RJ 1

S o ci a l I d e n t i t y

Kay Deaux City University of New York

I. Conceptions and Definitions II. Types of Social Identity

III. Multiplicity and Intersectionality IV. Aspects of Social Identity V. Assessing Social Identity

VI. Development and Change VII. Negotiating Social Identities

Encyclopedia of Women and Gender, Volumes One and Two Copyright © 2001 by Academic Press. All rights of reproduction in any form reserved. 1

Gl o s s a r y

Intersectionality The condition in which a person si- multaneously belongs to two or more social cate- gories or social statuses and the unique conse- quences that result from that combination

Minimal group paradigm An experimental proce- dure for creating social identity conditions in which participants are arbitrarily assigned to one group or another.

Social representations Commonly shared and collec- tively elaborated beliefs about social reality con- sensually held by members of a culture or subcul- ture.

Stereotypes Organized, consensual beliefs and opin- ions about specific categories or groups of people.

SOCIAL IDENTIFICATION is the process by which we define ourselves in terms and categories that we share with other people. In contrast to characteriza- tions of personal identity, which may be highly idio- syncratic, social identities assume some commonali- ties with others. This chapter introduces several key issues surrounding social identity, including form and content, assessment, development and change, and identity negotiation.

I . C o n c e p t i o n s a n d D ef i n i t i o n s

“Identity” is a term that is widely used and, as a consequence, can mean many different things to dif- ferent people. Identity is sometimes used to refer to a sense of integration of the self, in which different aspects come together in a unified whole. This in- trapsychic emphasis is often associated with Erik Erikson, who introduced the term “identity crisis” as part of his stage model of psychological develop- ment. Another common use of the term, particularly in contemporary times, is identity politics, where the reference is typically to different political positions that are staked out by members of ethnic and na- tionality groups.

In this article, the term “social identity” refers specifically to those aspects of a person that are de- fined in terms of his or her group memberships. Al- though most people are members of many different groups, only some of those groups are meaningful in terms of how we define ourselves. In these cases, our self-definition is shared with other people who also claim that categorical membership, for example, as a woman, as a Muslim, as a marathon runner, or as a Democrat.

To share a social identity with others does not nec- essarily mean that we know or interact with every

other member of the designated category. It does mean, however, that we believe that we share nu- merous features with other members of the category and that, to some degree, events that are relevant to the group as a whole also have significance for the individual member. As an example, a person who de- fines herself as a feminist is more likely to be aware of legislation regulating abortion, more likely to have read books by Betty Friedan or bell hooks, and more likely to be aware of salary discrepancies between women and men than is a person who does not iden- tify as a feminist.

I I . Ty p e s o f S o ci a l I d e n t i t y

Many forms of social identity exist, reflecting the many ways in which people connect to other groups and social categories. In our own work, we have pointed to five distinct types of social identification: ethnic and religious identities, political identities, vo- cations and avocations, personal relationships, and stigmatized groups (see Table I). Each of these types of social identification has some unique characteris- tics that make it somewhat different from another type. Relationship identities, in particular, have some special features. To be a mother, for example, can imply a sense of shared experience with other peo- ple who are mothers. Sometimes particular aspects of these experiences can be defined even more finely, as in Mothers Against Drunk Drivers (MADD). At the same time, the identity of mother implies a spe- cific role relationship with another person, a rela- tionship that is unique and grounded in one’s own personal experience with that other person.

Other social identities can be defined more gener- ally, tied not to any individual but to a generic group. Thus to identify as a doctor, for example, implies a shared definition with countless others, many of whom you may not know anything in particular about. Another defining characteristic of occupa- tional identities is that they are chosen by the person (what is sometimes called an achieved status). In contrast, social identities such as ethnicity or gender are ascribed categories, given to one at birth. Social identities also differ in the status or value that is at- tached to them. In Table I, for example, the stigma- tized identities stand apart from the other types of social identity, all of which are typically regarded more positively.

In the original study that defined the categories presented in Table I, gender was clustered together with other relationship identities in the final statisti-

cal solution. Certainly it is true that many relation- ships are gendered in their definition and implica- tions (as are many occupations as well). However, because of the importance and centrality of gender in our lives, it is often considered as a category in it- self. Similarly, sexual orientation can be classified as one form of a relationship identity, but it often has greater prominence than other relationship identi- ties. To understand more about the nature of social identity, let us consider three identities in more de- tail: gender, ethnicity and nationality, and sexual ori- entation.

A. GENDER IDENTITY One’s gender—most typically as a man or woman— is one of the most frequently mentioned identities

2 S o ci a l I d e n t i t y

Ta b l e I Types of Social Identity

Ethnicity and religion

Asian American

Jewish

Southerner

West Indian

Political affiliation

Feminist

Republican

Environmentalist

Vocations and avocations

Psychologist

Artist

Athlete

Military veteran

Relationships

Mother

Parent

Teenager

Widow

Stigmatized identities

Person with AIDS

Homeless person

Fat person

Alcoholic

when people are asked to describe themselves, and it is also one of the categories most often used by oth- ers to describe us. Similarly, the development of gen- der identity (see Section VI) has been a central topic for developmental psychologists. Because gender is such a fundamental category, it is perhaps not sur- prising that a great many meanings and implications are associated with gender. Personality traits (e.g., being competitive or being aware of the feelings of others), role behaviors (e.g., taking care of children or assuming leadership roles), physical characteris- tics (e.g., having broad shoulders or a soft voice), and a host of other associations can be linked to gen- der categories.

At the same time, many investigators believe that it is not useful to think of gender as a single social category. Rather, many have argued for a concept of gendered identities, which recognizes the multiple social identities that may be influenced by one’s gen- der. As noted earlier, both occupations (e.g., nurse) and relationships (e.g., wife) often have gender im- plications. Similarly, a person’s identity as a woman may differ radically depending on whether she views herself as a feminist or as a more traditional type of woman. Thus, in adopting a perspective of gendered identities, one acknowledges that multiple identities are shaped by one’s gender, and that social identities can intersect and overlap with one another.

B. ETHNIC AND NATIONAL IDENTITIES For many people, ethnicity is a central element of self-definition and becomes an important social iden- tity. In the past, social scientists categorized human beings in terms of basic racial categories, such as Asian, Caucasian, and Negroid. With increasing awareness of the arbitrary nature of the social con- struction of race, these categories are less frequently used. More common today is categorization on the basis of ethnicity, defined in terms of culture, lan- guage, and country of origin. Works by theorists such as William Cross on African American identity exemplify the approach to this form of categoriza- tion and identification.

Nationality can be closely linked to ethnic identity, but it often represents a distinct way of identifying oneself. In Finland, for example, being ethnically Finnish and being a citizen of Finland are highly overlapping bases of identification. In contrast, in the United States one can have an identity as an American and at the same time hold an identity (of- ten hyphenated) as an African American, an Asian American, a Latino, or a West Indian. Like most

identities, national identities are flexible and subjec- tively defined. People choose both whether to have an ethnic identity at all, and, if so, what identity to claim. Often second-generation immigrants, for ex- ample, feel a pressure to choose between maintain- ing an identity with their country of origin and de- veloping a new identification with the host country. These two bases of identification can have quite dif- ferent meanings for friendship networks, social and cultural activities, and even marriage and family. Yet at the same time, it is increasingly recognized that people are not necessarily required to choose be- tween one of two mutually exclusive identities, but may instead maintain dual identification or may use the two sources of identity as the basis for a new emergent form of social identification, for example, as a biracial person. Like gender, the analysis of eth- nic and national identity is more complex than it sometimes first seems.

C. SEXUAL ORIENTATION Many people use sexual orientation as a central cat- egory of social identification. As is often the case, members of the minority group—in this case, gays and lesbians—are more likely to give prominence to this social identification than are members of the dominant majority group—in this case, heterosexu- als. Many analyses of gay and lesbian identification have posited stage models of development, describ- ing the processes by which people come to recognize and then to endorse their sexual orientation. These models take into account the evidence that many in- dividuals do not become aware of their sexual pref- erences until adolescence or later. At the same time, stage models are often criticized, both for assuming invariant sequences in the development of the gay/ lesbian identity, as well as for assuming that the process works in the same way for gays and for les- bians. Far less work has been done in defining a het- erosexual identity, in part because it is less frequently referred to by those who might see themselves that way (although heterosexuality per se has certainly been studied widely).

A critical aspect of the gay and lesbian identity is that it is, in some segments of society, a highly stig- matized identity, a characteristic that is shared with some ethnic and religious identities. The experience of prejudice and discrimination that gays and les- bians face make the process of social identification a particularly difficult one at times, as the positive val- ues that one typically associates with one’s own group are not shared by the society at large. Identification

S o ci a l I d e n t i t y 3

in terms of sexual orientation also illustrates well the overlapping nature of identity categories, particu- larly with gender.

I I I . Mul t i p l i ci t y a n d In te rs e c t i o n a l i t y

It is easy to talk in terms of multiple identities, for example, having separate identities as a woman, a lawyer, a spouse, a mother, a roller blader, and so on. In fact, several theoretical traditions within psychol- ogy and sociology, including role theory and sym- bolic interaction, encourage us to think in terms of these distinct groups. In contrast, theories emanating from personality psychology, such as that of Erik Erikson, focus on the possibilities for integrating multiple identities into a single identity. Indeed, within that particular tradition, the successful reso- lution of potential conflicts among identities is seen as a criterion of the healthy personality. [SEE SOCIAL ROLE THEORY OF SEX DIFFERENCES AND SIMILARITIES.]

Each of these positions involves its own concep- tual challenges. For those who favor the position that there are a set of distinct identities, one must consider how and when these identities relate to one another. Are there points of overlap among identi- ties? Can identities be represented in some form of hierarchy, with more important or more encompass- ing identities at the top and other less central or more specific identities at the bottom? For those who favor the integrative position, the questions concern how integration is achieved and whether a single identity, defined as the integrated sum of various component identities, can be predictive of more domain-specific behaviors.

Another perspective on this issue is to consider the intersectionality among various social identities. “In- tersectionality” is a term introduced by critical legal theorists to refer to the specific conditions that exist when one holds two or more social statuses. Often discussion has focused on the intersections of race and gender, exploring what it means, for example, to be a Black female as opposed to being a Black male or a White female. Gender, it is argued, does not necessar- ily carry the same meanings for members of different ethnic groups. Similarly, ethnicity may be experienced differently for women as compared to men. At the same time, proponents of intersectionality suggest that it is not possible to clearly distinguish between expe- rience that is related to race and experience that is re- lated to gender. Rather, the conditions are inextrica- bly bound together in the individual’s life.

Many investigations have shown the importance of considering different configurations of social cat- egories. Often, the particular configurations and the importance of one versus another identity may change over time as well, reminding us that identity is a dynamic rather than static process. Taken to the extreme, of course, the notion of intersectionality could be problematic if all possible intersections needed to be considered at all times. More likely, however, there are a limited number of key identity categories whose influence is sufficiently strong to combine with others and it is those intersections that investigators will want to study most closely.

I V. A s p e c t s o f S o ci a l I d e n t i t y

A social identity is first of all a label or a category, a way of grouping a number of people together on the basis of some shared features. Beyond the labeling, however, social identity has many more implications, both for the persons who claim the social identity and for others who see them as members of particu- lar categories. Thus the category label can in a sense be considered the frame for a painting that is rich in cognitive beliefs, emotional associations, and behav- ioral consequences.

A. COGNITIVE ASPECTS The cognitive aspects of a social identity can be ex- tensive and varied, including personality traits, social and political attitudes, and memories for identity- related events. Because social identities are devel- oped and defined within a social world, many of these cognitions are shared. Indeed, some investiga- tors talk in terms of self-stereotyping, suggesting that when one views the self in terms of a particular so- cial category, one takes on the stereotypes by which society has defined that category. Another way of talking about these shared definitions is to refer to the social representations of salient categories. Groups defined by gender, age, ethnicity, and na- tionality are all represented in the culture at large. There is often a consensus as to what best charac- terizes boys and girls, for example, or people from Australia or Turkey or Senegal.

Consider the stereotype of woman, for example. Traits typically associated with the category of woman include being emotional, kind, understand- ing, and helpful to others. More specifically defined types of women, such as a businesswoman, a femi-

4 S o ci a l I d e n t i t y

nist, or a housewife, carry other associations. These societally shared beliefs about a category can be- come part of one’s own social identification with the category. However, people do not necessarily take on the whole set of associations that consensually define a category. From the general set of societal represen- tations, people may adopt some aspects as relevant while not accepting others. In addition, people often create their own idiosyncratic definitions of what it means to be a particular type of person. Thus, the cognitive contents of a social identity are best con- ceived as a combination of socially shared beliefs and other attributes based on personal experience. [SEE GENDER STEREOTYPES.]

B. EMOTIONAL AND MOTIVATIONAL ASPECTS In many cases, social identities include not only “cool” cognitions, but “hot” emotions as well. Thus to be a feminist or an environmentalist, for example, may entail strong, affectively based feelings about social equality or the preservation of the environ- ment. Similarly, ethnic and national identities often carry deep emotional meanings. Consider the recent conflicts in the former Yugoslavia, among Serbs, Croatians, and Muslims; or the killings in Africa of Hutus and Tutsis; the troubles between Catholics and Protestants in Northern Ireland; and the contin- uing conflicts in the Middle East between Palestini- ans and Israeli Jews. In each case, identification with the ethnic group has a strong affective element that underlies the cognitive meanings associated with the identity.

Eva Hoffman, a Polish writer who emigrated to Canada, conveys the intensity of affect that can char- acterize an ethnic identity in the following passage:

The country of my childhood lives within me with a pri- macy that is a form of love . . . All we have to draw on is that first potent furnace, the uncompromising, ignorant love, the original heat and hunger for the forms of the world. (Lost in Translation: A Life in a New Language, 1990, pp. 74–75)

The recognition that identification has an emo- tional as well as a cognitive basis has a long history in psychology. Sigmund Freud, for example, de- scribed identification in terms of the emotional ties one has, first with a parent and later with members of groups (and especially with the group leader). Subsequently, social psychologists such as Henri Tajfel included the emotional significance of mem- bership as part of social identification.

Social identities also have a motivational basis. Particularly in the case of identities that people choose or achieve, specific functions are believed to be satisfied by the choice of identification. Although the variety of functions served by social identities are numerous, it is possible to think about a few general types. First, social identity may serve as a means of self-definition or self-esteem, making the person feel better about the self. Second, social identification may be a means of interacting with others who share one’s values and goals, providing reference group orientation and shared activity. A third function that social identification can serve is as a way of defining oneself in contrast to others who are members of an- other group, a way of positioning oneself in the larger community. This functional basis of identifi- cation can both serve as the impetus for joining a group, as well as become a defining agenda for group activity.

C. BEHAVIORAL ASPECTS One reason why social identification is a topic of such high interest is because categorizations have im- plications for behavior. To the extent that one de- fines oneself in terms of a particular group, it affects the behaviors one enacts for oneself and the way one interacts with others who may be members of dif- ferent groups. Early research on social identity by Tajfel and his colleagues emphasized the intergroup aspects of social identification. His research, which used a paradigm known as the minimal group, showed that it takes very little to create a sense of identification with one group and a consequent dis- favoring of another group. In these simple experi- ments, people were assigned to be in a specified group on the basis of a preference for one painter over an- other, or on a bogus distinction between preference for green or blue, or even just an arbitrary assign- ment as an X or a Y. With even this minimal and highly artificial basis for group identification, people will allocate rewards in such a way as to favor their own group and to disadvantage the other.

But the behavioral implications of social identifi- cation go far beyond these simple experimental demonstrations. An increasing body of research shows that group identification has important moti- vational consequences, and that the identifications that one is assigned or chooses leads to relevant ac- tions in a variety of domains, from volunteering for an organization to participating in social protest to choosing a mate. Not surprisingly, people who are

S o ci a l I d e n t i t y 5

more strongly identified with a particular group are more likely to carry out actions that are supportive of that group. [SEE INDIVIDUALISM AND COLLEC- TIVISM.]

V. A s s e s si n g S o ci a l I d e n t i t y

Given the complexity and multifaceted nature of so- cial identity, the question of how one assesses a so- cial identity is important. Perhaps not surprisingly, given the ingenuity of social scientists, many differ- ent methods have been developed over the years. Disputes have also developed as to which approach is the best.

Probably the simplest way to designate a social identity is to assign it arbitrarily, as is typically done in the minimal group experiments. Almost as simple is an approach in which group membership per se is the basis for assuming social identification. Thus, if one can determine that a person is a woman, a pro- fessor, or an Asian American, it is possible to assume that the social identity is present. A problem for this assumption, however, is that social identity is more appropriately viewed as a subjective, rather than ob- jective, state. Thus, while every student at a univer- sity can reasonably be called a student, it is not nec- essarily true that every student feels strongly identified with that category. Knowing how impor- tant or central an identity is to the person is neces- sary in order to predict how much the identity will influence the person’s beliefs, emotions, and actions. To deal with this potential problem, many measures of identification have been developed in which the respondent is asked to indicate how important or unimportant a particular identity is.

Social identity involves more than just categoriza- tion, however. As suggested earlier, key features of social identification include sets of beliefs, emotional associations, and motivational considerations. Some investigators have developed more extensive ques- tionnaires to tap a variety of aspects of social iden- tification. One issue in developing such measures is how generic versus how identity-specific they should be. A generic measure is one that can be used to as- sess any social identification, and thus it allows in- vestigators to make comparisons between different social identities in terms of their strength or central- ity. An item on this type of generic scale could be the following (this one taken from the Collective Self- Esteem Scale developed by Luhtanen and Crocker): “Being a member of a social group is an important

reflection of who I am.” In this case, any specific so- cial identity group could be substituted for the gen- eral term “social group.”

A somewhat more specific form of identity assess- ment is the measure of ethnic identity developed by Jean Phinney. In this case, the scale was designed specifically to assess ethnic identity, as evidenced by items such as the following: “I have a lot of pride in my ethnic group and its accomplishments” and “I participate in cultural practices of my own group, such as special food, music, or customs.” Phinney suggests that this measure can be used to assess any ethnic identification, thus it would be equally ap- propriate for an African American, a Cuban Ameri- can, or a Vietnamese American.

General measures such as these have the advan- tage of allowing the investigator to make compar- isons between groups, using a common metric. At the same time, these all-purpose measures have been criticized because they do not get at the specific fea- tures of a specific identification. Within the area of ethnic identification, for example, the beliefs and ex- periences associated with being a Black American are probably different from those associated with be- ing a Latino (or, more specifically, a Mexican Amer- ican, a Cuban American, a Puerto Rican, etc.). To capture these more unique aspects of social identifi- cation, some investigators have developed measures that are specific to a particular group. As one exam- ple, Robert Sellers and his colleagues created a mea- sure of African American racial identity that includes both general measures of centrality and salience, as well as specific questions about racial ideology that are based on the African American experience (ex- pressed in ideological philosophies of nationalism, oppression, assimilation, and humanism).

Other quantitative measures of identity reflect dif- ferent theoretical traditions. From the perspective of Eriksonian models, for example, identity is assessed in terms of the attainment of integration among iden- tities. Other theories, such as those that assume con- tinuing multiplicity, suggest assessment techniques that speak to the structure and relationship among various identities. Seymour Rosenberg and Paul de Boeck have developed procedures that yield a visual representation of identity structure, showing how a person’s identities are positioned relative to one an- other. Figure 1 illustrates such a structure. In this particular method, people are asked to list both the identities that are important to them and the attri- butes that they associate with each identity. Based on the degree to which identities are characterized by

6 S o ci a l I d e n t i t y

common attributes (and attributes are similarly ap- plied to identities), a structure is determined in which some identities are more encompassing or superor- dinate (such as woman in this example) and others are more distinctive and limited (such as daughter or volunteer in this example). Daughter and sister ap- pear together in this figure because they are described by the same traits, i.e., comfortable, relaxed, etc. Other identities are described in other ways, as the figure shows. This method of identity assessment uses sophisticated quantitative methods, but at the same time yields very individualized portraits.

A quite different approach to assessing identity re- lies not on quantitative measures, but rather on var- ious qualitative forms of data analysis, including narratives and open-ended interview material. In this approach, the investigator is more willing to let the person herself define the domains in which identity is relevant and the dimensions by which it is charac- terized. Qualitative methods are generally more suc- cessful than strictly quantitative methods in provid- ing a context for identity, allowing the respondent to relate themes of self to the historical and social events in which they developed and are played out.

Qualitative assessment of identity has been partic- ularly successful in exploring and highlighting con- ditions of intersectionality. By turning to the partic- ipant’s own narrative, the investigator is better able to appreciate the complex ways in which various identities may combine and overlap, as well as gain a sense of the ways in which those combinations

may shift over time and place. Qualitative approaches are also a particularly useful way to enter worlds that may not be recognized and represented in more traditional approaches, which are often based on im- plicit norms that do not generalize.

V I . D e vel o p m e n t a n d Ch a n g e

The sense of oneself as belonging to a particular cat- egory of people, or of being characterized by partic- ular labels, begins quite early in life. In terms of gen- der identity, most investigators believe that between the age of two and three years children have a sense of their gender. Gender identity is often defined as a “fundamental, existential sense of one’s maleness or femaleness.” Some assume this fundamental sense of gender is biologically determined and unchangeable; others suggest that cultural norms may allow for greater or lesser variations in people’s gendered def- inition.

Certainly the forces of socialization act on the in- dividual to define gender. Studies have shown, for example, that parents describe their newborn chil- dren differently as a function of gender—daughters are seen as more delicate and sons as better coordi- nated. Similarly, teachers as early as preschool en- gage in behaviors that shape the behaviors of boys and girls in different ways, such as more often ask- ing girls to be quiet or to speak softly. Thus, gender identity develops not in a vacuum, but in a social context in which representations and beliefs about gender are well established and actively fostered. [SEE GENDER DEVELOPMENT.]

In thinking about gender identity, it is important to recognize that the concept entails more than a simple label of female or male. Gender identity casts a net far wider than the biological features, includ- ing activities and interests, personal and social at- tributes, social relationships, communications styles, and values. Thus, a girl’s gender identity might en- compass playing with dolls, being encouraged to care for younger children, and smiling frequently; a boy’s gender identity might emphasize sports, an emphasis on self-reliance, and a reluctance to cry. Certainly not all girls and boys adopt these or any other gender- linked characteristics. In fact, individual repertoires of gender-related behaviors can vary widely from person to person. Yet most people do maintain some sense of what it means to be a man or a woman, and what they share with others who are in that same identity category.

S o ci a l I d e n t i t y 7

Partner hardworking

driven successful

Volunteer concerned

helpful

Daughter Sister

comfortable relaxed

immature spoiled

Woman smart active

confident

Lawyer

Fi g u re 1 Example of an identity structure.

Racial and ethnic identity is also learned quite early. Like gender identity, racial identity is heavily influenced by the stereotypes and belief systems prevalent in the society. Early studies by Kenneth and Mamie Clark argued that segregated schools were one environment that shaped the self-images of African American children, placing a greater value on White than on Black. William Cross and others have suggested that the development of African American identity can be conceptualized as a series of stages, beginning with what is termed preen- counter, when people do not believe race is an im- portant aspect of their overall identity, to later stages of internalization and commitment, when racial iden- tity is both central to self-definition and a source of positive regard.

Not everyone agrees that the development of ei- ther racial or gender identity can be characterized by a set of invariant stages, leading to a single end point. On the one hand, there are often predictable events that shape the course of identity development. In the case of gender, for example, puberty acts as an im- portant marker and makes sexuality more salient. At the same time, educational systems often channel the experiences of boys and girls in predictable ways. On the other hand, people take various paths within these broadly defined settings and may, over the life course, negotiate and renegotiate what it means to be a particular gender or race. People who engage in sex change operations later in life serve as a dramatic example of the potential flexibility in identity defi- nition, but there are many less vivid examples of shift and change as well. When one becomes a par- ent, for example, the new identity of mother or fa- ther is added to one’s self-definition. Similarly, be- ginning a career typically involves the acquisition of a new and often very important social identity.

Long-term development of social identities can take a number of forms. In the examples provided here, identity change involves the addition of a new identity or the deletion of a previously held identity. In the case of a sex change operation, as perhaps the most dramatic example, the previous gender identity is abandoned and the other gender identity is as- sumed. In the examples of parenthood and occupa- tion, the change is primarily one of adding on a new identity. One can also conceive of instances within these same domains when an identity could be dropped: when one divorces a spouse, for example, or retires from an occupation.

Other forms of identity development and change are more subtle. Change in the meaning of an iden-

tity, for example, involves shifts in the attributes and behaviors associated with an identity, while the claim to the identity continues unchanged. An attorney, for example, might shift areas of responsibility or type of legal practice, continuing to define herself or him- self as a lawyer but seeing the implications of that identity differently.

Still another form of identity change is a shift in the importance or centrality of an identity. Thus, one might persist in defining oneself as an attorney over several decades, but the importance of that identity, relative to other identities and interests, might shift to become more or less central to self-definition. The importance of ethnic identity can change over time, as Kathleen Ethier and Kay Deaux showed in a study of Hispanic students who entered university and were followed through the course of the first year. Their findings showed that ethnic identity increased in im- portance for some students, while others gradually decreased their identification with their Hispanic her- itage. Studies of immigrants show similar patterns, as the new arrivals deal with the meaning of their national identity of origin while often taking on a new identity as a resident of the country in which they now live.

In thinking about patterns of identity development and change, it is important to recognize that changes are more than intrapsychic. In other words, shifts in identity require changes in the relationship to one’s social and physical environment as well. Some people have used the term “ecological self” to refer to this grounded aspect of identity. Others talk about “place identity” as a way of relating to a particular physical environment that holds meaning. These concepts rec- ognize that where an identity is enacted is important. Although a social identity is rarely totally dependent on a particular physical setting, it is easy to think of examples in which identity and location are linked, such as occupational identities or athletic identities.

Equally important to the maintenance and devel- opment of a social identity is the social environment. By definition, social identities are ways in which we relate to a group or aggregate, and these social con- nections are critically important for defining and sus- taining the identity. When a person relocates, for ex- ample when a student moves from home to university, it becomes important that social supports be devel- oped in the new location to replace those supports that are no longer part of the immediate environ- ment. This process of grounding an identity in a new social environment has been termed “remooring.” As an example, when people who hold a strong eth-

8 S o ci a l I d e n t i t y

nic identity change their environment, as in the case of immigration, they will often establish connections with neighborhoods, stores, and organizations in the new country that reflect and promote the ethnic iden- tity of origin. [SEE SOCIAL SUPPORT SYSTEMS.]

V I I . Ne g ot i a t i n g S o ci a l I d e n t i t i e s

In addition to the long-term shifts in social identities, which develop over time and often change quite slowly, the expression of social identities can fluctu- ate considerably. If we assume that people have mul- tiple social identities, each of which may be charac- terized by distinct attributes and behaviors, then we need to consider the ways in which people may shift from one identity to another. Such fluctuations in identity, rather than evidence of instability or whimsy, provide evidence of the ways in which people re- spond to their environment and can make choices that seem most appropriate to that setting.

Identity negotiation is influenced by a variety of factors: the repertoire and importance of social iden- tities that a person has, the setting in which one is lo- cated, and the actions and influence of other people in those settings. Something as simple as the number of like people in a room can affect gender identity— not only the salience that gender has, but also the be- liefs and experiences that come to mind. Being the only woman in a group of men, for example, will make gender identity more salient. Similarly, being the only Caucasian in a group of African Americans will make race and ethnicity more salient. As these con- texts shift, so it is likely that one’s awareness of a par- ticular social identity will shift as well (though the sta- ble repertoire of identities is likely to remain the same).

It is also easy to imagine how specific situational cues can bring a particular social identity to the fore- ground. Comments by others, for example, that con- vey their perceptions can make an identity salient, as when an observer comments directly on one’s gen- der, age, or ethnicity. Physically being at a university may make identities as professor or student salient for those who hold those identities. Although such environmental cues are not, in any rigid sense, de- terminants of one’s identity, they have the ability to influence salience in a temporary sense. At the same time, it is important to recognize that people often choose their environments. The self-defined book- worm will find a library and the political activist will

find a rally, thus selecting environments in which a favored social identity can best operate.

Although the optimal strategy might be to find a niche in which one’s favored identities can best be enacted, circumstances sometimes create less desir- able environments. Thus, for the person whose so- cial category is to some degree stigmatized, threats to identity may be posed that require the develop- ment of strategies to cope with those threats. Some- times this may mean negating the identity or tem- porarily diminishing its importance. In other cases, threat may lead to a more active search for environ- ments in which the identity will be more favorably regarded, as when people engage in civil rights ac- tions to promote the legitimacy and entitlement of their group.

Looking more broadly, one can see how different cultures influence the definition and choice of a so- cial identity. The category of feminist, for example, was unrecognized in some countries until very re- cently. Some people have argued that even the notion of identity itself is historically bound, a product of the Renaissance period. Similarly, ideas of multiplic- ity, or what Robert Lifton has called the “protean self,” may reflect a set of historical conditions char- acteristic of post-industrial societies. In short, social identity is, as the term suggests, an inherently social phenomenon that must be understood as a product of both individual and contextual-historical forces.

SUGGESTED READING Capozza, D., and Brown, R. (eds.) (2000). Social identity

processes: Trends in theory and research. Sage, London. Deaux, K. (1996). Social identification. In Social Psychology:

Handbook of Basic Principles (E. T. Higgins and A. K. Kruglan- ski, eds.). Guilford, New York.

Deaux, K., and Stewart, A. J. (in press). Framing gendered iden- tities. In Handbook of the Psychology of Women and Gender (R. K. Unger, ed.). Wiley, New York.

Hogg, M. A., and Abrams, D. (1988). Social Identifications. Rout- ledge and Kegan Paul, London and New York.

Ruble, D. N., and Martin, C. L. (1998). Gender development. In Handbook of Child Psychology (W. Damon, ed.), 5th ed., Vol. 3: Social, Emotional and Personality Development, N. Eisen- berg, ed., (pp. 788–827)]. New York: John Wiley.

Skevington, S., and Baker, D. (eds.) (1989). The Social Identity of Women. Sage, London.

Stryker, S., Owens, T. J., and White, R. W. (eds.) (2000). Self, Identity, and Social Movements. University of Minnesota Press, Minneapolis.

Waters, M. C. (1990). Ethnic Options: Choosing Identities in America. University of California Press, Berkeley.

S o ci a l I d e n t i t y 9