When the strike encounters the sans papiers movement: the discovery of a workers’ repertoire of actions for irregular migrant protest in France

Introduction

Between April 2008 and June 2010, France witnessed an unprecedented wave of strikes by irregular migrant workers claiming work and residence permits. At its height in October 2009, some 6800 workers occupied enterprises, employer federations and temporary work agencies in the Greater Paris region (Île de France). Trade unions coordinated this movement. They engaged in negotiations with local prefectures over issuing these permits for striking workers and with national government over criteria pertaining to regularization. By going on strike, irregular migrants took an enormous risk. Not only would they lose their income during weeks and even months to come due to the lack of a strike fund. They also deprived their families of remittances and ran the risk of being expelled from France.

One of the most remarkable phenomena associated with this movement was the central role played in it by the CGT trade union. The rapprochement between migrant protest and trade unions is not a novelty as such. An emerging body of literature has shown that in a number of countries trade unions have been participating actively in social movements addressing the situation of migrants, both ‘regular’ and ‘irregular’ (Milkman, 2006, 2011; Voss and Bloemraad, 2011; Tapia and Turner, 2013). In the past, French unions had supported several major mobilizations in favour of the regularization of migrants, e.g. during the 1980 hunger strikes of Turkish workers in the Parisian garment district Sentier (Gastaut, 2000) or the 1996 occupation of the Saint Bernard church in Paris (Haus, 2002). In the US, the 2006 marches for immigrant rights mobilized millions of migrants and organizations such as trade unions, community organizations, and churches. This observation has led Milkman (2011) to identify the immigrant rights movement as a third register of the immigrant-trade union nexus besides workplace organizing drives and individual worker support.

Yet, within such movements – in France and elsewhere – trade unions are typically part of a larger alliance of organizations operating in the ‘pro-immigration field’ (Mathieu, 2009). Furthermore, the modes of action to which these movements recur (marches, occupations of public spaces and hunger strikes) are principally associated with civil rights movements, ‘even if the underlying thrust is to improve the economic opportunities available to immigrants, especially the unauthorized’ (Milkman, 2011: 369) and may therefore be apprehended through the lens of worker protest (Milkman, 2011; Iskander, 2007). By contrast, in the movement under scrutiny the union not only took the lead over other groups (autonomous sans papiers collectives; migrant advocacy groups), but also imposed its specific mode of collective action: the strike, taking the form of work stoppages, occupation of enterprises and negotiations with employers.

The use of ‘industrial action’ in this movement is all the more surprising as the strike represents a central piece in the repertoire of collective actions in industrial relations. The right to mobilize collectively in the workplace has been recognized by national and international law as a means to address the power disequilibrium between workers and employers. At the same time, legislators have sought to channel the expressions of overt conflict by dense juridical prescriptions which in institutionalist analyses have served as an explanation for the decline of strike activities, in France and elsewhere in Europe (Groux and Pernot, 2008). If French strike law has been characterized as comparatively liberal – the strike is an individual right; trade unions do not have the strike monopoly; there are no social peace clauses or majority ballot requirements – courts have limited the strike to ‘job-related’ issues and declared certain forms as illicit. Notably, it excludes political and solidarity strikes.

This article seeks to account for this puzzle by asking: which adaptations have occurred in the strike repertoire in order to allow for deflecting the obstacles to collective action typically associated with sans papiers ¹ protest? How can we account for the key role played by the trade union in this movement? Resource mobilization literature (Oberschall, 1973) has developed the idea that social movements are not the more or less mechanical outcome of brutal outbursts of discontent, but require first and foremost the appropriation and mobilization of political resources, notably through pre-existing organizations, allowing individuals to transform their situation of accepted marginalization into collective action. If we can easily suppose that irregular migrants have reasons to mobilize, they are weakly equipped to do so. In a typology considering the obstacles to collective mobilization, sans papiers protest would undoubtedly figure amongst the most improbable (Collovald and Mathieu, 2009) due to stigmatization, the lack of militant experiences and competences, the weakness of organizational support, the danger of collective mobilization due to migrant ‘deportability’ (De Genova, 2002) and the absence of collective identity. However, contrary to other marginalized groups such as prostitutes (Mathieu, 1999) irregular migrant protest has developed its own specific repertoire of actions – the hunger strike (Siméant, 1998). Yet, as Terray (2006) has analyzed for France, such movements are plagued with specific difficulties such as attracting media attention, ethnic divisions and, most importantly, having an impact on national politics due to the essentially decentralized nature of this mode of collective action. It is only on few occasions that hunger strikes have spilled over into mass demonstrations mobilizing a wide range of people and organizations outside the core of supporters, thereby triggering (mass) amnesties and obliging the legislator to review criteria pertaining to regularization.

As Voss and Bloemraad (2011) have pointed out, the standard behavioural political sciences’ accounts of political mobilization (voting, campaigning, political party engagement…) are not well suited for capturing the processes of irregular migrant mobilization as they take for granted a basic condition for ‘normal’ politics: citizenship and opportunities for formal political participation. In order to take into account the group-related particularities of such mobilizations and to capture the specific political constraints which flow from them, the social movement literature is more adapted as it provides tools for understanding how social groups that are characterized by a large distance from the political system and the social classes which constitute it nevertheless manage to engage in ‘contentious politics’ (Tilly and Tarrow, 2007).

This article analyses this movement primarily through the notion of the repertoire of collective actions. This concept seeks to account for the observation that agents involved in social movements tend to act in accordance with a limited number of standardized scripts linking at least two actors (in the case of the employer-worker pair these would typically be strikes, slowdowns, lockouts, contract negotiations or grievance hearings). For a given set of agents and objects of claims, modes of action change relatively little from one contestation to another. Externally, change depends upon wider economic and demographic evolutions and, more specifically, changes in political opportunities. Internally, change in repertoires relies upon iterative, but mostly small-scale improvisations in the learned and historically grounded claim-making performances (Tilly and Tarrow, 2007; Fillieule, 2010). It is on the latter aspect that this contribution focuses. By showing that shifts in repertoires of collective actions can occur before actual changes in the political opportunity structures, it seeks to go beyond visions of changes in repertoires as being the mechanical outcome of shifts in macro-structures (Però and Solomos, 2010).

This article responds to the above research questions by pursuing a mostly sequential account of the movement based on the identification of decisive events and the underlying mechanics of mobilization (discovery, diffusion, adaptation and collapse of the strike). Thus, it shifts attention away from the causes of mobilization to that of its underlying processes in order to explain its emergence and dynamics (Oberschall, 1973). Analysis is situated rather at a micro-level of observation in order to cater for the logics, perceptions and practices of trade union actors in their interactions with allies and opponents. In order to account for their engagement and the complexities which are involved in the choice of an adequate mode of collective action of a population weakly equipped in political resources, I look at activists, contexts (time and place), performances in repertoires and political opportunity structures rather than considering union ideology or ‘master frames’ about immigrants and immigration as a driving force for mobilization. The reconstruction of the logics of collective action in the making allows for (1) identifying a series of adaptations occurring in the strike repertoire by which the movement responded to change in the legal framework pertaining to regularization and catered for the heterogeneity of the irregular worker population; (2) highlighting the role of the strike as a practical, expressive and cognitive matrix of collective action; and (3) specifying the role of the trade union as a constituent component of the strike repertoire.

Empirically, this article draws on some 15 semi-structured interviews which were conducted in the second half of 2011 with union leaders of the strike, migrant advocacy activists who had participated in its coordination and elected strike delegates of several occupied workplaces. Contacts with the workers had been provided through the Parisian CGT union; at the time of the interviews, all interviewed workers had been regularized. For the reconstruction of the chronology newspaper articles have been evaluated. A third – invaluable – source is the ethnographical account of the movement by Barron et al. (2011), based on the collection of different types of documents, interviews and participant observation on picket lines, demonstrations and assemblies during the movement.

A prelude and a blueprint: the discovery of the strike repertoire as a means for worker regularization

The origins of the 2008–2010 strike movement reach back to a series of rather isolated strikes in the Greater Paris region in the middle of the last decade. The first incidence of this kind was the occupation of Modeluxe, an industrial laundry situated in Chilly-Mazarin. In September 2006, the enterprise was on sale. The management wanted to dismiss 22 irregular workers in order to avoid a costly redundancy plan. The CGT local in Massy had for a while been in contact with these workers. It was prepared to riposte. Its leader, Raymond Chauveau, was an ex-public transport mechanic in his 60s with a marked background in revolutionary Communism. He is a prime example of a ‘conscience constituent’ (McCarthy and Zald, 1977): a supporter who is unlikely to obtain any direct benefit from the possible success of the mobilization. As a somewhat marginal figure in CGT, he was not a novice to sans papiers movements: the CGT local in Massy had already been supporting them for several years. As soon as the first notification of redundancy arrived, the 22 workers gathered in a corner of the enterprise and declared that they were on strike.

The local CGT took a calculated risk in this operation. It counted on a fact well-known by specialists of French labour law. Even if the employment of irregular workers is illegal, once the employer has committed this infraction (knowingly or not), irregular workers benefit from almost the same industrial rights as regular workers, including the right to strike. Unionists believe that in order to make effective these rights (e.g. protect workers against unfair dismissal), their regularization is necessary.

Already at that time, our logic was: they are the same as other workers. They have been working in the enterprise for many years. Therefore, they must be treated like the others. In order to exercise their rights, they need to be regularized. ²

The Modeluxe episode had marked a major discovery. This was the potential of labour law to protect (temporarily) irregular migrant protest against police intervention. The traditional sans papiers repertoire of the hunger strike caters for this vital necessity by seeking sanctuary in churches. As long as the clergy does not call the police the church provides a degree of protection (Siméant, 1998; Mathieu, 1999). Contrary to the workers’ strike, it also may attract other migrants to join in. While collective action in churches may mobilize ‘conscience adherents’ (McCarthy and Zald, 1977) from the parish, the labour strike potentially mobilizes the union by bringing the employment relation into play.

Another effect of such an occupation is that it renders publically visible the presence of irregular workers in an ‘ordinary’ enterprise and allows the pinpointing of an injustice that otherwise would remain unperceived. Hence, it provides for a ‘scene’ that is essential to any form of contentious political action (Cingolani, 2003). Prefectures are obliged to react to the revelation of the double infraction committed by employers and workers. Massy unionists knew from previous experiences that in such a situation, public attention is crucial for pressuring prefectures to act in favour of the workers. Therefore they sought to extend support to civil society. Unionists organized demonstrations and petitions and approached local politicians to sponsor irregular migrants from the community (parrainage républicain). Within the enterprise, the strike presupposed the constitution of an otherwise unlikely collective of workers. Three days after the beginning of the Modeluxe strike, regular co-workers joined in and brought the laundry to a standstill. In January 2007, all workers obtained their residence permits from the prefecture, but were obliged to leave their job.

The Modeluxe experience had established the tactical principles that would guide future collective action. ‘Everything we know by now, we experienced in this particular strike in a condensed fashion. We built on these very intense experiences especially as we could easily theorize them’, underlined Raymond Chauveau. ⁴ On 29 May 2007, he intervened in a similar situation. Twenty-six workers of a Buffalo Grill restaurant in Viry-Châtillon had contacted him. They were threatened with dismissal as police controls had revealed the presence of numerous irregular migrants in the workforce. At the press conference to launch the strike, workers brought forward a new type of justification. They referred to their situation as (exploited) workers: ‘The boss has employed us because we are undocumented; today public authorities must regularize us!’ (Barron et al., 2011: 30). Rapidly, more than 60 workers participated in the strike. Workers in similar situations from other enterprises showed up at the picket line and subsequently spread the word. Migrant rights activists engaged in sans papiers struggles also visited the picket line in the restaurant’s carpark. They felt that the strike could offer new perspectives for overcoming activists’ frustrations associated with the dominant mode of sans papiers action:

I encountered Raymond Chauveau… in Viry-Châtillon, the Buffalo Grill conflict. I had a revelation; in any case, I said to myself that something was happening there. Until then, we were constantly on a conflict that repeated itself endlessly; I mean the occupation of symbolical places. We had entered in a kind of routine that failed to draw the attention of the media. ⁵

Taking on board trade unionists: framing irregular migrant protest as a workers’ cause

In cognitive terms, the discovery of the resourcefulness of the strike repertoire for the claims of irregular workers allowed for the revelation of ‘evidence’ to participating trade unionists and then to the CGT as a larger entity. This evidence proceeded by a series of deductions which underpinned the appropriation of the regularization claim by unionists. (1) Because the irregular migrants make use of their right to strike, they are workers. (2) Because these workers make use of actions associated with and regulated by the strike repertoire (strike assemblies, picket lines, negotiations with employers etc.), they are involved in a labour conflict. (3) Because it is a labour conflict, the union has to feel concerned about it and act on behalf of the workers.

These logical deductions contained a diagnostic, but also provided motivations to actors. Both were basic ingredients of the emerging collective frame of shared understanding (Benford and Snow, 2000) that became crucial for the mobilization of unionists. It displaced the significance of the regularization claim into the universe of the labour movement and activated the dimension of (class) solidarity (see Connolly et al., 2014). Thus, unionists could perceive their engagement as a way to connect to CGT’s past, present and future. ⁶

The question of work, immediately we perceived it as determinate. For a trade union, this question necessarily goes beyond manifestations of solidarity or support, things we have done in the past in our relationships with the sans papiers collectives. This means that we enter a dimension that is completely linked up with our traditions as a trade union that is to defend workers’ rights. Besides, we saw this when we went to live the strike with these workers. We immediately perceived the solidarity. It was of another nature. These were workers, on strike, you know. They displayed the determination to win their struggle for dignity, for rights. We perceived this immediately… From the very beginning, this was more understood as a trade union thing’. ⁷

Importantly, the framing as a worker protest was conducive to letting unionists know what was to be done. The strike allowed them to deploy tactics, principles and knowledge acquired and tested throughout their careers as activists. Unionists gathered local political and financial support for picket lines; they organized demonstrations, strike assemblies and interaction with the media; they negotiated with individual employers and the prefectures over the regularization of striking workers.

In the course of the movement, the competences that unionists brought to the table would be clearly recognized and valued by employers and government officials alike. They provided them with a clear advantage over other external groups (autonomous sans papiers groups; immigrant rights and nationality groups) operating in the field. The latter lack technical and tactical competences to pursue a labour conflict and are considered as unreliable by the authorities. In other words, the strike repertoire clearly built on public authorities, employers, irregular workers and the union as its constituent components. To justify their privileged status in the negotiations, the union in turn would have to prove that it could guarantee a minimum degree of discipline of the movement through an effective leadership. Significantly, and unlike in preceding sans papiers movements which witnessed the emergence of migrant leaders, the leadership of Raymond Chauveau and the CGT’s immigration officer Francine Blanche remained uncontested.

The framing of the movement as a workers’ struggle marked a rupture with the hitherto dominant mode of sans papiers protest, the hunger strike and the occupation of public places. It justified the resources dedicated to the movement against critics within CGT who conceived the regularization struggles as a ‘societal’ and not a ‘union’ issue. Throughout the strike movement, unionists sought to distance themselves from this repertoire. They tended to conceive it as a sordid, individual mode of action, based on an apolitical human rights discourse. To mark its difference, from a very early stage CGT used the term sans papiers only in combination with the qualifier ‘worker’ (travailleurs sans papiers). Claims were thus based on the valorized and positive figure of the worker, striking a contrast with the figure of the ‘excluded’ of ‘humanitarian’ migrant protest. ‘There are no more have-nots, there are workers without permits, but who work. Their point of departure is where they are strong, i.e. they exist in society’. ⁸ Conceiving irregular migrants as subjects of (labour) rights and as parts of a larger politically recognized collective (the union movement) facilitated migrants’ public affirmation of their claims (see Anderson, 2010b) and their ‘deservingness’ (Chauvin and Garcés-Mascareñas, 2012) as candidates for regularization. Implicitly, it also enabled the union to distance itself from the highly contested issue of immigration policies. By insisting on the nature of the claim as an ‘immigrant policy’ issue (Hammar, 1985), CGT reconnected with a historical preference of favouring the extension of rights of resident migrant populations over calls for more liberal admission policies (Penninx and Roosblad, 2000), typically voiced by more radical supporters of sans papiers movements.

Amplifying the strike movement by diffusing the repertoire

The considerable media coverage of the Buffalo Grill strike had revealed to unionists and activists another strategic advantage of this framing. Journalists were intrigued by the images (workers holding their payslips, social security cards and work contracts before the cameras) and accounts (of working conditions and the role of the employers) of irregular workers on strike. By highlighting their degree of inclusion in the host society, the strike offered ‘something new’ that stood in contrast with recurrent, ‘well known’ incidents of sans papiers occupations and hunger strikes. In the sans papiers movements of the 1990s, working and employment conditions had remained in the background in order to avoid negative repercussions on participating workers (ASPLAN, 2009). Unionists however were keen to exploit this precious resource in further mobilizations, recognizing that the media represented a major vehicle for transmitting a social critique into the public sphere (Neveu, 2005).

Other, yet less mediatized strikes followed the Buffalo Grill movement. They confirmed that the strike repertoire – as a ‘module’ (Tilly and Tarrow, 2007) – had the potential to be transferred to other worksites, provided that the conditions were sufficiently similar (an existing workplace collective of irregular workers, holding payslips and labour contracts; the presence of activists to support the occupation of the enterprise). Its effectiveness was however still limited as prefectures were only hesitatingly regularizing workers on strike.

This changed when a major change occurred at the legal level. The 2007 Immigration Law (loi Hortefeux) had introduced the employer as a centrepiece of the case by case admission on economic grounds. ⁹ From then on, the attribution of the newly created ‘salaried worker’ permit crucially relied on the support of the individual employer who could ‘propose’ his or her candidate(s) to the prefectures for admission. On the eve of the economic crisis, this law displayed the Sarkozy government’s preference for economic over rights-based immigration (asylum and family migration). Unionists and migrant activists asked themselves how they could exploit this new regulation. Had the introduction of the employment relationship into the regularization procedure opened the door for pressuring employers to support the candidature of their workers? Under this hypothesis, the strike would not only mobilize public attention, but also introduce the employer as an important strategic ‘ally’ into the relationship with the prefectures and the government. A political opportunity (Tilly and Tarrow, 2007) had opened in which the already established, yet highly unreliable, practice of the sans papiers strike could be possibly of use.

Arguably, a major test took place on 13 February 2008 when Raymond Chauveau and Paris CGT launched the occupation of La Grande Armée, the favourite restaurant of France’s First Lady, Cécilia Sarkozy. Contrary to previous operations which took place after the dismissal of workers, this time CGT wanted to take the employer by surprise. The manager quickly gave up resistance and supported his workers’ claim at the prefecture. After only six days of strike and negotiations, the Paris prefecture backed down. For the first time, irregular workers on strike were regularized on economic grounds. They obtained the salaried worker permit introduced by the 2006 Immigration Law (loi Sarkozy II). ¹⁰ The strike had proven to be a means to obtain the regularization of workers outside the list of 30 jobs established by the Ministry.

When the conflict appeared on the national evening news, CGT activists felt that the general public might be ready to take on the new language of the movement. They identified hundreds of workers eager to go on strike. The moment had come to amplify and scale up the movement to the regional level, thus moving beyond the typical local character of sans papiers protest. In order to maximize the number of permits, pressure on local prefectures and, indirectly, the government had to be stepped up. In order to do so, Chauveau needed the political support of his confederation and contacted CGT’s national level. Together they decided to coordinate a strike movement in the Greater Paris region, home of the major part of French recent immigrant populations.

The tactics were to centre the movement, at least initially, on cases which were most likely to satisfy local prefectures and to generate sufficient pressure on employers: long-term resident (male) workers with permanent full-time employment contracts in enterprises characterized by the presence of several irregular workers. This decision would soon provoke a major crisis in the movement. The autonomous sans papiers group CSP 75 accused CGT of wanting to exclude its ‘isolated’ ¹¹ members. The conflict led to the occupation of the Paris Union House that lasted for over a year and was ended by CGT forcibly expelling the collective.

For three months, a very restricted circle of CGT and political activists had prepared the movement in secret. On 15 April 300 workers set out to occupy 18 workplaces. Amongst them were construction workers, cooks and janitorial workers. Rapidly, the movement became an issue for national politics. Raymond Chauveau and Francine Blanche were received by the Ministry of Immigration. In the talks CGT obtained the promise that the applications of the striking workers would be ‘regarded positively’ by the Greater Paris prefectures. However, the ministry categorically refuted the idea of a mass regularization of striking workers. Each case would continue to be considered individually by the prefectures, granting a maximum control over the number of permits issued.

The strike provoked immediate, yet contrasting reactions amongst employers’ organizations. SYNHORCAT (Syndicat national des hôteliers, restaurateurs, cafetiers et traiteurs) and UMIH (Union des métiers et des industries de l’hôtellerie), the two main hotel and restaurant employer federations, called for the regularization of some 50,000 workers in their industry, thus admitting that the employment of irregular workers was common. The principal employer confederation MEDEF (Mouvement des enterprises de France), tightly linked to the government and French big business, however, refrained from commenting on the issue. The construction employer federation FFB (Fédération française du bâtiment), representing a major industry for (male) migrant employment, publicly opposed any legislative changes. Individual employers hit by the strike were initially outraged, but most of them conceded under pressure.

To maintain pressure on the prefectures, on 20 May 2008 CGT launched a second wave of strikes. Another 321 workers occupied 23 enterprises. In total, between 600 and 700 workers were on strike. In June 2008, CGT announced that out of 1400 applications, 400 had already been treated favourably. In an interview with Le Monde on 16 June, Chauveau suggested the possibility of a third wave of strikes. However, within CGT resistances had built up. Unionists participating in the strike were afraid of becoming overwhelmed by an extension of the movement. Since April, day and night unionists and activists had been present on the picket lines in order to avoid manipulation by bosses; others were busy filing applications and accompanying workers to the prefectures. Their engagement collided with other events on the union agenda. Internally, alternative activist resources were rare as the CGT industry federations officially did not participate. To lessen the strain on the organization, in June 2008 CGT decided to take a break from launching new strikes and to observe the developments within the prefectures. A serious brake was thus put on the desire of numerous workers waiting impatiently to be ‘launched’ by the CGT. A phase of ‘cold regularization’ set in (Barron et al., 2011: 146). The CGT opened permanencies in local union offices in which dedicated full-time officers filed and followed up applications and accepted new ones. Was this the end of the mobilization?

Adapting the repertoire to new categories of workers and scaling up the movement

Increasingly faced with the unwillingness of prefectures to concede any further permits, unionists started discussing in spring 2009 the need to reestablish pressure on the authorities. The profile of the majority of the workers who contacted the union as well as the occupation of the CGT Union House had revealed the diversity of sans papiers employment (ASPLAN, 2009). Even if a significant number carried out a job corresponding to the standard employment norm (declared, stable and full-time), the majority was concerned by casual and informal employment. ¹² This imposed a new question: how could the movement take account of part-timers, temporary agency workers, housemaids, isolated or undeclared workers?

These workers not only lack leverage in the workplace. Many of them also fail to meet the criteria for regularization as they cannot produce a consistent ‘paper identity’ (Chauvin and Garcés-Mascareñas, 2012), obtained by accumulating formal proofs of presence (tax notices, rent receipts, employment contracts etc.) in order to be able to prove their past existence. Their employment characteristics also do not meet the exigencies of front-line bureaucrats who are instructed to valorize formal and continuous full-time employment as an asset if not a condition for granting (and renewing) permits. The decision was therefore taken to scale up the movement to the national level by explicitly targeting change in the criteria for regularization. Against calls for extending the movement beyond the scope of the Greater Paris region, CGT leaders decided to continue to rely on the proven mobilization structures (McAdam et al., 1996) to confront the government on its home turf.

If a future movement was to broaden its social base and address the legislator, the CGT was convinced that its organizational foundations would have to be enlarged, too. In September 2009, CGT’s Francine Blanche invited four union confederations (Solidaires, UNSA, CFDT, FSU) and six migrant support groups for discussions. At the meeting she announced that the CGT’s national executive board had finally approved the launch of a new major strike wave. She invited the other groups’ representatives to participate in it. In order to open the way for the ‘regularization of all sans papiers workers’ (and not only the striking workers), the CGT sought to go beyond the informal arrangements with the Ministry and prefectures followed in the preceding waves. It called for a government circular to establish simplified and improved eligibility criteria applicable to all irregular workers, independently of their nationality, employer or the place of their prefecture. The participants of the meeting approved this call and signed a letter to the Prime Minister. As it remained unanswered, the newly constituted Group of eleven took the decision to launch the mobilization and to enter directly into conflict with the government.

On 11 October 2009, a preparatory assembly for the strike wave was held at the CGT national headquarters in Montreuil. A total of 2000 migrant workers attended the meeting. Their profile was more heterogeneous than in 2008. Temporary agency, undeclared and recently dismissed workers were well represented. Part-time workers, mostly women, were also present. Some of the participants had already participated in the previous movement, but failed to obtain their permits. Alongside workers from francophone Africa, for the first time a sizeable population of Chinese origin participated. The next day, more than 1300 workers went on strike. By 24 October, 4000 irregular workers were occupying some 40 sites. At its peak, CGT registered 6800 striking workers. Notwithstanding these impressive numbers, in the context of the economic crisis the movement this time had many more difficulties to capture the media’s attention and obtain civil society support.

In order to cater for the more heterogeneous worker profiles and the call for new criteria for regularization there were several innovations in the repertoire of collective actions. First, each picket line had to elect strike delegates who would take on a major coordinating role. In part, this measure responded to the unions’ difficulties of providing sufficient activist resources. From now on, regular assemblies were held with delegates and representatives of the Group of eleven. The delegates had to report back to their picket lines all the information provided on these occasions. Delegates were crucial for communicating and explaining the shift in the movement’s objectives and tactics to the striking workers. Contrary to preceding waves, application files were not to be handed to the prefectures until the government had issued a national text with improved criteria for regularization. Collective discipline was paramount. Moreover, from now on, each striking worker held a nominative strike card to protect her or him against police arrests. The card also allowed delegates to check individual participation in the picket lines and to increase discipline by threatening to take back the card (Barron et al., 2011: 199).

Secondly, the participation of casual workers entailed new ways of conducting the strike. While most of the enterprises were occupied by their own employees, several collective sites provided a means for isolated workers who normally would not have the opportunity to participate in the strike. On 12 October, almost 400 workers occupied the national headquarters of the public works employers’ federation FNTP (Federation nationale des travaux publics). Amongst them were undeclared, temporary agency and isolated workers from small-sized construction firms. About 1000 cleaners occupied the headquarters of SAMSIC, a major player in the French office cleaning market. The training fund of the hotel trade FAFIH (Fonds d’assurance formation des industries de l’hôtellerie) was chosen as a site to gather several hundred isolated restaurant workers. Three agencies served as a picket line for temporary agency workers. The numbers at these collective sites were massive, but quickly employers successfully contested the tactics in court on the grounds that the striking workers did not have an employment relationship with the employer or organization in question.

On 28 October, the movement was hit for the first time. Police evicted 650 construction workers from the FNTP headquarters. In November, the SAMSIC site was equally threatened by an eviction procedure. Meanwhile, the government still had not issued a text. Temporary agency workers who were expelled from an agency organized the occupation of another agency. The ‘travelling strike’ was invented (Barron et al., 2011: 211).

The collapse of the strike movement: the renaissance of the excluded migrant

On 16 November, one month after the beginning of the wave of strikes, the Minister of Immigration issued a circular that merely repeated existing regulations and affirmed the discretionary powers of the prefectures over admission. The Group of eleven and the strike delegates unanimously rejected the text. At a press conference the same day, the Minister of Immigration declared that only ‘between 500 and 1000 workers’ will benefit from the circular, much fewer than the 2800 regularizations obtained in the 2008 strikes. The position of the government had hardened. In the context of the economic crisis, the opening towards ‘economic immigration’ no longer featured on the government’s agenda.

Despite enormous disappointment amongst the workers, the decision was taken to continue the movement in order to obtain an improved text. Also, it was agreed that workers were to set out to obtain a binding offer of employment in order to comply with the basic eligibility criteria imposed by immigration law. However, from this point, the movement started to decline. On 2 February 2010, the last remaining major strike site for isolated workers, the training fund of the construction industry FAF-SAB (Fonds d’assurance formation des salariés de l’artisanat du BTP), was evacuated by the police. Striking workers and activists were exhausted after months of occupation; some workers had restarted work without admitting it. CGT sought to compensate the loss of the major strike sites by mobilizing the support of artists as well as employers. In March 2010, a ‘common approach’ to regularization was signed between several minor employer organizations and CGT. Yet, the government failed to consider the initiative.

CGT still had one ace up its sleeve. On 27 May 2010, about 1000 irregular workers participated in the Parisian trade union demonstration against the government’s pension reform plans. Upon its arrival at the Place de la Bastille, the symbolic epicentre of the French revolution as a plebeian upheaval, Raymond Chauveau guided the irregular workers’ cortege to the staircase of the Bastille Opera. He publically declared that workers would occupy the place until the government conceded precise eligibility criteria. The CGT leadership argued that the signature of the common approach with employer organizations ruled out new occupations of enterprises. Certain unionists contested this position and tensions rose within the Group of eleven. The government’s refusal to negotiate had worsened the conditions for the movement and brought back to the forefront the figure of the excluded migrant.

The ultimate Bastille occupation took into account this situation. It publically exposed the desperation of exhausted workers and thereby denounced the violence exerted by the authorities. In the direct confrontation with the government, employers had disappeared from the scene of contention. The images revealed by the occupation of the Opera esplanade were disturbing. Was this still a workers’ strike or the occupation of another public place by a sans papiers collective? (Barron et al., 2011). Participating migrants, most of them from previously evacuated picket lines, unfurled CGT banners and put up plastic covers in order to protect themselves against the bad weather. For almost a month, during night and day up to 1000 workers would occupy the esplanade. By resorting to the occupation of this public place CGT had implicitly acknowledged that the strike had become insufficient to move the government.

On 2 June, police forces tried in vain to remove the occupants. CGT strike leadership sought an honourable way out of the situation. On 18 June, Raymond Chauveau announced to the Bastille crowd that the CGT had obtained an addendum to the circular from 24 November. The eligibility criteria listed in the text were the most favourable the movement had ever obtained and were, in principle, applicable to all workers. ¹³ Yet, the text had no legal value; it was officially intended to ‘guide’ prefectures. The CGT announced the end of the remaining picket lines.

Throughout the summer, activists were extremely busy preparing applications and submitting them to the prefectures. Yet, the results were disappointing. By the end of 2011, out of 4000 applications processed by the Group of eleven, only 800 had been followed by a permit. In particular, undeclared workers, temporary agency workers as well as workers who could not prove five years of continuous residence were refused permits. The main protagonists of the CGT defended the addendum as a major success. Other participants of the Group of eleven and Paris CGT were more critical. A Solidaires union officer affirmed: ‘We still do not have the final results, but a priori these will be relatively disastrous. We did not obtain the text we wanted’. ¹⁴

Conclusion

This article has focused on an incidence of the enlargement of the group-specific repertoire of French sans papiers protest. The strike appears as sufficiently specific both with regard to its means of collective action (occupations, work stoppage) and the interacting groups which characterize it (employers, prefectures, irregular workers, unions) to be considered as a significant addition to this repertoire. In order to allow for this extension, several innovations were necessary in the strike repertoire. They have been traced through a mostly sequential account of the movement, highlighting the uncertainties and complexities which characterize the choice of a mode of collective action of a population weakly equipped in political resources.

Emphasis has been put on the discovery of the significance of the strike for irregular migrants’ claims in a single event, the occupation of the Modeluxe laundry. As a legal resource it allowed for the protection of workers against immediate police arrest; as a scene for contention it highlighted that irregular workers are employed by ordinary businesses and that employers seek to benefit from workers’ vulnerability by dismissing them without further notice; as a cognitive resource it allowed for linking the regularization claim with the need for protection of otherwise ordinary workers; finally, as a practical resource it provided a place for unionists in the mobilization due to the expertise in ‘their’ repertoire of collective actions. Building on these initial experiences, a group of dedicated local CGT unionists and social movement activists strove to diffuse this blueprint to other worksites.

The innovations in the strike repertoire which were necessary to cater for the claims and the diversity of the irregular worker population were the result of an iterative, decidedly decentralized process of experimentation in the evolving confrontation with the authorities and the demands of migrant workers. It took place on very uncertain and risky grounds as every step had to be carefully considered in order not to expose workers to the danger of detention. However, these initial experiences only revealed their potential for giving way to a larger dynamics when a change occurred in the legal framework in early 2008. It was the introduction of the employer to the regularization procedure that allowed activists to submit the modularity of the strike to a serious test. This confirmed the importance of a leadership capable of recognizing a political opportunity and highlighted, more generally, the reactive nature of sans papiers mobilizations (Siméant, 1998). By 2009, the political situation had changed as the economic crisis had convinced the government to abandon its cautious opening towards ‘economic immigration’.

The leading role that CGT played in this movement fundamentally stemmed from its recognition by the constituent actors of the strike repertoire. As experts for initiating and resolving labour conflicts, unionists were considered as competent and reliable by workers, authorities and the employers alike – if only in comparison to other possible interlocutors. As the movement extended, from such basic relations of recognition power relations arose. They comforted the union’s ‘intermediary’ (Müller-Jentsch, 1982) role in the movement. While employers and authorities valued the capacity of the union to exert a degree of control over the movement and to provide a public service by selecting applicants and filing applications, striking workers could see in the unions’ straightforward relationships with prefectures and the government an efficient means to become regularized and obtain legislative change.

These virtual mechanics only broke down when the procrastinating tactics of the authorities as well as the limits of the strike repertoire to offer sufficient leverage to those who could not pressure their employer directly took their toll. This not only revealed the – well-known – difficulties of casual workers to go on strike (Barron et al., 2011). It also shed a light on the criteria pertaining to ‘migrant deservingness’ for accessing formal civic status. Holding a regular (full-time, single-employer and permanent) job has increasingly become valorized by contemporary migration policies as a condition for both granting and renewing permits (Chauvin and Garcés-Mascareñas, 2012). These criteria clearly operated against the more precarious workers participating in the 2009/2010 strike waves. As the French example has suggested, this emerging moral economy of the ‘good illegal’ (Chauvin and Garcés-Mascareñas, 2012) not only allows for the sponsorship of workers by their employers. It is in the interplay between irregular migration as an increasingly durable mass phenomenon due to restrictive immigration policies and its fuzzy openings for workers that new opportunities emerge for trade unions and other labour-related organizations to act as an ‘immigrant rights movement’.