The argument between Penny and Callinicos is illustrative of much of the public debate about the impact of social media on contemporary social movements. In Penny’s interventions we encounter a series of key ideas which have come to characterise the way in which contemporary activists understand the shift from ‘old’ to ‘new’ forms of collective action facilitated by technological innovation. Because of the availability of contemporary social media, activists like Penny argue, social movements can finally become leaderless, horizontal, and spontaneous. They no longer have to face the question of unity so obsessively important for dinosaurs like the Socialist Workers Party, which seems to be stuck in the era of Gutenberg. Much of the scholarship produced in the last few years about the impact of new media on contemporary society, and on activism in particular, has followed a similar line of reasoning. It asserts that the internet allows for more flexible relationships, enabling individuals to interact without the need for central coordination or a sense of unity in the display of collective action. This narrative is also exemplified by Paul Mason’s account of contemporary protest movements in which he describes them as ‘networked’ and activists as ‘horizontalists’. In his book Why It’s Kicking Off Everywhere (2012), there are 118 instances of the words ‘network’ or ‘networked’. There is little doubt that the idea of networked protest is reaching its climax in the debate about contemporary activism. But is this really the best image with which to capture the dynamics of the mobilisation of contemporary social movements and the role played by social media in this process?

As someone who considers himself a leftist libertarian and who has participated in direct action campaigns, I have little doubt over whom to sympathise with between Penny and Mason or Callinicos. Yet, after years of hearing the terms ‘horizontal’, ‘open’ and ‘networked’ floated around in activist circles, I have grown increasingly suspicious not only of them but of the whole ideology of ‘horizontalism’ (Juris, 2008). I am convinced that this idea tends to obscure the forms of organising underlying contemporary collective action and the forms of hierarchy, or the ‘hierarchy of engagement’ (Haunss and Leach, 2009), which continue to exist also within informal organisations like contemporary social movements (Freeman, 1972). The ideology of horizontalism obscures the fact that the process of mobilisation is constitutively ridden with imbalances and asymmetrical relationships between those who mobilise and those mobilised, between those leading the process and those following (Melucci, 1996a: 345). Moreover, this idea returns to an image of collective action as a static process, and thus overlooks the dynamic character which we associate with the concept of social movements, the fact that they are ‘things that move’, as evident in the etymology of this sociological notion across a number of languages from Arabic to German and English.

But there is a more fundamental critique which needs to be directed at the imaginary of horizontalism with its emphasis on de-centralisation and irreducible multiplicity. The critique turns on the fact that the process of mobilisation chiefly involves a process of gathering or assembling of individuals and groups around something they share in common. While this feature is arguably common to all social movements, it is particularly important in the case of popular movements, given their attempt to mobilise a diverse and dispersed constituency under the name of the ‘people’. This aspect has been spectacularly illustrated by the movements of 2011, which have constructed long-term mass sit-ins resembling rituals of popular reunion, in which a dispersed constituency is ‘fused’ (Alexander et al., 2006) into a collective actor. While emphasising multiplicity, network theorists have neglected the continuing importance of the question of unity and togetherness among participants. It is my contention that this question continues to be as relevant as ever in the era of social media. If anything, the multiplication of communicative channels and the individualisation of our mediated interactions, epitomised by the popularity of social media, make this question all the more urgent.

In this chapter I want to engage critically with this libertarian discourse of ‘horizontalism’, while developing a conceptual framework for analysing the role of social media in the process of mobilisation. The starting point is a discussion of the two main concepts which have informed this discourse: the metaphors of ‘network’ and ‘swarm’, as employed by, respectively, Manuel Castells (1996, 2009) and Hardt and Negri (2000, 2004, 2009). These authors correctly identify a condition of dispersion as the fundamental feature of spatial experience in post-industrial societies. They are opposed to the industrial imaginary of the mass and the crowd, and aim at advancing new subjectivities which escape the reduction ad unum: the fusion of individuals into a collective actor. The limit of this theoretical lineage is that it accepts dispersion and individualism as constitutive dimensions of contemporary society, rather than as the point of departure in the process of construction of collective action. The risks we face in a society of network and multitudes are made visible by the dispersion and seclusion which dominates the urban landscape, and by the danger of isolation inherent in social media, with their tendency to exacerbate the dynamics of social fragmentation. In and of themselves social media do not automatically allow for collective action to unfold without becoming channels for the construction of common identities and thick networks of solidarity and trust.

Building on the work of an array of authors including Zygmunt Bauman (2000, 2001), Hannah Arendt (1958), Alberto Melucci (1996), Ernesto Laclau and Chantal Mouffe (Laclau, 1996, 2005; Laclau and Mouffe, 1985), I develop an alternative understanding of the process of mobilisation, based on the notion of ‘assembling’ or ‘gathering’ rather than ‘networking’. Nevertheless, the spatial dispersion which characterises contemporary social space with its ‘fear of crowds’ (Davis, 1992a) makes this process of gathering particularly problematic and requires complex practices of symbolic and technological mediation. In conceptualising this process of mediation of physical assembling through the notion of a choreography of assembly, I put forward the hypothesis that collective action is never completely spontaneous given that pure spontaneity does not exist (Gramsci, 1971: 196). Rather, in the absence of a formal organisational structure, collective action is always structured by the forms of communication responsible for ‘setting the scene’ for its display.

In essence, the thesis proposed by Castells asserts an historical shift from the pyramidal structures characteristic of bureaucratic organisations – the company, the party, the state – to networks. For Castells the ‘solid’ and ‘rigid’ economic, social and political institutions of mass society, well described by Max Weber, have given way to more flexible and adaptable structures. This is first and foremost the consequence of technological innovation. The revolution in micro-electronics, beginning in the 1960s, created the necessary conditions for new forms of communication and cooperation which no longer required central coordination (Castells, 2000). Such societal shifts invest different social activities: from the economy, to social movements, to drug trafficking, the whole of society is restructured after the model of networks (Castells, 1996, 2000).

This account of the development of network technologies also has a bearing on the working of the so-called web 2.0. Social media in particular are characterised by a high degree of interactivity, and by a focus on user-generated content. Practically speaking, this means that users are also to a great extent ‘producers’ in communicative interactions. Social media typify the nature of the ‘participatory culture’ Henry Jenkins suggests is an underlying feature of the contemporary media landscape, in which people are no longer simply positioned at the receiving end of processes of communication (Jenkins, 2006). Castells has described this media landscape as dominated by a paradigm of ‘self mass-communication’ in which individuals and groups can broadcast their messages to large audiences (Castells, 2009: 416). For Castells, the advent of mass self-communication carries the promise of autonomy from bureaucratic structures and increasing scope for political and social engagement from below.

This evolutionary narrative is evidently coloured by an anti-authoritarian spirit. Castells’ discussion resonates deeply with the emphasis on self-determination and self-management put forward by the cultural movements of the 1960s and ’70s. Castells himself notes that, apart from the rise of new technologies, the ‘networking paradigm’ was informed by the libertarian and participatory culture inaugurated by new social movements such as environmentalism, feminism, and the student movement (Castells, 2004). Crucial in this context is the emphasis on horizontality and decentralisation, since ‘by definition a network has no center’ (Castells, 2000: 15). For Castells, these new forms of networked cooperation emancipate social groups from the top-down logic of command and from the need for leaders.

Given this anti-authoritarian twist, it is not surprising that the language of networks came to be enthusiastically adopted by activist groups within the emerging anti-globalisation (or more positively ‘alter-globalisation’) movement. David Graeber was probably right when he observed that even though many in the movement would not have defined themselves as anarchists, nevertheless ‘anarchism is the heart of the movement, its soul; the source of most of what’s new and hopeful about it’ (Graeber, 2002: 2). Since its inception with the protests in Seattle in 1999, the anti-globalisation movement was marked by a libertarian emphasis on self-organisation and direct action. For these activists, the imaginary of networks came to provide a useful term of reference for defining flexible and anti-hierarchical forms of organisation, at a time marked by the diffusion of the internet as a major platform for protest communications.

Emails, listservs, Indymedia websites, and web-forums became the communicative toolkit of a ‘new way of doing politics’, whose fundamental logic was ‘networking’ (Juris, 2008). In a 2002 article published in the New Left Review, the Canadian campaigning journalist Naomi Klein described the relationship between the movement and the internet in the following terms:

The level of popularity of the network paradigm within the anti-globalisation movement can be appreciated by reading the ethnographic account produced by Jeffrey Juris, himself a student of Manuel Castells. Developing his analysis from the standpoint of activist groups in Barcelona, and analysing the protests in Prague and Genoa, Juris affirms the ‘networking logic’ at the core of the politics of these new movements, who practice a leaderless politics based on consensual decision-making and participation. Crucial for sustaining these values is an investment in the production and circulation of information, as testified by practices such as ‘Indymedia, culture jamming, guerrilla communication, and electronic civil disobedience’ (Juris, 2008: 284). For Juris, ‘expanding and diversifying networks is more than a concrete organisational objective; it is also a highly valued political goal’. Net...