to communicate beyond the groups to other groups in the network (outreach e-mails). ‘Reinforcement’ e-mail lists continue to be central forms of communication for most social movement groups, despite the advent of social media and the integration of other forms of internal communication (e.g. Telegram, WhatsApp) (Gillan et al., Reference Gillan, Pickerill and Webster2008; Mosca, Reference Mosca2008; Askanius and Gustafsson, Reference Askanius and Gustafsson2010). E-mail, therefore is a key mechanism of communication in social movement groups, but one that is routinized and habitual, and I will suggest, much less subject to reflexivity and strategic thinking than those forms of ICT use that are used during ‘mobilization’.
Reinforcement e-mail is more private and, in theory at least, should serve to reproduce ‘an activist’s network centrality and pre-existing levels of involvement, belonging and commitment’ (Horton, Reference Horton2004: 743). Clearly, e-mail is extremely useful for activists and can also be an essential mechanism for reinforcing existing ties. Nevertheless, as I will show, selecting on more negative or unsuccessful cases, reinforcement e-mail can have negative and unintended consequences for social movement groups that work against feelings of involvement, commitment, and belonging, instead increasing feelings of dissatisfaction with decision making, existing groups power relations, and feelings of marginalization.
Despite a vast literature on collective identity processes in social movements, very few studies explore the relationship between collective identity processes and use of ICTs, with few exceptions (Cronaeur, Reference Cronauer2004; Kavada, Reference Kavada2007, Reference Kavada2009). Crucial to collective identity formation at group level are internal processes that develop a sense of common belonging, reciprocity and trust, and a shared commitment to common goals and actions, which nurture internal group cohesion and activist commitment to groups over time (Melucci, Reference Melucci1995; Polletta and Jasper, Reference Polletta and Jasper2001; Flesher Fominaya, Reference Flesher Fominaya2010a, Reference Flesher Fominayab). Communication plays a key role in building trust and solidarity, key components of collective identity processes. Sufficient information and trust in information exchange are crucial for creating bonds between participants (Etzioni and Etzioni, Reference Etzioni and Etzioni1999; Kavada, Reference Kavada2009) and a positive evaluation of deliberation processes has implications for decision-making legitimacy (Stromer-Galley and Muhlberger, Reference Stromer-Galley and Muhlberger2009), which in turn affects participants’ desire to continue to participate. Given the importance of e-mail to communication in social movement groups, the paucity of studies that examine the impact of e-mail on collective identity is surprising.
Although largely neglected by social movement scholars, the fields of computer science, organizational studies, information systems, and political communication have studied the dynamics of e-mail use on decision-making processes in organizations, and deliberation processes online. Given how underutilized this research is in analysis of ICTs and social movements, I first survey the relevant literature, highlighting those findings likely to impact the realization of the ideals of the groups studied here (participation, horizontality, inclusivity, consensus decision making), or to affect the subjective experience of e-mail as it could impact feelings of commitment, belonging, solidarity, and trust. As the realization of group ideals is likely to impact on feelings of satisfaction and commitment, these impacts are in fact reciprocal, and can be expected to affect group cohesion. I then discuss existing research on e-mail use in social movements, again focusing on its expected impact on collective identity or group cohesion and sustainability over time, before turning to the methods and case studies.
Research on e-mail use in organizations
In a comprehensive 30-year review, Ducheneaut and Watts (Reference Ducheneaut and Watts2005) survey key findings on e-mail use and computer–human interaction. This research is important in distinguishing between the intrinsic characteristics of e-mail as a tool and the ways that it is used in human interaction in organizational settings. Logically, the majority of the research on the use of e-mail in organizations refers to contexts where there are disciplinary mechanisms regulating participation and use, a clear chain of hierarchical command, costs for not participating (e.g. not checking or answering e-mails) and clear expectations of accountability and responsibility. None of these mechanisms apply in the case of voluntary, theoretically horizontal (non-hierarchical) groups with low costs of entry and exit and little accountability beyond status or reputation within social movement networks. Nevertheless, the findings highlighted below are highly relevant for research on e-mail use in social movement groups, with implications at the individual, communicative, and socio-organizational levels.
E-mail malleability and sensitivity to context
With regard to individuals, particularly notable is the striking diversity in patterns of individual use of e-mail systems and e-mail management strategies (e.g. archiving and retrieval), and the fact that information overload leads to missing or partially read documents. Therefore, the widespread assumption underlying routine practices of social movement listservs, that e-mail’s intrinsic capacity for enabling information transmission to many users will necessarily result in effectively informing users, either at the time of receipt or later through retrieval, is problematic.
Regarding e-mail and collaboration, which like information exchange is a key activist use of e-mail, research shows that ‘collaborative decision making through email places great strain on the participants’ ability to maintain the thread of their arguments’ (Ducheneaut and Watts, Reference Ducheneaut and Watts2005: 16). Systems that bundle conversations (like Gmail) mitigate this problem, but not all activists are using them. In addition, ‘unlike many other forms of communication, email keeps statements alive and subject to continual modification’ (Ducheneaut and Watts, Reference Ducheneaut and Watts2005). These factors mean that when decision making is shifted from the face-to-face assembly onto e-mail, participants will face problems of incomplete information, difficulty in following arguments and in closing debates, exacerbated by asynchronicity and unequal access.
As an effective tool for work production in groups, e-mail is useful in contexts where activities are fairly repetitive and well defined and much less so when more flexibility is required (Duchenaut and Watts, Reference Ducheneaut and Watts2005: 17) as with the non-institutionalized groups discussed here. In addition, e-mail is highly malleable and influenced by organizational context, with factors such as social influence, interpersonal relationships, organizational power structures, group perceptions, socialization, and social control mechanisms being crucial in shaping its use and effectiveness in organizations (Lee, Reference Lee1994; Markus, Reference Markus1994; Garton and Wellman, Reference Garton and Wellman1995; Ducheneaut and Watts, Reference Ducheneaut and Watts2005). This raises questions about e-mail’s ability to alter the social and power dynamics of face-to-face groups through status equalization or to alter the effectiveness of decision making.
Status equalization?
A key claim in the literature on online deliberation (below) is that status inequalities that hamper equal participation in face-to-face settings can be diminished online because of the absence of social cues to participant identity/status (race, gender, class, etc.), a claim that fails to be borne out when research is done within organizations, where status differentials remain robust (Saunders et al., Reference Saunders, Robey and Vaverek1994). In social movement group settings where participants know each other and are working together offline, therefore, such status differentials are likely to remain in place.
Lack of inhibition and conflict
Another key finding is that while e-mail can produce less inhibited, non-conformist, and combative exchanges (Kiesler and Sproull, Reference Kiesler and Sproull1992), uninhibited behaviour is quite infrequent within organizations, and ‘decreases with time, group history, and anticipated future interaction’ (Ducheneaut and Watts, Reference Ducheneaut and Watts2005: 21). As most e-mail lists in (autonomous) social movement groups are set up almost immediately after the first constitutive assembly (Kavada, Reference Kavada2007), the nature of e-mail exchanges in the early stages of group consolidation, such as the levels of conflict or flaming, can be expected to have important consequences for it. This played a key role in the second case discussed here, the Disobedience Lab.
Online deliberation and political communication
The largest body of literature examining the relation between the internet and deliberative democracy has centred on online political forums, much of it concerned with the extent to which the virtual sphere can strengthen and support deliberative processes (Dahlberg, Reference Dahlberg2001; Albrecht, Reference Albrecht2006; Freelon, Reference Freelon2010). Idealized criteria of deliberation include reflexivity, reciprocity, and inclusiveness (Janssen and Kies, Reference Janssen and Kies2005).
Plurality and full participation vs. status inequalities and skewed participation
In theory, the internet should be able to increase plurality by lowering barriers to participation, and increase the richness, transparency, and availability of information, thereby improving decision-making quality and legitimacy (Weare, Reference Weare2002; Janssen and Kies, Reference Janssen and Kies2005; Albrecht, Reference Albrecht2006), yet research yields contradictory findings. While there is evidence that under certain conditions the internet can serve to provide a forum for robust political debate (Dahlberg, Reference Dahlberg2001; Papacharissi, Reference Papacharissi2004; Dahlgren, Reference Dahlgren2005; Freelon, Reference Freelon2010), research also points to important barriers to realizing the democratic deliberative ideal. With regard to the impact of diminished social cues online, the benefits of status equalization and the favouring of rational critical deliberation over personalized exchanges (Dahlberg, Reference Dahlberg2001) are offset by heightened conflict, misunderstandings, and aggression (Kiesler and Sproull, Reference Kiesler and Sproull1992; Albrecht, Reference Albrecht2006).
In addition to problems of unequal access caused by the digital divide (Norris, Reference Norris2001), Dahlberg (Reference Dahlberg2001: 623) highlights a trend towards lack of reflexivity in cyberdeliberations, the failure to achieve respectful listening or commitment to difference, the dominance of discussion by few individuals and groups, and exclusions because of social inequalities. This lack of reflexivity will be highlighted below in relation to the case studies.
Online deliberation research is often based on controlled experiments, random samples of participants with no offline ties, and laboratory conditions (Albrecht, Reference Albrecht2006). Nevertheless, studies on online deliberation have found trends with relevance for social movement groups. Despite hopes that online forums would facilitate many-to-many exchanges, evidence shows a strong tendency for a strong concentration of a very small number of core users actively contributing to discussion (Janssen and Kies, Reference Janssen and Kies2005; Albrecht, Reference Albrecht2006). In addition, despite some evidence that status inequalities can be diminished online, there is still evidence of strong gender imbalances in contributions to online deliberation (Albrecht, Reference Albrecht2006; Taylor, Reference Taylor2014), a finding even observed in social movement groups holding a strong egalitarian ideal (Cronauer, Reference Cronauer2004).
Online decision making, satisfaction, and continued commitment
Stromer-Galley and Muhlberger’s work (Reference Stromer-Galley and Muhlberger2009) found that positive evaluations of the deliberative process by participants increased decision legitimacy, decreased decision ambivalence, and increased expectations of future participation, a finding with relevance for social movement group cohesion, continuity, and collective identity. Issues of satisfaction with deliberation can be expected to be especially relevant for groups seeking to make participatory deliberation a key element of their collective identity, as with groups committed to ‘horizontalism’. This expectation plays out in the case studies discussed below.
E-mail, social movements, and collective identity
Research on Global Justice activist use of e-mail shows that activists are reflexive about e-mail use and recognize both its potential benefits and costs. E-mail clearly provides important benefits for social movements, potentially improving efficiency, expression, and understanding, but also increasing emotional distance that could either benefit interaction or escalate conflict (Gillan et al., Reference Gillan, Pickerill and Webster2008). E-mail can be useful in building networks of solidarity and trust and generating new groups, but also can have negative or unintended consequences, such as saturation, problems with access and computer skills, and electronic surveillance (Ganesh and Stohl, Reference Ganesh and Stohl2010: 69).
As the organizational literature on e-mail shows, patterns of communication shift depending on the social context of use, the design and size of the e-mail lists, the nature, function or purpose of the organizations, and the geographical dispersion of the participants (Sproull and Kiesler, Reference Sproull and Kiesler1986; Ducheneaut and Watts, Reference Ducheneaut and Watts2005). Whether lists are open or closed, temporary or ongoing, moderated or ‘free’, and whether participants interact exclusively online or not, all have implications for their use and impact (Cronauer, Reference Cronauer2004; Kavada, Reference Kavada2007, Reference Kavada2009). Horton (Reference Horton2004) usefully distinguishes between ‘outreach’ and ‘reinforcement’ e-mails, highlighting the crucial differences between e-mails intended for communication with outsiders vs. those used to reinforce existing strong ties. However, he emphasizes the positive role of internal lists in reinforcing commitment and feelings of belonging. In what follows I will focus on the ways such lists can work against these mechanisms of reinforcement.
The strong tendency for domination of discussion by few participants noted above has also been observed in research on social movement use of e-mail discussion lists (Cronauer, Reference Cronauer2004; Grignou and Patou, Reference Grignou and Patou2004; Kavada, Reference Kavada2007; López, Reference López2008),Footnote 1 Grignou and Patou (Reference Grignou and Patou2004: 175) found that in participation on ATTAC discussion lists (with hundreds of subscribers) quasi-experts dominated discussions and those familiar to each other responded to each other but not to newcomers, despite the list being ‘open’ to all. This tendency was very marked in the case studies I present below.
Two important studies treat the relationship between e-mail list use and offline collective identity in the Global Justice Movement.Footnote 2 Cronauer (Reference Cronauer2004) analysed two lists, one of which was used by a group of subscribers who also met face to face. Combining e-mail and survey data, she analysed factors that hindered subscriber’s participation on the lists and in mobilization. Despite the list coordinator’s intentions, neither list effectively informed subscribers about group views, aims, tactics, issues, or how to become active. In general, Cronauer (Reference Cronauer2004) found that greater involvement offline and online were correlated. E-mail lists served to retain and further mobilize already committed activists but did not have the same effects on those less involved.
Even on the list most similar to Horton’s (Reference Horton2004) ‘reinforcement e-mails’, Cronauer (Reference Cronauer2004) found that collective identity was neither attempted nor developed through list use. Confrontational debates, and lack of mechanisms to foster mutual support and trust contributed to this. She suggests that face-to-face meetings to negotiate collective identities could serve to generate better prospects for collective identity formation. The case studies I present below, however, show that face-to-face meetings alone are insufficient to mitigate the potential negative effects of e-mail on collective identity formation.
In contrast to Cronaeur (Reference Cronauer2004), Kavada’s (Reference Kavada2007, Reference Kavada2010) research on the use of e-mail lists in the 2004 European Social Forum is more positive in its evaluation, despite highlighting some problems. Kavada’s e-mail lists differ in key ways from those studied here in that they were open, primarily connected geographically disperse activists, and were created to facilitate offline meetings to prepare for a particular event. Kavada focuses on the functional aspects of the lists (their usefulness for coordination, and how effectively they complement face-to-face organization), given how integral these are considered for sustaining consensus-based decision-making practices in many networks. She finds that e-mail contributes to movement-level collective identity processes by fulfilling essential coordination functions, allowing the movement to grow, and defining membership, but that e-mail listservs are not very effective in generating trust and solidarity between users compared with face-to-face interactions.
Methods and cases
The data analysed here is drawn from two case studies that formed part of a larger ethnographic study on the practices of autonomous ‘horizontal’ Global Justice Movement groups in Madrid, Spain from 2002 to 2005.Footnote 3 The research involved participant observation at all the assemblies of the groups studied, in-depth interviews with 19 activists, subscription to the e-mail lists of the groups, and field observations of a wide range of movement activities. In all, seven of the activists participated in both groups allowing them to contrast their experiences in one group with the other.
Despite encompassing a range of ideological orientations (e.g. anarchist, feminist, anti-militarist, squatter) both groups were explicitly committed to creating new groups based on a commitment to the principles of horizontality, openness, deliberation, fostering increased participation, and inclusiveness. They combined elements of both participatory and deliberative democratic ideals. Both groups were actively seeking to counter the hostility, authoritarianism, and hierarchies they felt were endemic to institutional left-controlled social movement groups in Madrid. Although never establishing explicit methodological protocols, they operated primarily on ‘practical’ consensus decision making (i.e. voting was not used but neither was unanimity a requirement for action), and considered face-to-face assemblies to be the sovereign decision-making space.
Assemblies were open, and group e-mail lists were open only to members of the face-to-face assembly, which met weekly [European Social Consulta (CSE)] and bi-weekly (Disobedience Lab). Therefore, these were not virtual communities, and e-mail was used primarily as a complementary means to ostensibly enhance participation and intra-group communication between assemblies and to provide continuity for participants who might miss an assembly. The CSE also had a wider ‘international’ list that connected the different promoter groups in the CSE network. As activists knew each other, the effects of anonymity on either enhancing or inhibiting participation were not at play. Typical assemblies ranged from 7 to 20 activists, with assemblies of up to 60 people in periods close to mobilizations. Most activists were aged 20–30, university educated, and worked in a range of blue- and white-collar jobs.
The use of three sources of data (interviews, participant observation, and e-mail list subscription) allowed me to compare the dynamic of participation in face-to-face assemblies (de facto leadership dynamics, interaction patterns, etc.) with the e-mail exchanges and to explore how activists interpreted the exchanges and their impact. This allowed me to examine the interaction between e-mail; face-to-face group dynamics in assemblies; and activists’ interpretations of the impact of the use of e-mail on group cohesion and their sense of identification and commitment.
At the time of this study, few of the activists in these groups had regular (daily) access to the internet, a situation that according to my current research (2014) on the 15-M movement in Spain continues to be an issue, despite a much less prevalent one. Increased access, however, does not mitigate the effects of e-mail asynchronicity, unequal distribution of time online, or media knowledge in the activist population (Mattoni, Reference Mattoni2012).
The Madrid CSE
The CSE hoped to create a group that would foster democracy in society and can be seen as a precursor to the 15-M movement. It was an experimental project in participatory democracy that hoped to collectively create a network that would critique the lack of ‘real democracy’ in representative democracies through its own example of radical democratic participation and direct action (i.e. consultas-illegal popular referendums). Its goal was to become a transnational network, and some promoter groups (GPs) were established in different European countries, but the two main nodes were in Madrid and Barcelona.
The Madrid GP, which I analyse here, was quite sceptical of the potential of internet and e-mail in developing the project, and opposed in principle their use as deliberative fora. They argued strongly that no decisions or deliberation should take place in cyberspace, because of the risk of exclusion owing to unequal access and time. They argued for the sovereignty of the assembly as the legitimate decision-making arena. Yet, even with this critical awareness, the use of the e-mail list exacerbated existing problems and created some new ones.
Like most ‘horizontal’ groups, the CSE Madrid had informal leaderships, and one of the founding members, Mario,Footnote 4 clearly had a leadership role and influenced the development of the assembly in concrete ways. This de facto leadership in the assembly was mirrored in cyberspace, as Mario also dominated the e-mail lists both in terms of frequency and length of contributions. His frequent submission of dense proposals for consideration by the group acted to inhibit participation from other members, as this quote from Fritzi, an experienced activist and long-term member, shows:
Every week I would read the mailing list and I would read these long super complicated dense documents and you think ‘Gee, if somebody’s come up with a three-page proposal what the heck am I going to propose?’ I know that it doesn’t make a lot of sense but that’s what happens, you feel like you can’t contribute anything that would be up to scratch.
I think the fact that few people have spoken out and the fact that it’s always the same person who writes the documents has been a big problem…
Although Mario surely felt he was making an important and positive contribution to the group (after all, active participation is a goal of the group), in fact his frequent e-mails were reducing participation by other people. The problem was not limited to the length and frequency of his e-mails but also to their style and content, as Fritzi’s reference to ‘super complicated dense documents’ shows. Mario used a writing style that did not reflect the tone of the discussions in assembly and which for some, lent his missives weight and authority. The content of the documents also tended to reflect his views more than those of the group as a whole. Interviews revealed that as time passed, participation in the list dropped off as people were either intimidated (‘I could never write anything that eloquent, so why should I give my opinion’) or overwhelmed (‘I just don’t have the time to write all that’) or alienated (‘I don’t feel that that document represents the discussion we had’), demonstrating also the effects of decreased satisfaction with deliberation on continued participation as noted by Stromer-Galley and Muhlberger (Reference Stromer-Galley and Muhlberger2009). These responses are strikingly similar to those offered by participants in Cronauer’s (Reference Cronauer2004) study above, suggesting that inhibiting factors can transcend group size and geographical dispersal and remain robust even in small groups that also meet face to face.
To the outside world (activists or potential participants in the Consulta), the person who sent out the documents became the ‘representative face’ of the Consulta, and the documents and messages sent in his style came to represent the Consulta as well. One document sent to the international Consulta list, ostensibly a concrete proposal to move forward with the project, was characterized in the following way from an activist in Ireland, who sent an e-mail back saying:
Re: Block proposal
This discussion about the Consulta is hopelessly academic and abstract. It will get us nowhere. We have a war right ahead of us – timed for somewhere between the middle of January and the middle of February. Can we focus and test ourselves on that – in the spirit of the last day in Florence? We need community and regional organisations and networks all over Europe – converging as they grow, from below – NOT from top-down.
Activists involved in the Consulta process in Germany also commented on the dense and jargon-laden style of the block proposal documents, which they found to be a ‘turn off’, despite being committed to the principles that motivated the Consulta project. As the e-mails were the only contact they had with the Madrid GP before a single meeting with a few members in Germany, they had no other means of countering this negative impression.
Ironically and frustratingly for the activists involved, a project that had bent over backward to be radically democratic was being dismissed as ‘top-down’. Mario’s dissemination of jargon-laden documents, which failed to reflect the vision of the group as a whole, increased feelings of alienation within the group and hampered the Consulta ‘cause’. Therefore, while political ideology (i.e. commitment to ‘horizontality’) did not appear to influence the way posters used the e-mail lists, it did affect the interpretation of the messages, with ‘horizontal’ activists outside the promoter group being ‘turned off’ by the content.
Another serious problem involving the use of e-mail was last-minute unilateral decisions communicated by e-mail. On one occasion, during a meeting of the Madrid GP, the assembly decided to take the momentous decision to dissolve the GP and merge it into a larger group. Long-term members of the GP, not present at the meeting, found out about this decision via an impersonal e-mail, which was the means Walter (who had pushed for the merge) used to convey the decision to the rest of the group. The decision was presented as a ‘done deal’ despite the fact that only six people had been at the meeting and two had not been in favour of the decision. One GP member, Darla, who had not been at the meeting, explained her reaction:
I was just getting back from giving a talk at the local neighborhood association about the Consulta. I was there as a member of the GP and I had been invited as such. I got home and found that the group I had just been representing no longer existed. I felt like an idiot. Not only that, I felt like my opinion had not even been taken into consideration.
As Darla’s response shows, the fact that e-mail was used to convey the decision had a specific impact on the way it was received. Seeing the decision in writing lent it a certain authority and finality, reading the e-mail alone increased her sense of isolation and feeling that she was not connected to the group, and not having the opportunity to discuss it face to face increased her disappointment and frustration.
Fritzi had also missed the meeting and said:
I was really surprised by the decision and I felt that I had really been left out of that process and quite a few other people felt the same way-as if my opinion counted for nothing after all that time… After all we have spent over a year on a process of establishing contacts with people in Europe and in the rest of Spain. Do we just blow them off because now we created a new group in Madrid? Something definitely got lost along the way.
Last-minute changes about assembly meeting times, communicated by e-mail, also took a heavy toll on participation. Again, it was Mario who generally felt entitled to take these unilateral decisions. The problem was that not all members of the group checked their e-mail every day or even had regular access to the internet, and even if they had, some needed prior warning to adjust their schedules to accommodate the assembly. The result was that some people would show up at the meeting at the prearranged time only to find out it had been cancelled, sometimes after travelling a long way.
One long-term member, Carolina, talked about the effect this had on the group:
… there were four or five disastrously uncoordinated meetings that dealt the death blow to the whole thing. People like Xurxo and Susana decided to stop coming, because Mario arranged a meeting, and didn’t turn up where he said it would be. Then there were some last minute cancellations of meetings by email, which half of us wouldn’t see… People get pissed off because it is a waste of time, and they have other priorities. I was ready to hit Mario myself at that point. We have this vibe like ‘Ah, we have to be non-bureaucratic’ but coordination is crucial. We need to be in this place at this time, because people get very burned out. There is this assumption that everyone’s connected to the Internet all the time. That’s ridiculous.
While such continual problems on the ‘reinforcement’ list contributed to a slow erosion of the Madrid group, a single e-mail cancellation of a European meeting on the ‘outreach’ list had even greater implications for the development of the network in Spain and Europe. The Madrid GP, who was coordinating the meeting, had delegated the task to Walter, who had volunteered. The meeting date had been arranged in advance and the group assumed Walter was coordinating the details. A few days before the agreed date, they discovered that Walter had unilaterally decided that the meeting should be postponed indefinitely and that he had sent a message to the international lists to this effect. He listed four arguments against maintaining the date of the meeting, mostly involving conflicting political actions in Madrid against the invasion of Iraq, which, he felt, took clear precedence over the Consulta.
His decision infuriated activists in the Barcelona GP, who logically assumed it was a Madrid GP decision. They insisted that the date be maintained and pointed out that Madrid’s local political reality could not be used as a rationale for postponing a European-wide meeting. Meanwhile, activists from Ireland wrote to say that they already purchased their tickets and would be unable to get a refund. The confusion and indecision generated by this unilateral proposal meant that activists from Italy, Belgium, and France who were interested in attending were left with a margin of so few days to purchase their tickets that it would be very unlikely for them to attend even if the date were maintained. Activists in the Madrid GP later joked that the War on Iraq’s collateral damage had included the European CSE meeting. The effect of Walter’s decision to propose a date change very late in the game dealt a serious blow to the possibility of advancing the process on the European level for the foreseeable future and significantly increased tensions between the groups in Madrid and Barcelona. Some people in Madrid privately expressed their anger at Walter’s unilateral decision, and felt that he had seriously damaged the CSE’s credibility in Europe.
Unilateral decisions by e-mail are in fact a common occurrence in social movement groups, even those with a commitment to the sovereignty of the assembly, and are sometimes compelled by changing circumstances or pressures. Mario and Walter’s actions are not exceptional, nor were they malicious in any way. On the contrary, both are highly committed activists who work in good faith to advance movement goals. This highlights the loss of control over e-mail that the medium’s malleability itself facilitates in the absence of social control mechanisms and hierarchies of responsibility and accountability (Ducheneaut and Watts, Reference Ducheneaut and Watts2005). Scale shift increases the impact of individual actions in smaller group settings (Albrecht, Reference Albrecht2006). The case highlights the importance of content and not just structure in the effectiveness of brokerage strategies (Tarrow and McAdam, Reference Tarrow and McAdam2005) between groups. The interpretation of the style of writing of CSE documents – an interpretation shaped by culture and language – also affected transnational network expansion.
In the CSE, far from increasing participation, e-mail use facilitated its decline, despite the fact that the environment in the face-to-face assemblies was on the whole considered to be overwhelmingly positive by activists (Flesher Fominaya, Reference Flesher Fominaya2007). Consistent with research on online deliberation (Stromer-Galley and Muhlberger, Reference Stromer-Galley and Muhlberger2009), activist evaluations of online deliberation affected activist perceptions of decision ambiguity and legitimacy. Activists highlighted the emotional impact of e-mail (frustration, alienation) and its effect on their sense of collective belonging and their desire for continued participation. These impacts were specifically related to the style, content, frequency, and timing of e-mails, but significantly not to conflict, hostility, and flaming, which is more commonly highlighted as a primary problem for e-mail use in social movements and a key deterrent to collective identity formation (Cronauer, Reference Cronauer2004; Kavada, Reference Kavada2007). A commitment to horizontality and a techno-sceptical approach did not appear to distinguish patterns of use (i.e. dominance by few, selective responses, and the gender gap) from those found in research on ‘ordinary’ citizens, nor increase group reflexivity about the potential impact of e-mail on group dynamics or cohesion. Because internal ‘reinforcement’ e-mail was seen as simply a routinized form of communication between assemblies, there was no collective reflexivity about its use and impact, and even the outreach e-mail fiasco was not collectively addressed explicitly in assembly beyond trying to take some actions to mitigate its effects after the damage was done. ‘Backstage’ or out of assembly activists did discuss the outreach e-mail and collectively perceived it as a great problem, highlighting the different levels of reflexivity around strategic outreach e-mails directed to people outside the group vs. those routinized e-mails used to ostensibly complement internal communication. Striking too was the inhibition women felt to participate in the online exchanges, despite being very active within the group overall.
The Disobedience Lab
The Disobedience Lab hoped to create an assembly that would generate theory-driven direct action in Madrid. Although this assembly was also in theory horizontal, a group called the Nomad University was the driving force behind the group’s creation and also acted as a hegemonic nucleus within the assembly. As with the CSE, the dynamics of e-mail communication reflected the dynamics in the assembly. The use of the e-mail in the Disobedience Lab reduced participation with even more immediate and clear results than the CSE. However, the problems caused by e-mail cannot merely be attributed to greater hostility on and offline. E-mail exchanges not only mirrored but magnified problems in the assembly, and also created new ones.
One recurrent problem in the Disobedience Lab assembly was the use of abstract theoretical jargon by members of the Nomad University both on and offline. This practice would alienate some members of the assembly who felt intimidated or turned off by this discourse:
They would circulate documents that said things like ‘we want to be a space of less to become a space of more’. And I would think, well you are going to be a space of less because no one knows what the fuck you’re talking about! (Maria).
Written documents circulated via e-mail lent themselves even more to abstraction, and the authority of the written texts sent by core members of the assembly gave them even greater impact, as they seemed to be setting the agenda for the assembly as a whole. However, this was not the only problem in reducing participation. Added to this was the highly conflictive tone of e-mails sent by some members of the Nomad University, which mirrored but was more uninhibited than their interventions in assembly. In this way conflict extended from the assembly into cyberspace as well, where acrimonious e-mails further heightened tensions. These e-mails caused other key people to leave the group, taking their sub-groups with them, debilitating the possibility of establishing a sustainable new project within the network. Such was the case of experienced activist Adela:
I was tired of being patronized and ridiculed. I remember in particular the email suggesting I sit at my computer and think things through 100 times before writing. Their arrogance was really amazing, especially since they did not exactly have much experience and their ideas about security were totally off base, they didn’t have a clue about direct action (Adela).
Darla talks about her reaction to another e-mail, which also highlights issues related to differences in computer skills:
I once made the ‘terrible’ mistake of sending an email to the list in Microsoft word. I got a very snotty email back from [a Nomad member] telling me that Microsoft was a very nasty company and that I should be using free software! I would have loved to use free software but I had no idea how to use it. I felt like saying, ‘Fine! Come and install Linux on my computer because I don’t know how’. Needless to say I never bothered to send another email.
If the face-to-face comments of some of the Nomads in assembly could be interpreted by some activists as insulting or provocative, their impact was magnified in their written e-mail communications. In the cases of Adela and Darla, their response was to either leave the group or to stop participating in e-mail exchanges. Cronaeur (Reference Cronauer2004) found that less integration on and offline increased the impact of negative online exchanges. While Adela and Darla felt alienated from the group, they were both experienced, well integrated, and respected in the wider network.
Problems noted in the literature about the dominance of few participants, selective responding, and power imbalances related to list control and moderation (Cronauer, Reference Cronauer2004; Janssen and Kies, Reference Janssen and Kies2005; Albrecht, Reference Albrecht2006; Kavada, Reference Kavada2007) were also present. Access to the e-mail list was controlled by members of the Nomad University. There were a few bitter discussions over access on the part of the moderators (who wanted to limit access to those people who had been formally introduced and vetted by other members of the assembly who were known to them personally and could vouch for them). This raised the intimidation factor significantly, and some people did not feel comfortable asking for access to the list. Therefore, the list became an exclusive forum for those with access, creating a group within the larger group. But even those with access were frequently treated to e-mails with a hostile content, which, even when not directed specifically at the reader, spilled over to contaminate the atmosphere of the assembly.
As free software ‘experts’, the Nomads also had a technological expertise that somewhat ‘legitimized’ their control of the e-mail list. This kind of hierarchical nuclei developing around technological expertise has been noted in other social movement groups (Pickerill, Reference Pickerill2004), as well as the genuine frustration experienced by more technologically savvy activists (Juris et al., Reference Juris, Caruso, Couture and Mosca2013), which was likely to have affected those Nomads strongly committed to free software. As with the CSE, in the Disobedience Lab those with more time and access were able to participate more actively in the e-mail lists, and elaborate proposals that would be circulated before meetings, which would effectively ‘set’ the agenda in a framework they had designed a priori.
E-mail research on organizations also shows that ‘email has the potential to allow various strategic and political manipulations of information thereby affecting the organizations’ structure in terms of power and control’ (Ducheneaut and Watts, Reference Ducheneaut and Watts2005: 21). One Nomad strategy for hegemony, acting in concert within assembly, was also pursued electronically, where one person would write an opinion relating to a decision taken in assembly that would then be supported by three or four other ‘individual’ opinions, thus shifting the discussion out of the assembly and into a forum they controlled (as moderators and active users of the list). The e-mail list was thus used to create false consensus around issues that had already been discussed in assembly.
One discussion related to security issues for a direct action being planned by the group against the Telefónica Corporation in December of 2002. In the assembly, one member experienced with direct action, Adela, had advised taking extra security precautions and had been supported by a number of other people. One of the Nomads then sent an e-mail ridiculing Adela, arguing that her proposed security measures were ‘worthy of a Chechen commando unit’ and characterizing her as ‘hysterical’. This e-mail was then supported by a number of e-mails from friends of the dissenter, leading to a false impression that most people did not agree with the more cautious approach. Adela (and others) felt the sovereignty of the assembly had been overridden illegitimately and that the attack on her had been sexist as well. She wrote back expressing this and asking ‘Would you have called a male colleague hysterical?’. When she received no reply she decided to leave the group. The incident caused one of the first major excisions from the Disobedience Lab, as 10 people (about half of a typical assembly) left with her.
It is tempting to conclude that because the Disobedience Lab was perceived as a combative space this has exceptional implications for the nature of the findings of the impact of the e-mail exchanges. Yet, problems with e-mail use were not limited to negativity or flaming, but also included the use of abstract language, strategic manipulation, false consensus, and agenda setting. Crucially, e-mail caused problems for both groups despite the differences in the perceptions of face-to-face assemblies. Although the Disobedience Lab was indeed a conflictual space, it was not unusually so for the Madrid network, and as Kavada (Reference Kavada2007: 235) argues, internal conflicts are almost unavoidable in social movements. If anything, the CSE was remarkable for the positive and non-confrontational atmosphere it created. As with the CSE, there was also a gender dimension to the perception of e-mail being problematic, although in this case there was also a greater imbalance within the assembly, with more men dominating discussion and informal leadership and with female activists interpreting certain behaviours as sexist. In neither group were the problems generated by reinforcement e-mails ever addressed collectively in assembly.
Discussion
Most e-mail research centres primarily on the content or frequency of online posts, causing the methodological problem of being unable to capture non-participation (Tufekci, Reference Tufekci2014), a problem mitigated somewhat by supplementing the online data with interviews or survey data (Cronaeur, Reference Cronauer2004; Kavada, Reference Kavada2007, Reference Kavada2009, Reference Kavada2010; Wall, Reference Wall2007; Gillan et al., Reference Gillan, Pickerill and Webster2008). The ethnographic method employed in this study allows for general observations in the literature to be subjected to closer scrutiny, and a stronger focus, as well as allowing for a study of e-mail’s effects on group cohesion over time. What these case studies suggest is that broad generalizations about ICTs and activism need to be contextualized significantly, and are likely to be more contingent than universal, despite some inbuilt affordances inherent to e-mail design. Nevertheless, some possible effects seem to transcend types of groups and lists, and should be noted, given their importance for group cohesion and collective identity.
E-mail, collective identity, and participation in face-to-face assemblies
My findings differ from a series of key claims made in one of the only major studies on e-mail and on and offline collective identity in social movements. Kavada’s work is much more positive and optimistic about the effect of e-mail on participation and collective identity than either Cronauer (Reference Cronauer2004) or the case studies I present here would lead us to expect. It is therefore worth reflecting on why this might be so.
In her study of e-mail use in the Global Justice Movement, Kavada (Reference Kavada2007: 210) found that e-mail facilitates the integration of activists who are less committed or miss face-to-face meetings, and ‘help to open up offline meetings to the direct or indirect participation of activists who would otherwise be excluded from decision making’. This was clearly not the case in the groups I studied, where e-mail exacerbated a sense of marginalization and alienation related to missed meetings. Kavada (Reference Kavada2007: 234) also argues that ‘face-to-face contact serves as an antidote to email conflict since online disputes tend to be settled when activists meet face-to-face’. I did not find this to be the case in either group, where people did not raise e-mail-related issues in the assemblies, or seek to resolve them there directly, sometimes opting to leave the group instead.
Kavada (2006: 235) also suggests that ‘online conflicts can in fact be beneficial… because they constitute a way of letting off some steam which may prevent conflict from developing face-to-face where it is much more hurtful’. My study suggests the opposite. Whereas face-to-face conflict allows for a process of dialogue and exchange in real time and affords the sociability cues that serve to save face, e-mail removes all of those elements, and the social isolation of the recipient may make personal attacks feel more hurtful rather than less hurtful. My findings are more consistent with research on e-mail use in organizations and online deliberation (Kiesler and Sproull, Reference Kiesler and Sproull1992; Albrecht, Reference Albrecht2006) and on the effects of social isolation (Lea and Spears, Reference Lea and Spears1991).
Reinforcement e-mails and routinization
Horton’s (Reference Horton2004) distinction between reinforcement e-mails and outreach e-mails enables us to distinguish between e-mails used in a routinized way that is integrated into the daily routines of activists and outreach e-mails that are directed to outsiders of the groups and, therefore, much more likely to be thought of reflexively. Reinforcement e-mails continue to be crucial mechanisms for internal communication in social movement groups. Although basic forms of netiquette or rules for their use are established in early assemblies, rarely do activists reflect deeply on the potential impact of their use. Instead they are viewed as simply a tool for coordinating or communicating between assemblies. In these cases, reinforcement e-mails were reflected upon in interviews, but never raised as an issue in assembly. This pattern of non-reflexivity regarding taken for granted or routinized exchanges in activist communities has been documented in other areas of interaction and communication, such as assembly methods and protocols (Flesher Fominaya, Reference Flesher Fominaya2005, 2014Reference Flesher Fominayac) and with regard to the integration of new technologies into daily life (Lovink, Reference Lovink2011). Of the many ICTs used by activists, e-mail is the one most integrated into routines outside the activist milieu, increasing its normalization and routinization, and thereby likely decreasing reflexivity about its use (Lovink, Reference Lovink2011). E-mail (unlike Twitter, for example), especially reinforcement e-mail, is not seen as a special or strategic form of CMC but rather forms part of activists’ habitus.
Movement effects vs. group effects: scale matters
While I do not argue that they account for all the differences, the nature of the e-mail lists and the groups I have studied here undoubtedly affect the difference in these findings, as organizational research on e-mail suggests (Ducheneaut and Watts, Reference Ducheneaut and Watts2005). Kavada’s lists were large, open, and the users geographically dispersed, and the socio-organizational contexts in which they were embedded were also different, with only some participants infrequently meeting face to face. The lists also had a clear raison d’être fixed in time (coordinating a meeting). The synergistic effects she notes between e-mail and offline participation may indeed play out at movement level. In smaller groups the impact of individual actions becomes greater than in large-scale communicative processes (Albrecht, Reference Albrecht2006).
While Kavada (Reference Kavada2007) identifies a series of potential power imbalances related from e-mail use and recognizes these as problems for particular groups, she does not see them as a problem for the movement as a whole because of the ease with which new lists, groups, and informal leaderships spring up. Kavada (Reference Kavada2007: 207) argues that ‘email lists tend to facilitate the polycentric character of the movements as they do not support long term, unitary and centralized leadership’. While this may be true at transnational/movement level, it is not the case at local network level. Movements as a whole are made up of groups often embedded in particular local networks, where informal leadership structures may be quite stable, and where the costs of creating new functioning groups can be high in terms of trust, energy, and other resources.
In a network like Madrid’s at the time of this study, resources were limited and starting a new group required considerable effort and personal investments in a process fraught with challenges. Group failures can also have lasting and damaging effects on networks, with implications for effective mobilization at movement level. I would therefore suggest that e-mail-related problems are potentially more damaging for movements than Kavada’s more optimistic assessment suggests. Likewise, Horton’s (Reference Horton2004) useful classification of internal e-mail in social movement groups as ‘reinforcement e-mail’ that stresses the positive reinforcements of pre-existing ties, commitment, and belonging needs to be opened to the possibility of its role in reinforcing negative dynamics in face-to-face groups, and possibly magnifying them.
Is smaller and face to face better?
Despite her optimism, Kavada, like Cronauer, recognizes that e-mail can work against collective identity formation owing to problems with trust, solidarity, and conflict. Both suggest that smaller lists and face-to-face contact might offset these problems. In this case, I started with two groups in the same network that we could expect to be the most resilient to the negative effects of e-mail on participation, collective decision making, and collective identity formation, given their explicit political commitment to combating hierarchies, power imbalances, and vertical decision making, and their pre-existing face-to-face ties. Kavada (Reference Kavada2007) and Cronauer’s (Reference Cronauer2004) hypotheses that smaller face-to-face groups with strong ties were likely to offset the negative impacts of e-mail use on collective identity formation would lead us to expect the impact of e-mail on the CSE to be negligible and the impact on the Disobedience Lab to be marked, but this is not the case. It is precisely the difference between the groups’ face-to-face dynamic (one overwhelmingly perceived as positive and one overwhelmingly perceived as negative) that makes the commonality between the effects of e-mail across the two cases so striking. Why then did e-mail work against the realization of group ideals and group cohesion, despite a firm commitment to participation and horizontality?
Increased participation by the few leads to decreased participation by the many
Contrary to what Kavada and Cronauer suggest as a possible solution to the problems they found, the use of smaller closed ‘reinforcement’ e-mail did not resolve the parallel between limited offline and online involvement, nor that those less-involved offline were also more alienated by hostile online interactions (Cronauer, Reference Cronauer2004). Cronauer (Reference Cronauer2004: 78) found that e-mail participation was highly skewed, subscribers interacted selectively, and subscribers whose posts received few responses were reluctant to post, which was perceived as problematic by the other subscribers but not by the posters. Subscribers attributed more weight to posts from people they respected or viewed as authorities (Cronauer, Reference Cronauer2004: 78). Subscribers’ reluctance to contribute to discussions was owing to the fact that they did not consider themselves to be authorities; involved enough for their opinion to ‘count’; what they wanted to say had already been said; and intimidation owing to responses to earlier posts that ‘shot people down’. These responses are strikingly similar to those I found on these lists. Mario’s helpful contributions to the list actually decreased participation, despite his good intentions. These participation effects were not mitigated by strong face-to-face ties in the CSE case.
Reflection and magnification
Consistent with Kavada (Reference Kavada2007) and Cronauer (Reference Cronauer2004) the dynamics of participation online closely mirrored those of the social relations in the assemblies. In general, those most active in the assembly were also most active on the lists; the people who exercised de facto leadership within assembly also felt entitled to unilaterally change meeting times, or take other decisions that supplanted the sovereignty of the assembly; and those who engaged in conflictive exchanges in assembly also did so online. The argument that the internet reflects power imbalances in ‘real’ life has been made persuasively (Lovink, Reference Lovink2011; Taylor, Reference Taylor2014) and status equalization is unlikely to take place in organizational settings (Saunders et al., Reference Saunders, Robey and Vaverek1994). Even more striking than this important claim is my finding that the impacts and consequences of these behaviours were magnified when conducted via e-mail, rather than in assembly, for a number of reasons that are worth highlighting, as they lie at the heart of e-mail communication challenges, and are particularly relevant for groups seeking to implement ‘horizontal participatory’ practices. Insights from research on e-mail use in organizations and online deliberation is helpful here in examining the specific problems faced by these groups.
Applying insights from research on e-mail use in organizations and online deliberation to explain magnification effects
Social isolation and asynchronicity
Social–psychological research shows that social isolation can reduce the external restraining or mediating cues, increasing flaming and uninhibited behaviour online (Lea and Spears, Reference Lea and Spears1991; McKenna and Bargh, Reference McKenna and Bargh2000). Yet, social isolation can also increase the impact of the interpretation of e-mail messages by the recipient, as Darla’s response above shows. Social isolation is further affected by the time/space disjuncture or asynchronous nature of e-mail communication. A message can be sent and not receive a reply until days later, if at all. The person(s) to whom it was addressed (in the case of a personal attack, for example), may not even have seen the message or been able to respond to it before it has been seen and reacted to by others on the discussion list, as happened with Darla and Fritzi. The absence of social cues, including body language and tone of voice, can also increase misunderstandings and heighten feelings of hostility. While e-mail’s asynchronicity can facilitate more measured, polished responses, it clearly does not do so in all cases (McKenna and Bargh, Reference McKenna and Bargh2000). Asynchronicity is often stressed as a crucial benefit for e-mail collaboration in social movements (e.g. Earl and Kimport, Reference Earl and Kimport2011), yet studies such as this, that integrate insights into the subjective experience of ICT use, suggest we need to re-evaluate claims based primarily on technological affordances.
The authority of the written word
As we can see from the effects of Marios’ e-mails in the CSE and the e-mails sent by the Nomad University’s hegemonic nucleus, informal leadership, and power imbalances are affected by the text-based nature of e-mail, owing to the power and permanence of the written word and the authority it conveys, and unequal access, skills, and control of electronic communication. While the issues of impact of unequal skills are hotly debated within some movement communities, including within the Global Justice Movement (Juris et al., Reference Juris, Caruso, Couture and Mosca2013), rarely do these debates actually incorporate insights about the subjective nature of the forms of communication, focusing instead primarily on technical aspects of their use or the division of labour (i.e. mastery of technology itself and how this is tied to which roles are adopted within groups).
E-mail can facilitate false consensus
The ease with which the public sphere can be extended into the virtual sphere, seen as facilitating democratic participatory processes in research on online political forums (Dahlberg, Reference Dahlberg2001), can be problematic for groups committed to the sovereignty of the face-to-face assembly. As demonstrated in both of these cases, particularly problematic is the ease with which false consensus can be reached via e-mail, a problem facilitated by the trend towards dominance of few participants and selective responding in online communication (Cronauer, Reference Cronauer2004; Janssen and Kies, Reference Janssen and Kies2005; Albrecht, Reference Albrecht2006).
The gender gap
The gender gap noted in other studies (Gürer and Camp, Reference Gürer and Camp2002; Cronauer, Reference Cronauer2004; Albrecht, Reference Albrecht2006; Cooper, Reference Cooper2006; Hargittai and Shafer, Reference Hargittai and Shafer2006) was also borne out here in terms of frequency and type of participation. Like Cronaeur (Reference Cronauer2004) I found that women in the group were less likely to feel confident about their own authority with regard to posting or objecting openly to decisions they felt badly about, instead being reluctant to participate online and internalizing feelings of frustration and alienation. Such gender disparities clearly work against the goals of groups committed to egalitarian participation.
All of these e-mail-related factors work to magnify the impact of face-to-face behaviour and power imbalances, to diminish the positive evaluation of the group experience by members, and to impede the realization of the goals for increased egalitarian participation, which in turn can affect collective identity processes, group cohesion, and sustainability over time.
Conclusion
Research shows that e-mail is highly malleable and its use will be significantly affected by the characteristics of the socio-organizational context in which it is embedded (Ducheneaut and Watts, Reference Ducheneaut and Watts2005), yet it does have specific characteristics with important potential implications for its use. Research on e-mail use in specific organizational settings shows that scholars (and activists) should be attentive to the specific properties of e-mail and how they can affect communication processes, yet without falling into technologically determinist assumptions about its potential benefits or limitations. Instead, insights from research on computer–human interaction, especially research conducted in organizational settings, can only enrich our understanding of the relation between communication media and social movement dynamics.
Building on earlier work (Cronaeur, Reference Cronauer2004; Horton, Reference Horton2004; Kavada, 2006, Reference Kavada2009) this research has drawn a connection between e-mail use in social movement groups and its unintended impact on group cohesion and collective identity processes. Central to collective identity processes at group level are reciprocal exchanges that develop trust, solidarity, feelings of belonging, and commitment to shared goals and actions (Flesher Fominaya, Reference Flesher Fominaya2010a). Research shows that participant evaluations of online deliberations have important implications for perceptions of decision ambiguity and legitimacy, which in turn conditions expectations of future participation (Stromer-Galley and Muhlberger, Reference Stromer-Galley and Muhlberger2009). Crucially, e-mail communication takes place as a series of social acts that are regulated by organizational norms and take on meaning within organizational contexts (Ducheneaut and Watts, Reference Ducheneaut and Watts2005). For groups in which ‘horizontal’ practices are central to collective identity, evaluations of online exchanges will be conditioned by perceptions of the degree to which such deliberations meet the group ideals for inclusiveness and participation, as well as respect group norms (such as respecting the sovereignty of the assembly).
These cases and others (Cronauer, Reference Cronauer2004; Grignou and Patou, Reference Grignou and Patou2004; Kavada, 2006) show that despite a commitment to ‘horizontalism’ autonomous groups still reflect trends in the wider population of the dominance of participation by a small core of users selectively responding to each other, and significant gender differences. This suggests that political ideology might have a limited effect on participant use of lists (e.g. posting or lack of posting, patterns of exchange) but is likely to affect the evaluation of such exchanges as problematic if they fail to live up to perceived shared political ideals, which is likely to be detrimental to collective identity processes.
Like Dahlberg (Reference Dahlberg2001), I found a low degree of reflexivity about the impact of e-mail in these two groups, despite one group holding a techno-sceptical view overall. It may be that the routinization of e-mail as a means of communication in activists’ daily lives and a presumed shared commitment to horizontal principles actually works to decrease reflexivity about its use. I suggest that the more integrated ICT’s are in our daily lives, the less they will be seen as explicit tools for activism and the less reflexive activists will tend to be about their use. Just as the white goods that once revolutionized daily living continue to be used but cease to be the subject of conversation (Lovink, Reference Lovink2011: 36), the e-mail lists that I focus most of my discussion on here, reinforcement e-mails, are used routinely but not strategically or reflexively. Just as we only notice the washing machine when it breaks down, reinforcement e-mail only becomes ‘visible’ to the group when there is a crisis or a rupture in its use, and by then the damage may be done.
Routinization is not the whole picture, however, because Facebook and Twitter, for example, are also increasingly used in ways that are integrated into daily routines outside of the activist milieu. Equally important is the distinction I make here between social movements’ internal ICT-facilitated communication and external communication use of ICTs for outreach and mobilization. As Horton (Reference Horton2004) argues, reinforcement e-mail lists used for internal communication are different by nature and intention to e-mail and social media that are used to communicate with audiences beyond the group. The latter likely form part of an explicit communication strategy and are therefore much more likely to be subjected to debate and reflexivity about their use.
Although conflict and flaming are understandably highlighted as being among the most negative problems for e-mail use in social movement groups, the CSE case shows that even in the absence of conflict other factors such as the length, frequency, style (e.g. ‘dense’, ‘abstract’, ‘elitist’), and timing of interventions, had unintended negative consequences. Likewise, negativity was by no means the only e-mail-related problem in the Disobedience Lab. These aspects of e-mail use were not only problematic from a rational perspective but carried deep emotional weight, adding to some activists’ sense of marginalization and alienation and affecting their desire to continue participating, negatively affecting group collective identity.
These findings highlight the need to distinguish between structure and content (Downing, Reference Downing2008) and content and meaning (Ganesh and Stohl, Reference Ganesh and Stohl2010). As Ganesh and Stohl (Reference Ganesh and Stohl2010) point out, many studies on ICTs and social movements focus on the frequency, sources, and reception of information, relying on closed quantitative data that ignores the interpretation of that information by activists. In a study on internet use and individual participation, Bimber (Reference Bimber2001) found that theories linking information availability and levels of engagement were not substantiated, and concluded that affective, social, and cognitive aspects would have a greater impact on participation, including social networks and mobilization. Without information about how activists themselves interpret and struggle with ICT use, certain assumptions about the ways ICTs facilitate or hinder social movements, such as participation, cannot be evaluated. In order to evaluate online participation, for example, it is also necessary to pay attention to factors that inhibit participation, something not possible to determine from a structural and content analysis of web posts or e-mail lists themselves. This is particularly relevant for those ‘horizontal’ social movements, such as 15-M, Indignados, or Occupy-style movements, in which participation in itself (as means and end) is central to collective identity.
As the organizational literature suggests, the effectiveness of e-mail for reaching consensus or taking quality decisions is limited in the absence of clear roles of authority, responsibility, accountability, and in the absence of mechanisms of social control (Ducheneaut and Watts, Reference Ducheneaut and Watts2005). This is also consistent with Dahlberg’s (Reference Dahlberg2001) suggestion that e-mail lists for political discussion are most effective when clear rules about their use and (self) moderation are in place. The extent to which autonomous ‘horizontal’ groups are willing and able to adopt effective rules is an open question. Greater reflexivity and more explicit rules about deliberation and decision making could have helped lessen the unintended negative consequences. Yet, even when specific protocols are in place, e-mail exchanges can still have significant impacts on group collective identity, contributing to decline or dissolution.
Given the strong emotional impact of some e-mails on individuals in the groups discussed here, and their consequent effect on group participation, social movement scholars would benefit from paying more attention to the social-psychological aspects of CMC and the relation between ICT use, emotions, and collective identity processes within social movement groups.
As research methods to study online activism become increasingly sophisticated, many scholars understandably embrace the possibilities that large network-based studies offer to examine new media, connectivity, and online participation. However, given the continued importance of face-to-face groups for key aspects of social movement mobilization and cohesion, qualitative studies that examine the relationship and interaction between ICT use, face-to-face group dynamics, and the implications for democratic deliberation, participation, and collective identity are also needed. While analysing the strategic uses of new media is an important task, scholars should not overlook the importance of internal routinized forms of communication for social movement dynamics.
Finally, given the interpenetration of e-mail use in the daily communication of a plethora of civil society groups today, it is likely that the findings highlighted here have implications beyond autonomous social movement groups and may be useful in other studies of organizational communication to better understand the interaction between technology use and processes of group cohesion and commitment in organizational settings. I am not suggesting that e-mail use will necessarily lead to these negative or unintended consequences. Lack of awareness and reflexivity about these possible effects, however, will certainly increase that likelihood.
Acknowledgements
The author would like to thank all the activists involved in this research. The author also thanks the editors and five anonymous reviewers of this article who greatly improved the final product. This research was funded by the German Marshall Foundation Fellowship, The John L. Simpson Foundation Fellowship, and the European Union Marie Sklodowska Curie Senior Fellowship, for which I am grateful. Thanks also to Bart Cammaerts and Kevin Gillan for suggesting relevant literature, to Antonio Montañés Jiménez for help with literature searches, and to Laurence Cox and Graeme Hayes for feedback.
