Less repression in Western democracies?

14McCarthy and McPhail, in their study on Washington, observe “the continuing institutionalization of protest”: [35] interactions between police and protesters have become increasingly routinized and predictable, which diminishes repression. These authors describe a shift away from the use of force towards a public order management system characterized by negotiation between affected parties, planning by the authorities, and the incentives given to protestors to plan their actions. Yet, according to these authors, this system that depends on the negotiation between police and protestors remains a fragile one, as evidenced by the many disruptions of negotiations that have occurred since the early 2000s. Indeed, transformations in policing are contingent, since they are primarily triggered by changes in activist and political strategies, as shown by the way in which states have sought to adjust their doctrine and their practices as a response to increasingly numerous anti-globalization demonstrations and to the repetition of confrontational situations. [36]

15While comparative research usually distinguishes two main policing styles – one more opportunistic, tolerant, soft, selective and flexible, the other legalistic, repressive, hard, diffuse, and dissuasive [37] – more recent research on European countries points to a degree of unification in the way protests are handled. In the contemporary period, protest policing seems characterized by three tendencies: bargaining, with the rise of negotiation techniques and the increasingly important role played by mediators between demonstrators and the forces of order; which results to some extent in under-enforcement of the law, as the eagerness to compromise and negotiate requires a degree of toleration of law-breaking, [38] and in the largescale, sophisticated collection of information. Although the use of intelligence is not a new trend, [39] the availability of new techniques, as well as increasing professionalization, have reflected an always increasing attention to the collection of information [40] – as is indicated, for instance, in the control of football stadiums.

16This trend towards pacification in the policing of conflicts can be related to the increase in the education levels and political competencies of the populations. Likewise, the demilitarization and professionalization of the police are reflected in recruits from a higher-class background, and increasing integration in society. Political reforms, as in the South African case, set the ground for a (slow) change in police attitudes. [41] More broadly, reforms introducing unionization and opening the police to women may have contributed to change. [42]

17Lastly, the literature often evidences a link between pacification and professionalization of police and the legal definition of demonstrating as a civil liberty. In many cases, the police focus on violent small groups in order to guarantee the safety of peaceful demonstrators. In countries experiencing a regime change – despite the fact that some practices, owing to the continuity in membership, resist reform and endure – new rules of the game are discussed and negotiated, and, with strict orders not to have recourse to violence, police culture and methods change – for instance in South Africa, [43] or, on a smaller scale, in Mexico City. [44]

18However, the effects of the professionalization and specialization of police are complex and research conducted in the last three decades often points out a number of ambiguities. If on the one hand specialization implies the development of negotiating skills, on the other, the deployment in public order policing of special ant-riot or anti-terrorism units, and/or special units to counter organized crime, have an escalating effect, as the evolution of public order police in Denmark shows. [45] Last but not least, professionalization brought about an increasingly legal definition of police intervention that sometimes constrains “tough” styles, but at other times reduces the possibility for the police to implement de-escalating strategies by “under enforcing” the law.

Legitimacy of the actors and definitional struggles: varying and selective pacification

19While the trend towards the pacification of conflicts in democracies seems to be a deep-rooted and widespread one, it is worth noting that policing styles remain selective and dependent on several factors, starting with the perception of marginalized groups by the authorities. According to Piven and Cloward, [46] protests emanating from marginalized groups – such as ethnic or religious minorities, lower-class people – are particularly targeted by repression; a similar argument is made by Stockdill in his study of minority activism against AIDS, [47] and Mathieu in his work on prostitutes. [48] Monjardet, in his research on the professional ideology of the French CRS [riot control forces] [49] and Donner, on the police repression of unrest during the 1968 Chicago Democratic Convention, [50] have likewise pointed out the differences in the perception of activist groups by the police. New movements addressing new causes are also not as well tolerated as others enjoying longstanding legitimacy; the anti-globalization movement is a case in point. Escalation is particularly likely in ethnic conflicts, especially when police forces come mainly from the same community. In Northern Ireland, for instance, the Royal Ulster Constabulary – a police force essentially made up of Protestants – because of its blatantly colonial policing style, has for a long time fuelled the radicalization of the Catholic community. [51] Likewise, the stigmatization of specific causes (deemed “anti-patriotic”, “communist”, etc., by political authorities and the forces of order in the street) stimulates repression. In the early 2000s, in Turkey and Tunisia, [52] Islamists were among those most often repressed, whereas conventional actors such as unions and political parties faced less repression. [53] The form repression takes can also depend on a “patrimonialistic handling of conflicts”, leading “civil authorities [to] define acceptable targets of violence while ruling out others” [54] such as the surroundings of presidential residences or the occupation of public buildings. [55] Lastly, groups that use a non-conventional or innovative action repertoire are more subject to repression than those whose repertoire is routinized. [56] This is particularly true of the recourse to guerrilla tactics, for instance, [57] and more broadly of any kind of “radical” and “revolutionary” tactic. [58]

20The contingent character of the evolutions of conflict policing in democracies also depends on a range of contextual events that prompt more or less durable backtracking, such as the Vichy regime, the multiple operations aimed at “restoring order” and “pacifying” the French mainland and the colonies under the pressure of national liberation wars, the May 1968 unrest, etc. [59] In his “historical anthropology” book Charonne 8 février 1962, Dewerpe points out that “democratic massacres” are far from infrequent, [60] as he brilliantly demonstrates by analyzing political, administrative, police and media mechanisms. The theory of pacification therefore appears to be subject to criticism; the return of the “repression and criminalization of mobilizations” [61] within the framework of the emergence and growth of the anti-globalization movement [62] are further evidence of this. Indeed, several elements seem to indicate an inflection, both in principle and in practice, in the control of mobilizations: on the one hand, in the street, public order polices are increasingly militarized and heightened security zones are created; [63] on the other hand, most importantly, the will to anticipate disruptions of public order is liable to elicit violations of civil liberties, in the broader context of the penalization of social movements. As King, [64] for instance, explains regarding Canada, the increase in the number of anti-globalization demonstrations has evidently prompted an extension of police surveillance to non-criminal populations, a new development that seems particularly well suited to infringing freedom of assembly on occasions where foreign demonstrators, though they may constitute a minority, are among the most hardened. [65] A number of violations of individual fundamental rights have been observed, [66] both in terms of freedom of movement (denial of the right to cross borders, banning of demonstrations, removal and deportation orders) and in terms of right to privacy (compilation and exchange of databases on demonstrators with no possibility of ex ante judicial review). [67]

21All of the above observations on the contingent character of conflict pacification in democracies usefully complement the very numerous studies in comparative politics relying on the quantitative analysis of aggregated databases, which tend to unambiguously link democracy and low levels of repression. Undeniably, in a democratic regime, the existence of alternatives to repression, its cost and negative consequences, particularly in electoral terms, and the existence of opposing forces make the recourse to repression less likely than in an authoritarian regime. However, several questions remain unanswered.

22On the one hand, the recognition of a given form of opposition as “political” is a constant subject of definitional struggles in democracies: to name just a few examples, we may think of the inner-city riots in France and the UK, [68] the outbreaks of violence in black US ghettos in the 1960s [69] and of the lootings in Argentina. [70] Depending on the prevailing interpretation, repression of the participants is either considered as the implementation of ordinary crime control policy, or as protest policing.

23On the other hand, there are many examples of states that, faced with growing protest, have issued ad hoc laws to criminalize protest actions or deprive specific categories of individuals of some of their political rights. For instance, as a response to the successful Montgomery bus boycott, a ban on carpools was voted. Likewise, several US cities have in the past few years banned all public gatherings around clinics to prevent disruptions from anti-abortion groups. More subtly still, in a paper where she argues that social movement repression and ordinary crime control policies should be considered jointly, Oliver [71] points out that in the US, the state also responded to the riots in the 1960s by implementing a fairly systematic policy of criminalization of ghetto populations. After Nixon’s 1968 victory (with 13% of votes for Georges Wallace, an openly racist candidate) and the launch of the “war on drugs” policy, resources for surveillance and repression in the ghettos increased steadily, and the rates of arrests and prison sentences for black citizens in underprivileged neighborhoods rose dramatically. Olivier argues that this systematic policy of “surveillance and prevention of crime” led to the selective “incapacitation” of the African Americans’ potential for activism, and may have been a contributing factor in the decline of the civil rights movement from the early 1970s. [72]

24This last point calls attention to a problem that is rarely addressed in the literature: it is impossible to restrict the definition of repression to the effective and most visible forms of coercion exerted on protestors, as do most studies on policing and the comparative literature relying on quantitative databases. [73] In the same way that power cannot be assessed on the basis of its actual application, [74] any consideration of repression must also take into account deterrence, incapacitation and intelligence. [75] In her study of the 1968 Chicago student movement, Rolland-Diamond emphasizes the role of “hidden repression” – “surveillance, infiltrations, actions of agents provocateurs, efforts at sabotaging actions and alliances” – in the demobilization of the movement. [76] Deterrence refers to all measures aimed at preventing protest by raising the cost of participation: we may think for instance of the 1970 French anti-wreckers law, the martial law declared in 1980 by General Jaruzelski in Poland, or of the red zones delineated during international summits. Incapacitation refers to techniques enabling a massive or selective return to a previous prohibition regime through the limitation of demonstrators’ rights. Stopping demonstrators at borders is de facto a breach of freedom of movement, [77] without there necessarily being a direct link between the reasons cited for the restriction of movement or the pre-emptive arrest and the event in question. The growth in intelligence brings us back to the continuum between ordinary crime control and political activity, as the means and techniques implemented in the struggle against hooliganism in sports and transnational protest so eloquently show. [78]

The influence of cycles of mobilization on the effects of repression

27Tarrow, [81] in his research on the proliferation and radicalization of social conflicts and repression in Italy in the 1960s, was the first to introduce a dynamic approach to the relationships between protest and repression. He showed that in the mid 1960s, the context was particularly ripe for the emergence of protest movements, prompting an ascendant phase of protest (1966/mid-1968) and subsequently a phase of radicalization (mid-1968/1972), after which the state launched a wave of repression that put an end to the protest cycle. In the last phase, in part as a result of repression, strongly radicalized new organizations seceded from reformist groups that had been brought back into line and within which disengagement grew. [82] Brockett conducted a similar type of research on the basis of case studies in Guatemala and Salvador. The cycle-based approach enabled him to explain the different effects of repression in the two countries. [83] He showed that the use of repression by authoritarian regimes restricts access to the polity and generally maintains protest at a minimal level. A movement can therefore only emerge in an initial context of relative opening up of opportunity. Conversely, when repression happens pre-emptively and routinely, there are few opportunities for protests to develop. In ascendant phases of protest, Brockett suggests that repression does not discourage protest, but instead stimulates it. During the upswing of a cycle, “a violent attack by state agents against a member of the group (such as a parent, a close friend, a village elder, or factory union leader) might provoke anti-regime activity from other group members, not necessarily out of self-defense but out of rage and a desire for revenge”. Accordingly, emotion is “a crucial component of individual and group motivation”. Lastly, in the last phases of the cycles, mass and sometimes indiscriminate repression of both activists and civil populations accused of covering for them, such as farmers in rural areas, ultimately leads to a radicalization of some groups – which take up armed struggle and go into clandestinity – and the demise of reformist organizations. [84] When regimes trained to use intelligence techniques and to do fieldwork have the ability to quash these movements, they generally do so. On the other hand, if repression subsides and an opening up of political opportunity is on the cards, popular movements emerge again. [85]

28In a book mainly focused on Algeria and Egypt that also includes attempts at comparisons with Morocco, Tunisia and Syria, Hafez [86] picks up Brockett’s distinction between pre-emptive and reactive repression to account for the trajectories of Islamic movements. At odds with the many studies that explain the emergence and radicalization of these movements through relative deprivation indicators only, and in line with Singerman [87] and Burgat, [88] the author shows that pre-emptive repression contributes to preventing the lasting emergence of resources, the constitution of activist networks and the gathering of popular support. [89] On the other hand, when thanks to a degree of toleration or even of tactical support from a regime busy repressing other sectors of society, Islamist movements manage to collect resources and form support networks, subsequent confrontation with repression is very likely to prompt heightened mobilization and radicalization in certain sectors. This observation ties in with numerous studies on protestors’ perceptions of chances of success [90] and with the idea that the strong ties forged within activist networks become strengthened rather than weakened in the face of repression, especially when the latter involves prison sentences and torture. [91]

A few blind spots of macro-social approaches

29Studying the effects of repression at the macro-social level of political opportunity yields insights when attention is paid to the moves and countermoves of the state and protest movements over time. Yet, opting for this level of analysis entails a number of blind spots, a few of which we outline below.

30First, as we suggested above, only taking into account visible forms of repression means disregarding the fact that in reactive phases of repression, what initially appears as the end of protest is sometimes actually the result of organizational networks going on standby, temporarily renouncing open protest and instead favoring more discreet or less visible forms of action. Many recent examples illustrating this point can be found in the literature, especially in Taylor’s work on feminist movements. [92] Osa, in her book about the advent of Solidarity in Poland, [93] provides a striking case in point: the author combines the analysis of events with an effort to identify the “networks of opposition” that were formed in Poznan during the 1956 movement, and went underground while still developing until the emergence of Solidarity. The dense network of organizations created in 1956 was soon crushed by repression and by the party’s stranglehold. A period of great political stability ensued, as all visible protests had disappeared. Yet, when the strike broke out in Gdansk in 1980, the protest grew so quickly that the government was temporarily paralyzed, and had to give in and officially recognize the Solidarity trade union. Osa argues that three factors explain the success of 1980: first, the political context changed; [94] second, the Catholic church strongly supported Solidarity: within the context of John Paul II’s election, these ties strengthened the movement considerably in terms of popular support and legitimacy; lastly, the strike spread to other Polish industrial sites and rural areas with remarkable speed. The latter two factors, according to Osa, result from the rebirth of the network of organizations that had led the protest twenty years earlier. Osa shows how over the preceding 24 years, these organizations kept developing and recruiting activists around a few groups of central organizations. Admittedly, these “dormant” groups were not responsible for the Gdansk strike, but they were in a position to build on it very quickly. When martial law was declared, this network made it impossible for the outcome to resemble 1956; in 1989, the union was powerful enough to immediately launch a new wave of strikes and defeat the ruling party in the first elections. This case is an instructive one in several respects. First, understanding the emergence and success of Solidarity entails putting it in perspective with past struggles. Had the author stuck to the study of Solidarity’s visible actions, she would have addressed separately things that are actually related, neglected the movements’ standby phases and underground activities – all the mobilization process that made the 1980-1989 movement possible without a single public demonstration. More importantly still, Osa deftly shows that the effects of the 1956 repression are more complex than they seem. While, in the short term, the regime quashed all visible forms of mobilization, it actually paved the way for the slow and progressive construction of a very dense network of opposition.

31As they are limited to a macro-social level of analysis, protest cycle approaches are not well suited to investigating internal dynamics, both in the diverging strategies devised by states to deal with opposition movements and in the relationships between loose networks and formal organizations within the same group of movements.

32Boudreau [95] shows that “weak states” with few resources for repression seek to concentrate their repressive policies as efficiently as possible, which means that the analysis of repression targeting a given group and of its effects cannot be dissociated from that of broader relationships between other dissenting groups, both synchronically and diachronically. Hence, in the 1980s, the Indonesian government let the student movement develop to a surprising extent because it was chiefly focused on subduing the Acehnese and Timor resistance movements. Likewise, in Turkey after 1980, as well as in several North African countries in the preceding years, the concentration of repression on far-left organizations was a contributing factor in the emergence and growth of Islamic movements, before the “Islamic threat” became the priority. Also in Turkey, the civil war between the state and Kurdish separatists practically determines the way all protest movements are perceived and handled by security institutions, in the name of a “meta-ideology of national security”. [96]

Repression and multi-organizational fields

33On the meso level of “multi-organizational fields”, [97] we can consider repressive policies in the more precise context of the space of mobilizations, constructed dynamically. This entails making a distinction between “initiator movements” that “signal or otherwise set in motion an identifiable protest cycle” and “spin-off movements” that “draw their impetus and inspiration from the original initiator movement”, [98] as well as identifying the interactions of counter-movements and potential standby phases, and conversions of material and human resources, of skills, from one cause to the next. As Sommier rightly points out, [99] such an approach makes it possible to “retrace the evolution of the cycle through a sequence-by-sequence breakdown made on the basis of responses to protests, of the reactions they elicit, for instance from counter-movements, but also via the relationships between organizations: mutual borrowings, competitions, sets of alliances and oppositions, etc.” She goes on to explain that “the radicalization process that characterizes the Italian cycle (and led to its exhaustion due to defections and a rise in the cost of engagement) is for instance closely linked to competitions between far-right groups as well as to their confrontation in the street with far-right activists, who were also used, as part of the strategy of tension, for counter-mobilization purposes by a fringe group of the military secret services”. [100]

The reorganization of activist work

34Looking at the meso level of organizations, we may also observe the multiple effects of repression on the internal functioning of activist organizations, as the literature shows. In his research on the Chilean Communist Party under the Pinochet dictatorship, Rolando Alvarez [101] shows that first, the party’s membership dramatically decreased. The most exposed Communist leaders were sheltered in casas de seguridade and kept away from the new party leadership. Following numerous deportations, arrests and deaths, the party went into clandestinity. “Going underground was like learning a new language”, [102] making everyday domestic life very difficult. Alvarez argues that this experience of the underground, this daily resistance to repression, was the impetus for a key turn for the Chilean Communist party in the early 1980s: the creation of a guerilla group, the Frente Patriotico Manuel Rodriguez, that attempted to assassinate Pinochet in 1986, whereas the Communist party had traditionally remained moderate and perceived such actions as leftist excess. In other words, “the subjectivity of the underground was an indispensible condition for ‘naming’ a new way of experiencing Communist activism, by then increasingly linked to the military theme”. [103] Lastly, going underground contributed to transforming a party that had been until then “inclusive” into an “exclusive” organization. [104] Indeed, “the main features of clandestine activism have resulted in a trend of extreme control and disciplinary vigilance in the way the Party operated; this was seen as the only viable way of resisting the dictatorship’s new ‘scientific’ methods of repression”. [105]

35Likewise, in the Arab world, there is a wealth of literature on the impact of repression on the internal organization of protest groups. [106] Regarding Jordan, for instance, Wiktorowicz explains how the Salafists, faced with repression, recruit and collect part of the financial resources necessary for action through informal networks in face-to-face interactions. This decentralized and segmented network structure, combined with the lack of a well-established religious hierarchy at the helm of the movement, contributes to protecting activists from repression. The arrest or killing of a religious notable would thus not mean that the organization became a headless body; the ties forged and maintained within the organization are hard to detect and punish by a regime constrained by its own “defensive democratization” policy and therefore unable to revert to the arbitrary policy of terror that had prevailed in the 1960s. This strategy led Salafists to concentrate their activities in contexts of micro-mobilization. Unable to lecture at the mosque, the preachers organized home lectures and the believers were brought together by word of mouth, from acquaintance to acquaintance, which makes it more difficult to have recourse to undercover agents. As it was impossible to partake in “moments of fervor” such as Ramadan, when the surveillance of public areas is heightened, they organized iftar (fast-breaking) gatherings within family circles, where friends were also invited. This was the opportunity to combine the temporality of ritual performance and the temporality of indoctrination and “physical or spiritual revolt”. [107] As they intercut networks of acquaintanceship and friendship, these activist networks were particularly opaque. [108]

36The Jordanian case calls attention to an all-too-often neglected dimension in the study of the effects of repression: the reconsideration of the division of labor within activist groups, particularly as far as gender relations are concerned. [109] First, going underground and taking up armed struggle to oppose repression favors the strengthening of roles generally fulfilled by men, with the pre-eminence of security forces, and the value placed on readiness to fight and physical strength effectively contributing to excluding women from power. More subtly, repression can conversely strengthen women’s strategic positions in the mobilization process, though this may not always be acknowledged. The literature has shown the importance of the role played by women in creating and maintaining activist networks. [110] These networks, because they are inscribed in the private and domestic spheres, are sometimes the only ones to escape state surveillance. Clark, for instance, shows that in Yemen, the mobilization of women is primarily accomplished though Qur’anic study groups – nadwas – intermediary institutions where women activists pursue goals of individual mobilization and propagation of the faith (da’wa). [111] Conversely, it is clear that as organizations go into clandestinity, resort to armed violence and become increasingly militarized, the redistribution of leadership positions and of the most valued roles is made at the expense of women. The literature on guerilla movements in Latin and Central America shows this consistently, [112] despite some slight differences between movements, as do studies on the place of women in the Black Panther Party [113] and the African National Congress in South Africa, [114] which show that women are frequently used to provide sexual services to fighters or limited to their role as mothers. Two more important effects of repression on the internal functioning of protest groups deserve a mention. On the one hand, several authors have raised the question of the specific effects of the repressive strategy of decapitating the leadership of protest groups. Studies particularly focused on the temporal scale generally point out that while repression can initially cause disarray, it can in the long run result in the proliferation of small-scale leaderships, as in the Algerian and Palestinian cases, or in the rise to the top of organizations of new generations of activists. [115] On the other hand, the tactical changes brought about by the burden of repression (withdrawal, standby or armed struggle) generally do not benefit the different factions that exist within the organizations in the same way. It is thus not surprising that, when facing repression, some groups whose material and ideological resources are best adjusted to a change of context manage to reinforce their prevalence over less well-endowed ones. This can be observed in the trajectories of revolutionary movements in the 1970s in Italy, Germany and Japan. Sommier, for instance, points out that the competition between left-wing organizations was a crucial factor in the Italian escalation, with the militaristic watershed of Potere Operaio and Lotta Continua starting in 1972, and then the game of higher and higher stakes first between Brigade members and independent actors and then amongst independent actors. [116] In Japan, the competition was even more heated, to the extent that the internal conflicts (uchigeba) caused more deaths than the repression. [117] A similar situation occurred in Mexico in the 1970s: the hunt for “petit bourgeois” in some armed groups and internal disputes resulted in executions of activists. [118]

Interactions and mutual adjustments of repertoires

37These remarks on the effects of tactical reorientations bring us to another element that can be observed at the meso level of organizations: the effects of repression on changes in modes of action as well as the way in which the repressive repertoire adjusts in response. There is a wealth of literature on these issues, exploring both the trajectories of radicalization we discussed earlier and forms of retreat into roundabout, covert modes of resistance and opposition. This includes the vast body of work on the “weapons of the weak”, [119] which we will not address here, although it is worth noting that it has recently experienced interesting developments, with the publication of studies on the modalities and conditions of protest in Eastern European countries. [120] There have been however fewer studies on the cross-effects of moves and countermoves of agents of repression and protest movements; those that do exist mostly stick to the observation of violent demonstrations [121] and riots in democratic countries. [122]

38All of the above, however, suggests that it is worth conducting further research on these “strategic interactions” (in the sense of Goffman). According to Goffman, the interaction order does not boil down to a structural order, which means that the determinants of the way action unfolds are also rooted in interactional logic, even if the protagonists are “caught” in the situation and interpret it on the basis of structural interactions. Additionally, following on from this first point, the protagonists are related to one another through a system of interdependences, and accordingly their decisions, anticipations and possible calculations are informed by the decisions, anticipations and calculations of both their partners and their opponents: Goffman calls this “mutual anticipation”. The individuals and groups involved co-produce the event, in a structurally determined normative frame, which means that the relationships established between the actors are also cooperative ones, since the actors negotiate a practical agreement on their perception of what is at stake, on rules of good conduct, and on boundary crossing.

39At this level of analysis, the impasses of the structural sociology of mobilizations and the methodological issues discussed earlier are resolved. Admittedly, the literature has progressively acknowledged that tactical repertoires are the product of a permanent construction between state structures and protest movements, [123] and Tilly [124] has pointed out the need to start from the observation of the actual events and simultaneously analyze the ways in which state actors define the environment in which individuals move, the place occupied in this environment by the media and target audiences, as well as the ways in which protest groups contribute to altering the conditions in which individuals act. Yet this acknowledgment has yet to result in a properly structured research agenda. “The attention to individual episodes, innovations, successes and failures in claim making, which affected subsequent claim making with respect to similar issues” [125] still retains a strong objectivist bias. [126]

40We argue that understanding the mechanisms at the core of repressive episodes entails paying heed to the rationales of interaction and to the performance of agents of repression and protestors alike, to the conditions of their adoption, their transformations and their extension. In other words, it is necessary to focus on the consistency of the means available and effectively used and on the uncertain relations between structural constraints, practical dilemmas and strategic interactions. [127] This approach holds that actors rely on their experience to settle on a course of conduct and decode current situations, but also to decide what needs to be done; as the action unfolds, the discrepancies between anticipation and reality leads them to make strategic adjustments. This model has two main benefits in that, first, it includes a theory of the actor, and secondly, it brings a more precise specification of the relation between structure and event to the table – in other words, of the dynamic of social processes. Methodologically speaking, this implies combining the macro and meso-sociological dimensions with the micro-sociological one, which means raising the question of the individual effects of repression, of the way individuals invest modes of action with their own attitudes and knowledge, their motives and perceptions.

For a career approach

42Pursuing approaches of individual activism in terms of “career”, [129] in the sense of Becker, [130] offers a way out of this quandary. Without going into detail on the theoretical and methodological requirements of this perspective here, we will emphasize a few aspects of it that bear directly upon our concerns.

43As it takes into consideration processes and the permanent dialectic between individual history, institution and situations, the career approach enables us to understand how, at each stage of an individual’s biography, behavior is determined by past behavior and in turn informs the field of possibles to come, thereby situating periods of engagement in their entire life cycle. Hence, the notion of career gives us the opportunity to consider together the questions of predispositions to activism, of introduction to activism, and of the differences and variations over time of the forms of engagement (such as radicalization and withdrawal). This marks a clear break with the majority of studies that stick to correlating the individual’s social properties (sex, age, class membership, degree of relative deprivation) with attitudes towards repression, neglecting the fact that these properties only make sense within the configuration in which they are actualized.

44Being mindful of the configurations in which actors move at the various stages of their trajectories, leads us to formulate a few general remarks.

45First, individual changes of attitude and behavior due to repression do not follow an action/ reaction pattern; they are part of long and complex processes, as evidenced by a host of recent research on radicalization, introduction to armed struggle and clandestinity. [131]

46Second, attempting to model changes of behavior on the basis of a rational-choice conception of action does not make sense. However refined the explanatory models of the individual effects of repression in terms of costs/benefits may get, they are particularly inadequate to explain sets of decisions that we can only begun to understand through a comprehensive approach paying heed to the motives put forward by actors, and relating them to their structural positions. [132] It should be noted, moreover, that the available literature is disappointingly inarticulate on this point: researchers have focused their efforts on defining and measuring positive and negative incentives, with findings that were often interesting in and of themselves, but of little use in terms of modeling.

47Lastly, the configurational rationales that account for the heterogeneity of the individual effects of repression must be interpreted in the context of interconnected levels of organizational contexts and rationales. First, behaviors in response to repression must be related to the perception of repression in a given space and in a given context (particularly in reference to the memory of past events), as well as to the value placed on a given type of response to repression in a given region of the social space (honor, revenge…), considering that individuals have diverse pre-existing allegiances (brotherhoods, guilds, clans, community networks). Second, at the meso level of organizations, we need to be particularly aware of the transformations of opportunities for engagement under the effect of repression (disappearing, breaking up, going underground, or conversely intensifying action and recruiting massively), while keeping in mind that they are also informed by the way in which groups encourage or discourage engagement through their public image and a range of tools for selection, used to filter potential members in or out and orient their activities. We also need to relate individual behaviors to the modalities of organizational socializations, i.e. “the multiple socializing effects of activism, themselves in part determined by organizational rules and modes of operation, understood as a set of constraints (status, proposed or reserved activities, leadership, and so on)”. [133] In what follows, without laying claim to exhaustiveness, we list a few mechanisms useful to better grasp the individual effects of repression.

Individual trajectories and repression

48In the literature, we find scattered comments on the different effects of repression on individual trajectories depending on socio-biological criteria. Differences in social backgrounds and the lateral possibilities they allow are predominantly covered. Combes, [134] for instance, shows that in Mexico, while it is always possible for members of the elites and of the middle class to leave hotbeds of repression, this is rarely the case for the most socially underprivileged who, once they have been identified and targeted by the repression, often have no choice beyond radicalization and exit. More broadly, levels of education, of unemployment or professional insertion are sometimes considered as factors explaining a frustration that is conducive to radicalization, but these findings are strongly nuanced by several studies that show that levels of relative frustration have ultimately little relation to trajectories of radicalization. [135] Regarding gender, based on the observation that repression takes on different forms depending on whether it targets men or women, some studies emphasize the gender determination of modes of response to repression. Thus, it seems that the “mothers of the disappeared” movements managed to grow in Latin America first and foremost because they were perceived as unthreatening by the authorities. The case of the Mothers of the Plaza de Mayo, in Argentina, is very much a paradigmatic one in this respect. [136] When women started organizing illegal demonstrations in April 1977 to protest against the “disappearance” of their children, husbands and friends, the government initially reacted by calling them crazy, and only made a few arrests, a response that appears particularly moderate considering the frequent recourse to torture and assassination at the time. [137]

49Findings on the influence of pre-existing networks are more consistent. The existence of groups and networks structured prior to the beginning of repression seems in most cases to favor trajectories of radicalization and make withdrawal and defection more difficult. [138] In the Southern Cone (Argentina, Chile, Uruguay), Loveman stresses the importance of the activist networks that existed before authoritarian regimes came into power to explain the emergence – or lack thereof – of human rights organizations. [139] In a study on anti-nuclear movements in Germany, Opp and Roehl show that while repression does have a negative effect on popular mobilization, it encourages the development of micro-mobilizations led by the people who had ties to the victims of repression through their social networks. [140] Lastly, and most convincingly, in societies with deep-rooted clan or caste solidarity, repression generally prompts phenomena of “block recruitment”, both offering material resources for effective action (learning of violence, mentoring, armament) and drastically reducing the field of what is conceivable. Grosjean and Dorronsoro [141] show this aptly in the case of the PKK and of the Kurdish Hezbollah: these organizations recruit blocks of new members, mobilizing entire families from the same villages and associations, not unlike the role of qawm in Afghanistan described by Dorronsoro earlier. [142]

50It appears, to be more precise, that the impact of pre-existing social networks varies depending on whether the repression is targeted or indiscriminate. [143] When the leaders and the most active members of a movement are targeted, the demobilization of the less committed activists and of simple sympathizers is more likely. However, when repression indiscriminately affects activists and sympathizers alike, or even members of the population suspected of supporting them, the extension of the mobilization and radicalization become more likely. France is a case in point: the “collective responsibility” policy applied to the FLN arguably played a significant role in the engagement of large sectors of the population in the war; so is Algeria with its more recent response to Islamist movements. [144]

51We also need to consider the existence and the succession of “micro cohorts” of activists, joining and sometimes leaving organizations at various stages of repression. Many studies on trajectories of radicalization in revolutionary movements in the 1970s under the effect of repression argue that radicalization more easily affected those who did not experience the initial phase, but joined the movement afterwards, at the peak of the cycle of mobilization. This seems to be corroborated by a rise in the levels of violence with the second or even third generation of activists. [145] The reasons are legion. Here we will refer especially to the research of Steinhoff and Zwerman, demonstrating that in Japan, as well as in the USA, the first generations of activists generally withdraw at the moment when repression leads to the development of the clandestine armed struggle; they are replaced by others, younger and with different backgrounds. In the US, the second cohort includes more African-Americans, a substantial proportion of working-class members and various minorities, recruited from public and community colleges, organizations fighting against poverty, and gangs. In Japan, while the movement started amongst university-educated elites, it has spread to young workers, marginalized members of society and Korean residents born in Japan, who are deprived of their civil rights. For both authors, these social differences are most likely a factor of radicalization.

52More generally, this effectively illustrates that, contrary to a homogenous vision of collectives, it is necessary to pay attention to two interconnected dimensions to understand the diversity of the individual effects of repression within a single movement: on the one hand, the succession of activist generations brings to the forefront of the analysis the internal dynamics of recruitment and selection, the transformations of collective identities and the resulting organizational and ideological changes; [146] and, on the other hand, this ebb and flow of militants must be correlated with a historical period that includes a succession of repressive events. Traumatic episodes, such as the killing of Fred Hampton and Mark Clark by the Chicago police in December 1969 or the “battle of Valle Giulia” in Rome on 1 March 1967, constitute “socializing events,” the weight and individual consequences of which depend, in fact, on earlier socializing events and on the particular generation of activists. Furthermore, the specific forms repression assumes for a given cohort produce a range of socializing effects that lay the foundations for generational phenomena. Analysts for instance often consider the experience of prison and torture crucial: they serve as incubators of activism and of intense forms of socialization, and may indeed redefine identities. [147]

53Lastly, the importance of the succession of generations of activists in understanding the individual effects of repression also raises questions about transmission of the memory of struggles, which could be disrupted and facilitate withdrawal when repression decimates an entire generation, as Bennani-Chraïbi shows with respect to the disappearance of militants from the left and the far-left on Moroccan campuses in the 1970s. [148]

54The reference to socializing experiences in prison brings us to the impact of institutions on individual trajectories. The influence of activist organizations, particularly in repressive contexts, is particularly important here. This secondary socialization may sometimes assume the form of explicit inculcations. Their goal is to homogenize activists’ categories of thought and their way of acting within the organization and on behalf of it, but most of the time they relate to a “logic of practice” that operates beyond the level of consciousness. [149] If, as Gerth and Wright Mills put it, “institutions imprint their stamps upon the individual, modifying his external conduct as well as his inner life”, [150] then we need to examine both the content and the methods of the process of institutional socialization, which can be broken down as follows: [151] on the one hand, the acquisition of resources ( “know-how” and “wisdom”), which obviously depend on the other capital held prior to engagement; on the other, the acquisition of an ideology. Lastly, in order to consider the degree of redistribution of activists’ networks of sociability, both in the activist sphere and in other social sub-worlds, we need to study both individual and collective identities.

55Here, research in social psychology helps us greatly to understand the mechanisms of attachment to a group and continued faithfulness to a cause in the face of repression. The maintenance of attachment is based on two mechanisms: sacrifice and investment. The higher the sacrifice made to join a group and remain committed to it, the higher the cost of defection; the cost of activism, in this sense, determines its price. The more efforts have been made, the more it is difficult to admit that they have been vain. The notion of investment, on the other hand, brings us back to the existence of alternatives. The more individuals are caught up in a system that is the only one to distribute rewards and costs, the more they remain committed. Attachment breeds cohesion; cohesion, defined as the range of affective ties that exist between individuals, is based on the two mechanisms of renunciation and communion. [152] Renunciation refers to a withdrawal from all social relationships outside the group. Communion designates all the ways to reinforce the feeling of belonging to a community brought together by a dialectic of unanimity-exclusion. Ensuring cohesion requires means and techniques of control, from the most subtle to the most extreme, such as mortification and denial. Mortification is renouncing one’s desires and interests, giving up one’s private identity to identify with a group. Denial, on the other hand, refers to unconditional dedication to an authority, and to the members’ internalization of what the group says and wants. The more exclusive the organizations, the more these mechanisms of attachment are likely to impact individuals and cause phenomena of conversion and alternation, in the sense used by Berger and Luckman. [153]

56As it happens, repression mechanically encourages the development of exclusive organizations. Faced with the risks of infiltration, arrest, and the dismantling of activist networks, they become progressively cut off from the outside world and adopt strict codes of conduct that often leads them to a state of isolation that is conducive to the construction of tight-knit emotional communities. [154]

57This final observation is another reminder that, in order to understand the individual effects of repression, we need to consider the contextual, organizational and individual levels jointly.