Conceptualizing the context of collective action: an introduction

ABSTRACT

What is ‘context’ and why does it matter? What are the different ways of conceptualizing the context of collective action? What are the trade-offs involved in choosing one conceptualization over another? With which explanatory strategies are these conceptualizations associated? This introduction addresses these questions and outlines the three conceptualizations of the relational context of collective action discussed in this special issue: field, space, and arena. In spite of some critiques, it argues that these three meso-level conceptualizations depart from a purely ‘movement-centric’ perspective and offer promising leads for analyzing the relationship between collective action and its environment.

On 6 January 2021, after listening to President Trump and several of his associates denounce the 3 November 2020, presidential election as fraudulent and illegitimate, several thousand Trump supporters marched toward the United States Capitol, in Washington, DC, chanting ‘Stop the Steal’ and demanding that the result of the election be denied certification so that Trump could remain president for a second term. The protesters quickly breached the fragile security perimeter established by the police, climbed walls, broke doors and windows, and managed to enter the Capitol. Inside the building, while some protesters seemed surprised to be there and walked around like tourists, others apparently had a plan and searched for specific Democratic representatives and senators to detain them. They were eventually forced out of the building by the police and hung around until the curfew went in place, when they finally dispersed. In addition to leaving five people dead and many others injured, the event, that took place on live TV and social media, shook the foundations of the American political regime to the extent that many commentators called it a ‘seditious insurrection’ and ‘a coup attempt.’ A week later, on January 13, the United States House of Representatives voted to impeach President Trump on grounds that he had incited a violent insurrection and violated the Constitution.

Is it possible to explain this event in abstract terms and on the basis of general causal laws, without referring to the context? Can we do so without reference to the polarized political and cultural context generated by the Trump presidency? Is it appropriate to ignore the social media environment, from Twitter to Parler, that has allowed disinformation and conspiracy theories to spread like fire and protesters to coordinate at a distance? Do we need to consider the pandemic as well as the economic and employment crises it has produced in 2020? Inversely, if we do take into account such contextual features, where do we stop? Do we run the risk of drawing an ad hoc picture so idiosyncratic that we will not be able to compare it to other violent protests or insurrections? And ultimately, if the context matters so much to explain the event, is it still just ‘context’ or has it become a cause in its own right?

The papers assembled in this special issue address some of these questions by introducing different ways of conceptualizing the relational context of collective action and social movements and assessing their value added. Most of the papers were presented at a workshop that I organized in Montreal in October 2017 and which brought together social movement scholars from Canada, France, and the United States.¹ In this sense, this special issue literally lies at the crossroads of transatlantic debates.

In this introduction, I first explain what ‘context’ is as well as why and how it matters. I then present different ways of conceptualizing the ‘context’ of collective action, namely, in socio-economic, cultural, and historical terms. Finally, I outline the main conceptualizations put forth in this special issue – field, space, and arena – and explain in what respect they converge and diverge. Although they could be compatible in some regard, they rely on different premises and point to distinctive accounts. At the same time, however, all the papers assembled in this special issue go beyond ad hoc descriptions of the context and provide us with insightful conceptual tools for analyzing in systematic terms how social movement dynamics are embedded in their environment. In this respect, this special issue ultimately aspires to contributing to a broadening of existing debates beyond movement-centric perspectives and building bridges to other sub-fields in sociology and political science.

What is context?

As abstract categories, concepts guide our gaze and have a heuristic value (Siméant-Germanos, this issue). They allow us to classify and order different aspects of social reality and to communicate to others fundamental properties of this reality in synthetic terms. They are the basic tool of social sciences and the building blocks of social theories. It follows that conceptualizing the context of collective action necessarily involves a series of ontological propositions about the world and leads us to raise certain questions rather than others. However, ‘context’ is also one of these folk concepts that we commonly use in social sciences without much thought. We often refer to the ‘context’ of an event or ‘contextualize’ that event without explaining on what grounds we highlight certain features and omit others, and without spelling out the analytical status of this ‘context.’

In many respects, ‘context’ is ambiguous and vague. Ambiguous because it lacks a clear meaning and vague because it lacks clear boundaries. ‘Context’ generally means the setting, environment, or surroundings in which a given event or phenomenon takes place. However, this environment can be defined in many terms and involve several different dimensions: organizational population, institutions, geography, culture, circumstances, historical period, etc. Regardless of the dimension we decide to emphasize, we assume that such emphasis is necessary to understand what is going on. It follows that ‘contextualizing’ entails providing enough background information to grasp the meaning and significance of a given event or phenomenon. Such work of contextualization requires that we clearly distinguish the event or phenomenon itself from the context in which it is situated (R. B. Collier & Mazzuca, 2006, p. 474). This specification is necessary to identify the main relevant properties of the context in question, recognize it when we see it, build comparisons, and determine whether it is unique or widespread as well as associated with particular patterns and outcomes or not.

In conceptualizing the context, we often presume that other contexts are similar to our own. As Tilly and Goodin (2006, p. 21) put it:

In trying to make sense of the social world, we tend (at least as a first approximation) to impute to others broadly the same sort of psychology, broadly the same sort of beliefs and desires, that we ourselves possess. Not only are we ‘folk psychologists’ (…); we are also ‘folk situationalists,’ assuming (until further investigation reveals otherwise) that the context in which others are acting is broadly the same as our own.

It follows that we need to distance ourselves from our own context and problematize it to avoid behaving like ‘folk situationalists.’ For example, conceptualizations of context in terms of field, social space, or sector, are not applicable to all places and historical periods insofar as they presume a high level of differentiation between relatively autonomous social spheres generally associated with Western modernity. When thinking about the context of collective action and social movements, we should thus try to avoid as much as possible the pitfalls of both ethnocentrism and ahistoricism.

When doing so, we face the challenge of determining the right level of abstraction for sound and effective comparisons across time and space. If we conceptualize the context of collective action in very specific and concrete terms, to the extent that it becomes an idiosyncratic description, the concept will neither be applicable to other cases nor yield significant epistemic gains. If we try and apply it to a wide range of cases anyways, we will engage in what Sartori has called ‘concept stretching’ (Sartori, 1970, p. 1034). To avoid ‘stretching’ the concept to fit in more cases, we need to move up the ‘ladder of abstraction’ and make our conceptualization of context more abstract, with fewer properties or attributes (Sartori, 1970, p. 1041), so that it can ‘travel’ and even become ‘universal,’ that is, applicable to any time and place. However, if we climb up too high on the ladder of abstraction, we may end up with a conceptualization of context that has become so thin that it does not provide us with enough relevant background information to grasp the meaning and significance of the event or phenomenon we are trying to explain. What we gain in scope, we lose in precision and relevance: ‘It appears that we can cover more – in travelling terms – only by saying less, and by saying less in a far less precise manner’ (Sartori, 1970, p. 1035; see also D. Collier & Mahon, 1993; Mair, 2008). Part of the labor of conceptualization involves thus addressing such trade-offs and trying to reach the right balance on the ladder of abstraction.

Why and how does context matter?

Conceptualizations of context are always intertwined with particular explanatory strategies, whether it is the search for general laws, propensity accounts, interpretivism, systemic explanations, or mechanism-based accounts (Tilly & Goodin, 2006, pp. 12–13). For example, from the standpoint of scholars seeking general laws, context is noise: ‘In that view, we must clear away the effects of context in order to discover true regularities in political processes’ (Tilly & Goodin, 2006, p. 7). This perspective assumes that context is either causally irrelevant or remains constant.² From this standpoint, the Capitol protests of 6 January 2021, could be explained by looking at individual properties (education, employment, income, interests, preferences, etc.), organizational resources, mobilization networks, interaction patterns with the police, etc. The Trump presidency, the pandemic, and other contextual elements would not be deemed relevant to account for the event and formulate a parsimonious, linear hypothesis such as ‘if x, then y,’ with perhaps a few complementary necessary and sufficient conditions.

In contrast, claiming that context matters implies contending that contextual variation is part of the explanation of the event or phenomenon we are analyzing. In this sense, context matters in two ways. First, the context shapes the meaning and significance of our data (whether statements, claims, actions, or chains of events) and our interpretations are thus dependent on it. The ‘same’ phenomenon can mean different things in different contexts and we ought to take into account such variation. For example, ‘political parties’ …

do not mean the same thing in continental Europe, with its many, often ideological parties, the United Kingdom with its very few leadership parties, and the United States, where parties are perhaps best thought of as loose pre-electoral government coalitions. Party membership may therefore well have to be thought of in different ways in these three cases. (Hancké, 2009, p. 90)

Similarly, contextualizing an event or phenomenon can provide us with information that helps us to solve a puzzle: ‘We “make sense” of an otherwise puzzling phenomenon by finding some special feature about it which, when taken into account, allow us to assimilate that case to our standard model of how the world works’ (Tilly & Goodin, 2006, p. 21). It follows that removing our data from their context can lead to errors of interpretation and evaluation, and we can end up unwittingly comparing apples and oranges (see Locke & Thelen, 1995).

Second, context can be part of the explanation because it shapes not only our data but also the actual manner in which social and political processes occur (Tilly & Goodin, 2006, p. 6). But if context matters so much, is it simply ‘context’ or is it a cause? According to R. B. Collier and Mazzuca (2006, p. 474), contextual analysis is thus ‘a “self-liquidating” enterprise’: ‘By underestimating context, analysts run the risk of misunderstanding the political phenomena located within it. By attributing direct causal power to context, on the other hand, analysts annul it, since it thereby becomes “part” of the phenomenon of interest, rather than its “surrounding”.’

One way to address and potentially solve this paradox is to treat context neither as a mere surrounding nor as a direct cause, but as a set of conditioning factors that shape the outcome without actually causing it.³ Indeed, as the papers in this special issue illustrate, the context constrains, enables, channels, etc., collective action but does not cause it. The key lies in the interaction between causal mechanisms and the context in which they operate. As Falleti and Lynch (2009, p. 1151) explain: ‘Given an initial set of conditions, the same mechanism operating in different contexts may lead to different outcomes.’ Similarly, according to Goertz (1994, p. 28, as cited in Falleti & Lynch, 2009, p. 1151), ‘context does not cause X or Y but affects how they interact.’ For example, techniques, practices, and policies often generate a particular outcome not in and of themselves but thanks to the context in which they are embedded; their effects depend on certain institutional complementarities and, therefore, they are likely to generate a different outcome in a context where these complementarities are absent (Ancelovici & Jenson, 2013, p. 296).

The political opportunity structure and beyond

In order to grasp how context can matter and shape social and political processes, it is helpful to look at a few examples. In social movement studies, the most influential conceptualization of context is undoubtedly that of the political opportunity structure (POS) (as defined by McAdam, 1982/1999, or Meyer, 2004). For example, one of Goodwin and Jasper’s (2012) edited volumes, entitled Contention in Context, essentially focuses on political opportunities and does so even though both authors have been forceful critics of this perspective for a long time (see Goodwin & Jasper, 2004).

The main dimensions of the POS highlighted in the abundant available literature are societal cleavages, political (re)alignments, elite cohesion/division, the presence/absence of influential allies, state configuration (centralized/decentralized), state strength (weak/strong), and prevailing state behavior (facilitation/repression, inclusion/exclusion) (Kriesi et al., 1995; McAdam, 1982/1999; McAdam & Tarrow, 2019; Meyer, 2004; Tarrow, 1996, 1998).

This perspective has been often criticized for being too state-focused and too structuralist, ignoring culture, relying on a reductionist conception of ‘politics,’ proposing an invariant model, and engaging in conceptual stretching (Ancelovici & Rousseau, 2009; Armstrong & Bernstein, 2008; Goodwin & Jasper, 2004). Some have also pointed out that institutions ‘produce actions, sensibilities, and ideas – they do not merely constrain them’ (Goodwin & Jasper, 2004, p. 12, emphasis in original). Similarly, Dufour and Ancelovici (2018, p. 169) argue that the POS perspective ‘takes interests as given and treats institutional configurations as arenas for social action rather than constitutive factors behind the very actors doing the action’ and propose instead to look at ‘regimes.’ Others even argue that we should abandon the terms ‘opportunity structure’ and ‘political opportunity structure’ and speak instead of ‘political context’ (Amenta & Halfmann, 2012, pp. 231, 234). According to Amenta and Halfmann (2012, p. 231), one of the advantages of this alternative is that ‘it leaves it up to the scholar to define these contexts – what they are expected to influence and why. It seems less likely to provoke needless debate over ontological issues.’ Putting aside the relevance of the terms ‘opportunity structure,’ that I address somewhere else in this issue (see Ancelovici, this issue, pp. 157–158), I would argue that ‘leaving it up to scholar to define these contexts’ simply defers the problem rather than solves it. If we define the context on a purely ad hoc basis, we run the risk of engaging in conceptual stretching and thus contribute to deepening theorical divides and debates rather than resolve them.

If there is a benefit in speaking of the ‘political context’ rather the POS, it lies on the renewed explicit emphasis on context. Such emphasis is necessary because, as McAdam and Tarrow (2019, p. 33) point out, the field of social movement studies has become primarily concerned with internal factors and dynamics (identity, framing, emotions, organizations, etc.) at the expense of macro political and economic, external factors (electoral and political systems, capitalism, world economy, etc.). As a result, ‘the field is now far more “movement-centric” and less focused on the relationship between movement and context’ (McAdam & Tarrow, 2019, p. 32).

From politics to the economy, culture, and history

In addition to taking into account the political context as the POS does, we can move away from a purely movement-centric perspective by looking at, for example, the socio-economic, cultural, and historical contexts of collective action. The socio-economic context refers to the economy and the class structure. In their reflections about ‘the strange disappearance of capitalism from social movement studies,’ Hetland and Goodwin (2013, p. 91) argue that capitalism and class relations matter insofar as they inhibit or facilitate the emergence of new collective identities and solidarities, penetrate and potentially fracture movements from the inside, and shape movement resources, strategies, and goals, as well as the movements’ likelihood of making some gains for their constituents. Similarly, della Porta contends that the state of the economy at the time of the emergence of the mobilization matters and she distinguishes movements of crisis, driven primarily by the victims of the crisis and often more spontaneous and violent, from movements of affluence, primarily composed of conscience constituents, better organized, and less violent (based on Kerbo’s analysis, Kerbo, 1982, in Della Porta, 2015, p. 10). In the specific case of the anti-austerity protests that shook Europe and North America in the early 2010s, she builds on Guy Standing and points out that a new social class, the ‘precariat,’ made of unemployed or part-time employed educated youths with no or few social protections, was the main actor (Della Porta, 2015, p. 4).⁴ Nonetheless, others have stressed the need to avoid sweeping, general claims because, even in times of crisis, the ‘“material bases” of [anti-austerity] mobilization are not the same everywhere’ and because the actual reality of capitalism (how goods and services are produced, how labor markets operate, etc.) varies across countries (Dufour et al., 2016, pp. 297–298).⁵ As Piven and Cloward (1977, p. 20) famously put it:

People experience deprivation and oppression within a concrete setting, not as the end product of large and abstract processes, and it is the concrete experience that molds their discontent into specific grievances against specific targets. Workers experience the factory, the speeding rhythm of the assembly line, the foreman, the spies and the guards, the owner and the paycheck. They do not experience monopoly capitalism.

Therefore, the socio-economic context matters and its specific form varies across time and space. Here, we face again the challenge of reaching the right mix of concreteness and abstraction along Sartori’s ladder of abstraction.

The cultural context, defined as the shared beliefs, understandings, symbols, and meanings commonly found at a place and time, matters insofar as it shapes people’s interests, motives, tactics, and strategies (Jasper & Polletta, 2019, p. 64). Accordingly, the cultural context both constraints and enables. To make such an argument, we need to go beyond general, macro-cultural claims about, say, the weight of ‘post-materialist values’ à la Ronald Inglehart and focus instead on more concrete and specific elements. One way of doing so is to combine discourse and the political opportunity structure to analyze the ‘discursive opportunity structure,’ that is, the prevailing or institutionalized (generally national) categories, legal definitions, norms, and master frames, that (a) shape how activists frame issues and make claims in the public sphere, (b) determine what symbols resonate, and (c) ultimately favor certain discourses over others (Ferree, 2003; Koopmans & Olzak, 2004). Ferree (2003) sums it up: ‘Discursive opportunity structures are institutionally anchored ways of thinking that provide a gradient of relative political acceptability to specific packages of ideas’ (p. 309; emphasis in original). But as Nicholls and Uitermark (2019, pp. 5, 3) point out, ‘not all advocates respond uniformly to a discursive opportunity structure;’ ‘the framing process is riven by disputes and is by no means a mechanical response to a discursive structure.’ The cultural context thus conditions the mobilization process but does not cause it.

Finally, the historical context can refer, broadly speaking, to everything that it is specific to a time and place, thereby encompassing almost all dimensions of social life. However, treating it as such meta-context is way too vague and not helpful to explain particular phenomena. The concept of historical context describes both a period and a conjuncture (‘a temporal coincidence of a potentially limitless number of forces, actors, structures, and events, including the accidental and the contingent,’ R. B. Collier & Mazzuca, 2006, p. 473). But more specifically, it can allow us to emphasize the way in which the sedimentation of cumulative cultural, socio-political and geographic processes shapes the form of collective action (Tilly, 2006, p. 421). Conceptualizing the context of collective action in historical terms also involves situating it relative to other historical contexts and, potentially, locating the phenomenon under scrutiny in a sequence of events that condition one another chronologically. Such path dependency arguments, according to which initial conditions and events affect subsequent sequences and possibly lock them in a particular pattern or path that can become self-reproducing over time (Mahoney & Schensul, 2006, p. 457; Pierson, 2004), can help us solve analytical puzzles, but they also imply that the context has become a causal factor in its own right and, therefore, is no longer a context as such.

Field, space, arena

This special issue builds on the aforementioned dilemmas and debates, but focuses primarily on three different ways of conceptualizing the relational context of collective action. Speaking of relational context implies granting primacy to relations and transactions as opposed to individuals and groups (Bourdieu & Wacquant, 2014, p. 289). Investigations about fields, spaces, and arenas begin thus with the reconstruction of relations and systems of relations (cooperation, interdependence, conflict, etc.) rather than discrete and bounded entities.

Field, space, and arena are by no means the only ways to conceptualize the relational context of collective action. The most obvious option is the concept of social network (Diani & McAdam, 2003). However, social networks and related causal mechanisms, such as brokerage, are an intrinsic part of the explanation rather than the context. Becker’s concept of ‘world,’ understood as networks of people cooperating and guided by shared conventions (Becker, 1982/2008; Bottero & Crossley, 2011), and Somers’ concept of ‘relational setting,’ which includes relations but also narratives (Somers, 1994), are important contributions as well, but they are far less influential in social movement studies.

Field

The concept of field posits that ‘society’ is divided in a multitude of relatively autonomous social spheres that are characterized by distinctive principles, logics, and interests. These spheres are constituted by objective relations rather than personal or concrete relationships. They may thus include social networks, but also go beyond them⁶ and cut across institutions.⁷ Although all perspectives stress that fields stem from, and generate, hierarchical and asymmetrical power dynamics, some essentially focus on conflict (Bourdieu & Wacquant, 2014) whereas others build on symbolic interactionism and acknowledge the importance of cooperation as well (Fligstein & McAdam, 2012). But in all cases, the explanatory logic at the heart of the concept of field implies that the actors’ interests, practices, and strategies are dependent on their respective social location in the field.

Although Fligstein and McAdam’s concept of ‘strategic action field’ offers a promising path for studying social movement dynamics (see Fligstein & McAdam, 2012), in this special issue Krinsky and Ancelovici build primarily on the sociology of Bourdieu to think about fields. Krinsky returns to the classical structure vs. agency debate and puts Bourdieu and Marxism in dialogue to unpack the internal dynamics of fields in terms of contradictions and dialectics: ‘where field theories can take the cohesion of fields for granted, Marxism poses this cohesion as a problem or an accomplishment to be explained’ (Krinsky, this issue, p. 175). Ancelovici puts Bourdieu in dialogue with the political process model developed by McAdam, Tarrow, and Tilly, and proposes to reformulate the concept of political opportunity structure as a field opportunity structure: ‘It follows that the political opportunity structure is one type of field opportunity structure amongst others, rather than the main, or only, opportunity structure relevant to the study of social movements’ (Ancelovici, this issue, p. 162). From this perspective, fields condition, constrain, and channel mobilizations but do not cause them in and of themselves. However, the concept of field faces several limits and does not apply to all times and places, particularly historical periods where there was little social differentiation. Moreover, even in modern societies, we should not assume that all social activities happen inside fields. Some of them take place off fields (Lahire, 2012) or in-between fields (Eyal, 2013).

Space

Some contributors to this issue argue that the concept of field is too restrictive for making sense of social movement dynamics. Even though he thinks that social movements constitute a microcosm somewhat similar to fields, Mathieu (this issue, p. 199) points out that this microcosm ‘does not possess a degree of objectivation, structuration and institutionalization sufficient to correspond to what Bourdieu considers a field.’ Instead, Mathieu (this issue, p. 196) introduces the concept of space of social movements, that is, ‘a universe of practice and meaning that is relatively autonomous from other social microcosms (…), and within which mobilizations (…) are linked by various relations of mutual interdependence.’ This space is thus understood in social rather than material or geographic terms. According to Mathieu (this issue, p. 202), it is also characterized by a distinctive set of acquired cognitive and practical skills that are necessary for acting.

Building on Mathieu, in this issue Bereni analyzes the gender parity law in France on the basis of the concept of ‘space of the women’s cause,’⁸ which refers to ‘the relational structure of groups mostly devoted to advancing women/challenging the gender order in a variety of social settings, cutting across the line between civil society and political institutions’ (Bereni, this issue, p. 209, emphasis in original). This space includes women not only in feminist organizations, political parties, and parliament, but also in the state bureaucracy and public administration as well as in universities and intellectual spheres. It cuts across institutional boundaries and develops at the interstice of adjacent fields such as the academic field, the electoral-partisan field, and the state field (Bereni, this issue, p. 217).

One of the virtues of the concept of space is to be more malleable and applicable than that of field. It also questions some of the traditional dichotomies of political sociology (for example, between movements and institutions or society and state) and allows for a finer grained analysis of power dynamics and conflict outside, inside, and across the state. However, it is not always clear whether the space of social movements is the context wherein a given social movement emerges and evolves or the movement itself. On the one hand, Mathieu (this issue, p. 196, emphasis added) contends that ‘social movements develop within’ this microcosm, thereby suggesting that the space of social movements is the context of collective action. On the other hand, Bereni states that she uses ‘the concept of [space] to emphasize the relational structure of collective mobilizations converging around the women’s cause’ (Bereni, this issue, p. 209, emphasis in original), thereby implying that the space of social movements is just another way to talk about the movements themselves. It follows that the concept of space, in spite of its heuristic value, does not always allow us to clearly distinguish the phenomenon we wish to explain from the context wherein this phenomenon takes place. It tends to conflate context, actors, and action.

In her article in this issue, Dufour builds on classical political sociology and critical geography to put forth the concept of ‘space of protest’ and account for local social forums in France and Quebec. Her conception of space differs from Mathieu’s insofar as it combines a social dimension with a geographic dimension. It includes three components: a social cleavage (cf. Lipset & Rokkan), political action poles (electoral, protest, concerted and participation), and a geography of protest (based particularly on place and scale). According to Dufour (this issue, p. 230): ‘With these three components (…) the concept of “spaces of protest” offers an analytical tool that respects the structuration both of relations and interactions and of contingencies and spatial dimensions.’ Although ‘spaces of protest can be systematically compared across times and/or places’, this concept ‘does not offer any explanation in itself; [it] is not a “theory of action”.’ (Dufour, this issue, pp. 230, 237). It is a context, not a causal factor.

Arena

According to Jasper (this issue, p. 244), the concepts of field and space do a poor job of describing both agency and constraints. As an alternative, he proposes the concept of arena, that is, ‘a bundle of rules and resources that allow or encourage certain kinds of interactions to proceed, with something at stake’ (Jasper, 2015, p. 14). The concept of arena goes hand in hand with the concept of player, that is, ‘individuals or groups who have some shared identity, some common goals, and who operate in at least one arena’ (Jasper, this issue, p. 244, emphasis in original).⁹ Players can themselves be treated as arenas if we look at their internal dynamics (Jasper, 2015, p. 12). Arenas are linked to one another and players often participate in several arenas. It is worth pointing out that Jasper’s conception has slightly changed over time: while he used to describe arenas as a bundle of rules and resources and claim that they are similar to institutions and ‘capture most of what has gone under the banner of structure’ (Jasper, 2015, pp. 15–16), he now stresses instead that arenas are ‘physical places where players interact’ (Jasper, this issue, p. 244, emphasis in original). As he remarks, ‘we can see an arena rather than having to construct it in our minds or in our pages’ (Jasper, this issue, p. 253, emphasis added).

This slight shift could allow for the concepts of arena and player to be compatible, to a certain extent, with the concepts of field and space. According to Jasper (Jasper, this issue), ‘as long as we define a space clearly, it deserves a role in our vocabulary as a way to talk about aggregations of linked players, especially around a set of issues’ (p. 251, emphasis in original). Similarly, ‘we can preserve field as a reasonable word to get at aggregations of arenas’ (Jasper, this issue, p. 252, emphasis in original). Therefore, we may ask, together with Siméant-Germanos (this issue, p. 144), whether it is really necessary to choose between concepts. Field, space, and arena could indeed serve complementary purposes and be applied in conjunction.

However, Jasper’s olive branch, as constructive as it may seem, involves treating the concept of space no longer as a transversal context but explicitly as a set of actors. Similarly, it implies treating the concept of field no longer as a set of objective relations but as an assemblage of physical places where interactions unfold. Moreover, whereas field and space build on a sociological tradition that emphasizes lasting dispositions, arena and player draw primarily from symbolic interactionism and emphasize strategic interactions.

Conclusion

The articles included in this special issue propose different avenues for thinking about the relational context of collective action and social movements. On the one hand, the concepts we favor depend on the questions we ask, and we should not preclude certain analytical tools on purely theoretical grounds. Moreover, these conceptualizations of context can point to different dimensions or layers that interact with one another and thus contribute to shaping together the phenomena we want to explain. Their combination could allow for a more nuanced and complex understanding of context as multilayered, as Falleti and Lynch (2009) propose. On the other hand, however, these conceptualizations build on different sociological traditions, implicitly rely on conflicting ontological premises, and lead to different research methods. It is therefore difficult, if not impossible, to reconcile them without sacrificing coherence or making contradictory claims.

Having said that, the ambition of this special issue is precisely to generate rich and constructive debates between these conceptualizations of context so that we, as a community of scholars interested in collective action and social movements, are forced to spell out our premises, better justify our choices, and acknowledge the lacunas and blind spots of our respective preferences and practices. The papers in this special issue respond thus to one another and, in spite of their disagreements, make a cogent case for the need to rely on precise conceptualizations of the context of collective action.

Acknowledgments

I wish to thank Montserrat Emperador Badimon and Clare Saunders for comments and suggestions. The usual disclaimer applies.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Funding

This work was supported by the Social Sciences and Humanities Research Council of Canada [Canada Research Chairs Program].

Notes on contributors

Marcos Ancelovici

Marcos Ancelovici (PhD, MIT, 2008) is Canada Research Chair in the Sociology of Social Conflicts and Associate Professor in the Department of Sociology at the Université du Québec à Montréal (UQAM). He has published numerous articles and chapters on the global justice movement, anti-austerity protests, and housing struggles, and co-edited Street Politics in the Age of Austerity: From the Indignados to Occupy (Amsterdam University Press, 2016; available in open access).

Notes

1. In addition to the contributors to this special issue, I wish to thank Barry Eidlin, Neil Fligstein, and Emanuel Guay for participating in the workshop and the discussions. Barry Eidlin also deserves special thanks for welcoming us all at McGill University, where the workshop took place.

2. However, attempting to eliminate contextual variability by ‘controlling’ for context still requires that we know what aspects of context matter and need to be controlled for (Tilly & Goodin, 2006, p. 24). In other words, we still need to take context seriously.

3. At a more general level, the claim that context conditions social action, mechanisms, and processes, is reminiscent of Marx’s classical statement in the opening of The Eighteenth Brumaire of Louis Napoleon: ‘Men make their own history, but they do not make it as they please; they do not make it under self-selected circumstances, but under circumstances existing already, given and transmitted from the past.’ See Karl Marx, The Eighteenth Brumaire of Louis Napoleon, 1852, retrieved from https://www.marxists.org/archive/marx/works/1852/18th-brumaire/.

4. Although she emphasizes the leading role of the “precariat,” Della Porta also notes that anti-austerity protests were made of cross-class coalitions: “These included emerging precarious youth, especially with high levels of education, but also previously protected groups (…) that felt less and less secure about their job and life chances” (Della Porta, 2015, p. 66).

5. In her book, Della Porta acknowledges the importance of varieties of capitalism but does not integrate this dimension to her analysis. She stresses instead “similarities in trends of mobilization” (Della Porta, 2015, p. 66).

6. See Bottero and Crossley (2011) for a nuanced critique.

7. In this respect, it is mistaken to treat ‘fields’ and ‘institutions’ as ‘roughly interchangeable,’ as Armstrong and Bernstein (2008, p. 82) do.

8. In her article, Bereni actually uses the word ‘field’ instead of ‘space’ because she thinks that ‘In sociological English, the word “field” is much less associated with the “strong” conception of the field from Bourdieusian theory’ (Bereni, this issue, pp. 220 n2). Nonetheless, here I prefer to use the word ‘space’ to avoid any confusion and clearly distinguish her perspective, which builds on Mathieu’s concept of space of social movements, from Fligstein and McAdam’s or Bourdieu’s. As Bereni (this issue, pp. 220 n2) herself points out: ‘In French, I use the term “espace” (space) rather than “champ” (field).’

9. Jasper (2015, p. 14) actually sees the concept of players as an alternative to that of social movement: ‘Talking about players allows us to avoid the term “social movements.”’