Who Is This “We” You Speak of? Grounding Activist Identity in Social Psychology

Abstract

What is an activist identity? Prior answers have focused almost exclusively on collective identity, without (a) considering the possibility of role-based identities or (b) grounding collective identities in broader social-psychological theories. The present study investigates activist identity through the lens of role-based and category-based identities and reports two major findings. First, there is a distinct role-based activist identity, one that involves internalizing role responsibilities and the expectations of friends and family. Second, collective identity represents a relationship between a social identity and an injustice frame; it either involves incorporating an injustice frame into a preexisting social identity or using the injustice frame to create a new in-group. The present findings help to illuminate the processes underlying collective identity, indicate that a great deal of role-based activist identity is mistaken for collective identity, and suggest new directions for the study of micro-mobilization and organizational forms and tactics in social movements.

What is an activist identity? The answer is important to several areas of social movements research as activist identity has a strong influence on micro-mobilization (Kelly and Breinlinger 1995) as well as organizational forms and tactical choices (Jasper 1997). But most social movements scholarship reduces activist identity to collective identity—broadly defined as the sense of “we-ness” that binds activists together in a social movement (Levitsky 2007; Owens, Robinson, and Smith-Lovin 2010; Valocchi 2009)—to the exclusion of other useful theoretical constructs (Stryker 2000; Viterna 2013). The present study demonstrates that activist identity is considerably more complex than standard treatments of collective identity by showing (a) how collective identity is the intersection of a social identity and injustice frame and (b) how activist identity is sometimes based upon a social role (e.g., “activist”) rather than a social category (e.g., female).

This study’s first contribution is providing evidence that collective identity actually consists of two separate components. First, collective identity demarcates who belongs to “us” and “them” (Taylor and Whittier 1992); in other words, it includes a social identity where a person recognizes they belong to a larger category of individuals (Ashforth and Mael 1989; Tajfel and Turner 1979). Second, the social identity is politicized so that belonging to the group entails adopting grievances and working for change (Simon and Klandermans 2001; Taylor and Whittier 1992). As a result, collective identity involves a relationship between a social identity and an injustice frame (Gamson, Fireman, and Rytina 1982), and the relationships between the social and political components vary. Activists can build on preexisting social identities or create new ones, highlight injustice done to themselves or to others, and emphasize either a shared history of resistance or the willingness to serve in a new conflict. Because the individual’s connection to grievances and action involves belonging to a category, collective identity is a type of category-based activist identity.

This study’s second contribution is distinguishing the category-based identity from an alternative role-based activist identity. As Stryker (2000:29) notes, people “do not live in [social] categories”—an individual may recognize they are female or an environmentalist, but the individual’s day-to-day experience takes place in a set of social relationships that reinforce certain types of behavior. As a person interacts with others, they take on certain social roles—such as a parent or an activist. Actions in those roles elicit messages from others about the appropriateness of performing the social roles. For example, one parent may encourage a spouse to spend more time with their children while one activist may encourage another to attend a protest. Over time, a person internalizes the expectations of others so that valued social roles become identities (McCall and Simmons 1978; Stryker 2008). In this case, the individual’s connection to movement behavior is due to a social role rather than a category and is therefore a role-based activist identity.

Unfortunately, scholarship on activist identities overwhelmingly focuses on collective identities, largely ignoring role-based explanations in favor of categorical approaches (for exceptions, see Viterna 2013; White 2001, 2010; White and Fraser 2000). Empirically, it is easy to mistake role-based activist identity for collective identity because relational support for role-based activism often takes place in a group setting (e.g., Nepstad 2004). However, an activist’s interactions in a group setting are not equivalent to identification with category-based membership. An environmental activist could be motivated by the expectations of friends, affinity with a broad class of “environmentalists” that he or she has never met, or both.

Furthermore, the scholarship on collective identities has largely developed in isolation to other category-based approaches to identity (Owens et al. 2010). Collective identity involves both a social and political component, and this intersection has been largely neglected in scholarship on social movements; our understanding of the social component can be improved by relating insights from social identity theory. Furthermore, understanding the relationship between the social and political components can help us understand why collective identity appears to affect so many social movements phenomena (Polletta and Jasper 2001).

The present study shows how insights from social psychological research can improve our understanding of activist identity. There are distinct role-based identities that can reinforce movement behavior; they are not necessarily embedded within a social category and are forged by social support for beliefs and task responsibilities as well as a sense of obligation to others. Additionally, collective identities are more complex than previously portrayed, with the relationship between social identity and injustice frame defining many different variations in collective identity. Ultimately, each type of activist identity is the result of a different process, involves different types of subjective experiences, and could potentially lead to different outcomes in micro-mobilization as well as organizational forms and tactics.

Category-Based and Role-Based Activist Identity

All research on identity investigates how people answer the question “Who am I?” (e.g., Ashforth and Mael 1989; Melucci 1996; Stryker 2008). However, while collective identity theorists draw from a European tradition where people derive their identities from their social category membership (Polletta and Jasper 2001; Simon and Klandermans 2001; Taylor and Whittier 1992), role-based identity scholars draw from an American tradition in which people derive their identities from internalized social expectations (Owens et al. 2010; Stryker 2008). Although there is disagreement over whether integrating the two approaches is desirable or possible (Hogg, Terry, and White 1995; Stets and Burke 2000), there is broad scholarly consensus that each theoretical tradition captures different empirical phenomena and constructs.

Category-based Activist Identity

Collective identity theories help to explain the pervasive “we-ness” that binds activists together in a social movement (Polletta and Jasper 2001; Taylor and Whittier 1992; Valocchi 2009). Different scholarly conceptions of collective identity exist; it is defined as understanding the difference between “us” and “them” (Gamson 1991; Guenther, Mulligan, and Cameron 2013; Taylor and Whittier 1992) and an “individual’s cognitive, moral, and emotional connection with a broader community, category, practice, or institution” (Polletta and Jasper 2001:285). However, all definitions of collective identity combine acknowledgement of belonging to a social category as well as a political interpretation of what it means to be a group member (e.g., Simon and Klandermans 2001; Taylor and Whittier 1992).

The first component of collective identity is membership in a social category (Polletta and Jasper 2001; Valocchi 2009); in other words, collective identities include a social identity. Social identities involve a person recognizing that they are in a group with other similar people, even if he or she has not met them previously, and that all members of the group have a shared fate (Ashforth and Mael 1989; Tajfel and Turner 1979; see also Polletta and Jasper 2001; Simon and Klandermans 2001; Valocchi 2009). People who share the social identity are said to be part of the “in-group,” while those who do not share the social identity are part of an “out-group.” For example, a person who identifies as an African American can quickly recognize other African Americans as part of his or her in-group and that there are many social and institutional arrangements that affect every group member (e.g., affirmative action). ¹

Social identity motivates people to take action to protect their self-esteem (Hogg et al. 1995), and the desire for positive in-group identities is a recurring goal in many social movements (see identity “negotiation” in Taylor and Whittier 1992). Those who hold a positive image of their social identity tend to align their behavior with group standards, while those who hold a negative view of their social identity often distance themselves from the in-group (Ashforth and Mael 1989; Blanz et al. 1998; Hogg 1995). For example, gays and lesbians who hold negative in-group valuations may attempt to “pass” as heterosexual. In some cases, a negative social identity leads to a “splitting” of the in-group, with one subset claiming that they are part of an elite or select group (Blanz et al. 1998; see also Taylor and Whittier 1992). For example, men may distance themselves from violent masculinity by proclaiming themselves “pro-feminist men.”

The second component of a collective identity is a political interpretation of the social identity, sometimes forged during intergroup conflict (Gamson 1992; Guenther et al. 2013; Simon and Klandermans 2001; Taylor and Whittier 1992). Simon and Klandermans (2001:324) suggest this political interpretation involves “shared grievances” and the designation of a common enemy. For example, labor unions may work to fuse grievances about working conditions and pay onto a preexisting working-class social identity; Kelly and Breinlinger (1996) also report that activists in women’s groups feel that they have been treated unfairly as women. This political interpretation may be what separates the individuals who hold category-based identification (e.g., gay/lesbian) from those who are active in social movements.

However, Simon and Klandermans’s (2001) conceptualization of shared grievances assumes that the in-group has suffered injustice; sometimes, the injustice is against an out-group, which offends the conscience of the in-group (White 1992). For example, the presence of white protestors in the civil rights movement was not because white individuals suffered directly from discriminatory laws but rather because they were upset at the treatment of others (Von Eschen, Kirk, and Pinard 1971). To account for the collective identity of “conscience constituents” (McCarthy and Zald 1977:1222), it is easier to conceptualize shared grievances as a shared “injustice frame” (Gamson et al. 1982:14), which is the sense that current social and institutional arrangements produce injustice.

Role-based Activist Identity

While collective identity defines the individual as a member of a social category, many modern symbolic interactionists conceptualize identity as the internalization of social roles. Stryker (2000, 2008) and McCall and Simmons (1978) emphasize that individuals hold semipermanent social roles (e.g., parent, activist), which are either encouraged or discouraged through interactions with friends and family. When a role is encouraged, the individual internalizes the expectations to perform it as an identity, and they are more likely to perform associated behaviors in the future.

The first proposition of role-based identity approaches is that there are relatively permanent social roles in a society (McCall and Simmons 1978; Stryker 2008). A person may hold a role as a professional, parent, and/or activist; all of these roles are relatively permanent parts of the American social structure, even as the definition of each may change over time. In the context of social movements, an individual may take the role of an activist (Corrigall-Brown 2012), which involves working for some kind of social or institutional change.

The second proposition of role-based identity approaches is that friends and family may encourage or discourage participation in a social role (McCall and Simmons 1978; Stets and Burke 2000; Stryker 2000, 2008). Friends and family who share the activist’s political beliefs may provide psychological support to activists (Nepstad 2004), while friends and family who share childrearing responsibilities may reinforce parental roles instead (e.g., Klatch 1999; White 2010). Friends and family sometimes share the activist role and reinforce the task behaviors through shared commitment (White and Fraser 2000); in other cases, friends and family are the targets of activism and reinforce behavior through acceptance of the activists’ political beliefs (Burke and Stets 2009). Over time, individuals internalize the expectations of others; these identities are relatively strong, while identities with less internalized social support are comparatively weak. This process is similar to how scholars envision interaction within social networks in movement contexts, with community members providing support for the role performance of activist behaviors (Friedman and McAdam 1992; White 2010). ²

Finally, the strength of an identity—sometimes characterized as prominence or centrality—guides individual choice to either persist or disengage in a given activity (McCall and Simmons 1978; Stryker 2008; Stryker and Serpe 1994). Individuals with a strong role-based activist identity are likely to persist in movement-related activities, even at great personal cost to themselves. Thus, Nepstad’s (2004) peace activists are willing to engage in high-risk activism despite substantial obstacles, in part because of the social support they receive. However, leaving a movement group may undermine this role-based identity; even though a person may still sympathize with the goals of the movement, the expectations of participation are more distant and abstract without active recruitment (Corrigall-Brown 2012). The relationship between roles is also frequently explored in role-based identity theories; these theoretical propositions are discussed in Appendix B.

Activist Identities: A Summary and Implications

There are two distinct social psychological traditions, one that emphasizes belonging to a social category and another that emphasizes how friends and families encourage the internalization of social roles (Hogg et al. 1995; Stets and Burke 2000). When people define us or we in political terms, they are expressing solidarity with people who they may not have met. However, when a person says they will participate because they do not wish to let their friends down, they are expressing solidarity with people with whom they have interacted. In other words, category-based identity (including collective identity) is a generalization where a person becomes one with the group, while role-based identity is rooted in ongoing relationships that define a set of tasks.

Stets and Burke (2009) argue that the distinction between role- and category-based identity parallels the difference between organic and mechanical solidarity (Durkheim 2014). Mechanical solidarity is similar to category-based identity, where individuals feel a strong emotional connection to all others in a nation or tribe. Meanwhile, organic solidarity is defined by the web of tasks associated with a role—and that others rely on you to perform. ³ And while most researchers argue that all individuals simultaneously hold both role- and category-identities, there may be cases where only one identity is the primary driver of activist behavior (Burke and Stets 2009; White 2001).

How would we distinguish role-based activist identities from collective identities? Role-based activist identities are difficult to spot because they are often fostered in a group context. The clearest examples of role-based activist identity occur when others praise the activist’s movement-related task behaviors or express emotional support for a shared ideology. For example, Nepstad’s (2004) activists support each other emotionally and validate the decision to undertake high-risk activism. In addition, activists who feel that they are not living up to their obligations to their fellow activists have often internalized the expectations of others. Finally, we should expect little variation within role-based activist identity; Stryker (2000, 2008) and McCall and Simmons (1978) suggest the internalization of social role expectations is a universal process.

While role-based activist identity is the result of a similar process across different groups of people, collective identities should vary depending on the relationship between the social identity (which itself consists of a clearly defined in-group/out-group) and the injustice frame. In some cases, activists will politicize a previously existing identity by linking it to an injustice frame and political conflict. For example, African American political leaders often invoke the civil rights movement to argue for social change and have a tendency to situate their efforts within the context of hard-fought historical gains (Polletta 1998). These accounts play a vital role in establishing a collective identity even though the actual events may be disputed by historians and social scientists. By recounting the oppression of the past and the continued necessity to fight, activists can politicize a preexisting social identity. At the same time, the activists are able to cast the in-group as a team that is standing on the side of justice and equality, which should enhance self-esteem.

On the other hand, some social identities do not exist prior to social movements. Some movements start with an injustice frame and then either construct new social identities or adjust the boundaries of old ones. For example, Taylor and Whittier’s (1992) radical lesbian feminists actively created a new social identity to deal with anti-lesbian and anti-woman discrimination. While rooted in previous social identities (lesbian, woman), the radical lesbian feminist identity is defined in opposition to a wide variety of out-groups, including men, heterosexuals, and liberal feminists. Rather than accept a negative in-group valuation of them as women, lesbians, and feminists, radical lesbian feminists created new boundaries that separated “us” from “them.” This is an example of how grievances can lead individuals to split their in-group into several smaller in-groups and out-groups (Blanz et al. 1998). Rather than importing grievances into a preexisting social identity, they created a new social identity to match their grievances.

Methods

To investigate role-based and category-based identities, the present study involves semi-structured interviews with 27 activists. The interviews produced “narrative” accounts of participation in activism and the relationship between their activist role and other parts of their lives. I analyzed the data through open coding, merging similar codes, and splitting codes that describe different identities (Corbin and Strauss 2008).

Sampling

This study investigates activist identities and therefore requires a substantial number of activists, defined here as people who are involved with creating some kind of social, structural, or institutional change. In other words, an activist is not necessarily a person who provides “services” or “helps” the needy but rather a person who attempts to change the underlying conditions that produce injustice or inequality. For the purposes of this study, I first sampled by finding a population of campus-based organizations that attempt to change social, structural, or institutional conditions that produce injustice or inequality. Then, to ensure that I was obtaining activists—rather than individuals who only sporadically attended meetings and were not involved in actual change efforts—I obtained member lists of every graduating senior who had either run a meeting or planned an event as a member of the group.

By defining activists behaviorally rather than by “membership,” the present sample excludes individuals who have “paper memberships” or have signed up to be on listservs and includes individuals who have actually worked toward social change. Additionally, this sample involves many activists who are excluded from social movement scholarship because they do not hold formal roles in an organization; many groups eschew formal leadership roles, and defining membership behaviorally helped to identify activists who would otherwise be excluded because of their ideological stance on formal leadership. Furthermore, by sampling individuals across organizations, I can also make comparisons across different movement groups and diverse personal backgrounds, which should permit comparisons between different types of identity (for similar approaches, see Valocchi 2009; Viterna 2013).

First, I defined a population of organizations by identifying all “social justice organizations” at 15 colleges and universities across diverse regions and institutional types. For the purposes of this study, I define a social justice organization (SJO) as an undergraduate student group that seeks to change the social, structural, and/or institutional relationships that perpetuate inequality and injustice. As a result, the organizations are fairly heterogeneous. Many of the groups focus on traditional left-wing issues—including but not limited to feminist, environmental justice, and labor rights groups. However, many traditional left-wing movement organizations, such as those advocating for more stringent gun laws, do not qualify under this definition. Other SJOs do attempt change in social or institutional relationships but insist that they are apolitical to attract support across partisan lines—groups that campaign to end sexual assault often fit this definition. This sample also includes Evangelical Christian groups that combat human trafficking, and there is no reason to assume that the participants in these groups hold stereotypically left-wing beliefs on other issues (e.g., on feminism, labor rights).

I then requested that the SJOs at each school nominate all individuals who have organized meetings or events for the SJO in the past and who were graduating in spring of 2012. A total of 161 out of 222 student groups responded with a full list of activists. ⁴ I then contacted all of these “core activists” directly and informed them about the study and the email-based questionnaire. A total of 192 of 341 (56.3 percent) of the sample completed the online questionnaire. The only compensation provided was to receive ongoing information about the project; currently, 88 participants have elected to receive periodic updates and results.

I then selected a subset of 31 participants to engage in a face-to-face, semi-structured interview. The full sample of participants spanned multiple geographic regions, and to perform all of the interviews face-to-face, ⁵ I selected the 31 students who attended a large public university in the southeastern United States. Overall, 27 of the 31 participants agreed to the interview. However, this sample is extremely heterogeneous with a mixture of individuals from different movement groups (e.g., anti–human trafficking, feminist, civil rights) and who differ on a number of ascribed characteristics (e.g., sex, race).

This sampling strategy has several advantages over alternative approaches. Probability sampling of activists is not viable because the percentage of activists in the U.S. general population is relatively low. A more common approach is to sample within organizations, but probability-based approaches are still inappropriate because it is unclear whether most of the people on membership or email lists are actually involved in social change efforts. Additionally, in loosely organized groups, many of the activists who are heavily involved with the organization may not actually be listed as members.

A third alternative approach utilizes single-organization case studies. A participant observer in a social movement organization can often identify who is an activist and who is not (e.g., Nepstad 2004), but this makes it difficult to study variations in activist identity across personal backgrounds. Furthermore, since many members of the group have similar grievances, it may not capture variations in the relationship between social identity and injustice frames. To make cross-movement or cross-organization comparisons, many social movements researchers limit their study to activist leaders (e.g., Levitsky 2007; Valocchi 2009). However, formal leadership positions are also formal roles, which means that role-based activist identity may not be comparable between formal leaders and other activists.

This study resolves these issues with a heterogeneous sample of activists defined behaviorally rather than by formal leadership role. However, this study’s approach to sampling has a few potential pitfalls. First, this sample includes student activists only, and it is possible that activist identity develops differently on college campuses than with older participants who have jobs with more decision-making authority or spouses and children and who have lived in a community for decades as opposed to four years. Second, because of the need to perform face-to-face interviews, all of the participants attended a single school. There are a few differences in the interview subsample and the larger study; more information on the differences between the overall sample and the subsample are discussed in Appendix A.

Finally, while this sampling approach captures the majority of “core activists” in an organization, there are likely some individuals who contributed to events without necessarily planning them or were attendees at protests but who were uninvolved with the organization’s planning efforts. That said, I had the opportunity to discuss the criteria for inclusion with many of the SJO spokespersons over email, and many of them indicated that this definition included all individuals who were consistent group members. Furthermore, this sample included both individuals who held formal leadership roles and those who did not; the results in this study do not differ between those holding formal positions and those who do not. Therefore, while the conclusions in this study do not necessarily extend to protest attendees who are uninvolved with organizations, the findings can be extended to activists in ongoing relationships with organizations.

Data

I conducted all of the interviews, which generally lasted between 60 and 90 minutes. I had no prior relationship with any of the participants in the study, and I offered to conduct the interview at any quiet location where the participant could speak freely without friends or family overhearing. ⁶

I began each interview with a set of starting questions for each participant, which I expected would help illuminate activist identity. These questions included (1) how the participant became involved with their group, (2) their present involvement with the group, (3) information about relationships with friends and family, and (4) how they saw their lives changing over the next few years. I used several open-ended prompts to encourage the participants to speak more about these issues, including (1) “Can you tell me more about . . . ”; (2) “Can you give me an example of what you mean?”; (3) “What do you mean when you say____?”; and (4) How did that make you feel?”

As a result, the interview process produced a narrative about participation in activist communities and how it relates to the rest of their lives. Narratives are accounts of events, arranged in a particular way to convey a message to an audience (Polletta et al. 2011). The narratives are similar to life history interviews and involve the subjective reconstruction of the past, which is particularly well suited to investigating identity (Andrews 1991). The narrative has characters that are either people (e.g., friends, family) or groups (e.g., ethnic groups); the narrative defines the relationship between the narrator and other characters in the story. This type of narrative has been used repeatedly in social movements research, although it may not be explicitly recognized as a narrative (e.g., Guenther et al. 2013; Levitsky 2007). Recognizing this sort of interview as narrative acknowledges that the interview data are not an objective retelling of how events actually happened but rather a subjective construction of the relationship between the participant and other important people and groups (Polletta et al. 2011). And by defining the relationship between the narrator and other characters, the narrator helps to answer the question, “Who am I?” (Ashforth and Mael 1989; Melucci 1996; Stryker 2008).

Analysis

To analyze each narrative, I followed several analytic steps suggested by Corbin and Strauss (2008). After each interview, I wrote an analytic memo outlining my initial impressions of the interview. Then, I performed open coding. While there are many different ways to perform open coding, open coding generally consists of (1) finding “strips” of interview text that make up units of analysis in later stages of coding and (2) repeatedly asking open-ended questions that force close reading.

While Corbin and Strauss (2008) note that there are many different ways to identify relevant strips of text, I repeatedly used four strategies to identify important passages. First, Corbin and Strauss (2008) suggest looking at how people discuss time since people often organize their stories to reveal theoretically important information around the passage of time. Second, I paid close attention to any part of the interview that could answer the question “Who am I?” to see how participants were representing themselves. Third, I looked at how the participant arranged the characters in their stories and how they used those characters to make a point. Finally, I looked for moments where people discussed reasons for their continued involvement in activism. I analyzed the data by asking whether the participant was developing distinctions between constructs, a relationship of some kind, or otherwise expressed how they interpreted the world.

After open coding, I performed merging and splitting of the codes. When two codes represented a larger, superordinate phenomenon, I merged similar codes into common themes. However, when a single code represented distinct phenomena, I split the theme into two separate codes. After several iterations of merging and splitting, I performed member checks by sending the initial findings to all participants who had elected to receive publications. Participants did not object to any of the core findings in the paper, although their responses helped clarify some of the analysis that they thought did not accurately capture their experiences.

Findings

Through analysis of the interviews, I uncovered four different ways that participants represent their role-based and category-based activist identities. Role-based and category-based identities are widespread in the narratives, and most participants express both types of identities in the interview. However, while the role-based activist identities are remarkably consistent across participants, category-based identities depend on how the participants construct the relationships between their social identities, inequality, and the injustice frame. I only found limited evidence for relationships between different roles represented either in a role hierarchy or as role conflict; I discuss this nonfinding in Appendix B.

The role-based identities are expressed through the theme community, in which friends help participants internalize the expectations of the activist role. First, friends provide emotional support for shared beliefs and also encourage the participant to take action on behalf of those beliefs. Additionally, friendships with other activists create a sense of obligation—that others depend on the participant fulfilling the activist role. The task-based role responsibilities take center stage during these interactions, with a strong relationship between role performance and the support of others. The participant embeds the activist role within the context of relationships, and interpersonal interaction communicates the expectation that the participant will perform certain activist tasks as well as the best ways of fulfilling the role responsibility.

However, there are three different and distinct themes that merge a social identity with an injustice frame to create collective identity. The first theme (legacies) begins by explaining how the participant’s social identity in-group (e.g., African American) has suffered immensely. However, this in-group has fought and alleviated some of this oppression and must continue to fight on behalf of themselves and others. The theme fuses a preexisting social identity with an injustice frame (Gamson et al. 1982) and therefore defines a collective identity.

The second theme (boundary adjustment) also begins by explaining how their social identity in-group (e.g., women) has suffered immensely. However, the participant does not perceive that their in-group has fought against oppression as a unified group and therefore holds a negative in-group valuation. The negative in-group valuation spurs the participant to redraw the boundaries of their social identity category so that the participant is now in a smaller in-group (e.g., feminists) that does provide resistance. While the legacies theme relates an injustice frame into a preexisting social identity in-group, the boundary adjustment theme uses the injustice frame to support redrawing the boundaries of the social identity in-group. By creating a new social identity to match the injustice frame, the storyteller generates collective identity.

The third theme (conscience constituents) differs from both legacies and boundary adjustment because it explains how an out-group has suffered immensely. The storyteller does not suffer from oppression but constructs an in-group of conscience constituents (McCarthy and Zald 1977), which contains people who fight oppression on behalf of a specific out-group. This is unlike most forms of collective identity, which draw a contrast between an in-group that suffers and an out-group that is responsible (Gamson 1991; Simon and Klandermans 2001). Instead, the conscience constituents theme uses an injustice frame to define “them” (an out-group) who suffer, which is contrasted with “us” (an in-group) who fight on their behalf.

Relationships and Roles in a Community

Participants expressed role-based activist identity when they spoke about community—the network of interactions they had that reinforced the legitimacy of their belief systems and activist role responsibilities. Interactions with friends in their movement groups provided emotional support for activist beliefs and behaviors as well as a sense of obligation to continue performing the activist role. For example, Tom—a white male activist in a multi-issue social justice group—notes that he draws a great deal of emotional strength from interacting with other members of his group:

It’s sort of a reinforcement that [activism is] okay, and that there are people who share those ideas . . . and [are] able to better organize and sort of make what can be more jumbled notions, or I guess, desires or pictures of justice that you might have into a reality . . . it can be, like make you feel a little less crazy or something. It can make you feel. . . . It makes you feel that there is support . . . being with people who are not deterred by [anything], or they won’t put their beliefs in a box and be deterred . . . [from] acting on them.

Tom’s experience in the group directly communicates to him that he should continue with his activist work; it legitimizes beliefs and behaviors related to the activist role. Internalizing the expectations of others creates an activist identity, and the strength of this identity is reinforced by the emotional bonds with those individuals (Stryker 2008). However, Tom’s emotional connections are not to a generalized in-group category, as in collective identity; it is with members of his group. While Tom doesn’t have a generalized emotional connection to a larger social category (e.g., sex, race), it is relatively easy to misclassify this type of identity as collective identity because the reinforcement takes place in a group-based setting; but interactions in a group setting are not the same as belonging to a category-based group.

However, reinforcement for activism does not always come in a group-based context. Cora—an African American activist in a civil rights activist group—provides an example of encouragement for social roles in the context of one-on-one interactions:

When I was a first year I met this guy who was a junior . . . he was definitely really good about like reaching back and saying, “Hey, come here, you should apply for this . . . ” or “You should do this . . . if you’re ever interested in these kinds of things, let me know.”

In this case, Cora’s activist identity is directly fostered by a longer tenured member who encouraged her to participate in certain activities and take on formal roles within the organization. By encouraging her to enact this social role, he directly communicated his expectations for her. But while the expectation may have extra legitimacy coming from a longer tenured member, expectations can be fostered by peers as well. Ellen—an African American woman involved with a different civil rights organization—noted that enthusiasm from peers can also help develop activist identity:

I actually went to that first meeting with a girl that I had met last year in class. And so, we went to that first meeting and we were really, really excited. And her being there with me, and me being there with her, kind of encouraged the both of us to dive in and really get involved. And we were able to maintain that interest and that fire throughout the rest of the school year and we are still a part of the organization 2 years later. Still tackling the same issues.

Some activists also interacted with older, paid staff members during activist work. While paid staff played only a supporting role in their organizations, they directly reinforced the role responsibilities by suggesting ways of performing the role and validated the role performance by helping execute the responsibilities. One participant in a gender-based violence prevention campaign noted that paid staff members assisted her with grant writing. Meanwhile, an activist in a feminist group discussed how her faculty advisor gave her tips on how to treat an outside speaker and to network with other groups on campus. Sometimes, professionals outside the organization assisted directly with role responsibilities; one activist spoke with her pastor about the difficulties she had with retreat planning and then discovered that the pastor had helped arrange for a retreat location for her activist group. Finally, some college activist groups had a strong relationship with national organizations, and activists involved with national groups sometimes mentored participants in the present study on organizational decisions and tactics.

All of these interactions between older paid professionals and the activists were focused on the actual role responsibilities that participants engaged in as leaders of their organizations. As participants reached into their social networks for help with activism, older professionals provided advice and resources and in doing so provided support for engaging in their role-based behaviors. Notably, the interactions did not involve aligning behavior with an in-group of people they had not met or a contrast with an out-group, as you would expect in collective identity. Participants were being coached on how to perform the activist role in relation to other activists, potential targets, and potential funders—not to act as a representative of a larger in-group (e.g., black, woman, or even as an activist).

Throughout the interviews, I asked participants to discuss what their friends and families thought of their participation in activism. According to role-based identity theories, people learn appropriate role performance when others validate or sanction behavior. Statements from friends and family that expressed pride such as “you’re [a] servant of the people” were relatively common in interviews; these declarative statements express approval of the activist role. However, participants also said that when their friends attended activist events they planned, it demonstrated the friends approved of their activist role performance. For example, when I asked why a trusted friend supported their activism, I received responses such as:

[She] would always come to any of our events. (Claire, human rights activist)

I always knew he supported me because he came to my events. (Lucy, human rights activist)

If he doesn’t want to go to something, he would tell me that he’s not going to go because he’s not interested in it . . . so when we comes out to something to me, that does signal that he thinks it’s something important and wants to know more about it . . . [and another friend], she’s taking time out of her schedule to go and hear about this, and we’re all very busy. (Mary, labor rights activist)

Furthermore, failing to fulfill the expectations that others have laid out for you can damage relationships within movement organizations. Many activists noted that they continued doing work with their activist groups because of a sense of obligation. Cindy, a white female involved with anti–human trafficking campaigns, said that her friends would have noticed and challenged her if she stopped performing activism with the group:

Second semester of junior year [I was] wavering on how much I was going to be involved in [the group]. And I think if it had been composed of people that I didn’t really know, there’s this element of when you’re in a big classroom you can skip more easily because there’s no personal accountability. But when it’s like a class of 15 people, you have a bit more incentive to go because [your] presence will be noticed. So I think it’s one of those things so not only was it a small enough group, it was people who I need and we had both been talking to each other about our excitement about this issue. So there would be this personal inconsistency if I like didn’t show up to meetings kind of thing. So that kept me, I think kept me motivated to go.

While Cindy could have stopped working on behalf of the group, she felt she would have faced consequences in her personal life as a result. However, the obligation in this text is not to the broader class of anti-–human trafficking activists but to a specific set of friends who encourage and nurture her activist role. Additionally, while she felt that they would challenge her, she does not ever say that anyone did challenge her when she stopped participating. Neither Cindy nor anyone else who was worried about the personal consequences of not participating actually reported that anyone challenged her to do more work on behalf of the group. Instead, she internalized the role-based expectations of others as an identity, which she maintains.

While role-based activist identities are common throughout the sample, participants also frequently merge a social identity with an injustice frame, constructing a collective identity. This occurs in three different ways: by fusing an injustice frame and legacy of resistance to a preexisting social identity, using an injustice frame as a reason to redraw the boundaries of an existing social identity, and using the injustice frame to designate an oppressed out-group, thus creating a social identity in-group of conscience constituents to support them.

A Legacy of Resistance for a Social Identity

The first theme representing collective identity is the legacies theme, which begins by relating how the participants’ in-group (a social identity such as being African American) has suffered intense oppression. The story continues by stating that the in-group has managed to have some success battling for justice and that continued resistance is necessary to ensure equal justice for all. By fusing a preexisting social identity with an injustice frame and legacy of resistance, the storyteller politicizes the social identity while simultaneously highlighting positive aspects of the in-group.

The most prominent examples of the legacies theme occur in the narratives of the African American participants, who claim to represent the legacy of the civil rights movement. They characterize themselves as the stewards of a decades-long tradition of protest and change and proudly recount the work their organization performed in the past while deeming themselves ready to confront oppression in the future. For example, Maya casts herself as a “representative of 103 years of awesome activism” and further states: “[O]ur state president, has a metaphor that he uses that it’s not Jim Crow anymore, it’s James Crow, Esquire. . . . Jim Crow has changed and morphed itself into all these different things to meet different barriers to freedom that we’re seeing now, and different categories of oppression.”

Maya’s story first notes that while African Americans have managed to defeat Jim Crow, oppression still continues in different guises. According to Maya, Jim Crow is only one small part of a larger campaign against injustice and puts her own work within the legacy of the civil rights movement. Angela, who is involved with labor and feminist organizations, echoes many of Maya’s themes except she also casts her own family as actors in the story. She argues that she represents her family and others in the black community by attending a protest when she says:

[M]y dad, he went to segregated schools and things. So he grew up during the Civil Rights era. He and his family protested a lot. . . . So the stories I would hear of them growing up was like well, that much hasn’t changed so I still need to be speaking up for what I feel is wrong. My mom, she grew up after him but she talked about remembering the protests . . . I just grew up in a family that always talked about how you have to be an activist . . . I just think back to, I always go back to the civil rights movement. If they weren’t driven every day to go out there and march and protest, or to fight for equal rights then they wouldn’t be—or we wouldn’t be where we are now, where we wouldn’t still be achieving great milestones.

Angela starts by talking about the past but ends by discussing “milestones” that impact the future, thus emphasizing the continuity between past and current injustices. But Angela also places people that she cares about (her parents) as central actors in the story, which provides an example for how she should behave as an African American. The injustices that her parents’ generation fought against and that she continues to fight are part of what it means for her to be African American.

While African American participants presented the most straightforward examples of the legacies theme, one Jewish participant also drew on a legacy of in-group oppression. Al, an activist in multiple social justice organizations, argues that his Jewish ancestry motivates him to take action, just as it has motivated his parents and grandparents before him. The discrimination and genocide against Jewish people are why he believes that “we, as Jews, we have to [set] an example for everybody else.” He illustrates how a Jewish person should behave by citing his own grandfather:

I think I have sort of a family legacy—sort of like a story of what it means to be a Jew in the world . . . I grew up learning about my grandfather . . . who lived in [Southern City] and like was . . . beyond his time when it came to issues of racial justice in [Southern City]. Like wouldn’t stand up for, at the beginning of [College] football games, he was a huge—he played on the university football team and went to like every single home game his entire life—and they used to play, back in the day, the song “Dixie.” . . . Everybody stands up for it. He would never let my mother or her siblings or anybody in the family stand up for that, to pay respect to a song that was . . . celebrating a culture of slavery.

One of the defining characteristics of the legacies theme is the repeated use of we throughout many of the narratives; using we is a direct example of a collective identity (Gamson 1992) and further suggests that the storyteller has generalized a “cognitive, moral, and emotional connection” to a much larger community than people with whom they interact regularly (Polletta and Jasper 2001:285). While interpersonal relationships do appear in this narrative, they are largely present as exemplars for the participant to emulate as a member of the in-group. As a result, this provides a clear-cut example of collective identity: a preexisting social identity that the storyteller politicizes by incorporating an injustice frame and a legacy of resistance.

Boundary Adjustment of Social Identities

The second theme representing a collective identity is a boundary adjustment of an existing social identity in-group. Like the legacies theme, boundary adjustment begins the story by stating that the storyteller is part of an oppressed group. However, unlike the legacies theme, the storyteller does not see the group resisting oppression. As a result, the participant adjusts the boundaries of his or her in-group, drawing a contrast between those who are willing to fight (e.g., feminists) and those who are not (e.g., other women).

Amy, a white woman involved in a feminist group on campus who mentioned reproductive rights throughout the interview, stated that she works on reproductive justice issues because she is a feminist. She responded to the question “And why are you a feminist?” by stating:

Big question. I mean, obviously, I am a woman so I have a stake in it. Even though, not necessarily all women are feminists. That’s definitely not the case . . . you have like all these experiences. And then once, I guess the example of taking sex and gender, you like realize like that there’s a reason [for injustice], some of those experiences start to make sense and sort of like this system of how things work. And then you like realize—Oh, it doesn’t have to be like that. I can change that.

While Amy acknowledges that she is a woman and that women suffer from inequality, she also notes that not all women are willing to stand up for their rights. For Amy, this is a negative in-group valuation; adjusting the boundaries of the in-group so that she is a feminist woman as opposed to a woman lets her take part in resisting sex- and gender-based oppression. To contrast with the legacies theme: Where the legacies theme imports an injustice frame and history of resistance into a preexisting social identity, the boundary adjustment theme uses the injustice frame to alter the social identity from one that does not resist to one that does.

Another student, Margaret, contrasts her commitment to resisting patriarchy to that of her sister, who is part of the out-group of non-feminist women. While her sister holds relatively progressive political values, she is also a skeptic who questions everything—including feminist assertions.

[T]here’d be times when, especially as I was taking [women’s studies classes] and I was really refining my vocabulary about repression and inequality, then I’d be saying things . . . [about what] we’d read about in articles we were reading . . . and [my sister] would be like, “Well how do you know that. . . . What, where’s your proof? Where are the numbers on this?” And I’d be like, “Well some of it people just know from what they experience. You don’t, you can’t quantify everything.” It got to the point where she thought—she never said anything to me, but I could just tell that she thought I’d been brainwashed by all these radical [feminists] and going through this phase and maybe I’d come out of it someday.

Amy and Margaret’s quotes reveal an interesting dichotomy: While they define themselves as part of a larger group (being female), they do not hold particularly positive opinions of many members of that in-group. The boundary adjustment theme features the participant contrasting themselves to a set of people who, while like them in some ways, are uninterested or unwilling to resist. Amy and Margaret’s narratives suggest that a negative in-group valuation can spur a feminist to redraw the social boundaries that place both feminist and non-feminist women in the same social category.

One of the most common symbolic forms of boundary adjustment that the feminist women performed in this study involved decisions about physical appearance. In particular, many of the feminist activists in this study brought up their own decision to stop shaving their body hair. Laura, one of the feminists in this study, drew some similarities to her mother based on shared values. However, Laura’s mother still constructs femininity in a traditional manner and makes strong comments to her about decisions to have short hair and avoid shaving body hair. She says:

[My mother was the] lone female engineer in the many of the jobs she took part in. So she was always sort of, you know, encouraging me to not let anybody slow me down just because I’m female. So she’d always sort of impressed that on me . . . [but] she still engages in the social constructions of gender in a fairly traditional way. And she doesn’t necessarily mean to, because I know that what she’s really thinking is that she’s my mother, she wants me to be safe, she wants me to feel comfortable about myself, but occasionally you get comments.

The boundary adjustment theme resembles the legacies theme in that they are both social identities that are merged with injustice frames and are used as rationales for resistance. The legacies theme includes fairly stable in-group boundaries that incorporate an injustice frame and a history of resistance. However, the boundary adjustment theme uses the injustice frame as a reason to adjust the social identity; now those who resist are part of the in-group, and those who do not are part of the out-group.

Conscience Constituents and Oppressed Out-groups

Both the legacies and boundary adjustment themes emphasized that the participant is part of a social identity group that has suffered oppression in the past. However, not all collective identities are bound together by shared oppressions. McCarthy and Zald (1977) call individuals involved in social movements that are not directly affected by the movement’s success conscience constituents. In this case, an activist applies an injustice frame to an out-group that is suffering and needs help; their social identity as a conscience constituent is a positive identity that comes from working on behalf of others. Hillary, an anti-genocide activist who had recently returned from providing direct services in a conflict zone, said that her activism began because she felt terrible that she couldn’t do more:

I felt guilty leaving and not being able to stay with those people and help them. So I joined [the group] because I needed to feel like I was helping them in some way . . . I think maybe because you feel like what you’re doing matters, or that you’re making a difference, that you’re helping somebody. . . . I feel like it’s my calling in life or this is what I’m supposed to do is to make life better for people that it really sucks for.

Claire, an activist who works with an anti–human trafficking group, gives another example of this when she states:

I feel like if everyone could think about the first time they heard about the issue. It’s shocking. Because you think modern day slavery? What is that? That doesn’t exist. You’re trying to tell me that 12 year old girls are being forced into prostitution? . . . [I’m a] camp counselor. And to think about like my campers having to do something like that is outrageous . . . I think I have a natural instinct to want to protect them and I would never, ever want anything like that to happen to them. So when you see these documentaries and these videos of girls who can’t speak any English except for the fact they can tell how much certain things will cost in English—like, that’s disgusting. And you can’t just be like, “Oh, that’s not really happening.” Like, it hits too close to home. Even though it might be like, “Oh that’s in Cambodia—not really that close to home.” And then on top of that when you’re like, “THAT is happening in [my state]?!?!”

Neither Hillary nor Claire cast themselves as members of an oppressed group; instead, they use an injustice frame to define an out-group and then assign themselves the identity of conscience constituent. Furthermore, because the social identity is created in response to an injustice affecting an out-group, it is also inherently collective.

Hillary and Claire have another commonality: At various times in the interview, they note their devout Christianity, and they both indicated that many of their friends share their religious beliefs as well. It is not clear whether their Christian background is an important component of their conscience constituent identity. However, there are times at which the conscience constituent identity has a definite religious orientation and is rooted in a prior social identity of Christian. Evangeline, a white female who works with a Christian social justice group, provides one example:

I don’t really see it as an option to love Jesus and not to do what He has commanded us to do as believers. And there’s all through the Bible, there’s commands to seek justice and to bring justice to the oppressed and to free people from slavery and I, I can’t really get away from that. Because I don’t think I can look at one part of the Bible that says “Love your neighbor” and not do the rest of it. Like I can’t really pick and choose, what I can do. And it excites me because I think that’s, I think that Christians in the world are God’s tools to show His love to the world. And I think that that’s how, we’re God’s plan for freedom in the world. And so I mean, it’s a very clear statement to me.

Evangeline draws on her faith to construct a moral analysis, placing herself as one of “God’s tools” who is commanded to “bring justice to the oppressed.” Much like Hillary and Claire, Evangeline constructs an out-group of those who suffer and puts her in a group of individuals who fight for justice on their behalf. In contrast, Evangeline makes it clear that the in-group consists of Christians, whereas Hillary and Claire leave that part of the social identity open.

Discussion

What is an activist identity? Since activist identity plays a role in a large number of movement outcomes (Jasper 1997; Kelly and Breinlinger 1995; Polletta and Jasper 2001), the answer is important to several areas of social movement research, including micro-mobilization, organizational forms, and tactical choices. One the one hand, activist identity involves role-based expectations and responsibilities, which are supported and reinforced by relationships with friends and family (Stryker 2000). By providing support for the activist’s beliefs and actions as well as creating a sense of obligation due to personal loyalty, activist identity occurs by internalizing the expectations of a social role and its associated tasks.

On the other hand, activists ground their activity in a social category, a collective identity that involves defining a relationship between a social identity and an injustice frame (Gamson et al. 1982; Tajfel and Turner 1979). The relationship between the social identity and the injustice frame varies considerably across movements and personal backgrounds, and this study identifies three different narratives that represent collective identity. In the first story (legacies), the narrator tells how a group has suffered injustice but has fought and partially won; future resistance is required to ensure fairness. In the second story (boundary adjustment), the narrator tells how a group has suffered injustice but that their group has not necessarily resisted; the narrator belongs to a smaller, more select group that has decided to fight this inequality. In the third story (conscience constituent), the narrator draws a relationship between a group of “them” who suffer (the out-group) and a group of “us” who fight on the out-group’s behalf.

The findings in this study have major implications for the study of identity in social movements as well as micro-mobilization, organizational forms, and tactics. First, there are substantial differences in collective identity that revolve around the relationship between the social identity in-group and the injustice frame. While some scholarship has proposed a relationship between social identities and collective identities (e.g., Simon and Klandermans 2001), the existing literature does not provide many details. Decomposing collective identities into a relationship between social identities and injustice frames helps us to understand how collective identities form. ⁷ Additionally, collective identity plays an important part of organizational form, function, and tactics in social movements (Polletta and Jasper 2001). As scholars begin to unpack the ways that collective identity influence how groups make decisions, select targets, and deploy tactics, they should pay attention to the interrelationship of social identities and injustice frames; the combinations in which organizations define in-groups, out-groups, and grievances could play a crucial role in understanding their choices.

Second, because role-based activist identity is often reinforced in group contexts, it is easy to mistake it for collective identity. It is likely that this confusion is furthered by multiple meanings of the term group; a group can refer both to a set of individuals who all have relationships with each other or a category of people who share some characteristic. As a result, individuals who feel connected to their “social movement group” may really be expressing that their beliefs and actions have been validated by their friends. This differs from an expression of collective identity, which would involve acting as a “good” in-group member either by fighting injustice or advocating for certain tactics (Polletta and Jasper 2001). ⁸

Confusing role-based activist identity with collective identity obscures the impact of identity in micro-mobilization processes. For example, some scholars puzzle over how social movement groups manage their identities to both maintain and attract new members simultaneously (e.g., Friedman and McAdam 1992) when it is quite likely that different types of identity are responsible for different parts of micro-mobilization. Stryker (2000) argues that role-based identity helps to determine day-to-day task activity, including in social movements—for example, the decision to spend time planning a protest instead of going to the zoo. While collective identity can play a key role in initial recruitment, maintaining persistence over a longer period of time, and presenting a public image (Polletta and Jasper 2001), the actual day-to-day activities are probably a function of role-based identity. Alternately, White (2001) argues that in some cases, collective identity is unnecessary and role-based identity is sufficient for mobilization. Future scholarship should recognize the potential of role-based identity to shape activist behavior rather than relying on collective identity as the default theoretical explanation.

Third, the importance of role in social movements has largely been overlooked. When researchers investigate social movements, they tend to focus on activists in equivalent positions. This is likely because much of the early research on collective identity and activism was conducted on nonhierarchical social movement organizations; feminist collectives and radical environmental groups deliberately obscure individual roles to promote a sense of equality and collectivism, making role-based identity harder to observe and making collective identity more obvious (for examples, see Polletta and Jasper 2001; Taylor and Whittier 1992). And yet, it is impossible to distribute power, tasks, and resources equally across a volunteer group (Freeman 1972). This study investigated a group of individuals who had taken some degree of leadership over their activist groups. Future research should look at the actual tasks that participants perform within a variety of social movement organizations; different tasks may be associated with different role-based identities and may lead to differential patterns of mobilization.

Fourth, the present study looks at collective identity and role-based activist identity as discrete theoretical constructs, but virtually every participant expressed both a role-based and category-based identity in this study. As a result, there are still several unresolved questions about the relationship between collective identity and role-based activist identity. For example, Gamson (1991) asserts that collective identity is a necessary part of mobilization, Stryker (2000) asserts that role-based identity is required for activism, and White (2001) argues that either role- or category-based identity can motivate activism. Is it true that one type of identity is sufficient for activism, or is a combination necessary? Do activists negotiate between their roles and categories to create a unified activist identity, and how does the presence of a role-based identity change a categorical one? How much work does it take for movement groups to maintain collective identities as opposed to role-based activist identities? Is there a way in which these two identities interact over time—for example, how collective identities indirectly influence role-based identities via social networks? How do the roles that people take on (e.g., worker) lead to strengthened collective identities (e.g., union man)?

The relationship between role- and category-based identity is one of the more important frontiers in social psychology research (Burke and Stets 2009; Stets and Burke 2000). ⁹ Furthermore, understanding the nexus of role- and category-based identity would provide a core foundation for the social psychology of political life. Unfortunately, there are very few cases where participants discuss both types of identity in the same interview segment, and I am not able to observe changes over time. As a result, this study is unable to answer these questions. However, other research designs that look at changes in identity over days, months, or even years could potentially help to understand how combinations of people’s role-based and category-based identities work together. Looking at how these identities rise and fall over time would give a better sense of the combinations of category-based and role-based identities as well as the conditions under which activists can creatively use role-based identities as anchors for collective identities and vice-versa. Participant observation or intensive longitudinal designs would be ideal methods to observe whether people negotiate and integrate category-based and role-based identities.

Finally, the implications of this study speak to an issue raised by Polletta and Jasper (2001:284): Scholars have tried to make collective identity do “too much analytically.” Collective identity is the primary social psychological explanation of movement behavior, and as a result, it seems to operate everywhere and influence every aspect of a social movement. But acknowledging that collective identity does not do everything is only the first step; social movements scholars need to begin using other social psychological constructs to help explain movement behavior. This study begins to solve this issue by (1) providing an alternative social psychological explanation for some activist identities and (2) specifying how collective identity represents a relationship between two other well-studied phenomena: the social identity and the injustice frame (Gamson et al. 1982; Tajfel and Turner 1979). With substantial research on role-based identities, social identities, and injustice frames, using these constructs can help us understand how social movement behavior is similar to—and different from—individual action.