Introduction

Why do individuals care for the needs of others (particularly strangers)? This question is at the heart of a fairly extensive body of literature on caring and generosity. Particularly perplexing is the question of why individuals engage in acts of “altruistic” care. That is to say, why do individuals voluntarily act to promote the well-being of others without request for, or expectation of, reward. A substantial body of social psychological research has demonstrated that altruism is motivated by humanistic ideology, personality, affect, and social modeling (e.g., Batson and Shaw 1991; Johnson et al. 1989; Lee et al. 2005; Mattis et al., under review; Midlarsky et al. 2005; Oliner and Oliner 1988; Smolenska and Reykowski 1992; Underwood and Moore 1982). A smaller body of scholarship has highlighted the role of race, ethnicity, gender, and urbanicity in the development of altruistic forms of helping (e.g., Baron 1992; Becker, and Eagly 2004; Bridges, and Coady 1996; Butovskaya et al. 2004; Zaleski 1992). This latter body of studies has been compelling in that it has shifted the discourse about altruism away from a focus on individual-level—particularly intrapsychic—factors, and has attended instead to the ways that altruism is informed by social identities and broader social forces.

Scholarship on the role of race in altruism suggests that individuals are motivated to behave altruistically towards others who share a common racial or ethnic background because doing so promotes genetic survival (Butovskaya et al. 2004; Michalaski 2003; van de Berghe 1981). Class-based theories imagine altruism as a pattern of behavior that involves the downward flow of resources from more materially privileged to less privileged individuals (Michalaski 2003; Schervish 1990). These theories argue that individuals at the upper end of the socioeconomic strata often are motivated by moral or theological ideals to use their excess resources to benefit those who are in need (Schervish 1990). Gender informs the likelihood that men and women will help particular targets, the efficacy with which they judge the needs of others, the forms of help that men and women provide, and the relative risk associated with offering or accepting help from men and women. More specifically, men are particularly likely to provide forms of help that might be categorized as either heroic or chivalrous (i.e., helping acts directed towards classes of individuals who are perceived to be entitled to courtesy and protection; Becker and Eagly 2004; Eagly and Crowley 1986; Levine et al. 1994).

Finally, theorists argue that altruism is influenced by urbanicity. According to these theorists, urban centers are defined by the presence of a disproportionately high density of poor families, economic and housing instability, overcrowding, overexposure to crime, social isolation, and inadequate and unresponsive social and cultural institutions. These stressors are theorized to lead to a diffusion of social responsibility, an increased likelihood of anti-sociality, and a reduced likelihood that individuals will engage in behaviors intended to preserve the welfare of others (Amato 1983; Bridges and Coady 1996; Hansson and Slade 1977; Latané and Darley 1970; Levine et al. 1994; Milgram 1970; Zimbardo 1969).

Importantly, the extant work on the link between altruism, race, class, gender and urbanicity has been hampered by two distinct shortcomings. First, quantitative empirical work has deployed race, class, gender, and urbanicity as discrete independent variables in multivariate empirical models of altruism (Amato 1983; Becker and Eagly 2004; Bridges and Coady 1996; Eagly and Crowley 1986; Hansson and Slade 1977; Latané and Darley 1970; Levine et al. 1994). This treatment of markers social identity has failed to account for the reality that one is neither a singularly gendered, racialized nor classed being. As such, socially produced behavior (e.g., altruism) cannot reasonably be viewed as a product of either one’s gender, one’s race, or one’s class. Further, the lived meanings and ramifications of our various identities cannot be “controlled.”

Second, existing work has failed to account for the reality that identities are steeped in complicated histories and systems of meaning that become all the more complicated when they are woven together. As such, a serious effort to interrogate the relationship between altruism and social identity requires that we explore how these identities, taken in tandem, operate as historically and contextually situated constructs to shape peoples’ willingness to care for others. In this study we endeavor to achieve this end. More specifically, using intersectionality theory as a frame, we analyze data from a three year ethnographic study of altruism conducted in among African American adults from a low-income, urban, housing community in the effort to explore the ways that key markers of social identity (e.g., race, gender, class, urbanicity) work together to inform (1) individuals’ willingness to engage in acts of caring, as well as (2) the particular patterns of care in which individuals engage.

Four defining tenets of intersectionality theory are central to the present study. First, intersectionality theory asserts that social identities are neither exclusive nor discrete. Indeed, each person is the bearer of multiple identities (e.g., gender, class, ethnicity) that complicate each other (Crenshaw 1993; McCall 2005; Yuval-Davis 2006). Second, the theory holds that social identities are steeped in ideological and representational regimes that include beliefs, myths and stereotypes about the abilities, worth, and roles of people, and that profoundly affect how people live (Crenshaw 1993). Third, intersectionality theory argues that social identities and their associated systems of representation are historically and contextually situated (Crenshaw 1993; Yuvall-Davis 2006). That is to say, identities, and the beliefs, myths and affect (i.e., emotions) attached to them, function differently at different times in national as well as personal history, and in different geographic, institutional, and social organizational settings. Fourth, the theory holds that although identities are concretely attached to individuals, they do operate within, and are affected by, structures of power (e.g., the legal justice system, educational system, and political systems). In sum, as racialized, gendered, and classed beings, all people interface with systems that view or respond to them in ways that reify histories and patterns of marginalization, oppression, or privilege, and that either support or thwart efforts to survive and or thrive. In sum, these four tenets of intersectionality theory focus on the ways that identities and social structures change, and are changed by, each other, and the ways that those changes inform the behaviors, life conditions and life outcomes of individuals and groups at various times and in various contexts. Given our interest in the link between altruistic giving, social identities and social structures, intersectionality theory emerges as an ideal analytic frame.

The goal of this study is to examine the ways in which classed, racialized, gendered and urban identities work together to inform the emergence of altruism particularly for low-income, urban-residing, African Americans. Because we are interested in the interpenetration of identities and in people’s subjective understandings of how these identities shape behavior, we take a deliberately qualitative approach in this study.

Method

Participants

Participants were 40 adults (18 women, 22 men) who reside and or work in a low-income housing community (Hasseldorn Houses) located in New York City. We note here that the names of the housing community, the names of individual participants, as well as all identifying information (e.g., participants’ country of origin) have been changed in the effort to preserve the anonymity of participants. Hasseldorn Houses is a complex of federally subsidized housing units that spans several square blocks and that serves as home to more than 2,000 families. The median family size is four. Approximately 50% of the adults in this housing community have earned a high school diploma, GED or higher. More than 50% of women and more than 40% of men in Hasseldorn Houses are employed. The families who reside in the Hasseldorn Houses live at or below the federal poverty datum. Indeed, the median household income for families in the community is under $13,000. All of the participants were of African American, Afri-Caribbean, or of Afro-Latino descent. The sample ranged in age from 23 to 90 years, and all were parents to at least one child.

The Principal Investigator (PI) spent several months in the community prior to beginning of the study. This time was spent developing relationships with residents and community stakeholders (e.g., ministers, community activists). The research team worked with residents and community stakeholders to identify research questions and or community needs that might be addressed by the research team. The PI and the research team served as volunteers in a range of capacities including chaperoning youth on outings for a local summer camp, developing a library in a local community center, and leading workshops. The team’s participation in these community activities was not conditioned on community members’ decisions to participate in the study. Volunteering did, however, allow the research team and community members to develop relationships, and served as a context in which members of the community could assess the credibility and sustained commitment of the research team. Prior to the beginning of the study, community stakeholders reviewed study materials (e.g., the interview protocol) and assisted in refining interview questions.

Participants were recruited through a variety of means including through posting fliers for the study in public spaces throughout the community (e.g., on bulletin boards in local nail salons, barbershops, and the local community center). No a priori attempt was made to select participants on the basis of gender or ethnicity. Community stakeholders (e.g., residents, business owners, ministers, and individuals who had a long history of working in the community) were informed about the focus of the study and asked both to nominate individuals whom they viewed as altruistic, and for their assistance in informing individuals in the community about the study (e.g., distributing fliers about the study to community members especially to individuals whom they perceived as especially altruistic). This approach allowed us the benefits of self-selection as well as of community nomination. Residents as well as workers in the community were interviewed because conversations with members of the community revealed that community membership was determined by residency as well as by long-term presence in Hasseldorn Houses. As such, men and women who spent a substantial amount of time in the community, and who had developed significant, positive, and sustaining relationships with families there were considered to be members of the community. The five participants who worked but did not live in Hasseldorn had a minimum of 10 years of experience in the community. Fifty-two people contacted the PI and consented to be interviewed. Twelve of these interviews were not conducted either because of scheduling challenges or because the interviewees consistently failed to appear for interviews. The 40 individuals who completed the study were interviewed in a place of their choosing. Interview sites included private rooms in the local community center and private settings in the interviewee’s place of employment

Participants were interviewed by the Principal Investigator (PI) and a research assistant in private settings (e.g., a private room in the local community center). The PI conducted the interviews while the research assistant was responsible for taking detailed interview notes, asking any clarifying questions that might emerge, ensuring that probes were asked, and managing the tape recording equipment. Interview questions and probes inquired about participants’ experiences as witnesses of altruism, or as givers or recipients of altruism. More specifically, participants responded to two questions and their accompanying probes: (1) “Please tell me about a specific situation in which you saw (or heard about) someone in this community going out of their way to do something to help someone else,” and (2) “Please tell me about a time in the past year when you went out of your way to help someone who is not related to you.” The probes focused on the motivations for behaving altruistically were: “What do you think made them do __?” “What made you do __?” Responses were excluded from the analyses if participants indicated that their motivations were self-serving (e.g., that they were seeking to obtain material rewards). The interviews lasted 90 to 120 min and were audio-taped. Individuals who completed the interview received a $25 gift at the close of the interview. In order to ensure that prospective participants would not be coerced by the incentives, participants were not informed about the incentive payments until at the close of the interview. Each interview was transcribed verbatim. This study focuses on verbatim responses to the two core questions.

Two coders independently conducted line-by-line readings of all transcripts. To guard against bias in the analytic and interpretative processes, the two individuals who coded the data were not involved in either the data collection process or in conducting service work in the Hasseldorn community. An open-coding process (Strauss and Corbin 1990) was employed with phrases used as the units of analysis. Given that some individuals have a tendency to repeat ideas, we recorded the presence or absence (rather than the frequency) of thematic occurrence for each participant. This approach ensured that we would not yield artificially inflated data that would lead us to conclude that a particular theme was especially important because it was endorsed more frequently. In the first stage of the coding process each coder independently identified emergent descriptive themes from all of the transcripts. Consistent with Miles and Huberman (1994), the initial set of independently identified themes was compared and integrated into a single comprehensive list of coding categories. Coders refined the coding scheme by discussing the meanings of, and relationships between, each coding category, and identified rules for determining when a particular coding category should be assigned to a response. The two coders assessed the reliability of the coding scheme using randomly selected narrative samples from four of the forty interviews. After each effort to establish reliability the coders clarified the rules for applying the codes. Inter-rater reliability was calculated using the formula: inter-rater reliability = agreement/(agreement + disagreement) and a target rate of 85% reliability was used as the lowest acceptable level of inter-rater reliability for the study (Miles and Huberman 1994). The rate of reliability for this study was 87%.

Results and Discussion

Narratives revealed four overarching themes related to the ways that the intersection of identities informs altruism. First, narratives revealed that social identities work together to create different patterns of vulnerability and, consequently, different needs among particular groups of individuals. Second, individuals’ movement across particular lines of identity (e.g., class) helped to shape their subjective sense of responsibility to care for others. Third, intersecting social identities shaped individuals’ commitment to care for particular identity groups. Fourth, intersecting identities worked together to shape peoples’ perceptions of altruists. These themes are explicated in turn below. Narrative samples are offered in an effort to illuminate each theme. Bracketed text in these narrative samples reflects text that is paraphrased and included for the purpose of clarity.

Intersecting Identities Inform Patterns of Vulnerability and Inform How and What People Share

Race, gender, class, urbanicity and age, work in tandem to differentially shape individuals’ vulnerability to particular social forces. Men in the study, particularly those in their 20s, 30s, and 40s were especially explicit about having either witnessed or personally experienced assault, harassment, and or public humiliation at the hands of one particular structure of power: the police. These men reported intervening to protect other members of the community from police maltreatment. For example, Paul, a participant in his early 30s noted:

One day some cops came, they weren’t dressed like cops…They were picking the kids (boys) up and slamming them down on the concrete. So I called the cops and I told them they got a gun, and I thought someone was gonna get killed…I lied to make them come around. …You see, things like that is worrisome and bother you. You know, that could be me. I might be walking down the street coming from the store and got mixed up in that. They don’t know who they grabbin’. Except for the color. The kids they were beating up were Black.

Paul’s actions highlight two realities. First, age, race, gender, urbanicity work together to mark young, urban-residing men of color as particularly vulnerable to police (and other forms of institutional) violence. Second, when victims are from stigmatized social groups those who endeavor to care altruistically for their well-being sometimes are required to engage in transgressive behaviors (e.g., dishonesty). Paul, for example, asserts that a call to the police indicating that other police officers were assaulting young men of color would not likely inspire immediate, protective action. As such, he constructs a tale that he knows is more likely to get the attention of the police, and that will more likely result in positive outcomes for the young men (i.e., he tells the police that someone had a gun and that a murder might be imminent). Ironically, these transgressive acts may reify a view of members of this community as deceptive, immoral, and untrustworthy.

In contrast with men’s vulnerability to the police, women (particularly mothers) identified a particular vulnerability to child welfare agencies. Twelve women endorsed this point. For example, Darlene, a 38 year-old, single mother of three children, noted the tendency of outsiders [including the staff of Agency for Children’s Services (ACS)] both to essentialize low-income, urban residing individuals, and to superimpose on low-income, urban residing black women, images of social and moral pathos. These essentialized images inform the way that child welfare agencies (e.g., ACS) respond to the problems that emerge for parents and families in the community. Darlene stated:

When you say people who live here are drug addicts, and they are, you know—they just don’t care about their children, that’s not an accurate picture. And that’s how ACS see it. … The minute you have a problem with your child, they are ready to take it. And then they are ready to lump you into the people who… are uncaring and all those Black people who just sit there, and don’t do anything, and don’t want better for themselves, and that’s not true.

Consistent with the concerns raised by Darlene, adults in this study often engaged in acts of altruism that were centered on protecting children and mothers from the actions of the child welfare system. People provided food, clothing and other instrumental supports to parents (usually mothers) and children who were in need. They also provided (often unplanned) temporary and long-term informal care for children in the community as a way of buttressing against the possibility of having children taken by the ACS.

Nicole, a woman who is parenting her nephew (her only “child”), was one of nine women participants who reported that at various points in their adult lives they have spontaneously taken custody of at least one child who was in need of care. She stated:

We had a little girl. Her mother had gone away and left her at home alone and it had been like 24 hours since she had seen her mother …She asked someone to cross her across the street and she came immediately here. We could have, you know, did the legal thing and called the police and have them take her and do everything. But we took care of her for about two months … and once her mother was okay and back on her feet, you know. But, my son said, “Well, mommy why are you doing this?” I said, ‘Well, she can end up some place where she might be molested, she might be raped, the people won’t take care of her or things like that… we didn’t want her life to be turned up any more than what it was.

Nicole’s specific act of altruism, and the fears that prompt that act of altruism, are best appreciated when viewed against the backdrop of larger, historically situated intersecting relationships between race, class, womanhood, girlhood, innocence, and sexuality. Historically rooted representations of low-income black women and girls as loud, aggressive, assertive, and hypersexual place these women and girls outside of the boundaries of protections afforded by authentic and respectable femininity (Crenshaw 1993; Roberts 1997). Nicole’s rationale for taking custody of this child reflects a recognition that a young, poor, black girl can easily be preyed upon by individuals who inappropriately view her as a viable sexual target, or who fail to recognize that she is a child who deserves and requires protection. In contrast to these essentialized and devaluing representations of black women and girls, Nicole views this mother as a woman in the throes of extraordinary distress who, with time and support, has the capacity to change and care for her child. It is her ability to hold a humanistic and respectful view of both mother and child that sets the stage for Nicole’s altruistic action (i.e., her decision to take custody of a child she barely knew).

Intersecting Identities, Border Crossings and Subjective Social Responsibility

Sixteen participants reported that shifts in their place of residence (e.g., from the rural South and from the Caribbean to New York City) and in their class standing (e.g., the movement from extreme poverty to working poor or working class status) left them with a sense of responsibility to ease the plight of individuals with limited resources, limited privilege, and limited options. These sentiments were expressed by Enoch, a Caribbean man in his 80s who emerged out of an early life of poverty and, as a result of education and employment opportunities, was able to enjoy relative privilege as a working-class adult. He migrated in early adulthood to work in a destitute area of the world. This experience led him to a life of altruism that included paying for the education of indigent children. He explained:

I’m a strong believer in giving people opportunities. … I didn’t believe too much in the class structure… from the Bush’s standpoint, but I believe that if people have the education and opportunity, they could be anything. And, so I helped. You see, I worked in [farming in the islands] and so I had a lot of sympathy for all the tragedy. I was never very destitute in my… adult years, but I met a lot of people who I had to help to get out of it, to lift them out.

Enoch’s work, his travels, and changing economic status brought him into contact with people who were more disadvantaged, and forced him to confront issues of relative class privilege. This awareness of his relative privilege left him with a heightened sense of responsibility, and compelled him to behave altruistically toward those who were more economically disadvantaged than he.

As was the case for Enoch, for other participants the fluidity of class and the permeability of classed spaces (i.e., class contact) proved to be important in the production of altruism. Some participants (n = 5) noted that they grew up in extreme poverty and were better resourced now than during their early years. Others (n = 7) who noted that they came from working-class families and those who saw themselves as poor (n = 4) lamented their current SES. Movement across statuses gave people a sense of responsibility. Some men and women in Hasseldorn (n = 14) noted that over the course of their lives they have worked with or for middle-class or wealthy individuals or in middle-class or wealthy communities. Routine crossings of classed spaces (e.g., neighborhoods) and into and out of relationships with well-resourced others, gave individuals access to relationships and resources (e.g., clothes, food, knowledge) that they were able to use to the benefit of others. For example, Francine noted:

I had met this lady…she had five children… She told me that her kids were hungry and that they wanted something to eat and I got into my purse and …I went and I brought her some food, and I went and gave her some money. And … the lady that I work with, she has a little child. She normally gives me…you know, I have seen her with clothes like she wanted to take them out [and throw them away], and I told her to give them to me because I knew somebody that needs them. And, I take the clothes and [give them to the lady who needed them].

Although class and the crossing of classed borders were important in the social production of altruism, ethnicity and migration also emerged as important. Finally, Darlene describes the ways in which identity and relationships forged across lines of ethnicity influenced engagement in acts of care. Darlene is part of a trio of Caribbean women who hail from three different islands. These three women care for each other’s children, and provide social and material support to each other. Referencing a common set of images from America’s racialized and gendered narratives of womanhood, Darlene argued that she views African American women as “free,” “fearless” and therefore capable of advocating on their own behalf with institutions. Darlene suggests that this freedom and fearlessness means that African American women are more knowledgeable about the social service system, they are more willing to ask for the services and resources to which they are entitled, they are more willing to challenge abusive treatment at the hands of social workers, and they are at less risk because they are not immigrants to the USA (as such they cannot be deported and they have full protection under American laws). In sum, Darlene suggests that in contrast to African American women who can rely on established institutional systems to help to meet their needs, Caribbean women must rely more heavily on each other for their survival. Importantly, she and her Caribbean friends created an enclave of support through which they give to each other in ways that preserve each others’ dignity. For example, Darlene reported that her friend, Shirley, one of the trio of Caribbean women, will casually invite Darlene’s family over for dinner when the latter’s family is in need of food. Through these practice, Darlene and her friends meet concrete needs that might not otherwise be met. Darlene states:

I don’t know how she knows, but she will invite me to dinner. My whole family. “Tonight, Miss Anderson, dinner’s on me.”… If I don’t have food I know to call her. …. You know, I am Trinidadian. And, unfortunately, Caribbean people have a little bias against, I think, American people because, Black American, because they are so free. And, they are so independent. We are a little bit scared of them. I think I was because they seem so fearless in their little way that, I come from a country [where we are] a little bit repressed, you know, … it’s amazing to me when I look at an American woman.

Darlene reifies particular essentialist views of African American women (e.g., as strong and independent) and of women from her country (e.g., somewhat repressed). These images signal Caribbean women’s difference from their African American counterparts. These essentialized images are called up by Darlene to explain the seemingly insular nature of the altruistic giving that transpires between her and her Caribbean friends. Importantly, although Darlene invokes essentialist views of African American women as a way of explaining the need for Caribbean women to care for each other, neither she nor other Caribbean women participants in the study argued that this view of African American women would make them less likely to help African American women who might be in need.

Intersecting Identities Shape Commitment to Care for Particular Identity Groups

Social identities informed who people saw as worthy targets for altruistic action. This theme was evident in the narratives of 100% of interviewees (n = 40). For example, Mikal, a man in his mid-thirties from a working-class family who grew up around and now works in the Hasseldorn community, noted that although he mentors youth in Hasseldorn, he is particularly focused on supporting the welfare of girls. He stated:

I went to the neighborhood high school here. Yeah, so, you know, some of the girls who may have gotten pregnant in high school, you know, I know their children and grandchildren. …When I first came here, that was the first thing that really happened to me that really hit me, you know. …I favor females over males. … their patience—I guess that I’m a man or they’re a woman, or growing up to be a woman… I see that they need some of the more critical things because they’re more a victim than men are. Once a woman is in the gutter, she’s always in the gutter. A man can brush himself off and still be a man. You know what I’m saying?

Mikal’s decision to focus his energies on mentoring young women is rooted in a classed understanding of manhood/masculinity and womanhood/femininity. He suggests that men who succumb to addiction or premature parenting can, because they are men, eventually overcome the stigma of those experiences. However, he believes that the social construction of womanhood means that young women are more vulnerable to the effects of social stigma and that they cannot easily recover from violations of codes of femininity. Consequently, he directs his altruistic energies towards preventing women from succumbing to the forces that would lead them to “the gutter.”

While Mikal’s decisions about whom to help are rooted in a focus on gender, for Jude, a man in his mid-fifties, altruism is rooted in a focus on race. Jude reported that he selectively directs his altruistic energies towards Black men and women. Jude stated:

I was more of a Malcolm (X) man. I love the way he would look them in the eye and say what it was that he had to say… I am not hampered by the little bit of money I’ve accumulated or my position in the world. I love my people and when they’re offended I feel this need… I’m not gonna walk by a hungry Black kid. I am not going to let a Black man beat up a Black woman, or a White man beat up a Black woman. That’s family to me.

In his explanation of his particular pattern of altruism Jude invokes two images. First, he calls upon a specific racialized model of masculinity—a model embodied by Malcolm X. This model of masculinity is unapologetic and uncompromising in engaging with structures of power and privilege. This raced model of masculinity conveys his style of confronting those people or institutions that function in ways that are destructive to Black people.

Second, he uses race as the source of his sense of filial connection to and responsibility for the plight of Black people in this struggling community. Importantly, Jude acknowledges the class privilege afforded to him by his occupation (program director). However, he reports that his class position does not obscure either his sense of connectedness to, or his understanding of his responsibility to assist those who are less well resourced. Jude’s sense of responsibility is an act of resistance against the view that as African American people rise in class standing they become increasingly disengaged from others in the community.

Identity Shapes Perceptions of the Altruist

Altruism often requires individuals to engage with stigmatized others (e.g., addicts). This engagement with socially stigmatized others often made altruists targets of suspicion and rejection. Lemuel, a 27 year-old man, was one of 11 people who articulated this point. Lemuel expressed this point in his discussion of his experience of helping a male friend who was addicted to crack cocaine, and who had become homeless as a result. He noted,

I used to buy him new clothes… That was coming from the heart. Despite, there was a time— a young lady, I wanted to, you know, date. She thought I was on drugs ‘cause I was hanging out with him. She figured, I mean, “Why are you hanging out with him? You have to be doing the same thing that he’s doing.” But no, that’s out of the heart. To get him back, to get him back on his feet.

In sum, reaching out to aid marginalized members of the community came at a cost. Sometimes others could not be convinced that an act of altruism was simply an act “out of the heart.”

Implications

Research on altruism has highlighted the point that gender, race, class and urbanicity influence altruistic giving. However, in investigations of altruism these indices of social identity traditionally have been examined in isolation from each other. The present study used intersectionality theory as a lens through which to address a single organizing question: “What does a focus on situated social identities reveal about the expression of altruism?” Our findings point to four advances wrought by the use of intersectionality theory as an analytic lens.

First, attention to intersecting identities highlights the reality that different combinations of social identities create different patterns of vulnerability. Those patterns of vulnerability, in turn, inform the kinds of help that individuals receive. For example, young, low-income, urban residing men are particularly vulnerable to becoming victims of violence, and young mothers are especially vulnerable to losing custody of their children. Sensitivity to these patterns of vulnerability may shape the kinds of help that particular members of the community are likely to give to others in this community. Second, as a theoretical frame, intersectionality theory highlights the role of identity mobility (e.g., immigration, class mobility) in the evolution of altruism. Third, the theory highlights the fact that situated social identities lead people to behave altruistically toward particular groups within the community (e.g., youth, girls). Finally, this work highlights the point that members of the community who choose to behave caringly towards socially stigmatized others (e.g., addicts and mentally ill people) often pay costs for their altruism (e.g., they become objects of suspicion).

This study is not the first to take an ethnographic approach to the study of life in a low-income, urban, African American community. Nor is the study the first to examine the ways in which social identities combine to inform social behavior. Indeed, Stack’s (1983) study of kinship network development in The Flats, a low-income, urban African American community, and Kathryn Edin’s more recent ethnographic studies of economic survival and the meaning of marriage among low-income, urban residing families (e.g., Edin 1991; Edin et al. 2004) are critical examples of scholarship that links social identity, urbanicity, and psychosocial outcomes. A distinguishing feature of the present work is its explicit focus on theory—particularly intersectionality theory. That is to say, this study contrasts theories that consider gender, class, race, and urbanicity in isolation from each other with intersectionality theories that consider these identities in tandem. A second distinguishing feature of this work is its explicit focus on a broad array of social identities. In her work Edin elects class and gender as central points of focus; Stack centers her analysis largely on the intersection of race and class. In this study we explored the way that a range of social identities work together in various contexts to shape prosocial development among a group that has been excluded from much of the social science research on altruism. In the end, this work deepens existing psychology research on altruism (and on prosociality more generally) by attending to gender, race, class, urbanicity as substantive categories that work dynamically to influence behavior.

Extant theories of gender, race, class, and urbanicity, are not independently useful for explicating altruism among our participants. Gender theories have drawn our attention to the relationship between traditional and non-traditional notions of manhood and womanhood and altruistic giving. However, these theories have failed to consider the ways in which the relationship between gender and altruism is informed by race, ethnicity, and class. For example, gender role theory suggests that chivalry is important in the landscape of altruism (Eagly and Crowley 1986; Eagly and Wood 1991). However, chivalry is a gendered, classed, and racially-grounded construct. One cannot talk about chivalry without grappling with racialized and classed notions of manhood, ladyhood, respectability, and entitled protection. But, the findings of this intersectional analysis remind us of the limits of this chivalry argument because in America’s sociopolitical imagination, urban-residing men of color are rarely represented as chivalrous, and their female counterparts are popularly represented as neither ladies nor entitled to the protections and courtesies of ladyhood. African American men and women in low-income, urban communities operate within a complex matrix of representations some of which are consistent with traditional notions of chivalry, and some of which suggest that men and women in the community are incapable of living within accepted codes of ladyhood and manhood. In fact, the reality that Black women and girls in this low-income urban community are generally situated outside of the context of the protections of authentic ladyhood inspired fear in Nicole and motivated her to protect a child she barely knew.

This intersectionality approach points us to a novel way of understanding the link between social identity (e.g., class, ethnicity) and altruism. This novel understanding is effectively captured by the theme of border crossings. Two distinct kinds of crossings emerged from the narratives of participants in this study. Some participants experienced shifts in identity status (e.g., upward and downward class mobility, migration, and immigration) during their lifetime. With respect to class, these shifts sometimes brought people into contact with even more economically disadvantaged others. The confrontation with others who had greater material needs inspired some participants to view themselves as relatively privileged. Here, Cole and Omari’s (2003) work on class mobility among people of color is especially relevant. Their work highlights the fragility of class for people of color, and draws attention to the ways that upward mobility informs people of color’s sense of responsibility for the uplift of family and community. Importantly, urbanicity is also subject to these kinds of analyses. Although theories of the link between urbanicity and altruism imagine urbanicity as a stable identity, many people who reside in urban contexts were raised (or spent substantial portions of their formative years) in non-urban contexts. This is certainly true for participants in this study who mentioned that they were born in the Caribbean, as well as participants who were born or spent substantial time in the South (n = 12). Attention to movements into and out of urban spaces forces us to consider the reality that urban residing individuals who have lived outside of urban contexts may have encountered different representational regimes (e.g., ideas about race, class, gender) that ultimately inform the ways that they make meaning about needs and about our responsibility to behave altruistically toward others. The second form of border crossing experienced by some participants was a more localized and more routine movement across classed spaces during their daily lives. For most participants (n = 21) work served as the bridge between classed spaces. The permeability of classed spaces provided some participants with access to resources (e.g., clothes, food, knowledge) that they could share altruistically with others.

The fluidity of identity that is intrinsic to this notion of border crossings is worthy of attention. Theorists often think about social identities as fairly stable. This notion of border crossings forces us to consider that while some of the identities that we hold (e.g., race and the phenotypic characteristics that signal one’s membership in a particular racial group) are fixed, over a lifetime other identities (e.g., class) are much more mutable. Notably, both fixed and fluid identities have different meanings in different contexts and at different times in history. In sum, although an individual’s race does not change over his lifetime, the meaning of race will change as he moves from contexts in which his racial group is well represented into contexts in which his group holds marginal or minority status. In addition to our consideration of the functions of fixed and fluid identities, we must acknowledge that the nature of lived experience is such that people sometimes move into and out of contexts (e.g., workplaces, friendships) that bring them into contact with people of different social identities, and into contact with different systems of meaning. Social science researchers, including those who are interested in using intersectionality theories as lenses through which to examine behavior, may benefit from taking three steps. First, in addition to using conventional category-driven measures of race, ethnicity, class, and gender researchers may find it beneficial to employ scales that assess changes in the salience and meaning of various social identities across contexts and time. For example, a person of Caribbean descent who resides in a largely African American community may experience ethnicity/nationality as especially salient in the community where she resides. If this person works in a largely White and middle class setting, then race and class may be more salient in the hours when she is at work. Further, as each identity changes in salience, that identity may influence the meaning(s) of other related identities. As such, to the extent that gender and class are cultural constructions, then priming Caribbean cultural identity in neighborhood contexts may also prime individuals to ascribe culturally relevant meanings to gender and class. However, the priming of race and class in work contexts may lead to a different way of understanding and enacting gender, and to a different set of behaviors. By measuring the dynamic relationship between context and various social identities we will be better able to understand individual and group-level variations in social behavior.

Second, with respect to analysis, researchers should examine whether changes in the meanings and functions of various social identities can be modeled empirically with the use of emerging techniques for the analysis of change (e.g., growth curve modeling). Given that the salience, meanings, and functions of some social identities can change while others remain fixed across time and across contexts, researchers may benefit from using techniques of empirical analysis that account for both intra-individual growth and contextual as well as experiential stability over time. Finally, given that qualitative research is ideal for unpacking meaning, it will be useful for researchers to conduct qualitative studies (particularly longitudinal qualitative studies) that explicitly link behavior and psychosocial outcomes to the changing meanings and functions of interlocking social identities. Certainly, studies that critically link quantitative and qualitative approaches are likely to prove especially beneficial.

This study has important limitations. Although we endeavored to create a sampling strategy that would not depend solely on self-selection, we are aware that self-selected participants and peer-nominated participants may not be representative of the general community. In addition, because the study was designed to examine motivations for and manifestations of altruism, the questions and probes used in the study did not lend to a fully focused examination of social identities. Finally, we are cognizant that studies of positive social behaviors are likely to inspire a focus on socially desirable self-representations. We do not presume that social desirability bias can be eliminated, however, the fact that participants were more comfortable describing how they had been helped by others than describing their own acts of altruism suggests that social desirability may be less prominent and influential than we might expect. A study that deliberately endeavors to explore the role of social identities in shaping altruistic behavior (and that uses probes and questions intended to examine these identities) will certainly reveal a more detailed and complex view of the identity-altruism link.

The limitations of the study notwithstanding, our findings raise important questions about the ways in which identities inform the social ecology of care. Our findings suggest that identities help to shape how individuals construct advantage, disadvantage, and privilege. Interventionists may find it useful to explore the specific pathways by which situated identities inform how people understand and legitimize need, who is worthy of being helped, and who is responsible for helping. For example, it is important that researchers examine the ways in which the stereotyped representations that are embraced by African American and Caribbean individuals, and by men and women, inform these individuals’ willingness to engage in in-group versus out-group acts of altruism. As discussed earlier in this manuscript, researchers may find it valuable to explore the ways in which the priming of particular social identities (e.g., ethnicity or race) influences the meanings of other identities (e.g., class), and to examine the ways in which this dynamic influences behaviors in different contexts and at different points in development. Finally, the finding that some individuals were treated with suspicion by others raises questions about whether some individuals may opt not to help particular groups of people because of the social costs associated with being linked with members of those groups. If people are deterred from behaving altruistically towards certain stigmatized others because of the social costs of doing so, then interventionists may benefit from identifying strategies for helping potential altruists to negotiate these concerns. The survival of vulnerable individuals and groups may depend on us doing so.