Keywords

Among sociologists who study inequality there is a good deal of agreement on the broad conceptual and analytical parameters of the field. Social inequality is typically conceptualized as a relatively durable pattern of institutions and social relationships in which valued resources are distributed unevenly across social groups and social categories. Consistent with this definition, empirical analyses generally focus on (1) the processes by which resources obtain value; (2) the rules for allocating resources; and (3) the mechanisms linking individuals to resources (Grusky 2007). It is worth noting, however, that most theory and research in this tradition explores the causes and consequences of inequality without explicitly theorizing the self or identity (e.g., Neckerman and Torche 2007; see Hunt 2003 for a similar conclusion). Surprisingly, this is the case even for a large segment of social psychological research where the dynamics of social interaction are of central concern and the “mechanisms linking individuals to resources” are a primary focus (Hollander and Howard 2000). Given that significant advances in our understanding of social inequality have been achieved without explicitly theorizing the self, it is reasonable to ask if a more formal and deliberate incorporation of self-processes is even necessary. Might self and identity be tangential or peripheral to the production and reproduction of inequality?

There are at least three basic arguments in support of a more intentional examination of self and identity in the study of social inequality. I briefly summarize these arguments before developing them more fully in the review and analysis that follows. First, some sociologists study inequality because the uneven distribution of resources in society is assumed to be harmful to human dignity. While this reason is rarely explicit, there is a widely shared presumption that actual persons are injured both physically and psychologically by systems of inequality. To be sure, the question of human dignity is more commonly problematized in some theoretical traditions than others. Critical Theory and Marxist traditions, for example, are unequivocal in their advocacy of equality and concern for the preservation of human dignity (cf., Agger 1991; Wright 2010), and most feminist traditions also begin with the position that equality and dignity of persons is preferred over inequality and indignity (cf., DeVault 1996). But even in traditions that are more positivistic in orientation, strict ethical standards for conducting research are in place to protect human dignity, and most sociologists display a formal professional commitment to protocols that avoid the exploitation of human subjects. In other words, the stratification of people in society is widely assumed to be morally and ethically distinct from the layering of sedimentary rock, the stratification of basal cells, or the dominance hierarchy in a wolf pack. Thus, the self matters because it emerges from persons, and persons have a common interest in preserving human dignity and advancing the values of equality, justice and solidarity.

The second argument in support of a more intentional examination of self and identity in the analysis of inequality is that the meanings and social practices that frame and define interaction are expressed in terms of social identities, and many categories of identity are the product of inequality processes. This means that social inequality cannot be adequately or fully addressed without considering the meanings of value and power attached to person labels. Man and woman, black and white, gay and straight, employee and owner, are not simply categories of difference, they are the symbolic means for doing inequality (see Wilkins et al., this volume).

Third, the self matters in the study of inequality because it operates as a social process, or mechanism, that converts social interaction into higher order patterns of resource distribution. The self is more than an outcome variable that happens to be correlated with inequality; it is instrumental in the generation, reproduction, and alteration of the social structures that sustain inequality.

The following review and analysis of self, identity, and inequality is organized around these three arguments. In the first section, I clarify the conceptual boundaries that define person, self and identity, emphasizing the particular importance of personhood in the study of inequality. In the second section, I examine identity as a product of inequality at three different levels of analysis (person, interaction, and culture), and review research on value and power as dimensions of identity meaning. In the final section I continue to differentiate among three levels of analysis as I explore the self as a process in the production of inequality.

Person, Self, Identity, and Inequality

Person, self, and identity are interdependent concepts with a rich, complex, and sometimes messy intellectual heritage. Clarifying the boundaries among the three overlapping concepts is a necessary first step toward a more coherent explanation of their contribution to the study of inequality.

For most social psychologists, the concept of “person” is synonymous with “human being” or “individual” and is typically viewed as the corporal slate upon which self and identity are written. But this narrow view of personhood in relation to self and identity misses a key dimension of social life. According to Cahill (1998, p. 131) a proper sociology of the person is one that focuses on “the publically visible beings of intersubjective experience,” as well as the cultural interpretations of what it means to be a person in different societies. This emphasis actually has a long history in sociology and anthropology dating back to the work of Durkheim and Mauss (cf., Carrithers et al. 1985). Durkheim ([1915] 1965, pp. 305–306), for example, saw personhood as a collective representation of the individual; a social fact that reflects a shared understanding of what it means to be a human being in a particular time and place. Under this conceptualization, the definition of person is conditioned by the dominant folk psychology of the culture. Thus, the assumption that persons are unique, self-reliant individuals may be characteristic of modern, western representations, but it is not a definition consistent with representations of persons in pre-modern, nonwestern societies.

Goffman extended the Durkheimian approach to personhood by investigating the interactional process by which the specific cultural representation of a person is socially produced. Indeed, Goffman’s interaction order is primarily concerned with the collaborative manufacturing of persons. As Cahill (1998, p. 139) points out, for Goffman “the public person is not made in the image of a unique self; rather, an interpretive picture of a unique self is made in the image of the public person.” This is a distinction that is not always appreciated by social psychologists, and it is a limitation that is due in part to Goffman’s own inconsistent use of the terms individual, person, self and identity. Nevertheless, Goffman (1959, p. 253) was clear in asserting that the corporal body is simply a peg on which the socially manufactured person is to be temporarily hung. Social identities, on the other hand, are the means of categorizing persons in terms considered appropriate and consistent with the shared assumption of what is required to be a person.

Still, it would be a mistake to take Goffman’s peg analogy too far. It is one thing to recognize historical and cultural variation in the production of persons, and quite another to conclude that there are no essential qualities of human persons beyond physiology. As Smith (2010, pp. 277–314) has recently stressed with regard to the question of personhood , we must be careful not to confuse “how things happen” from “what things are.” The cultural beliefs about what constitutes a person, and the means of socially manufacturing the social category of person, are certainly a core concern of sociological social psychology, but the socially constructed category of person is not independent of the objective nature of what it means to be human. For Smith (2010, pp. 25–89), therefore, a conceptualization of personhood should also recognize that human capacities for consciousness and self-reflection are emergent from physical bodies that serve as the center of subjective experience and the hub of a coherent structure. Persons are inescapably social and subject to the power of social forces, but as human persons we are also agents who are (at least partly) responsible for causing our own actions. In this sense, a person is both a socially constructed category and an acting organism with uniquely human capacities. Self and identity are two uniquely human capacities that emerge from persons.

For social psychologists working in the tradition of symbolic interactionism , self refers to the unique potential of persons to engage in symbolic interaction, to take the perspective of other, and to produce a self-conscious object—an object to itself. Following Mead (1934, p. 140), the self is evident in the process of “responding to oneself as another responds to it, taking part in one’s own conversation with others, being aware of what one is saying and using that awareness of what one is saying to determine what one is going to say thereafter.” Identity, on the other hand, is a product, or outcome of the self-society relationship. Identities are the socially constructed categories that are used to establish meaningful understandings of persons—both self and other. As such, identities are not universal, but reflect particular historical and situational circumstances (Wiley 1994, pp. 1–3). Sociologists have employed a variety of different terms when referencing identity categories (e.g., label, role, status), have identified a range of different types of identities (e.g., personal, dispositional, situational, institutional), and have invented useful typologies for making conceptual distinctions among these categories (e.g., MacKinnon and Heise 2010). All of these schemes, however, share the core idea that identities are socially contingent constructions that depend upon the self-processes of persons engaged in symbolic interaction (e.g., Howard 2000). Linking personhood to the concepts of self and identity is particularly important for the study of inequality for two reasons. First, it reminds us that self and identity are ultimately embodied. Persons are biological systems, integrated into the natural world and subject to the laws of nature. It is the practical action of human bodies that gives rise to self and identity, and it is the practical activities of physical survival—finding ways to eat, shelter, procreate, and avoid harm—that develop into social structures of inequality. Thus, the material and corporeal reality of persons in community with one another is basic to both the emergence of self and the emergence of inequality. When material resources necessary for survival are unavailable, or are unevenly distributed—scarce for some and hoarded or controlled by others—there are physical and psychological consequences for actual persons.

The second reason for developing a conception of personhood is that it connects self and identity to the problem of human dignity. This is evident in both Goffman’s strong constructionist definition of persons, as well as Smith’s critical realist position. Take, for instance, the following statement from Goffman (1959, p. 13):

(W)hen an individual projects a definition of the situation and thereby makes an implicit claim to be a person of a particular kind, he automatically exerts a moral demand upon others, obliging them to value and treat him in the manner that persons of his kind have a right to expect.

Here we see Goffman’s assertion that establishing the dignity of persons is both a negotiated outcome of social interaction, and a fundamental prerequisite of interaction itself. Under ideal conditions, the joint production of personhood is mutually supportive and balanced. However, control of the means of person production is rarely shared equally among participants, and certain structural arrangements—prisons, mental hospitals, slavery, patriarchy—make it difficult for some individuals to claim personhood and experience dignity. For his part, Goffman was not concerned with the question of whether prisoners, mental patients, slaves, and women, are in fact persons deserving of dignity. Like many sociologists, he avoided questions of ontology—even though such assumptions are implicit in his work.

Smith (2010, p. 435), however, believes it is important to directly address the issue because:

Dignity inheres in the emergent constitution of human personhood, including in the personhood of people who are ignorant of or deny its reality. It is inalienable. It cannot be thought or wished away. It cannot be sold or negated by legal judgment. Dignity exists as a real and ineliminable dimension of human persons, just as liquidity does of water and growth and reproduction do of living organisms.

For Smith, dignity is not a social construction or cultural invention. It is an objective, ontologically real attribute of all human persons. When humans treat others as though they are things and refuse to recognize inherent personhood, dignity is denied. The assumption, therefore, is that personhood is not a matter of degree or a matter of capacity; those who are illiterate, have less ability to reason, or have limitations of sight, hearing or mobility, are still persons and still have dignity (for a related argument see Hodson 2001, pp. 3–21).

This particular conceptualization of persons as dignified, inviolable, and equal has an elective affinity with basic principles of democracy and ideal democratic institutions (Callero 2003; Habermas 1987; Wiley 1994, p. 11). Voting, citizenship, human rights of privacy, life, and liberty begin with an assumption of persons sui generis. Similarly, a deliberative democracy requires symbolic interaction, reason, and empathy—characteristic features of a pragmatic self (Talisse 2005). In contrast, reductionist theories of self, characteristic of postmodernism and much of psychology, struggle to justify democracy and equality on moral grounds. Mead and Dewey voiced a similar critique of reductionist theories from an earlier era and argued “German Idealism served to legitimate monarchy, aristocracy, and serfdom” (Wiley 1994, p. 227). Early American pragmatists battled against social Darwinists, eugenicists, and other biological determinists who used science and specious theories of personhood to justify racial segregation, the subordination of women, and the medical exploitation of physically disabled persons. Clarifying the relationship between person, self and identity is therefore a necessary step in understanding inequality.

Identity as a Product of Inequality

The whole (society) is prior to the part (individual), not the part to the whole; and the part is explained in terms of the whole, not the whole in terms of the part or parts. (George Herbert Mead 1934, pp. 7–8)

For Mead, the self is explained in terms of society, suggesting that the systemic patterns of social inequality observable in society can be used to explain identity. Common sense alone gives credence to this assertion. We experience inequality in and through categories, labels, and classifications that define individuals, groups and collectives, where rewards and resources are predictably and unevenly distributed. Identity categories such as gender, race, and class matter to sociologists because they are profoundly and unmistakably linked to social structures where the ownership and control of labor, land, machines, financial capital, communication media, and other material and symbolic resources are systematically stratified.Footnote 1

Table 12.1 presents an organizing framework for reviewing research on identity and social inequality at three different levels of analysis: culture, interaction and person.Footnote 2 While these three levels present a clear analytical distinction, they are not independent of each other. Persons are defined by identity categories , particular definitions of self and other are negotiated at the level of interaction, and a generalized meaning for an identity category is shared at the level of culture. On the other hand, the generalized cultural meaning associated with an identity category is reproduced and altered by persons engaged in face-to-face interaction.

Table 12.1 Identity meanings and forms of inequality at three levels of analysis
Full size table

Here I use the term identity to refer to all categories of social location that may be employed in the definition of self and other. The intention is to capture the full range of sociological approaches to identity and all possible ways in which identity expresses inequality. This includes traditional sociological categories associated with structural locations, group affiliations, and types of social relationships (e.g., class position, occupation, nationality, geography, religion, marital status, sexuality), categories linked to physical attributes (e.g., age, race, disability, sex, size), category labels that are more localized and defined by a person’s biography, skill, ability or individual characteristics (e.g., test scores, athletic prowess, criminal history), as well as dispositional categories referencing personality, stereotypes, or slang classifications (e.g., angry, extrovert, motherly, wimp). To the extent that a category label can be used to identify, classify, or indicate a person as a certain type, it has the potential to be used in the production and reproduction of inequality. As Table 12.1 indicates, the manner in which identity categories are linked to inequality takes a different form depending on the level of analysis. For this reason, the review of theory and research that follows is organized in terms of culture, interaction, and person.

The Level of Culture

Identities have meaning, and the meaning of any identity can be recognized, accepted, and documented at several different levels of analysis. At the cultural level, where language and common social practices are sustained, the meaning of an identity is abstract and generalized. What it means to be a man or woman, black or white, gay or straight, rich or poor, is part of a common lexicon, a cultural tool kit, and a recognizable status hierarchy (Heise 2007; MacKinnon and Heise 2010). Identity meanings at this level of generality have more or less settled over time, remain relatively durable, and are difficult to transform. Yet, the boundaries between identity meanings are never completely permanent or inflexible; they are usually intersecting, sometimes contested, and always open to the possibility of change over time.

The culturally settled meanings associated with an identity category can vary from a core set of generalized references to a highly specific and particular collection of relatively unique responses. But at both ends of this spectrum we can find theory and empirical evidence indicating that meanings are organized along a limited number of dimensions. Osgood et al. (1975), for example, have found three dominant dimensions of meaning (evaluation, potency, and activity) across more than twenty distinct cultural communities. Evaluation refers to meanings that offer an appraisal or assessment of an object, person, or event, as being either positive or negative. Potency captures meanings that refer to variation in strength or power, while Activity references meanings associated with levels of energy or relative liveliness.

Under Affect Control Theory (Robinson and Smith-Lovin 2006; Heise 2007; MacKinnon and Heise 2010; see also Foy et al., this volume), the same three dimensions of meaning have been used to produce “cultural dictionaries” for a range of identities. According to the theory, the culturally shared affective meanings associated with identity categories are experienced as sentiments that serve as a generalized standard for assessing the more particular affective meanings of situated interaction. Thus, Langford and MacKinnon (2000) found that meanings associated with male and female identities seem to reflect two different status hierarchies, such that men are seen as more productive and powerful (higher on potency), and women are viewed as more caring and positive (higher on evaluation), while the activity dimension was found to be less different. In a broader sociological context this is not a surprising result given that evaluation and potency are analogous to value and power ; two concepts that have a long and diverse history in the study of inequality (e.g., Castells 2010; Newman 2007; Sennett 2003; Thye 2000; Weber 1946). Indeed, it is safe to say that in the case of social inequality, the two most important meanings for any identity are those associated with value and power. For this reason, Table 12.1 limits the meanings associated with an identity to these two dimensions.

At the cultural level, the value of an identity is typically associated with differing levels of respect, prestige and honor. We see this, for example, in the unequal value historically associated with the categories of “heterosexual” and “homosexual”. Persons defined as “gay” or “lesbian” often receive less respect than persons categorized as “straight”. The power of an identity, on the other hand, is defined by different levels of authority and control. In the United States, for example, the racial category of white has historically been associated with more power than the racial category of “black.” In general, to be defined as white means greater authority and control in comparison to persons defined as black. While value and power are highly correlated, these two dimensions of meaning are sufficiently distinct to warrant separate analysis. For example, we can think of identities defined by occupational categories with different levels of prestige. Thus, when considered in the abstract (i.e., action and context is unspecified), the identities of teacher and farmer receive relatively high ratings in surveys of occupational prestige and are consistently ranked above the identities of banker and politician in this regard (e.g., Nakao and Treas 1994). On the other hand, banker and politician are under most conditions viewed as being more powerful and controlling.

Here it is important to emphasize the distinction between power as a dimension of meaning, and power as the actual accomplishment of dominance and control. (For a review of different sociological understandings of power, see Thye and Kalkhoff in this volume). As Table 12.1 suggests, the meaning of an identity (powerful to powerless), has implications for the actual control of valued resources; persons who hold powerful identities will be more likely to control and dominate. But an identity category has a degree of independence separate from any particular person. Institutional roles such as President, General, or CEO, are recognized as “existing” in an organization even when the position is vacant. Individual occupants of the position might display different styles of control and may even redefine the meaning of the position, but this does not erase the fact that there are clear institutional and cultural limits as to how the position is defined. The same is also true for less formal identity categories such as gender or race, where the meaning of an identity category can limit or enhance opportunities for exercising power. In this sense, identities are resources, or tools, for doing inequality. Power is not completely symbolic, but power has a symbolic component associated with identity categories. For this reason, it is important to recognize power as both a dimension of meaning associated with an identity category, as well as the accomplishment of control and domination.

Where do the generalized and relatively durable cultural meanings of power and value originate? For most identity categories the answer to this question is buried under layers of history, but Tilly (2005) offers a plausible theory. He suggests that when two different groups of people encounter each other for the first time, they each create labels to mark and identify the other and establish symbolic group boundaries. But these identity categories are simply indicators of difference and do not necessarily cause inequality. Inequality is produced when repeated transactions across group boundaries regularly advantage one side and at the same time reaffirm the identity boundary. In transactions of exploitation, for example, members of one side of the boundary enlist effort from members of the other side to secure a scarce resource, but fail to deliver in return the full value of the other group’s effort. The exploiters then use part of the surplus value to produce symbolic markers and material conditions that reinforce the original boundary. Think for example of the boundary between workers and owners of a factory. With the wealth generated by the factory, owners and managers can invest in expensive suits, office buildings, security guards, media messaging, and political relationships that serve to highlight the difference between the two groups, and at the same time mark one as more valued and powerful than the other.

Within social psychology, status construction theory has produced a more localized and interactional theory of how nominal categorical differences between people are converted into differences of status (value). Using a cumulative program of empirical investigations (mostly laboratory experiments), researchers have demonstrated that the structural conditions under which people encounter one another can shape beliefs associated with the value of different identity categories (Ridgeway and Nakagawa, this volume). In addition, when locally produced beliefs are carried into other situations, they may be diffused and become widely held cultural beliefs (Ridgeway and Erickson 2000). When this happens, status beliefs serve to reproduce inequality through processes that affirm the relative value of different identity categories. The end result is the establishment of status hierarchies and a form of cultural capital linked to identity (Johnson et al. 2006).

To be sure, the development of cultural meaning systems is a dynamic process that is framed by the higher-level constraints of macro-level social structures associated with the institutions and practices of a political economy, including legacies of colonial rule, forced enslavement, patriarchy, and the accumulation and concentration of capital by multinational corporations. But within these larger social arrangements the value and power of identity categories is learned, shared, defended, challenged, and altered in face-to-face encounters among people in identifiable social settings. This is identity at the level of interaction.

The Level of Interaction

At the level of interaction, the value of an identity is experienced in terms of the relative amount of deference granted to a particular actor in a face-to-face encounter. Power on the other hand is experienced in the relative ability of one actor to exert dominance over others in specific situations. Both identity meanings contribute to the establishment of an asymmetrical relationship where the holder of the less powerful and less valued identity is more likely to be ignored, intimidated, dismissed, and at the same time will be expected to produce an interactive demeanor that is obsequious, reverential, and submissive.

Specific examples of inequality at the level of interaction are well documented and diverse (Anderson and Snow 2001; Link and Phelan 2001; Major and O’Brien 2005). Derber (2000) shows, for instance, that the amount of attention someone receives in a social setting is a reflection of the relative power and value of their situated identity. Similarly, Snow and Anderson (1993) demonstrate in their study of the homeless that life on the street is associated with a type of social invisibility, where recognition is often in the form of negative attention from a disgusted or angry passerby. Identities with less value and power are also associated with more queuing and waiting (Schwartz 1975), greater emotional labor (Hochschild 1983), an increased likelihood of interruption (Smith-Lovin and Brody 1989), as well as intimidation and threats of violence (Anderson 1999).

The markers of identity, and the cultural meanings associated with identity categories , serve as an initial framework for locating and negotiating the relative power and value of self and other, but the context of the social encounter narrows the field of likely interpretive outcomes. Think of, for example, the different cultural meanings associated with the identity categories of heart surgeon and prison guard. At the generalized level of cultural meanings, the identity of a prison guard is associated with less value and power than the occupational identity of a heart surgeon. But if a heart surgeon is serving time inside a state penitentiary, the prison guard is more likely to enact dominance and achieve deference from the inmate, even if the heart surgeon identity is made salient. Here we can see that the resources associated with a particular setting are key to establishing how value and power are performed. And when the setting is lodged within a more encompassing institutional context, there may be more constraints on the culturally shared meaning of an identity (e.g., Gubrion 1997).

To be sure, the interaction level includes more than the identification of the relative value and power of identity categories. It is also the level at which identity categories are created, negotiated and reproduced. Goffman’s dramaturgical approach, for example, draws our attention to an interaction order that includes the rules and normative procedures that actors use to create authentic identity impressions—both valued and devalued, powerful and powerless. In other words, while the interaction level is the place where the meanings of identity are experienced, it is also where we find the “rules and procedural forms” that are used to sustain value and power in the doing of deference and dominance (Schwalbe and Shay in this volume).

The Person Level

The person level identified in Table 12.1 encapsulates the universal symbolic capacities for reflexivity and role taking that are uniquely human. Analysis at this level focuses on the more or less durable identities that are carried by persons across situations and help structure the meanings, motivations, and dispositions of individual actors. Here we find the identity categories associated with a person’s social biography, affiliations with different groups and institutions, and the unique intersections of categories such as race, class, gender, and sexuality (Howard and Renfrow in this volume). It is at the level of the person that the relatively enduring consequences of inequality are both inscribed and scarred onto actual bodies and minds with real, visible consequences for an individual’s physical and mental health (Thoits 2010). Thus, persons from lower class positions are generally sicker and stay sick longer than people in higher-class positions. And at all levels of class, African-Americans will on average experience worse health than whites, and women will experience higher rates of depression than men (McLeod et al., this volume).

In terms of identity meanings , most research at the person level examines the relative value of an identity as represented in the concept of self-esteem. Here self-esteem is typically defined as an overall assessment of worth, merit, or value—a continuum that ranges from positive valuations of the self to negative valuations of the self. A number of studies have found evidence that devalued identities are associated lower self-esteem, but the magnitude of the correlations are generally weak and findings are not entirely consistent (Wells 2001). While the lion’s share of this research focuses on a generalized self-evaluation (global self-esteem), researchers also recognize that self-esteem can be specific to particular dimensions or aspects of the self (specific self-esteem). For example, in one especially influential study, researchers found that global self-esteem had a relatively stronger relationship with psychological well-being, but specific self-esteem was a better predictor of actual behavior. Moreover, this same study also found that the relationship between specific “academic self-esteem” and global self-esteem was a function of how highly academic performance was personally valued (Rosenberg et al. 1995).

Still, it would be inaccurate to claim that a simple linear relationship exists between social inequality and self-esteem. Instead, reviews of this literature have concluded that the link between self-esteem and inequality is contingent on dimensions of measurement (Wells 2001), context (Crocker and Major 1989), life stage (Orth et al. 2010), as well as variation in historical and cultural discourse (Hewitt 2009). Given that the self is nested within both an interaction system and a cultural system, the fact that we find evidence supporting a conditional relationship between person-specific self-esteem and macro indicators of inequality is not surprising. And given that self-esteem is concerned with only one dimension of meaning, it is not surprising that the magnitudes of these same correlations are relatively weak.

The relative power of an identity at the person level is most commonly associated with the concept of self-efficacy . As originally formulated by Bandura (1977), self-efficacy refers to beliefs about one’s ability to execute a particular course of action. Put more generally, self-efficacy is an assessment of one’s level of competence, effectiveness, control, and agency—traits that are synonymous with powerful identities. Like self-esteem , self-efficacy is typically conceptualized as an overall personal assessment, generalized trait, or disposition, but researchers examine domain specific assessments of self-efficacy as well (Gecas 1989; Schunk and Pajares 2009). Also similar to the literature on self-esteem, research on self-efficacy has concluded that individual variation is a function of social context, institutional setting, as well as background identities such as gender and ethnicity (Usher and Pajares 2008). In other words, self-efficacy is not simply a personality trait, but rather a type of self-assessment that is very much tied to one’s social location. One study (Boardman and Robert 2000), for example, found that low levels of self-efficacy were associated with high levels of neighborhood unemployment even after controlling for individual level measures of socio-economic status.

We can say that social inequality has been internalized when individual assessments of value (self-esteem) and power (self-efficacy) become part of a person’s self-definition. Understanding the structural conditions under which this occurs has been the focus of a tradition of research categorized under the rubric of social structure and personality (McLeod and Lively 2003). Although not explicitly concerned with the self or identity meanings , this body of research has produced persuasive evidence that objective social circumstances can transfer conditions of inequality to the person with detrimental consequences for self and identity.

The most compelling research in this regard has examined the consequences of managerial control, routinization of labor, and limited autonomy at the work site. Kohn and Schooler (1969, 1983) and their colleagues (Kohn and Slomczynski 1990), for example, have shown how value and power in the workplace (one’s relationship to the means of production) has enduring consequences for particular types of psychological functioning (e.g., self-confidence and intellectual flexibility). This research tradition provides strong evidence that objective conditions of inequality in the workplace not only have negative effects on the self, but also have emotional repercussions that harm the dynamics of family interaction (DiTomaso and Parks-Yancy, this volume; Menaghan 1991). In addition, these adjustments and alterations of identity toward a less valued and less efficacious self may be transmitted across generations as children “inherit” less powerful and valuable forms of cultural capital from their parents (Lareau 2003). Additional research, more specifically focused on the development of self-definitions, also finds evidence that autonomy in the workplace is associated with higher esteem and higher efficacy for individual workers (Gecas and Seff 1989; Staples et al. 1984).

Still, we need to keep in mind that value and power have a degree of independence, and under certain conditions may be unrelated or may be shaped by different identities. For example, some research has found that while there appears to be little difference between blacks and whites in the U.S. in terms of self-esteem , blacks tend to report lower levels of self-efficacy . Hughes and Demo’s (1989) analyses suggest that although esteem and efficacy are positively correlated, they are the outcomes of different social processes. In a national survey of African Americans they found that religion, family, and friends were the most important predictors of self-esteem, while self-efficacy was more highly dependent on age, gender and socioeconomic status.

In sum, research on the relationship between identity and inequality can be understood as an examination of value and power at three distinct but interdependent levels of analysis. At the level of culture, inequality is reflected in differing amounts of value and power associated with generalized identity categories . This is most evident in patterns of resource distribution defined by status hierarchies that are often widely accepted as natural or just. At the level of interaction, inequality is experienced in face-to-face relationships where identity signifiers initiate asymmetrical relations of deference and dominance, resulting in patterns of positive and negative attention, fear and intimidation, and an interactive demeanor that is either poised or insecure. At the level of the person, value and power are internalized in the form of self-esteem and self-efficacy. These effects accumulate over the life course and are often associated with poor physical and psychological well-being.

Thus far I have primarily focused on identity as a product of inequality and as the consequence of a larger social system. But the self is also a social force and an instrument in the production and alteration of inequality. This is the self as a social process or mechanism.

The Self as a Social Process

A common assertion among so-called postmodern theorists is that the self is merely an effect or product of social practices, and that it has no objective reality outside of historically specific systems of discourse (e.g., Denzin 1992; Foucault 1988; Gergen 1991). In contrast, sociologists grounded in the tradition of American pragmatism and symbolic interactionism argue that the self emerges from cognitive capacities and social relationships to become a real, objective, and causal force in society. In this way, the self is said to be a subject and an object, a social product and a social force (Callero 2003; Owens 2003; Rosenberg 1979; Weigart and Gecas 2003). The distinction is important for the study of social inequality. When the agentic self is dismissed as a linguistic epiphenomenon, or as an artifact of discourse, it is difficult to theorize individual and collective resistance to systems of inequality (Best 1994; Collins 1997). As noted earlier, identities may be historically, culturally and situationally specific, but self-reflection is universal. Understood in this way we can appreciate the self as a social mechanism (Gross 2009; Smith 2010), or a set of relatively fixed processes—mostly unobservable—that convert social interaction into higher order patterns of equality and inequality.

Recognizing the self as a social process guards against the temptation of reducing the self to a simple predictor or outcome variable. This is increasingly the direction of some scholars (mostly psychologists), who have contributed to a catalog of self-dimensions that share much with personality theory (e.g., Swann et al. 2007). When treated as a variable, the self loses part of its complexity, explanatory power, and sociological significance. As Smith (2010, p. 289) notes, “Variables do not make things happen in the world. Human persons do.” And the self is the primary mechanism by which persons make things happen—both good and bad. This is not to say that the self is at all times and places operating as a social force, or that the self is the only social process in operation. Rather, under particular circumstances the self will operate as an instrument in the production of certain observable outcomes. Social inequality is one such outcome, and for this reason a complete explanation of the production, reproduction, and alteration of social inequality requires a theory of the self as a social process.

Table 12.2 presents a basic framework for examining the self as a social process at three levels of analysis. Here the focus is limited to either reproduction or resistance in the production of inequality. This is not a hard boundary; theories focusing on reproduction do not deny the possibility of resistance, and processes that explain resistance assume a high degree of social reproduction. Nevertheless, these same self-processes are distinct enough to justify a categorical separation.

Table 12.2 Reproduction and resistance at three levels of analysis
Full size table

Generally speaking, the self contributes to reproduction when stocks of habits (social practices, modes of response) go unchallenged as solutions for prior problematic situations. These solution patterns continue in the form of tradition, momentum, and unchallenged assumptions until exposed by new problematic encounters (Joas 1996, p. 126–144). Resistance is the visible response of individuals and groups struggling to resolve perceived problems of inequality. This can be evident, for example, in the street demonstrations of a political revolution or a brief objection and apology sequence between two actors.

To say that the self is a process or mechanism through which problem situations are encountered, interpreted, and collectively resolved, does not mean that inequality is essentially symbolic or that solutions to problematic events are always just and equitable. We must not forget that reproduction of inequality is often accomplished with brute force, torture, killing and imprisonment, and is frequently associated with intimidation, threats, and fear. Similarly, collective solutions are not necessarily just, and are often enabled by the control of material resources in the form of land, money, and weapons. As a consequence, the resolution of a problem will regularly favor those with more power. For example, the exploitation of workers in an office or factory might become problematic and lead to the formation of a labor union and collective bargaining. A new contract may improve benefits and working conditions—offering a temporary solution—but in the end, the managers and owners are still in control and continue to operate with more power.

Individuals, groups and institutions that are advantaged by the unequal distribution of value and power of particular identities have an interest in regulating and reinforcing systems of inequality, while those who are disadvantaged by the process have an interest in altering or erasing the meanings associated with the identity. Consistent with the organizing framework presented in Table 12.2, I will review research on reproduction and resistance at three different levels of analysis.

The Level of Culture

Reproduction at the level of culture occurs when the value and power of an identity category is processed by the self as natural, legitimate, or just. When inequality is either invisible, passed off as acceptable, or dismissed as an inconsequential difference, there is no pressure on persons to examine the meaning of an established identity category. Thus, when women and ethnic minorities are portrayed in film, television, and print as subservient and weak, and when these images go unchallenged, inequality is reproduced at the level of culture (Bullock et al. 2001). As Barker (2005, p. 503) notes, “Issues of cultural representation are political because intrinsically they are bound up with questions of power through the inevitable process of selection and organization that are part of the representational process. The power of representation lies in its enabling some kinds of knowledge to exist while excluding other ways of seeing.”

Lamont’s (2000) research reminds us, however, that reproduction is not a simple process of the powerful demeaning the powerless. In her study of workingmen she found that devalued meanings for identity categories are often reproduced within boundaries of the same oppressed class, and between boundaries of race, in a manner that prevents class solidarity. As a consequence “Workers often judge members of other groups to be deficient in respect to the criteria they value most” (p. 241). This discovery complements tests of social identity theory that find that actors experience positive self-esteem by conforming to their own group identity, while denigrating outsiders (cf., Scheepers et al. 2009). To the extent that this self-process leads to intergroup stereotyping, prejudice, and discrimination, it can reproduce already existing identity meanings associated with group membership (Riesch 2010). As a consequence, the cultural hegemony of various identity categories is not experienced as problematic. Instead, the injurious meanings and negative cultural representations associated with identity categories go unchallenged. The end result is the reproduction of inequality within the cultural system. Indeed, one could argue that under conditions of reproduction at this level, the self as a mechanism is not immediately engaged.

However, resistance at the level of culture is always on the horizon because reproduction cannot be maintained indefinitely under conditions of inequality. Persons who are categorized under a relatively powerless or devalued identity will eventually object to the inherent indignity and seek to alter their circumstance. Following Gramsci’s (1971) classic distinction between “war of position” and “war of maneuver,” some acts of resistance may be focused on altering the meanings of the identity categories (position), while others may be focused on altering the distribution of resources and the coercive powers that serve to legitimate the meanings (maneuver). But as Nagel’s (1995) analysis of Native American identity shows, the material and symbolic are intertwined. Collective acts of resistance are necessary to change the meanings of generalized identity categories, and collective identities are central to the mobilization of oppositional groups and oppositional social movements (Melucci 1996; Polleta and Jasper 2001). For resistance to succeed at this level, a collective identity must be politicized (Klandermans and de Weerd 2000; Snow and Owens, this volume; Taylor and Whittier 1992), and this means that the self must begin to process new identity meanings .

An evolving and particularly promising line of research on resistance to inequality can be found in the examination of symbolic boundaries as applied to identity categories (Lamont 1992; Lamont and Molnár 2002). In the case of collective identity, boundary work refers to the strategies and practices used to contest the cultural meanings of an identity (Owens et al. 2010). This includes moves to deconstruct boundaries as well as attempts to reinforce boundaries so as to mobilize on the basis of identity. For example, by redefining the meaning of “queer,” activists have sought to construct a more inclusive social movement of people with a wide range of sexualities. In this instance, resistance is not simply a political struggle for equal rights under the law, but also a cultural struggle to establish new meanings and new identities (Bernstein 2005, p. 26; Rimmerman 2002). On the other hand, some lesbian feminist communities have sought to reinforce oppositional gender categories and establish a privileged idealization of female as a way to challenge misogyny embedded in traditional identity meanings . Such a strategy can serve to enhance collective consciousness, but as Taylor and Whittier (1999, p. 178) conclude, it can also promote “a kind of cultural endogamy, that, paradoxically, erects boundaries within the challenging group, dividing it on the basis of race, class, age religion, ethnicity, and other factors,” a complication that may produce unintended consequences. Gamson (1998), for example, found that the representation of gay people in television served to legitimate middle-class gay people while simultaneously invalidating meanings associated with poor and working-class gays.Footnote 3

The Level of Interaction

Individuals cannot control the meanings of value and power in isolation; the value and power of an identity is an interactional accomplishment (e.g., Speer 2012). Some interaction work will reproduce inequality and some will contribute to resistance and the revision of identity meanings—both personally and culturally. In both instances we find evidence of the self operating as a social process or mechanism.

Reproduction of identity inequality at the level of interaction has received the lion’s share of attention. Thus, Goffman’s catalog of self-based strategies associated with the concept of identity work is principally focused on methods that reproduce the status quo. When actors work to avoid confrontation, seek validation for a devalued identity status, and participate in an interaction exchange that preserves the interaction order, they are indirectly reproducing social inequality. Here we find a self that surrenders to the dominant order and seeks to conceal oppositional or discrepant identity categories so as to avoid confrontation. This is evident, for example, when an actor strategically controls markers of a devalued identity. Goffman (1963) highlights two such self-processes in his analysis of stigma. Passing refers to the tactic of concealing or obliterating stigma signs, as when a member of a devalued ethnic group changes his or her last name or speaks with an intentional accent. Covering, on the other hand, refers to tactics used to keep obvious stigmatized identities from looming large, as when a blind person wears dark glasses for the sake of the sighted. Both strategies suggest a self that is aware of an interaction order, and is willing to cede ground to those with more value and power so as to avoid additional stigmatization (see Link et al., this volume).

To achieve a creditable self, to avoid being discredited, actors must not violate the structure of the social encounter. “The key factor in this structure is the maintenance of a single definition of the situation, this definition having to be expressed, and this expression sustained in the face of a multitude of potential disruptions” (Goffman 1959, p. 254). Preserving the structural requirements of the interaction order is often a collective priority. Indeed, Goffman argues that a temporary loss of face and the embarrassment that it engenders has a social function wherein identity is sacrificed for the moment in the name of the larger principles of system reproduction: “Social structure gains elasticity; the individual merely loses composure” (Goffman 1967, p. 112).

Cultural meanings for an identity category establish a set of generalized expectations that initiate identity work at the level of interaction. How these expectations shape behavior has been the concern of several perspectives, including labeling theory (e.g., Becker 1963), status construction theory (e.g., Ridgeway 2006), role theory (e.g., Biddle 1986), and several versions of symbolic interactionism (e.g., Stryker 1980). A particularly relevant program of research in this latter tradition is Affect Control Theory (ACT) . Affect Control Theory provides a formal model of the self as a mechanism at the level of interaction. This theory is principally concerned with explaining (1) how meanings at the cultural level are used to produce predictable patterns of interaction; and (2) how innovative solutions are produced when deviations from cultural meanings occur at the level of interaction (Robinson and Smith-Lovin 2006). ACT begins with the assumption that sentiments (generalized affective responses) attached to identity categories are used in the generation of transient impressions (situated meanings). Actors are motivated to maintain consistency between sentiments and impressions. If a discrepancy between sentiments and impressions occurs (deflections), the self will actively work to regain consistency of meanings, and in the event that deflections become too large, and attempts at realignment or readjustment are unsuccessful (i.e., the control system fails), conflict emerges and actors will attempt to redefine the situation, or in more extreme cases seek out different settings in an attempt to restore sentiments.

Compare, for example, the generalized meanings that are attached to the identities of “employer” and “employee”. In the context of a specific workplace, these sentiments should generate transient impressions consistent with the expectation that an employer will be more highly valued and should have more power than the employee. At the work site, for instance, it might be assumed that employers will “hire” and “direct”, while employees will “work” and “follow”, but in another context, say at the employee’s home, a different set of transient impressions would likely prevail. To the extent that sentiments and impressions of employer and employee match, situated behavior can be expected to reproduce the asymmetry of the relationship. When employers or employees deviate from these contextualized expectations, however, readjustments will occur. This might be a minor realignment of personal impressions (“overly demanding employer” or “insubordinate employee”), but it could also lead to a more collective readjustment that results in a structural alteration of the workplace or the generation of new generalized sentiments at the level of culture.

In general terms, ACT simply predicts expected behavior, and in this sense it is not an innovative approach to the study of inequality. On the other hand, it does offer a parsimonious cybernetic model that has the potential to account for both the reproduction of value and power, as well as the circumstances that may lead to resistance and change when these meanings result in repeated deflections. Promising lines of research in this direction have explored ACT in relation to gay-lesbian identities (Smith-Lovin and Douglass 1992), social movement solidarity (Britt and Heise 2000), and equity in marital relationships (Lively et al. 2010). At this point, however, most empirical research has focused on the prediction of emotional reactions and the relationship between identities and emotions; an explicit focus on the relationship between social inequality at the levels of both culture and interaction is mostly absent.

When inequalities at the level of interaction are opposed, resistance can be either individual or collective; in both instances the self is activated as a social process. Disrupting entrenched systems of inequality is difficult, especially if a single person initiates change. A lone objection to the presumed definition of the situation may temporarily disrupt interaction, or succeed in distancing an actor from a negatively valued identity, but individual strategies do not alter the interaction order. In fact, evidence suggests that identity refusal (negotiating a Not-Me) may actually serve to reinforce hegemonic definitions through an implicit acknowledgement and acceptance of the meanings associated with a less valued and less powerful identity (Killian and Johnson 2006; Pyke and Johnson 2003). Schwalbe et al. (2000) call this interactive process defensive othering because it involves accepting the devalued identity meanings in others, while at the same time working to distance the same meanings from one’s own identity.

For the most part, individual resistance to an identity meaning cannot succeed without the support of a team of allies. When collective resistance does occur at the level of interaction, it typically involves cooperation among persons who are similarly situated and who experience common deprivations and indignities. An extensive body of ethnographic research on exploitation in the workplace has documented the interpersonal strategies and tactics workers use to oppose and defy management control. Although this body of work is not explicitly concerned with the self as a social process or social force, the intervening operation of the self can be inferred. Hodson (2001), for example, shows that when basic human dignity is denied in the workplace through abuse and manipulation by managers, workers actively and cooperatively engage in four types of defiance: resistance, citizenship, the creation of independent meaning systems, and the development of social relations at work. As used by Hodson, resistance encompasses destructive acts, sabotage, and theft, as well as foot-dragging and the withdrawal of cooperation. Citizenship refers to actions that are intended to enhance self-worth despite the indignities of the immediate context. These are enterprising activities that serve as alternative sources of pride—doing a job well, perfecting a skill – and are largely independent of the formal demands of power. Autonomous meaning systems also emerge among workers who seek value, purpose, and control within an overarching system of supervision and management. This can involve personal rituals or the display of symbols that represent life outside of the workplace. Examples include engaging in games during lunch break, joking with coworkers, or personalizing a workspace by exhibiting family photos, sport memorabilia or hobby emblems. Finally, fundamental to all of these strategies is coworker relations—where collective identity is formed separate from management. Here informal ties and patterns of mutual assistance can develop into unofficial workplace roles and identities. When group values and leadership positions emerge from coworker relations, a united opposition to oppressive conditions is possible.

Other ethnographic studies have documented similar acts of collective resistance in schools (e.g., MacLeod 2009; Willis 1977) and neighborhoods (e.g., Anderson 1999; Harding 2010). In all of these accounts we find rich descriptions of situated resistance that often occurs behind the backs of the powerful, inside commanding institutional structures, and against the interests of state and market forces. This is a type of infrapolitics where opposition is often subtle and intentionally obscure (Scott 1990). As a result, evidence of structural change occurring as a consequence of these acts is difficult to locate. To be effective, strategic disruptions of the interaction order must be public (Schwalbe and Shay, this volume). In this way, refusal has an audience, solidarity has the potential to germinate, and sympathetic supporters gain confidence. When successful, disruptions that were once defined as personal affronts are redefined as threats to an established system of power. This is not power in the narrow sense of politics, but power embedded in structures of meaning. Altering political systems through revolutionary action can occur quickly, but changes to an interaction order are usually gradual and more difficult to achieve.

One reason it is difficult to change an interaction order is that every encounter engages multiple identities that interact in complex ways. For example, workplace resistance involves more than one’s work identity; ethnicity overlaps with religion; neighborhoods intersect with class; and gender lies behind all of these. Ridgeway and Smith-Lovin (1999, p. 193) make this point in their analysis of the gender system when they note: “the interactional conduct of gender is always enmeshed in other identities and activities. It cannot be observed in a pure, unentangled form. Gender is a background identity that modifies other identities that are often more salient in the setting than it is.” This suggests that a more complete understanding of the operation of the self at the level of interaction requires an examination of the self as a process or mechanism in its own right.

The Person Level

At the person level, the focus is on the internal structures and processes of identity construction. These are the cognitive processes and mechanisms associated with symbolic interaction. In the case of social inequality, most theory and research at this level tends to emphasize processes that contribute to reproduction. Considerably less attention has been paid to self-based processes associated with social change and struggles against inequality. I will first review models of reproduction at the person level before discussing corresponding explanations of resistance .

A common understanding of the self is that it is a reflexive process of adjustment and regulation. We see this, for example in Mead’s description of the I and Me as separate parts or phases of a unified process. By taking the attitudes of others, an organized social Me emerges, and the self reacts to this phase of the self as an I. Taken together, the reflexive process allows for conscious, socially adjusted action, as well as novel, unpredictable, and creative experience. Contemporary elaborations of Mead’s basic framework have maintained the emphasis on the self as a regulating mechanism. The most influential elaborations of the self as a mechanism emphasize processes and motives that result in self-consistency, congruity, and/or balance (Turner 2006, pp. 368–370). As a consequence, these models are well positioned to explain the reproduction of inequality.

For example, Burke and Stets’ (2009; Stets 2006) identity control theory postulates a cybernetic model in which actors are motivated to verify an identity standard—the relatively stable self-meanings associated with a particular identity category.Footnote 4 If feedback from reflected appraisals in a particular situation is inconsistent with the identity standard, actors will engage in action so that their perceptions are congruent with the standards. When an identity is verified in a situation, persons experience positive emotions, and when identity verification fails, negative emotions are produced. Taken together, the control process leans toward stability and reproduction of existing self-meanings and the dominant structural arrangements associated with these meanings. Identities with less value and power have a limited capacity for independence and will be subject to greater control. We see this, for example, in a study of newly married couples that found that spouses with less powerful and valued identities outside of marriage (less education, lower prestige occupation) were more likely to have their self-meanings influenced by a higher-status spouse. On the other hand, spouses with higher-status identities reported self-meanings relatively independent of their lower status spouse (Cast et al. 1999).

A similar specification of the self as a mechanism of reproduction is represented in self-verification theory (Swann 1983). The argument here is that people seek confirmation of already established self-views, including social identities, because of a preference for a coherent, orderly, stable and predictable social experience. As a result, individuals will choose to interact with others who see them as they see themselves, even under instances when the meaning of self is negative (Kwang and Swann 2010; Swann 1996). For example, Swann et al. (2002) found that college students with negative self-views were generally unsatisfied with roommates that provided positive appraisals. The inconsistency between self-perception and others’ feedback motivated students to seek alternative living arrangements. And in cases where ending the relationship is not possible, the evidence indicates that people will seek to withdraw psychologically by limiting their emotional engagement and expressions of commitment (Swann et al. 1994).

Both identity control theory and self-verification theory offer models of the self as a mechanism that facilitates consistency, predictability, and reproduction of the status quo. However, neither theory precludes the self from contributing to acts of collective resistance, and there is some evidence that the self-verification process can actually function to promote social change. Pinel and Swann (2000), for example, argue that under certain conditions the self-verification process can motivate the decision to become active in a social movement. This is particularly true for social movement identities that are consistent with already established self-views. McAdam and Paulsen’s (1993) study of the 1964 Mississippi Freedom Summer Project illustrates this latter point. They found that success in recruiting participants depended on (1) the occurrence of a specific recruiting attempt; (2) a successful blending of movement and identity; (3) support from persons tied to the identity; and (4) the lack of a strong opposition from persons associated with other salient identities. In other words, under certain conditions, social change movements may advance because the self operates in a manner that seems to privilege consistency and equilibrium (see Snow and Owens, this volume).

But the McAdam and Paulsen study is also noteworthy because it highlights the relevance of identity salience as a self-process in resistance movements. The idea here is that identities are organized in terms of a cognitive salience hierarchy where personal commitments are prioritized. Social movement success depends on activists with salient movement identities. This operation of the self has been described in different terms by a range of self theorists (McCall and Simmons 1978; Rosenberg 1979; Stryker 1980; Turner 1978), and although there are key differences in conceptualization, there is compelling empirical evidence supporting the basic idea that the likelihood of an identity being invoked in a particular situation depends in part on its relative position in an internal hierarchy of all identities (Callero 1985; Hoelter 1983; Stryker and Serpe 1982). The more salient a social movement identity is, the more likely it will be invoked, and the more likely it will guide action. McAdam and Paulsen found that a highly salient movement identity, combined with strong social support for activism in other identity networks, was a key predictor of participation in the Freedom Summer project. In other words, when identities are politicized and become salient within the self-structure, they serve as resources for resistance .

Recognizing the operation of a salience structure and the influence of salient identities is important to explaining movement participation, but it does not address the more fundamental question of how social movement identities become salient in the first place. While a salience hierarchy is a relatively stable and enduring structure, it may be altered over time as new identities are formed, social networks change, and new opportunities arise. Indirect evidence from studies of non-movement identities points to positive social ties, reoccurring actions tied to the identity, and intense relationships, as factors associated with identity salience (Callero 1985; Nuttbrock and Freudiger 1991; Stryker and Serpe 1982). However, forces initiated at a higher level of analysis also condition the development of a salient social movement identity. Economic booms and busts, war, disease epidemics, demographic shifts, migration patterns, governmental policy changes, environmental disruptions, all have the potential to alter social relationships and simultaneously impact the salience hierarchies of multiple actors who are similarly situated. This is what Klandermans and de Weerd (2000) found in a panel study of Dutch farmers who engaged in protests over a change in agricultural policy during the mid-1990s. Their findings show that over time the relationship between protest participation and group identification increased, a finding that they interpreted as evidence of the growing salience and political relevance of the farmer identity.

In sum, a salience hierarchy is a type of self-process that structures action and reflects social relationships. When outside forces threaten the relative value or power of an identity, relationships change, hierarchies are restructured, and behavior is adjusted. It is helpful here to think of identities as resources that enable action and therefore have the potential to enable resistance to structures of inequality. Sometimes the resistance is long-term and widespread, resulting in enduring changes for both self and society (e.g., Tilly 2004). Other times, however, the protest may be small, narrow, and quickly extinguished by powerful forces of the status quo (e.g., Callero 1995). While contemporary elaborations of Mead’s theory of self have tended to emphasize the manner in which self-processes serve as regulating mechanisms, we should not forget Mead’s corresponding assertion that novelty, change, and the emergence of new structures, is a fundamental characteristic of the social process.

Conclusion

In this chapter I have argued for a more intentional, extensive, and integrated examination of self and identity processes as they relate to social inequality. While multiple avenues of research at different levels of analysis point to the significance of self and identity in the production, reproduction, and alteration of social inequality, this body of work is not well integrated. Thinking across different levels of analysis has the potential to offer new understandings of social inequality and a more complex description of how the unequal distribution of material resources is accomplished, maintained, and altered. This is especially true for analyses of resistance and positive social change where theory and research are less developed.

It has become somewhat of a sociological truism to assert that structures of inequality are the consequence of human interaction, and that interaction is constrained by structures of inequality. But this basic principle should not be interpreted to mean that all structures of inequality are reducible to interaction. To understand this subtle but important feature of the micro-macro link requires an appreciation of the philosophical notion of emergence—an idea central to Mead’s theory of the self (cf. Mead 1932, 1934), and one that has more recently been developed by a number of contemporary sociological theorists (see especially, Porpora 1993; Sawyer 2001, 2002; Smith 2010). By emergence I mean a dialectical process whereby lower-level structures give rise to higher-level structures, and higher-level structures constrain the same lower-level structures from which they emerged. We can say, for example, that macro patterns of inequality (e.g., class divisions, institutional racism, relations of patriarchy) emerge from symbolic interaction and depend on persons with selves and identities. But, at the same time, these larger social structures (patterns of social relationships) possess a level of reality and coercive power over and above selves, identities, and the rules of interaction. To this point, Porpora (1993, p. 220) provides a relevant example:

Rules of allocation may make workers dependent on capitalists for jobs, but the dependency itself is neither a behavior nor a rule. There are no rules saying that workers are to depend on capitalists for their livelihood. Such dependency is a relationship. This relationship, to be sure, is a consequence of rules of allocation, but it is a consequence that itself has consequences. It enables the capitalist to coerce the worker into submitting to (among other things) the rules of authorization that obtain at the job site.

To be sure, the dependency relationship identified above can be categorized, known, and experienced through the identity categories of worker and capitalist. We can also assume that the relationship emerged from symbolic interaction, and that meanings associated with these two identities are tentative and negotiable. But this does not detract from the coercive efficacy of the relationship itself. The principle of emergence offers one way of recognizing different levels of social reality without being forced into a false choice between macro and micro explanations.

Consistent with the basic notion of emergence, I have argued here that identities emerge from selves, and selves emerge from persons. This means that identities have a level of independence over and above selves—even though identity categories cannot exist without the capacity for selfhood. This also means that the self has a level of independence over and above personhood —even though selfhood requires the corporal capacities of human persons. The same principle is at work when lower-level processes of self and identity assist in converting interaction into higher-order structures of equality and inequality. Relationships of inequality emerge from self and identity processes, but these structures also work back to enable and constrain human persons. Again, recognizing the principle of emergence can facilitate an integration of macro and micro accounts of inequality.

Human persons are like other living organisms in that they must navigate a material world to secure resources necessary for survival. And like other living organisms, this quest will usually produce patterns of stratification and relations of dominance. But the primary argument of this chapter has been that social inequality in human societies is unique in that it emerges from our capacity for symbolic interaction. Evidence presented in this chapter suggests that the rules for allocating resources, the processes through which resources obtain value, and the mechanisms linking individuals to resources, involve selves and identities at several levels of analysis. This does not mean that self and identity are always and everywhere actively engaged in the reproduction of inequality. Indeed, one of the objectives of a social psychology of inequality should be to explain when, and how, lower level processes of interaction contribute to the emergence of higher-level structures of inequality. More macro oriented researchers, on the other hand, should not dismiss the symbolic processes associated with self and identity, for it is through the creative and problem-solving capacities of the self that calcified social relationships are dissolved, and entrenched structures of inequality are disrupted.