Abstract
Formal organizations advancing the goals of identity-based social movements and identity groups have become increasingly interdependent. The former often lacks legitimacy in the eyes of stakeholders and the latter typically possesses insufficient organizational capacity. In principle, the transfer of ideas and resources between formal organizations in social movements and social identity groups can result in organizational innovation that revives the formal organization while at the same time enhancing the status of the identity group. But in practice, collaborations between formal organizations and identity groups often result in identity groups being overpowered by formal organizations. This article compares outcomes for identity groups in two cases of trade unions adopting the causes of identity-based social movements to examine the role of organizational processes in explaining outcomes for identity group members. The findings from the comparative cases analyzed here suggest that identity group members must be able to influence organizational processes in order to impact how they are incorporated into the formal organization.
Introduction
Formal organizations advancing the goals of identity-based social movements and identity groups have become increasingly interdependent in the past few decades. The former often lacks legitimacy in the eyes of stakeholders (Donaldson and Preston, 1995) and the latter typically possesses insufficient organizational capacity (Rucht, 2004). The transfer of ideas and resources between formal organizations in social movements and social identity groups can potentially revive the formal organization and enhance the status of the identity group. However, often collaborations between formal organizations and identity groups result in identity groups being overpowered by formal organizations (Giugni, 1998). Scholarly literature has documented many benefits for formal organizations that ‘borrow’ cultural codes, rhetoric, and protest tactics from identity-based social movements (Isaac and Christiansen, 2002). Scholars have also pointed out the resource benefits for identity groups allying themselves with larger organizations (Van Dyke and McCammon, 2010). Whaling industries’ symbiotic relationship with communities that depend on whaling as a lifestyle, or political parties and the farmers that they represent demonstrate the mutual reliance between social movement organizations and identity groups. But there is scarce empirical or theoretical work on the variability of outcomes for identity groups when formal organizations adopt their causes. Predictions of outcomes for identity groups have been either uniformly positive – as in the thesis that representation by formal organization leads to integration – or uniformly negative – as in the proposition that formalization of social movement organizations leads to their co-optation by donors and electoral politicians and distances them from the original beneficiaries. By formal organizations I mean functionally differentiated organizations representing the process of rationalization in society (Lawrence and Lorsch, 1967). An identity group is a collectivity whose self-concept is defined by membership in a relevant social group, such as nationality, ethnicity, race, sexuality, sex, or age.
This article compares outcomes for identity groups in two cases of trade unions adopting the causes of identity-based social movements. The first of the two cases is the adoption by the Service Employees International Union (SEIU) of immigrant movement goals in its Justice for Janitors campaign. The second is the Harvard Union of Technical and Clerical Workers’ (HUCTW) adoption of feminist identities and goals in its organizing. The two cases are analyzed as critical cases that assist in understanding different organizational processes that institutionalize social identity. Trade unions have lost legitimacy in the eyes of their stakeholders and the general public in recent decades. They are examples of bureaucratized organizations reaching out to identity groups and seeking organizational revival through identity-based social movements. Examining the integration process of identity groups into trade unions can enhance our understanding more generally of outcomes for identity groups when they collaborate with formal social movement organizations. Based on the findings, a process-based model is developed for examining outcomes of the adoption of identity-based repertoires by trade unions. General implications of the model for collaboration between formal organizations in social movements and identity groups are discussed. While earlier theories have been uniform in predicting either entirely positive or negative outcomes of collaborations between formal organizations and identity groups, this study advances a nuanced understanding of the conditions under which integrative outcomes prevail over differentiating outcomes. Integration is defined as the state where identity claims are no longer perceived as ‘special’; instead they become part of ‘normal’ organizational process (Frank and Meyer, 2002; Meyer, 1992). The findings from the comparative cases analyzed here suggest that integration is not automatic. In the short to medium-term, identity group members must be able to influence organizational processes in order to impact how they are integrated into the formal organization.
Formal organizations, social movements, and identity groups
Formal organizations are defined here as organizations whose operations are based on written rules (Blau and Scott, 1962), whose units are differentiated by function, and whose ‘functionaries’ work exclusively toward stated goals of the organization (Merton, 1940). Identity groups are groups of individuals who are members of a collective defined by its social identity, whether this identity is circumscribed historically, as in ethnicity, or naturally, as in sex (Melucci, 1989: 34). Scholars have argued that the erosion of previously important boundaries such as class and the nation-state after the Second World War has lent potency to claims based on social identity, noting that the emergence of ‘new’ social movements has replaced earlier claims-making patterns based on class and citizenship (Offe, 1985). Organizational theory has surmised that the adoption by formal organizations of identity-based claims results in the integration of the identity group into rationalized arenas of society (Scott and Meyer, 1994: 211–12). When ‘specialized’ claims are adopted by formal organizations, organizational theory predicts they will cease to remain specialized and become ‘normal’: ‘To the extent that the specializing identity claims are successful, they are absorbed into normal personhood, as ordinary individual needs, tastes, preferences, and choices’ (Frank and Meyer, 2002: 100). Examples of such integration can be found in the generalization of affirmative action as an outcome of successful identity claims from ethnic minorities and women, or legal abortions as the normalized form of specific claims around women’s rights to control their bodies. In this study, integration is defined as the normalization of identity-based claims as they are taken up by formal organizations operating in a social movement. Integration is understood not as an all-or-nothing outcome but one that varies in degrees and modes.
The literature on social movements leaves open the possibility that identity claims can be used instrumentally, in which case the kind of integration predicted by organizational theory cannot be presupposed. This literature has distinguished between instrumental and expressive movements, whereby movements are instrumental if they deploy collective action only as a means for policy change, and expressive if they value collective action and the construction of a movement community as goals in themselves (Bernstein, 1997; Williams, 2004). To date the strategic nature of identity deployment has mostly been discussed at the level of the movement, but a small subset of this stream of literature has examined the role of organizations. Bernstein (1997), for example, showed that the extent to which lesbian and gay organizations engaged in grassroots mobilization influenced whether at a specific time the gay and lesbian movement deployed identity instrumentally or expressively. The current study builds on the insight from recent work that has maintained that organizational processes influence logics of action toward identity deployment (Polletta, 2002). Logics of action are material practices and social constructions available to organizational actors to justify their actions (DiMaggio, 1997). When logics mature, they become institutionalized and less open to revision over time, thereby becoming durable influences on organizational outcomes (Bacharach et al., 2000: 727).
Formal organizations as carriers of social identity
Social movement organizations typically invoke the social identity of their members to elicit action from their members, or to achieve legitimacy in the eyes of stakeholders (Davis, 2000; Gamson, 1995). Organizational and social movement theory recognize that organizations are strategic about borrowing identities and cultural repertoires from other fields, modifying and re-framing them so that they resonate in local contexts (Benford and Snow, 2000; Campbell, 2005). Adopting new identities can have profound effects on ‘carrier’ organizations – organizations that transport identities and cultural repertoires from one environment to another (Meyer, 1996; Scott, 2001). Day-to-day practices and formal structures can also be impacted and processes may need to be developed to ensure that new goals are met (Sahlin and Wedlin, 2008: 220). If existing structures and practices of carrier organizations cannot adequately accommodate new identities, new organizational forms may be developed (Rao et al., 2000). For example, Rao et al. (2003) document the rise of nouvelle cuisine as a new organizational form created to accommodate the expanded role identities of chefs who rebelled against classical cuisine. The resulting organizational form is disseminated throughout the organizational field of the carrier organization (Sahlin-Andersson and Engwall, 2002). However, we know relatively little about how social identities enter social movement organizations and become institutionalized as part of organizational processes. The approach taken here is to examine the institutionalization of social identity as a potential mechanism explaining divergent outcomes for identity groups’ integration into formal social movement organizations.
Research context: Trade unions and identity-based social movements
Trade unions have experienced a progressive decline in power and influence in recent decades. The decline of trade unions in the USA has at least partially been attributed to their failure to incorporate identity groups – African-Americans, women, and recent immigrants – into their fold (Lichtenstein, 2002). Importantly, the organizational revival – defined here as growth in membership and the reversal of decline – of a select group of trade unions since the 1980s has been attributed in large part to their success in recruiting identity groups into their membership (Milkman and Voss, 2004). Conventional unions relied on institutionalized sources of power vested by the labor law and collective bargaining institutions such as mediation and arbitration (Barbash, 1984). They organized new members through formally sanctioned workplace elections, securing a stable membership based on the full-time and long-term nature of employment in unionized sectors. Unions that espoused social movements rejected the organizing process of conventional unions as they believed these were ineffective in motivating and mobilizing a demographically transformed workforce. The majority of this workforce was women, immigrant and ethnic minority workers who had been excluded from typical union jobs in manufacturing industries. These unions made workers’ social identities and social injustices central to their organizational rationale and strategies.
Scholarly attention has mostly examined the strategic innovations that led to some unions embracing social movements (Clawson, 2003; Milkman and Voss, 2004). With a few exceptions (e.g. Voss and Sherman, 2000), intra-organizational processes of change in these unions have not been examined, despite anecdotal observations suggesting that some organizations experienced more conflict than others in espousing social movements, and that in some change was top-down while in others change was driven from the bottom up (Tait, 2005). Furthermore, whether union revitalization has resulted in the integration of identity group members into these organizations has hitherto not been systematically addressed.
Method
Case selection
Case studies are appropriate for studying organizational change because the concepts involved – power, interests, and conflict – are complex, difficult to measure, and unfold over time (Greenwood and Hinings, 1996: 1044). A comparative case study (Eisenhardt, 1989) of two organizations is undertaken here. In addition to theoretical sampling based on the independent variable of interest (Strauss and Corbin, 1990), intra-organizational processes of change, other criteria for selecting the cases include the following. Cases were selected so that the organizations’ approach to identity groups was not a foregone conclusion. Both the SEIU and the HUCTW were committed to the causes of the identity groups they represented (Fantasia and Voss, 2004; Hoerr, 1997), and in the early years of adoption their divergent logics of action could not have been preordained. The cases are also contemporaneous such that the comparison is not jeopardized by differences in historical context. In addition, the organizational field was held constant by choosing two unions, rather than organizations operating in different fields. Finally, to control for organizational capability, both cases were selected from among unions that were successful not only in adopting new organizational forms based on social identity but also in disseminating these templates to other unions. The cases are treated here as illustrations of two contrasting processes, and not as matched cases. That is, the article aims to build theory, and not to test a causal relationship. Other factors that may help explain the different outcomes of incorporating identity groups are discussed toward the end of the article.
The SEIU was founded in 1921 as a decentralized and loose network (Piore, 1994: 525) of low-skilled service workers, mainly janitors and building maintenance workers. In the early 1970s the union experienced precipitous decline in its building services sector owing to subcontracting and the proliferation of part-time work. I focus here on the SEIU’s Justice for Janitors (JfJ) campaign, which was undertaken to re-gain the union’s foothold in the office cleaning occupation. Reflecting the workforce in this occupation, a majority of whom are recent Hispanic immigrants, the JfJ campaign placed the immigrant identity and the legacy of Latin American peasant revolutions at the center of its organizational reinvention. The campaign developed out of experiments in mobilization carried out by SEIU organizers in the early 1980s to organize immigrant janitors, who until that time unionists had widely believed to be ‘unorganizable’ (Milkman and Voss, 2004). The explosive success of these early experiments had led the SEIU to replicate the JfJ model in 21 other cities by the time of this study. The outcomes of this radical change (Greenwood and Hinings, 1996) have made the SEIU one of the most powerful unions in the USA. With 2.1 million members, it is now the second-largest union in the country that is highly influential in state- and federal-level politics (Fletcher and Gapasin, 2008). The SEIU’s JfJ has been a model to be followed by several other labor unions as well as non-labor activist organizations (Fletcher and Gapasin, 2008: 131; Manheim, 2001: 119).
Like immigrant janitors, female clerical workers were historically not considered organizable by industrial unions whose membership consisted primarily of male blue-collar workers (Feldman, 1980). The organizing of Harvard University’s clerical workers – including secretaries, laboratory technicians, and medical assistants – germinated in the early 1970s amid discussions of gender discrimination at Harvard’s medical campus. In the medical campus context, most technical workers were women whose career progression to professional research jobs had been stunted (Hoerr, 1997). The first 14 years of the Harvard workers’ organizing efforts were consistent with conventional unionism and proved difficult because existing union practices conflicted with gendered experiences at work. The first union election across Harvard’s three campuses in Cambridge was unsuccessful, dealing a major blow to organizers who were forced to re-think their approach. Major changes in the union’s approach to organizing resulted in a successful election in 1988, and the union has not lost an election since. While slower to grow in membership than the SEIU model (currently the union has approximately 4600 members), the HUCTW’s organizing model has been disseminated to unions at the University of Minnesota and the University of Massachusetts, with ongoing organizing at Tufts University.
Importantly, both unions’ workforces were members of identity groups since the union’s inceptions. Early SEIU janitors were of immigrant backgrounds and, in the 1960s and 1970s, of African-American background; the majority of HUCTW’s clerical workers were women. However, organizational processes in the two unions followed conventional unionism until the unions adopted identity as central to their organizational processes and doing so triggered radical changes in the respective organizations. Figure 1 provides a timeline of events in the two unions’ adoption of social identity.
Timeline of events.
Data collection
Data collection on the SEIU’s JfJ campaign proceeded as part of a larger dissertation project. Historical data on the SEIU’s organizational change process during the time of the campaign from 1980 to 1996 were collected from the SEIU’s official archives at the labor archives of the Wayne State University. These data included minutes of the union’s national convention (held every four years), internal documents of the Organizing Department, and correspondence between the Executive Office and local unions. The archival data were supplemented with data from the author’s fieldwork in four SEIU local unions (Boston, Houston, Los Angeles, and Washington, DC) in 2006, which included observations of staff and members’ meetings and interviews with 77 SEIU staff and 85 members. Of the SEIU fieldwork data, the parts analyzed for this article pertained to incidents that demonstrated how social identity was brought into organizational processes.
Data collection on the HUCTW included two interviews in 2006 with Kris Rondeau, a lead union organizer who has been with the union since 1977. In 2007, four additional interviews were conducted with union members who were part of the Harvard-HUCTW contract negotiating team, and an online survey was administered to union members as part of a study of the union’s contract negotiations with Harvard (out of 42 union members contacted, 23 answered the survey). In addition to primary data collection, in-depth interviews with two HUCTW lead organizers by Lisa Oppenheim (1991), interviews of HUCTW organizers by Lydia Savage (2007) as well as several empirical studies on the history of the HUCTW (Eaton, 1996; Hoerr, 1997; Kurtz, 2002; Savage, 1996; Valli, 1986) were analyzed in order to construct a historical account of organizational change and continuity in the HUCTW.
Data analysis
The cases were analyzed through an iterative process of examining the data in light of extant theories followed by refining the theoretical framework based on findings from the cases (Eisenhardt, 1989). A ‘first analysis’ (Van Maanen, 1979) of the data openly explored the relationship between the adoption of social identity and organizational processes. This analysis rendered the finding that bringing in social movement elements into the two unions had triggered organizational change, and that distinct logics of action emerged from experiences in previous stages of change. These logics of action were linked to different strategies in integrating identity groups into the organizational fold. Organizational change is a political process where relationships of power and control are re-configured as new practices and structures emerge to address newly acquired goals (Hargrave and Van de Ven, 2006). Thus, a second analysis asked of the data whether new power configurations rising in the process of organizational change favored the inclusion of identity group members in decision-making. Organizational forms that had developed out of the change process were reviewed in terms of their governance structures and key practices (Greenwood and Hinings, 1993). Additionally, I examined how social identity was institutionalized into four practice areas identified as key to trade union activity (Barbash, 1984; Bronfenbrenner and Hickey, 2004). Organizing pertains to the union’s interaction with workers – influencing their decisions to be represented by the union, and, once they are ‘organized’ as union members, to continue being engaged in the union. Bargaining refers to the union’s interaction with employers – negotiating a labor contract, and thereafter, dealing with day-to-day contract implementation. The union interfaces with its environment through recruitment, the selection of talent into the ranks of paid union staff (Tolbert, 1988). Internal development is a union’s work in building leadership among its members. Finally, following Meyer’s (1992) conception of integration, analysis of outcomes focused on the extent to which identity groups’ specialized claims had been mainstreamed into the goals and practices of the overall organization, thereby becoming ‘general.’
Adopting social identity into the organization
The adoption of social identity triggered organizational change in both unions and engendered logics of action that influenced the establishment of distinct organizational forms in each union.
SEIU
Upon observing initial successes in JfJ campaigns in Denver (in 1986) and Los Angeles (in 1987), the national union expanded the JfJ to other cities. Over the years, the national union deployed hundreds of organizers, developed a standardized approach to organizing, and coordinated market research, organizing, and bargaining across different markets (Manheim, 2001: 120). Local unions saw the introduction of new personnel and practices as an intervention by the national union in their affairs; they expressed their resistance in diverse ways ranging from refusing to cooperate to deliberately sabotaging campaigns (Yu, 2007). Both in interviews and in archival documents, countless references to the conflict between proponents of the new model and adherents of the old model were found. These appeared to stem from different world views, training, and career aspirations (Yu, 2011). From early on, the skills sets and outlook of existing union staff were seen to be irreconcilable with new practices. In a memo to the national president, a senior staff member of the Organizing Department stated:
Each of these locals can be seen to be overstaffed in one way or another. Often, it is that there are too many service reps for the current membership. At the same time there are new staff members hired to do organizing. It is rare that the local has been able to convert a service rep to an organizer under the new types of projects we help locals create. (SEIU, 1986–92)
The national union responded by establishing rules of conduct and making financial subsidies to local unions conditional on compliance with new rules (Voss and Sherman, 2000). Increasingly in the 1990s, organizing the social movement way was equated with building power across multiple markets. A focus on operating across different geographic markets justified increased centralized coordination by the national union. As a strategy document issued by the national union in the mid-1980s declared, ‘Communality of purpose will naturally require agreement on the part of participating locals to accept a central point of coordination and direction, the [national union]’ (Organizing Department collection, 1986–92). Thus, both resources (through increases in dues collected from local unions by the national union) and expertise (through movement of key personnel into the national union) flowed toward the center. Having brought in large numbers of young activists, and having once witnessed rapid decline, the SEIU developed a militant credo that prioritized building organizational power (Fantasia and Voss, 2004: 116). Social identity was appreciated chiefly for its effectiveness in galvanizing workers and adding flexibility to the union’s existing range of tactics. Hence, the SEIU developed an instrumental logic with regard to social identity.
HUCTW
In contrast to the SEIU, organizational change at the HUCTW after adopting identity as a central principle did not result in heightened conflict or large structural changes. The union remained decentralized, and the bulk of the union’s work was carried out by its members rather than by paid staff. When, after 14 years under the conventional approach of appealing to clerical workers’ sense of economic deprivation, the union’s plans did not bear fruit, organizers concluded that a stronger community needed to be built. Commitment to building relationships meant that the union had to resist pressures by outside unions seeking the Harvard workers’ affiliation to produce quick results in organizing. Hence even after affiliating with American Federation of State, County & Municipal Employees (AFSCME) in the late 1980s, organizers insisted on maintaining independence in managing union affairs (Eaton, 1996; Hoerr, 1997). The union kept a minimum number of full time organizers (there were five full-time staff at the time of this study), most of whom had been involved in early organizing efforts since the 1970s and had extensive knowledge of the membership. In contrast to the SEIU’s use of social identity to build market power, the HUCTW eschewed distributive tactics. Rather than seeking to beat management at its own game with decidedly fewer resources, the union chose to question the rules and assumptions behind the game:
To the extent that we organized around issues in the past, Harvard was always able to match us. We put out a button once that said ‘Dignity, Democracy, and a Dental Plan.’ Thought it was very cute. You know, the next day we had a dental plan . . . So what we said is, ‘This is about power and self-representation and we’re good enough that we can get in the room and represent ourselves.’ (Interview with Kris Rondeau, Oppenheim, 1991: 54–5)
The HUCTW’s approach to social identity was based on a transformational logic resting on the belief that both organizers and workers could be provided the support necessary to come out of their shells and stand up for themselves (Savage, 2007). Transformation was experienced as a relational process whereby workers, by mutually supporting one another, found the strength to build an organization that was unlike any other union in reflecting the identities of its constituents.
Institutionalizing social identity
This section explains how social identity was institutionalized into four practice areas of the two unions: organizing, bargaining, staff recruitment, and internal development of members. Table 1 summarizes the distinct organizational forms developed in the two unions that resulted from the institutionalization of social identity and provides supporting evidence in the form of quotes and field observations.
Supporting data for institutionalization of social identity in the two unions
SEIU
Following Jenkins (1977) and Osterman (2006) who studied bureaucratized organizations that nevertheless maintained a high degree of ideological progressivism, I characterize the ideologically progressive and highly efficient bureaucracy that the SEIU developed as a radical bureaucracy. This organizational form in the SEIU practiced strategic organizing, confrontational bargaining, and reliance on expert knowledge of markets and the legalities of protest.
Organizing
The aim of the SEIU’s organizing was to lift as many workers out of poverty as possible by unionizing their jobs. The union considered conventional union organizing slow and vulnerable to attack by employers (Lerner, 1996). In the union’s words, ‘strategic organizing’ sought to ‘organize at a fast enough pace that we increase the percentage of workforce that we represent’ (SEIU, 1991–2). To ensure success, the union prioritized organizing geographic markets in one industry where a few employers, preferably including those who already have a relationship with the union in other cities, held large shares of employment. Pressure on employers to show ‘neutrality’ while the union collected signed cards from workers was exerted by threatening protests, legal action, and ‘corporate campaigns’ (Manheim, 2001), and by mobilizing the union’s networks of elected politicians and community groups. Social identity was incorporated into slogans, worker testimonies and media narratives to motivate protest among immigrant workers, shame employers, and justify external stakeholders’ support. However, as explained below, worker mobilization was part of a multi-pronged effort at unionization rather than its central focus, and identity was primarily used to build more market power.
Bargaining
In the SEIU bargaining was built into the process of organizing: by negotiating with the largest employers first, the union could influence the actions of other employers in the market. Top executives believed that the union’s bargaining power stemmed from the size and scale of the bargaining unit – thus, they strove toward ‘rationalization,’ which was defined ideally as having a single contract for one occupation across an entire industry (SEIU, 1991–2). This approach to bargaining relied on sophisticated market research, specialized negotiation skills, and the ability to mobilize protests as an alternative to negotiations (Yu, 2007). Identity and identity-based narratives supported bargaining in two ways. They placed the union on higher moral ground (Chun, 2005), and motivated protests when negotiations failed.
Recruitment
Both top executives’ commitment to changing culture and practices in the union, as well as difficulties in transitioning existing staff and members to new roles, meant that most staff who came into the SEIU since the union adopted social movement practices were externally recruited. Out of the 77 staff interviewed as part of this study, 64 had never been a member of the union. Staff typically had backgrounds in related fields, such as student or community activism (Rooks, 2004). Staff, too, were part of the identity group the SEIU sought to represent: 58 percent of my interviewees were members of minority groups and 47 percent came from immigrant families. But contrary to the low educational levels of its janitorial members, staff typically possessed higher education degrees or commensurate experience in relevant fields: 74 percent of the 77 interviewees possessed a college degree or an advanced higher education degree, and two were PhD candidates. The union had also developed rigorous training for its staff, further increasing the gap in knowledge levels between staff and union members.
Internal development
Organizers at the SEIU believed that immigrant workers could develop personally and occupationally once their jobs were unionized. Developing union members so that they could transition to more advanced jobs, including jobs in the union, was a stated aim (SEIU, 1992). In practice, however, job mobility for immigrant workers in the cleaning industry was limited and it was rare that a member joined the ranks of union staff. Ironically, members of the identity group were not seen as qualified to do the work required to incorporate social identity into the union. Even routine work such as organizing protests required high levels of planning and legal knowledge. Professional staff worked under explicit instructions that distinguished legal acts based on social movement tactics, such as informational picketing and protests on public grounds, from acts that were illegal under current labor laws, such as picketing for union recognition. In press conferences and meetings observed by the author, such as a West Coast ‘members’ leadership conference’ in April 2006, members’ roles and speeches that incorporated themes of immigrant identity were heavily advised and in many cases scripted by staff. Planning and scripting public speeches was thought to be necessary in order to align them with broader strategies. One staff member who eventually left the SEIU explained her view that the union lacked a long-term view:
I think that when we were talking about leadership building, it was more like here’re your talking points in front of the media and here’s what you need to say. Or you need to get people to the meeting or to get these cards signed . . . But I think that there could have been also further investment in terms of actual leadership building and actually have them be part of the analysis in terms of how the campaign is going. (Personal interview, former staff in Houston, 2006)
Governance structure
The SEIU strove to ‘look like’ its members in its ranks, and, given equal merit, the union promoted people of color and immigrant members to elected positions. Immigrant members were also represented in formal democratic structures such as the national and local union executive boards. However, formal decision making structures were more often used as confirmatory vehicles rather than platforms for debate and discussion in the interest of accelerating decisions. Archival documents reveal that a form of governance developed whereby the national union’s technical departments drafted proposals that were typically ratified by the national convention with minor amendments, if any. This finding was supported by field observations, where a similar pattern of decision-making was operational. While in principle there were opportunities for members to have a say, few members possessed the knowledge or the political skills to move initiatives through the union’s democratic apparatus. In recent years the union came to favor enlarging local unions through mergers, which inevitably reduced member representation structures. In a union-wide review in 1991 of the ‘viability of very small locals,’ the national union made clear that small local unions that were not financially viable would not be supported with national funds (SEIU, 1991–2). In a now-famous article, Stephen Lerner, widely recognized as a thought leader in the SEIU, made a case for structural consolidation among national unions, stating that ‘If unions are weak, there is no democracy at the workplace’ (Lerner, 2003)
HUCTW
The HUCTW can be characterized as a participatory democracy that embodies both an egalitarian community and a decentralized bureaucracy (Polletta, 2002; Zilber, 2002). As an organizational form, the HUCTW’s participatory democracy combined relational organizing, a problem-solving method of bargaining with employers, a focus on internally developing staff, and an informal form of governance.
Organizing
Faced with a setback in unionization efforts, organizers developed the insight that administrative assistants who had limited contact with other workers needed self-empowerment (Eaton, 1996; Savage, 2007). Thus, building a community became the cornerstone of organizing. As one senior organizer said in an interview with Lisa Oppenheim, ‘In our model, the goal is to build a community of workers, not merely win an election. We seek to build a strong, committed, organization where everybody counts; where people create a vision of the kind of union they want’ (Oppenheim, 1991: 48). Overcoming social alienation in the workplace and finding one’s voice were challenges that organizers, who came from the ranks of the clerical workers, also had to face. Existing unions seeking affiliation criticized early organizers for the slow pace of their approach, but organizers believed it to be the only sustainable way. Kris Rondeau stated it succinctly: ‘If we try to move people too quickly, even if we succeed in getting them, we won’t be able to keep them. We begin the basic work of transforming the culture of the workplace, building the community in order to make the union able to survive through thick and thin’ (Oppenheim, 1991: 49). The HUCTW approach to organizing has widely been referred to by observers as a ‘relational model of organizing’, which Susan Eaton, who researched the union in the late 1980s, characterized as ‘an approach to organizing and bargaining embedded in personal relationships and interactions that are mutually empowering, both within and across union-management lines’ (Eaton, 1996: 291).
Bargaining
HUCTW’s approach to bargaining was similarly rooted in gendered experiences at the workplace. To engage female clerical and technical workers, the union eschewed a confrontational approach for one that built relationships with management and enabled the two parties to find joint solutions. As one senior organizer explained in an interview with Lydia Savage (2007: 127), ‘Problem-solving takes the adversarial nature out of the process and emphasizes union and management working together to solve the problem.’ Organizers learned that female clerical workers valued recognition and self-representation over distributive issues. Rondeau explained: ‘you really can’t sustain a negative campaign for very long before people get tired and bored and disgusted. You have to have other reasons for existing [. . .] The most important thing for us to know is that this predominantly female workforce deserves recognition’ (Oppenheim, 1991: 54). Relational bargaining was institutionalized into a system of ‘joint councils’ at all levels of the university, where labor and management representatives deliberated on key issues (Eaton, 1996). In the author’s observations of joint council meetings during the 2006 round of contract re-negotiations, statements made by representatives in meetings rarely were partisan. Instead, members suggested ideas based on what was desirable and practicable.
Recruitment
The HUCTW engaged in very little recruitment in the conventional sense. Most of the union’s work was done by members as part of their job with Harvard. This approach meant that full time staff needed to be deeply knowledgeable of, and able to identify with, members’ needs, and to have the ability to translate these needs for existing unions with which the HUCTW had affiliated over the years (first United Automobile, Aerospace and Agricultural Implement Workers of America (UAW), then AFSCME). It also accounted for why the few full-time staff had all started out as members and, in addition, had worked with the union since the early years. Thus, staff in the HUCTW did not only share the identity of members, as did the majority of staff in the SEIU, but also partook in the same social and economic conditions as members.
Internal development
Development of members in the HUCTW was rooted in the belief that members were jeopardized not by objective structures of oppression but by socially constructed beliefs of themselves that inhibited them from realizing their full potential. Member development thus permeated all domains of its practice in contrast to the SEIU, where member development was treated as a separate domain of work. Organizers explained the philosophy of HUCTW’s people development as ‘constant adult learning’ (Oppenheim, 1991: 48). Rondeau described it as supporting stages in a worker’s development, from finding one’s own voice to leading the organization: ‘The primary component of this model is that each person in the workplace has to go through a learning process . . . First they become pro-union; then they develop a sense of community, they know where to go if they have a problem; then they become active; they’re able to organize co-workers, make strategic decisions, and help create strong organizations’ (Oppenheim, 1991: 48). A union member on the receiving end of the union’s efforts to engage and develop members stated in an interview with Susan Eaton (1996: 319) that, ‘The union never leaves you stranded. It gives us guidance. It trains you [in] how to run a meeting, what to do when things are going wrong, how to persist if it’s difficult, how to work as a group. Every meeting has a preparation meeting.’
Governance
Participation was thought to be central to the transformation that the union sought to bring about in each worker, and it was treated as an end in itself as well as a means to building a stronger organization (Savage, 1996). Bill Jaeger, a full-time organizer with the union since the early 1980s, explained this to Susan Eaton (1996: 319):
Employee involvement is the defining aspiration of our union… Relieving the economic burden, while it is very important, doesn’t ever give a member of the union the feeling of progress in her soul and real improvement in her life, of having really gained something, in the same way that a rich, interesting successful participation in some sort of process does where power is shared in a new way.
Governance in the HUCTW was notably informal. In an interview with the author, Rondeau explained that in the HUCTW, the traditional role of the ‘shop steward,’ who represents the union at the workplace and oversees the grievance process, does not exist. Instead, the HUCTW has union ‘representatives’ in each department that Rondeau referred to as ‘more of a friend that you’d want to go with you when discussing your case to the supervisor.’
Integrating identity group members
The adoption of a social movement-based model into the SEIU had the effect of elevating the position and status of national and professional staff in a union that hitherto had been noted for the autonomy of its local unions (Piore, 1994). At the helm of strategy, professional staff prioritized expanding the union’s power over the industry to protect wages and benefits. Although they believed union democracy was important, professional staff tended to think integration of workers into the democratic process would ensue naturally from lifting them out of poverty. A common sentiment was expressed by one professional staff reacting emotionally to outside criticism of union democracy (or the lack thereof) in the SEIU: ‘Look, our industry is confronted with a vicious predatory industry such as janitorial work. Men and women can be thrown out of their jobs . . . It’s like, what’s democracy to someone who’s gone from 10 dollars an hour to 4.75 at that time. And we had the nonsense of that debate [about union democracy]’ (Personal interview, former national organizer, 2006)
The problem of internal democracy has been exacerbated since creating ‘mega locals’ out of mergers had the effect of reducing the number of executive board positions available to members. Members complained regularly about the lack of day-to-day representation in the workplace. As one member told me, ‘The union never has a plan for after the contract is gotten . . . They don’t have a plan about how to provide stability for the worker in the job. They just leave the workers there. So then the company backlashes’ (Personal interview, former executive board member in Los Angeles, 2006)
While both the SEIU and the HUCTW have transformed traditional unionism, the resulting organizational form in the HUCTW has gone further in incorporating the needs and reflecting the values of the identity group whose causes it adopted. Organizational goals were defined in terms of women’s needs rather than inserting the latter into existing practices in conventional unionism. Organizers focused on building a new organization based on members’ identities rather than adopting rules and practices from outside unions. In an interview with Oppenheim, Gladys McKensie, an organizer, eschewed the notion of the union as an objective structure that stood apart from members’ wishes:
People would come to us all the time and say ‘What is AFSCEM, what are you about?’ as if we were some ‘planet AFSCME,’ outside of their experience. We wanted to challenge that notion of the union as third party . . . We were building something. We could tell people why we chose to work with AFSCME, but a lot of that had to do with the fact that they would make room for us to do something special for ourselves here that was based on who we were and not who anybody else was. (Oppenheim, 1991: 51, emphasis in original).
During the author’s visit to the HUCTW offices in 2006, there was a constant flow of children on the premises because the union doubled as a child-care center for members. The union’s work was geared toward securing the kind of flexibility its members needed, for example, by bargaining for individualized work times to suit members’ care responsibilities instead of accepting ready-made alternatives. That is, the organization sought to amend bureaucratic rules to incorporate women’s values of fairness. One senior organizer elaborated this point in an interview with Lydia Savage (2007: 131):
Developing flexibility for people’s lives means changing the idea of what fairness is because fairness is often defined as doing the same thing for everybody. We often hear a manager say something like, ‘I can’t do it for you because then someone else will want it, too.’ As a union, we don’t buy that definition of fairness. In our view, fairness is trying to do the most important things for each person.
In a study of the HUCTW, Kurtz (2002) pointed out that when asked, union members did not explain what the union did in terms of feminism or gendered practice. This implies that women’s ways were mainstreamed in the union to the point that they were no longer ‘specialized’ demands; instead, they were the union. One of the HUCTW’s earlier organizers explained to Kurtz (2002: 162): ‘We didn’t want it to be seen as a women’s group, we wanted men to feel welcome . . . But it was never not seen as a women’s thing.’
Discussion
This article has examined the conditions under which the revival of formal organizations through the adoption of identities motivated by social movements can also result in positive outcomes for the identity group members. A process-based model contrasting the divergent modes of integrating identity groups into trade unions is developed as Figure 2. Not all formal organizations adopt social identity as their central tenet, and the two cases presented here suggest that antecedents for adoption include an organizational crisis, the resolution of which results in the emergence of actors that propel ideas for revival and the development of a consensus for change among those in power. These antecedents were common to both unions studied here, but the cases diverge in the nature of the ensuing organizational change, the logic of action that emerged based on collective experiences during the change process, and the resulting organizational form. The contrasting models of intra-organizational processes for adopting and institutionalizing social identity are explained below.
Two contrasting models for adopting social identity into formal social movement organizations.
In the first model, trade union efforts to incorporate the goals of the identity group are geared toward building power from the union’s market position. To counter market forces the union undergoes accelerated change with help from outside experts that proves highly contentious and necessitates centralized command and control. Rules are devised to ensure compliance with new standards. In the second model, the union orients itself toward building a community that will empower members of the identity group. Organizational change is slow and highly participatory, and relies on building strong ties. Rather than adopt formal rules, the union fosters members’ personal understandings about the necessity for change. In the first model, the logic of action developing vis-a-vis identity groups is an instrumental one which regards social identity as a source of union revival. The second model is based on a transformational logic of action that believes in the capacity of each person to grow with and build the union. The resulting organizational forms are characterized as radical bureaucracy and participatory democracy, respectively.
The radical bureaucracy exhibits a militant view of social justice whereby power is conceived as a zero-sum game. This type of union seeks to realize the goals of the identity-based social movement through efficient, standardized practices. It relies on expert knowledge to strategize victories, sourcing this knowledge from outside for expedited outcomes. Despite the trappings of a formal democracy, actual decision-making in this type of organization is likely to be exclusive, relying on professionals possessing expert knowledge. The participatory democracy holds a relational view of social justice. Unions of this type tend to view the problem as lack of self-empowerment among identity group members and as a product of structural factors rather than of direct injustice. Power is conceived as a positive-sum commodity that opposite groups can share. It seeks to realize the goals of the identity-based movement through emergent, localized standards and decentralized practices. The work of the union is done by people who identify deeply with the identity-based movement and who themselves undergo personal transformation. Understanding occurs through internal development, and not through recruitment of external talent. Decision-making in this type of organization is highly inclusive and participatory.
The present study has implications for understanding the broader relationship between membership-based formal organizations and identity groups, in the context of social movements. Integration is made more difficult in the radical bureaucracy that operates in social movements and reaches out to identity groups, arguably because this type of organization operates in expansive markets and therefore favors standardized strategies over building intimate communities. But it is also made more difficult because organizational structures emerge as a response to the nature of change. Where organizational change has been met with opposition and thus has been contentious, actors are likely to prioritize organizational survival and stability over encouraging open participation and self-expression of members. The contrasting models suggest that organizational theory’s key assumption – that the adoption of identity claims by formal organizations brings about integration – must be conditioned by an understanding of organizational processes that govern the collaboration between identity groups and formal organizations. Formal organizations produce integrated action through politics and coordination among functionally differentiated units (Lawrence and Lorsch, 1967), and identity group members must be able to influence these processes in order to achieve equal standing. The findings suggest that organizational actors can exercise discretion in prioritizing goals, determining the speed of change, and developing specific practices. A limitation of this study is that the comparison of the two unions was focused mainly on organizational processes and thus it is unable to make definitive claims on factors that cause an organization to develop one model over the other.
Although both of these models have been celebrated for their respective successes – the SEIU for its strategic victories and the HUCTW for its ability to transform existing institutions – scholarship has generally favored the HUCTW model for its democratic values. However, it would be hasty to pronounce one model ‘better’ than the other. Collaborations between formal organizations and identity groups in social movements can take many different forms in addition to the templates presented here, and the outcomes of collaboration should be studied for their equalizing or differentiating impact. Whether the hybrid model of participatory democracy studied here is likely to yield more positive outcomes than a boundary organization model (Guston, 2001; O’Mahony and Bechky, 2008), or a third-party’s brokering between movements and formal organizations (Rucht, 2004), for example, remains to be explored. A more comprehensive theorizing of organizational forms that render the best outcomes for identity-based social movements must take into account the goals of the social movement (self-expression or policy impact?) as well as the extent to which interests of the social movement and the formal organization that seeks to represent it are converging or diverging (O’Mahony and Bechky, 2008).
Previous scholarship on whether identities are deployed strategically or expressively has mostly examined factors that affect identity deployment at the level of a movement (Bernstein, 1997; Williams, 2004). By examining the processes by which formal organizations adopt identity repertoires, this article has contributed to shifting attention from the movement level to the organizational level of analysis in our conceptualization of identity deployment. Although the focus is on intra-organizational processes, the findings of this article have broader implications for field-wide dissemination processes. Scholars of dissemination have pointed out that carrier organizations modify the templates that they disseminate, but few have looked at intra-organizational processes that affect which organizational forms are ultimately disseminated. The present findings suggest that studying processes in organizations that lead the adoption of new templates in their respective field can be important for understanding the evolution and institutionalization of new organizational forms (Child and McGrath, 2001; Romanelli, 1991).
Footnotes
Acknowledgements
The author would like to thank Editor-in-Chief Paul Edwards and three anonymous reviewers for excellent feedback and guidance throughout the review process. Any remaining errors remain the author’s own.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.