The New Color Complex: Appearances and Biracial Identity

IDENTITY: AN INTERNATIONAL JOURNAL OF THEORY AND RESEARCH, 3(1), 29-52 Copyright © 2001 Lawrence Erlbaum Associates, Inc. The New Color Complex: Appearances and Biracial Identity David L. Brunsma Department of Sociology University of Alabama Kerry Ann Rockquemore Department of XXX University of Connecticut Ethnic identity research has largely focused on the identity choices of White ethnics (Alba, 1990; Ignatiev, 1995; Waters, 1990). One key factor in these choices is bodily appearance. We extend this research to Black and White Biracial individuals and examine the role that physical appearance plays in their “choices” of racial id entity. We test Rockquemore's (1999) taxonomy of Biracial identity using survey data from a sample of 177 Biracial respondents. The results indicate that Biracial individuals do make choices within circumscribed cultural contexts and these understandings are influenced not by skin color, but by an actor's assumption of how others perceive his or her appearance. Recent work on eth nic identity has focused on the construction and maintenance of White ethnic identities an d the choic es made b y individuals with m ultiple W hite Requests for reprints should be sent to David L. Brunsma, Department of Sociology, Morton Hall 344, University of Alabama, Huntsville, AL 35899. E-mail: [email protected] 30 BRUNSMA AND ROCKQUEMORE ethnic heritages (Alba, 1990 ; Ignatiev, 1995; W aters, 1990). As Am ericans’ ethnicities have become progressively intermixed, research has focused on why individuals with multiple White ethnic backgrounds choose to emphasize one ethnicity over others. Factors such as knowledge about an ethnicity, surname, physical appearance, and the general p opularity of eth nic groups emerge as the m ost salient factors influencin g individual p references. T his body o f research sup ports the notion that W hite ethnicity is largely “symbolic” (Gans, 1979) and has no real impact on an individual’s life chances o r everyday inte ractions. It is this symb olic basis of White ethnicity that differentiates it significantly from race. Although White ethnic identity remains an option, racial identities are neither situationally based nor symbolic. For members of racial groups, options are either nonexistent or function in a mo re tightly constrain ed manne r than for W hites because racial categories are socially and culturally stratified. Due to cultural norms like the “one drop rule” and the subseque nt racist social structure (Bonilla-Silva, 1997; Loveman, 1999), racial identity cuts across situations and im pacts individual life chances. Studies of ethnic identity o ptions pro vide a rich fou ndation for r acial identity research. The strength of this work is its fertile illumination of the complex link between social context, culture , and identity. The literature highlights the ways that individuals reflexively interpret their sense of self in specific social and cultural contexts by using various symbols , cultural understandings, and relative relationships between groups. In this article we focus not on W hite ethnic options, but on Black and White Biracial individuals and the “choices” they make about their racial identity. Because these Biracials (at least theoretically) straddle the color line, their marginality makes them a unique group for the exploration of the relationship between appearances and identity. Focusing on Biracial people enables us to explore how culture and social context influence identity construction through an understanding of how cognitive interpretations of differing racial categories interact with significant symb ols to influence the way that social actors understand themselves and their relationship to others. The most salient symbol representing group membership is bodily appearance. Appearances of various so rts (e.g., skin colo r, clothing, hairstyle, etc.) are important to the proc ess of human interaction be cause we ar e unable to directly observe the intentions and character of individuals (Goffman, 1959). It is through appearances that we evaluate both others and ourselves, and through which we present ourselves to others (Ichheiser, 1970). Appearances present our identities to others and allow us to infer the identitie s of others. Appearances help to define a situation and provide some cognitive context for all the individuals involved. In this sense, app earances c an beco me a reality in and of themselve s (Stone, 19 62). The empirical literature on the effects of skin color, the most salient of a ll bodily appearance characteristics for discussion s of race, illumina tes its importanc e. Skin color has been found to affect mate and friend selection (P orter, 199 1), a variety of life chances and opportunities (for reviews, see Hughes & Hertel, 1990; Porter & SKIN COLOR AND BIRACIAL IDENTITY 31 Washington, 1993), and mental illness (Boyd -Franklin, 19 89; Har vey, 1995; Neal & W i lson, 1989). The phenotype literature, however, has neither addressed the issue of how indiv iduals of colo r perceive th eir own racia l identity nor how phenotype might play a ro le in the racial self-understandings of persons of color. Biracial individuals’ appearances are often times ambiguous and they exist in a variety of different social contexts of differing racial composition. According to the common sense assumption that race is a biological, as oppo sed to a soc ially constructed reality, individuals’ physical traits determine racial group membership. Therefore, an individual’s phenotype should determine racial identity. The caveat to this system is the one drop rule that manda tes the assumption that, above and beyond phenotyp ic categoriza tions, anyone with a Black ancesto r is Bla ck (regardless of appearance). Exploring the influence of appearances on identity for Biracials allows us to join and extend the literature by understanding how presentations of self and the reviews of others are reflexive and function within a socially constructed set of parame ters. Using Rockquemore’s (1999) taxonomy of the racial self-understandings of Biracial individuals, we test how phenotype and social context affect the racial identities of Black and White Biracials. We review the literature on the relation between phenotyp e and racia l identity for multirac ial and Biracial individuals. Unlike ethnic optio ns, appea rances play a counter-intuitive r ole in the way Biracial people choose the ir racial identity. T hrough inve stigating the app earance– identity link with an emerging demographic group, we can grasp the power of cultural influences and social interaction over the process of constructing and understanding racial identities. Biraciality and Racial Identity Since the late 1970s, research on the racial identity of Black and White Bir acials has suggested that there exist two possi bilities for these individ uals’ racial selfunderstandings: Black or Biracial (Bowles, 1993; Brown, 1990; Cross, 1971; Field, 1996; Hall, 1980; Herring, 1995; Miller & Miller, 1990; Morten & A tkinson, 1983; Porter & Washing ton, 1993; Porterfield, 1978; Poston, 1990; Poussaint, 1984; Taylor-Gibbs, 1997). Early scholarship appears to have been based less on the results of empirical research and more on the historical and cultural norm of the one drop rule as the underlying assumption. This rule mandated that Biracial offspring adopt a Black racial iden tity. M ost research in the 1970s and early 1980s was interested in understand ing how these individuals d evelo ped a Black identity. By the mid 1980s and throughout the 1990s, the extant pathologies associated with Biracial individuals’ ma rginality drew th e attention of a new generation of researchers who sough t to explain psycholog ically, clinically, and d evelopm entally how these individuals developed a Biracial identity and how they could maintain 32 BRUNSMA AND ROCKQUEMORE a healthy, integrated sense of their Biraciality (Bowles, 1993; Brown, 1990; Daniel, 1996; Field, 1996; Hall, 1980; Herring, 1995; Johnson, 1992; Kerwin & Ponterotto, 1993; Kich, 1992; Poston, 1990; Taylor-Gibbs, 1997; Wardle, 1989 , 1992). These studies incorporated findings and discussions about the racial self-understandings of Biracials fro m earlier rese arch and c ontinued the develop ment a dichotomous typology of racial identity possibilities for Biracials: Black and Biracial—this time with an overt preference for the deve lopment o f a Biracial ide ntity. Such research set the parameters, within the scholarship, of the po ssibilities for racial identity among Biracial people. The 1990s brought yet another generation of scholars interested in the lives of multiracial people. These investigators utilized new analytical tools and incorporated interdisciplinary approaches. The work of several authors from this group (Davis, 1991; Rockquemore, 1999; Root, 1990, 1992, 1996; Spickard, 1989) began to illuminate for us that multiracial people had multiple understandings of their racial identities. Davis challenged us to consider the question Who Is Black? Root (1990) convinced us that these individuals had different negotiations and answers to the question of “how to resolve the status of the ‘other.’” This new work has set the tone for research on multiracial and Biracial individuals. Root’s (1990, 1992, 1996) work however, represented both the methodological limitations of the existing research (i.e., the exclusive use of clinical studies and small self-selective samples) while replicating the unidirectional, atheoretical, and ideologically biased work of the 1980 s. Rockquemore (1999) presented a comprehensive typology that integrates the voices of Biracial individuals with a variety of classic and conte mporar y identity theories (Cooley, 1902; Gergen, 1991; Giddens, 1991; Lifton, 1993; Mead, 1934; Turkle, 1997). She suggested that there are four types of racial identity options for biracial people: a) singular identity (singular Black or White) b) bord er identity (exclu sively Biracia l) c) protean identity (sometimes Black, sometimes White, sometimes Biracial d) transcen dent identity (no racial identity) It is this typology of ra cial identification s of Biracials that we use to explore the appearance–identity link using data from the Survey of Biracial Experience. Research o n Pheno type and Ra cial Identity Early research, built on the assu mption of the one drop rule, was inade quate in identifying the complexities of racial identity among Biracials in the United States. This being the case, the relation between phenotype and racial i dentity remained unquestioned and was rarely the focus of empirical investigation. Methodological problems in collecting data on Biracial individuals, coupled with samples revealing little variation in phenotypes, made exploration of the relation between phenotype SKIN COLOR AND BIRACIAL IDENTITY 33 and racial identity amon g Biracials d ifficult. The lack o f empirical stud y is associated with the paucity of theoretical development of the appea rance–id entity link. The research on Biracials in the 1980s never investigated the role of appearance in identity formation and maintenance because people assumed that regardless of your physical appearance, if you had a Black parent, you were Black. Several researchers have given the appearance–identity question passing comments that transcend the one drop assumption (Bowles, 1993; Brown, 1990; Daniel, 1996; Field, 1996; Hall, 1980; Tizard & Phoenix, 1995). Brown stated that appearance is an important factor in the emotional and cognitive attitudes toward oneself and discussed the importance of physical appearance for Biracials. He implied that the darke r the skin of the B iracial individu al, the more likely he or she would be to adopt and develop an exclusively B lack identity. However, he also theorized that factors such as the ratio of Blacks to Whites in given socialization contexts could influen ce this relationsh ip, but he did not specify ho w. Similarly, Daniel discussed how the intricate interrelationships between socialization experiences and the broader social and cultural context of race relations can influence the appearance–identity link; this was a step forward although he made no clear hypotheses. Bowles described the treatment of her Biracial clients toward the development of a healthy Biracial identity, found that women had a more difficult time accepting their Black physical features than men; thus, implicating a possible gender interaction with phenotype in the influence on identity. Tizard and Phoenix found that White-appearing Biracials were unlikely to identify ex clusively as Black. On the whole, although several scholars believe that physical appearance is crucial to the developme nt and maintenance of a rac ial identity among Biracials, the research has not rigorously investigated the appearance–identity link. Furthermore, there is little theory to guide investigation. The research on Biracial identity has often cited the reactions of W hites to Biracial individuals; it was The Color Complex (Russell, Wilson, & Hall, 1992) that really tackled the issue of the politics of skin color among African Americans. They considered the Black community’s love–hate relationship with the one drop rule and with skin color. T he argume nt is that a three-tiered society existed in the South with the following hierarchy from highest to lowest status: Wh ites, Mulattos, and B lacks. Mixed-race individuals often served as a buffer group between Whites and Blacks, through which cross-color interactions and b usiness transactions c ould occur. T his situation caused a preferential treatment of Mulattos by Whites and a generational advantage for Mulatto s—the fo undation fo r a socia l and cultural system of color classism within Black America was laid. The authors provided strong evidence that those members of the community with the lightest skin color and the most Caucasian-looking features have been allowed the greatest freedoms and achieved at higher rates. The argument goes further to display the ways that darker-skinned members of the Black comm unity discriminate against mixed-race individuals in the workplace, how patterns of dating within the community are tangled up with phenoty pe, how netwo rks are con structed or d ismantled o n the basis of co lor 34 BRUNSMA AND ROCKQUEMORE classism, and how culturally, Blacks use unique cultural coding (tertiary clues such as hair, first names, etc.) to distinguish between those who are Black and those who are not. How ever, The Color Complex does not discuss the relation between color and identity. It is clear that there is only speculation as to the relationship between appearance and racial identity among Biracial ind ividuals. Th ere has bee n little to no theoretical development on this relationship. One important distinction (Rockquemore, 1999; Root, 1990) is that color is both a personal and a social characteristic; that is, one perc eives their skin co lor, but one a lso interprets the ir appearance through the eyes of others within any given interactional sp here. Bo th Root and Rockquemore highlighted the inherent problem that exists when there is a mismatch b etween ap pearance and identity. In o ther words, it wo uld be extrem ely difficult for a person to c hoose an exclusively B lack, or exclu sively White , identity if their physical ap pearance s do not m atch their chosen identity. In the end, the literature on the appearance–identity link is sparse and seriously underdeveloped. We re medy this situatio n here. Conceptual Hypotheses Based on the previous discussions, we propose several hypotheses concerning the link between ap pearance s and the rac ial identity choice s of Biracial individuals. The literature seems to indicate that appeara nce is itself multifacete d. We conceptualize appearance to be comprised of two components: self-perceived skin color (how Biracial respondents place them selves on a skin color co ntinuum) and sociallymediated appeara nce (how B iracial respo ndents und erstand their own appearance based on how others interpret their appearance). Although previous work assumes that the one dro p rule would guide identity ch oices, w e feel, given the previous appearance distinction, that self-perceived skin color will not directly affect racial identity (H1). We expect this to be the case because a Biracial individu al’s understanding of their own appearanc e seems to be rooted more in others’ perceptions and assum ptions of a B iracial individu al’s appear ance and the link with identity (i.e., if others assume a p henotypica lly ambiguou s Biracial ind ividual is Black, that Biracial will adopt a Black identity). The variable we call “appearance” is distinctly more social than phenotype because it is cr eated by the Biracial ind ividual’s unde rstanding of the ir skin color as conditioned through the judgments of others in interactions. Thus, we expect that it is appearance, not skin color, which will influence the racial identification of Biracials, and that skin color works through one’s app earance to affect identity. Therefore, our secon d propo sition is that an actor’s socially perceived appearance will affect racial identity even after controlling for self-perceived skin color (H2). Finally, we hypothe size that a variety of social factors, including the racial composition of preadult social contexts, parental socioeconomic status, and SKIN COLOR AND BIRACIAL IDENTITY 35 negative treatment by Whites or Blacks will alter the appearance–identity link. W e are particularly interested in examining the effects of racial differences in social network composition on the appearance– identity link because it is within these social contexts that the interactional parameters for identity are established. From the review of the literature that has touched on the appearance–identity link, we expect that the strength of the link will fluctuate depending on Biracial individuals’ contextual experiences of race. Therefore, our final hypothesis is that the contextual experiences of race will moderate the effect of appearance on racial identity (H3). Method In efforts to collect data on Biracial people, researchers have encountered numerous unique challenges, including accurate identification of potential respondents and the ability to attract e nough respondents to draw meaningful generalizations. Numerous data sets have been collected in the pa st 5 years, however no published empirical studies have exceeded samples of over 50 respondents and most studies have included, within small university-based snowball samples, respondents with varying combinations of racial backgrounds. The most oft-cited research anthology of mixed-race people is Root’s (1996) The Multiracial Experience: Racial Borders as the New Frontier. Although it is generally considered the most comprehensive collection of research on the Biracial p opulation, the largest samp le used in any of the studies include d in the antho logy consisted of 31 resp ondents. M ost of the studies in this text generalize b roadly from anecdo tal, biograp hical, or small interview studies with fewer than 20 respondents. Small samples, which include large degrees of variance o n elements as fundamen tal as a racial–e thnic combination of parents make it extremely difficult to draw any meaningful generalizations whatsoever. To delve more deeply into the social processes that govern the meaning of racial identity and factors influen cing racial ide ntity construction for Biracia ls, we condu cted a surve y. Respon dents for the Survey of Biracial Experience were taken from two colleges in metropo litan Detroit, M ichigan: a priva te liberal arts college and a large commu nity college. The Detroit Metropolitan Area was selected due to the large African American population and the diversity of neighborhoods and social contexts. Both the community college and the university dra w students from all parts of the metropolitan area, allowing for respondents who live both inside and outside the racially homogeneous city limits. Both institutions have students who ranged in age, socioeconom ic status, and life experience, all of which were factors examined in the process of the research project. In sampling the college student population, we wanted to include respondents who were comparable along the dimension of education level, but purposively stratified by socio econom ic indicators, racial comp osition of their social networks, and other salient dimensions of interest. Using ins titutional lists of students, we so licited 4,53 2 students 36 BRUNSMA AND ROCKQUEMORE registered as “Black o r African Am erican”, “Other”, or those that left the race question blank on the ir college’s ad mission form s. The ratio nale for this decision was drawn from the sampling strategy used in Phase 1 of the research design where e-mail solicitation of African American students at the University of Notre Dame drew out the initial Biracial respondents. The pilot study reve aled that eve n students who most m ilitantly stated they were Biracial (i.e., they were “not Black”) indicated they had identified as Black or African American on their college admission forms. This was done for two reasons: (a) the desire for social or organizational inclusion and (b) perce ived individ ual gain. Students stated that they checked the “Black” box instead of “other” (and writing in “Biracial” or “mixed”) because they did not want to be excluded from activities that were targeted for the Black student population, such as solicitations for Black student organizations or announcements of special speakers. The second, and more salient reason, was that students felt that it couldn’t hurt their opportunities for admission or for financial opportunities that may be designated for minority students. Given this consistent response, soliciting the entire Black student p opulation at each institution seemed to be the most efficient way to draw a group of Biracial respondents. We contac ted these individuals, inquiring whether they had one W hite and one Black identifying parent and eventually we located and surveyed 250 individuals. After cleaning the data, identifying only Black and White Biracials, and removing cases with missing data, we w ere left with 177 cases for the analysis. Measu rements Identity. To assess the racial identifications of the Biracial survey respondents, we used a slight modification of Rockquemore’s (1999) typolo gy of the racial selfunderstandings of Biracial individuals. As mentioned previously, she found that Black an d Wh ite Biracials un derstood themselves in four basic ways: (a) singular identity, (b) border identity, (c) protean identity, and (d) transcendent. The responses to the following question in the Survey of Biracial Experience were used to measure id entity in this study: Which o f the following state ments best describes your racial identity? Respondents chose from the following options: (a) I consider myself exclusively Black (or African American ), (b) I some times consid er myself Black, sometimes my other race, and sometimes Biracial depending on the circumstances, (c) I consider myself Biracial, but I experience the world as a Black person, (d) I consider myself exclusively as Biracial (neither Black nor White), (e) I consider myself exclusively as my other race (not Black or biracial), (f) Ra ce is meaningless, I do not believe in racial identities, or (g) Other (fill in the blank). W e coded the singular Black identity as 1, the singular White identity as 5, the protean identity as 2, and the transcendent identity as 6. Fundamental social-psychological theory (e.g., Stone, 1962) highlights the important fact that identities, formulated SKIN COLOR AND BIRACIAL IDENTITY 37 within interaction with o thers, can be validated o r unvalidated by others in interaction. Therefo re, because these individu als in our samp le are all Biracials, we modified the border identity within Rockquemore’s typology to include both a validated border id entity and an unvalidated border identity. Hence, we coded the validated border id entity as 3 and the unvalidate d borde r identity as 4. T his distinction will become crucially impo rtant later. We do no t look at those responding “Other” in this study—only 4.2% of the sample c hose to ide ntify this way. Phenotype (skin co lor). Defined as self-perceived skin color, the survey asked respond ents to locate themselves on a scale from 0 to 12 that be st described their skin color. The c ontinuum used the following terms and codes: 12 = black, 10 = dark brown, 8 = medium brown, 6 = light brown, 4 = yellow, 2 = olive, and 0 = white. This resulted in an ordin al level phenotype variable (odd numbers were possible and indicate a phenotype between two categories). For the analytical purposes of this study (and to reduce em pty cells) we co llapsed this or dinal variab le into three categories, in which 1 was light skin (0 to 4), 2 was medium skin (5 to 7), and 3 was dark skin (8 to 12). Appearance. Although th e previous varia ble attemp ts to measure respond ents self-perceived skin color, the appearance variable measures respondents’ estimates of how other people categorize them based on th eir physical features. The distinction here is between the respondents’ self-assessment and their understanding of others’ assessments. We ask ed responden ts to sele ct which of the following statements best described their physical appearance: (a) “I look Black and mo st people assume that I am Black “; (b) “ M y physical feature s are ambig uous, peo ple assume I am Black mixed with something else “; (c) “My physical features are ambiguous, people do not assume that I am Black “; and (d) “I physically look White, I could pa ss.” We r everse-c oded th is item so that highe r values reflect a more Black categorization, or response from others , whereas low er values indic ate a more White response from others in interaction. Interactional Experience 1: Pread ult racial composition. Modifying Demo and Hughes’s (1990) interracial contact variable to assess the racial composition of respondents’ significant socialization contexts, individuals were asked to assess the racial composition of their: (a) grammar or elementary school, (b) closest friends in elementary school, (c) junior high schoo l, (d) high scho ol, (e) closest frien ds in high school, and (f) neighborhood while growing up. The responses were cod ed in the following way: 0 = most all W hite, 1 = mostly W hite, 2 = about half Black, 3 = mostly Black, 4 = all Blacks. From these items we constructed a preadult scale of 38 BRUNSMA AND ROCKQUEMORE social network composition (.93) ranging from 0 to 24. Fo r purpose s of the analysis in this study, we collapsed this variable into three categories in which 1 was predom inately Wh ite (0 to 4, 32.8% of ca ses), 2 was mixed B lack and White (5 to 10, 33.3% of cases), and 3 was predominately Black (11 to 24, 33.9% of cases). Interactional Experience 2: Negative treatment from Whites and Blacks. Another contextual experience of race that may alter the interactional experiences of Biracial ind ividuals is whether they experience negative treatment from Blacks or from W hites throughout their lives. Such negative interactions can serve as “push factors” that may lead Biracials to adopt a particular racial understanding of themselve s, and that ma y paramete rize the app earance– identity link. To assess the way that Biracials may experience race, we aske d respon dents whether they had ever experienced personal discrimination or hostility from either Whites or Blacks because of their mixed-race. We coded this variable in the following way in which 0 was no and 1 was yes. Results Phenotype , Appearances, and Identity This sample of Biracial individuals was varied on the dimensions of most interest to us in this study: phenotype, appearance, racial identity, and contextual experience of race. Before repo rting the results of the analysis, it was importa nt to indicate the degree of variation in this sample on the se key variables. Within our sample, there existed wide variatio n in self-reporte d skin color. For the original ordinal variable, which ranged from 0 to 12, the median score was 6, and the variation is striking. For the analyses presented later, we collapsed the original skin color va riable into thre e equal cate gories: light, (33.9%), medium (39%), and dark (27.1%) skin. Respon dents, by their o wn estimation , assessed their skin color as ranging across the phenotyp ic spectrum. T his variance w as critical to our analysis because it allowed us to explore the influence of skin color on the choices that o ur respond ents make a bout their rac ial identity. In addition, respondents differed in their answer to the questio n of their socially mediated appearance. Some variation in responses does exist. However, the majority of respondents (56.2%), despite reporting wide variation in skin color, described their appearance as “ambiguous though most people assume I am Blac k.” Seventeen percent (17.2%) stated that they “appe ar Black, m ost peop le assume I am Black,” whereas 16.6% stated that they were “ambiguous, most people do not SKIN COLOR AND BIRACIAL IDENTITY 39 assume I am Blac k.” Finally, 10.1% of this sample o f Biracials said they “appear White, I could pa ss as Wh ite.” An examinatio n of the desc riptive measu res reveals two crucial facts that ar e importa nt to our forthco ming analysis: (a) The sam ple includes variation on the two key measures of appearance set forth in our theoretical model, and (b) the d ifferences between these two variables suggests that they are, in fact, measuring two conceptua lly distinct elements. Contrary to much of the literature on the racial identification of Biracials, the individuals in this sample also vary in their racial self-understanding. Nearly two thirds (64%) of the respo ndents con sider themse lves to have b order ide ntities. Specifically, they understand themselves racially as neither Black and White, but as exclusively B iracial. How ever, most o f these border Biracials say they experience the world as a Black p erson; thus, their identity as exclusively Biracial is unvalidated by others in interactions—that is, the category of border or biracial simply is not recognized by their interaction partners. Fo rty percent o f the sample have these unvalidated borde r identities, whereas 23.6% indicate a validated border identity. Also contra ry to popu lar assumptio ns, only 13.7 % of our sample understand themselves as singularly Black. The same percentage (13.7%) understand themselves w ithout racial lab els, or with a transc endent ide ntity. Five percent see themselves as shifting between Black, White, and Biracial depending on the circumsta nces (i.e., pro tean identity), whe reas the rem aining 3.7 % understand themselves a s singularly White. T here is significant va riation with regards to racial identity among this samp le of Biracials. There is also significant variation in the contextual experiences of race o f these respondents. These Biracials came from a variety of prea dult socializatio n contexts with varying racial compositions. The original variable, ranging from 0 to 24, has a mean of 7.74 and a standard deviation of 5.90. We collapsed this variable into one with roughly equal proportions: predominantly White (32.8%), mixed Black and White (33.3 %), and predominantly Black (33.9%). Also, most of these Biracial individuals had, throughout their lives reported experiencing negative treatment from Blacks (60.5%) and even more from W hites (74.3%). 1 1 In addition to these dimensions of variation within the sample, it is important to note other vari ables that show good degrees of variation in the Survey of Biracial Experience: For example, most of their parents had at least a high school degree or some college, their parents held average occupations in terms of prestige rankings, their parents were, on average, not very politically active. These respondents range widely in age, but average in the mid-twenties. 40 BRUNSMA AND ROCKQUEMORE The Relationship Between Phenotype, Appearance, and Racial Identity Am ong B iracials H1 stated that we did not expect skin color to affect racial identity choice among Biracials. Table 1 presents the relationship between phenotype and racial identification among our B iracial respondents. There is no association b etween skin color and the way that Biracial individuals racially understand themselves, c 2(10, N = 161) = 8.28, p = .602. This provides supportive evidence for our first hypothesis. In fact, we discussed the idea that others’ interpretations of our appearances are more important in o ur self-understandings than our own perceptions of how we look. H2 proposed that it is socially-mediated appearance, or the way you believe others perceive yo urself, that we expected to influence racial identity among Biracials. Table 2 explores H 2—the categorica l association between appearance and racial identification among Black and White Biracials. There is, in fact, a strong association between ap pearance and identity in this sample, c 2(15, N =159) = 35.80, p = .002. The results can be most clearly understood by examining each of the various identity types and their unique link with appearance. Appearance, measured this way, operationalizes both the reviews of others and the racialized assumptions others make about Biracial individuals in interactions—it is clear that these reviews and assum ptions dee ply affect Birac ials’ self-understand ings of their own race. Singular identity. By exam ining the appearance–identity link within each separate identity choice, distinct patterns emerge. Biracials who adopt singular Black identities most commonly report (95.5%) that others “assume they are Black.” Even when they don’t “appear Black” and are ambiguous in appearance, others assume they are Black and these assumptions have lead most of these individuals to understand themselves as Black. In fact, none of the singular Blacks say that they “appear White and could pass,” even though according to data from Table 1, 27.3% of them have light skin color. The results for singular Whiteidentifying Biracial resp ondents are tenuous due to the low num ber of case s in this category. Protean identity. Some re sponde nts adopt a shifting, or pro tean, racial ide ntity, in which they understand themselves as sometimes Black, sometimes White, and sometimes Biracial. These respondents have a wide range of appearances as understood through the reviews of others. Most (75%) report their appearance as ambiguous (either “ambiguous, not assumed Black” or “ambiguous, assumed Black “). We are cautious however, in o ur interpretatio n of the app earance id entity TABLE 1 Relationship Between Phenotype and Racial Identifications Among Black and White Biracials Validated Unvalidated Singular Border ID Border ID Black ID Protean ID Phenotype % n % n % n % n 27.3 6 12.5 1 44.6 29 31.6 12 Light skina Medium skinb 36.4 8 50.0 4 38.5 25 22.6 14 36.4 8 37.5 3 16.9 11 27.9 12 Dark skinc Totals 13.7 22 5.0 8 40.4 65 23.6 38 Note. Percentages refer to those who identify as indicated who have that type of skin color. ID = p=.602. a 34.8%, n=56. b 38.5%, n=62. c 26.7%, n=43. TABLE 2 Relationship Between Appearance and Racial Identifications Among Black and White Biracials Singular Unvalidated Validated Black ID Protean ID Border ID Border ID Appearance % n % n % n % n Appears White: 00.0 0 12.5 1 4.7 3 15.8 6 could passa 4.5 1 25.0 2 10.9 7 18.4 7 Ambiguous: not assumed Blackb 50.0 11 50.0 4 62.5 40 60.5 23 Ambiguous: assumed Blackc 45.5 10 12.5 1 21.9 14 5.3 2 Appears Black: assumed Blackd Transcendent Singular ID White ID % n % n 33.3 2 27.3 6 33.3 2 40.9 9 33.3 2 31.8 7 3.7 6 13.7 22 identity. c 2 (10, N=161) = 8.28, Singular White ID % n 20.0 1 40.0 2 22.7 5 40.0 2 45.5 10 00.0 0 4.5 1 Totals 13.8 22 5.0 8 40.3 64 23.9 38 3.1 5 Note. Percentages refer to those who identify as indicated because they have that type of appearance. ID = identity. c 35.80, p=.002. a 10.7%, n=17. b 15.1%, n=24. c 56.6%, n=90. d 17.6%, n 41 Transcendent ID % n 27.3 6 2 13.8 22 (15, N=159) = 42 BRUNSMA AND ROCKQUEMORE link for this identity type because the number of respondents describing their racial identity as pro tean is small. Transcendent identity. A small number of our respondents described their racial identity as transcendent, or with no racial self-understanding (13.7%). O f these respondents, only 4.5% “a ppear B lack” and th e bulk of the transcendents (68.2%) are “ambiguous” in their socially-mediated appearance. In othe r words, one does not have the luxur y of being “race-free” if one’s phenotype is dark because for dark-skinned individuals, race impinges itself on their day-to-day existence. Border identity. The most substantial patterns highlighting the effects of appearance on racia l identity are among the respondents we categorize as having a border identity. These individuals see themselves racially as neither Black nor W hite. Their racial self-understanding is something new, not a combination of Black and White, but a new category of existence: Biracial. This identity, however, does not exist in all of these individuals’ social networks. If it does exist as a socially understood racial category then the likelihood of this identity option being validated by others is increased. If the category of Biracial does not exist, then these individuals’ racial identity is not validated by other people. Over 60% o f the Biracials w hose bor der identity is validated (60.5%), and those whose identity is not validated (62.5%) describe their appearance as “ambiguous but people assume I am Black.” However, this interpretation of others regarding these Biracials appearance is stronger for those choosing the unvalidated border identity because they are more likely to be assumed B lack than those whose border identity is validated by others. In other words, 34.2% of those choosing the validated border identity do not appear Black, whereas only 15.6% of the unvalidated borders’ a ppearan ces do. T he interpreta tions of Bira cial peoples’ appearance by others place serious para meters on the ir racial self-understandings. The question remains, whether the appearance–identity link is solely a function of the contextual experienc es of race thes e individuals e xperience or if it stands on its own. Th ese findings sup port our hyp othesis that ap pearance does affect id entity. The results indicate there is no association between self-perceived skin color and identity, yet there is a strong association between socially-mediated appearance and identity. H3 stated that appearances mediate the link between skin color and identity—that skin color influences othe rs’ interpretation s of appea rance and this ultimately influences the choice of identity. If this is true, we would e xpect skin color to be assoc iated with app earance. T able 3 pre sents the results that explore this idea. We find that there is a very strong association, within this sample, between SKIN COLOR AND BIRACIAL IDENTITY 43 skin color and they way these individuals u nderstand their appea rance, c 2(6, N = 169) = 34.50, p = .000. Among those who appear White and “could pass,” 64.7% of them have ligh t skin—the percentag e of respon dents with light skin c ontinuously decreases as we move across the table from ambiguous to appearing Black. Mo st (82.2%) of those who appear ambiguous and are no t assumed to be Black have light to medium skin. Of those who say they “appear Black” and are assumed Black by others, most (65.5%) perceive their skin color as dark. It does appear that skin color is associated with appea rance. W hat is most intere sting is that this association is not perfect. In other words, our respondent’s self-reported skin color does not completely determine their previews or ho w they perceive the reviews of others. What is less clear is whether these three primary associations of interest: skin color and identity, appe arance and identity, and skin color and appearance, are affected by contextual experie nces of race. Is the relationship between appearance and identity the same for Biracials from predominantly Black preadult contexts? Is the relationship between skin color and identity always insignificant, or does it become important if the Biracial experiences negative treatment from Blacks, Whites, or both? The se are but two questions exp lored in the following two sections. Table 3 Relationship Between Phenotype and Appearance Among Black and White Biracials Phenotype Appear White: Could Pass Ambiguous: Not Assumed Black Ambiguous: Assumed Black Appears Black: Assumed Black % n % n % n % n Light skina 64.7 11 42.9 12 32.6 31 6.9 2 Medium skinb 29.4 5 39.3 11 45.3 43 27.6 19 5.9 1 17.9 5 22.1 21 65.5 19 10.1 17 16.6 28 56.2 95 17.2 29 c Dark skin Totals Note. Percentages refer to those who have the appearance indicated and have that type of skin color. c 2 (6,N=169) = 34.50, p = .000. a 33.1%, n = 56. b 39.6%, n = 67. c 27.2%, n = 46. 44 BRUNSMA AND ROCKQUEMORE The Role of Phenotype and Appearance in Racial Identification Under Differing Contextual Experiences of Race An individual’s social context seems to be a crucial component in the relation between phenotyp e and soc ially perceived appeara nce as well as in how these two factors influence iden tity. Two important ways that these processes are parameterized are due to the influence of the racial composition of significant social networks throughou t Biracials’ soc ialization— particularly in the pread ult years—and their experiences of negative treatment from Blacks, Whites, or both. Is it possible that context influences Biracials’ very perceptions of skin color and appearance? Our results (not shown here) indicate that there is an association between Biracials perceptio ns of their own skin color and the racial composition of their preadult s ocialization c ontexts, P 2(4, N = 177) = 10.38, p = .XXX . Those Biracials who report having darker skin color came from contexts with more Whites, whereas those who perceive the ir skin color as lig hter, came fro m pread ult contexts of predominantly Blacks. This indicates that self-percep tion of one’s sk in color is affected by the range of skin tones one is surrounded by in their social networks. In addition, before lookin g at the associations of interest and how the racial composition of contexts affects th ose, it is interesting to n ote that there is also an association between the racial comp osition of these Biracials’ pr eadult con texts and their socially mediated appeara nce, P 2(6, N = 169) = 19.17, p = .XXX. Three findings are most in teresting in this regard: B iracials who sa y they appea r White and could pass came from predominantly Black contexts, those who are ambiguous and not assumed Black by others were more likely to come from predominantly White contexts, and those who report that they look Black and are assumed to be Black by others came from p redominantly Black so cial networks. These findings should be kept in mind as we look at the influence of context on the associations of interest, and they point to the fact that supports our expansion of Stone’s (1962) model: Percep tions and self-un derstandin gs are heavily influenced by context and interactional experiences. The Role of Early Contextual Experiences of Race: Racial Composition of Pre-Adult Social Networks Biracial individuals exist in a variety of social contexts an d are socia lized by a myriad of agents—some who are Black and some who are White (and some who are Biracial them selves). In this section we report on how these different contextual SKIN COLOR AND BIRACIAL IDENTITY 45 experiences of race influenc e the primar y processe s of interest in this study. Tab le 4 reports the tests of association (chi-square) for each of the processes under investigation broken d own by the ra cial composition of these respo ndents’ pre-a dult contexts. To reiterate, for the entire sample of B iracial individu als, regardless of context, there was no association between skin color and identity (see Table 1). This relation remains insignificant for both predominantly White contexts and contexts that had a mixture of Blacks and Whites. However, there is a significant relation between skin color and racial identity among Biracials who grew up and interacte d in predom inantly Black contexts. On closer look at the data, it appears tha t those with darker skin are more likely to choose a singular Black identity (65%), whereas 61.3% of those with lighter skin and 57.9% of those with medium sk in are more likely to adopt the unvalidated Border identity. This means that dark-skinned Biracial individuals who grew up in predominantly Black contexts developed a singular Black identity more than any other identity. Furthermore, those Biracial individuals whose skin ra nged from light to medium and who w ere socialize d in primarily Black co ntexts are more likely to say that they understand themselves as Biracial, but they experience the world as a Black person. The fact that the relationship between sk in color and identity is salient within individuals who grew up in predominately Black preadult contexts provides further evidence that the Black community sees shades of color whereas Whites see only two colors: Black and White (see Russell et al., 1992). The appearance–identity association was significant for the entire sample of Biracials (see Table 2). If we look at the second row of results—those for the association between appearance and identity—we see that, overall, the association is significant regardless of the racial composition of preadult contexts of these Biracials. Thus, it appears that the racial co mposition of Biracials’ e arly interactional contexts does not mediate the relationship between appearance and identity choice. However, our hypothesis was that context will moderate this relationship — meaning that the relation will be somewhat stronger or weaker depending on the context. This is in fact the case. T he appe arance–id entity link is strongest in mixed W hite and B lack pread ult contexts, P 2(X, N = XXX) = 49.70, p = .XXX. This seems to indicate that these contexts are more complex and more full of a variety of pr eviews and reviews that nee d to be taken into consideration. A closer look at the data indicate that in these mixed contexts: (a) those who “appear White and could pass” developed a validated border identity (80%), (b) many of those who are “ambiguous and assumed Black” developed an unvalidated border identity (40.4%), and (c) most of those who “appear Black and are assumed Black” develop ed a singular B lack identity (68.8%). The appearance– identity link is weakest in predom inantly Black preadult co ntexts, P 2(X, N = XX) = 25.38 , p = 46 TABLE 4 Relationships Between Phenotype, Appearance, and Racial Identity by Racial Composition of Preadult Contexts P Skin Color x Identity Appearance x Identity * p < .05. * * p < .01. * * * p < .001. 2 7.55 30.51** Predominantly Black Preadult Contexts Mixed White and Black Preadult Contexts Predominately White Preadult Contexts Relationship/Association df n P 10 56 15 55 2 df n 12.44 10 72 49.70*** 15 72 P 2 df n 22.15* 10 70 25.38* 15 68 SKIN COLOR AND BIRACIAL IDENTITY 47 .XXX, primarily because the bord ers’ experiences and self-understand ings are less likely to be validated in predom inantly Black social netwo rks (see Rus sell et al., 1992) . Finally, we demonstrated how the association between skin color and appearance for the entire sample was significant (see Table 3). The relationship between skin color an d appea rance app ears heavily m oderated by social co ntext. Looking at the Row 3 of Table 4, we see that skin colo r is not associate d with appearance in predominantly White contexts and the most strongly associated with appearance in predominantly Black contexts. Again, it appears that the Black commu nity can, and d oes, distinguish among the variation in skin tones among Biracials, whereas, possibly the White community sees only two “colors”—Black and White. The Role of Push Factors in Identity: The Experience of Negative Treatment from Blacks and Whites Most of the Biracials have experienced negative treatment from both Blacks and Whites. We wanted to investigate whether this particular contextual experience of race moderates the three associations of interest. Table 5 summarizes the findings regarding this issue. Experiencing a push factor of this kind from Blacks might lead a Biracial of darker skin not to adopt a Black identity. On the other hand, a light skinned Biracial who experiences negative treatment from W hites may cho ose to adopt a Black identity. We also look at that group of Biracials who report experiencing negative treatmen t from both groups. TABLE 5 Relationship Between Phenotype, Appearance and Identity by Experiences of Discrimination of Blacks and Whites Experienced Discrimination from Blacks Relationship/ Association P * 2 df n Experienced Discrimination from Whites P 2 df n Experienced Discrimination from Whites and Blacks P 2 df n Skin Color x Identity 24.13** 10 121 23.57** 10 152 19.13* 10 105 Appearance x Identity 34.73** 15 119 54.84*** 15 149 31.75** 15 103 Skin Color x Appearance 33.92*** 6 125 51.05*** 6 159 28.02*** 6 109 p < .05. * * p < .01. * * * p < .001. 48 BRUNSMA AND ROCKQUEMORE The association between skin color and identity for o ur sample of Biracials d id not exist in previous analyses (see Table 1 ); however, this relationship does exist for Biracials who have exp erienced n egative treatm ent from B lacks, P 2(X, N = XXX) = 25.13 , p = .XXX; from W hites, P 2(X, N = XX X) = 23 .57, p = XXX; and from both groups, P 2(X, N = XXX ) = 19.13 , p = XXX. A closer look at the data, illustrates the push factors. Over thirty percent (31.4%) of dark-skinned Biracial individuals who have experienced negative treatment from Blacks adopt a singular Black identity, compared to 42.6% of similar Biracials who have experienced such treatment from W hites. This kind of contextua l experienc e of race do es seem to modera te the appearance– identity association in our data— because the association is still significant (see Table 2), but stronger in the case of Biracials who have experienced negative treatment from W hites. The same app lies to skin color and appearance. Although the extent to which social context influence racial identity formation of Biracial respondents remains somewhat complicated, the fact that social context does indeed influence racial identity formation is undeniable. Discussion When focusing on racial identity, the use of appearances as signifiers of group membe rship is not always clear cut. T his relationship between appearances and individual identity choice becomes even more complex when we examine Biracial identities. We have presented the various racial self-understandings of a sample of Biracial people to explore how the appearance–identity link may function. It is clear that the basic processes of mutual identification and concomitant awareness help us to better understand two ou tcomes. The first is how people with the same parental background (one Black and one White parent) can make very different choices about their identities. The second is to try to understand why one’s appearance does not always predict that outcome—or why we find pe ople who are physically White who identify as Black and those who are physically Black who say they are not Black, but Biracial. Trying to interpret these cases becomes clearer if we focus on the basic pro cesses that take place betw een individu als and within the individual (in the proce ss of “self-ing “). The first process, mutual identification, is critica l to both iden tity construction and maintenance. If the individual actor (regardless of bodily char acteristics) exists within a social context where Biracial has a meaningful existence, he or she may cultivate a border identity. If this cultural category does not exist and one becomes accustomed to and ade pt at switching from Black to White, they will cultivate a protean identity. If their app earance is W hite, they may develop a transcendent SKIN COLOR AND BIRACIAL IDENTITY 49 identity, but only if their social context does not demand categorization. If none of these options are available to an individual, than the existing cultural no rms dictate the racial identity above and beyond (and at times despite) their appearance. The second process at work allows us to see the self-functioning as subject. How does the self reflect on o ne’s own ide ntity-in-context? In some cases (the protean identity), we saw the self as not only having the capacity for co ncomitant awareness, but also utilizing that capacity in all interactions as a survival tool. The individual was able to assess the anticipated interpretatio n from the app roaching o ther, to modify the program accordingly, and to continually monitor for necessary renegotiations or reformulations of the program. This intense activity requires an ongoin g subjective awareness of the self’s activities and much bicultural (or identity) capital (see Côté, 1996). We see this same process co ntinually at work in the other ide ntity options, ho wever, it takes o n a slightly variant for m of activity. A new body of literature focuses on the sp ace between the prev ious two processes of mutual identification (a fundamentally social process) and concomitant awareness (a process rooted in contextualized psychological phenomena) by acknowledging that identity is a dynamic journey not a stagnant dest ination. Our data indeed point in this direction as well. This growing collection of studies highlight that identity constru ction and m aintenance in volves strateg ies (e.g., Berzo nsky, 1999c; Blanz, 1998), negotiations (e.g., Gilliam & Gilliam, 1999; Swann, 1987), and tactics. Indee d, these authors approa ch this new area from several directions. Some focus on the ways that individuals “do identity” on an interaction by interaction basis through processe s of “identity work” (Storrs, 1999) and uses of “identity capital” within a variety of salient “identity markets” (Côté, 1996, 1997). Other investigators concentrate on social-cognitive models of identity formation where individuals strategize and make designed choic es as to whether they will engage in o r avoid alto gether the co nstruction or r evision of their se lfidentities (see Berzonsky, 1989, 1992, 1994, 1999a, 1999b). Research on Biracial and multiracial identity would also do well to understand the existence o f multiple identities within one individual, the functions each plays, the contextual enactment of these identitie s, and how th ey change o ver time (D eaux, 199 3; Deaux et al., 1995; Ethier & Deaux 199 4). These emerging theoretical approaches to the study of identity avoid a n over-soc ialized view o f the self and reflect the level of complex ity illustrated by data such as ours. All of these theoretical schemas may prove beneficial in our understanding of the appearance-identity link in particular and the formation and maintenance of identity in general. Those researching Biracial and multirac ial individuals w ould do well to ackno wledge th ese theories because they cross disciplinary lines and p aradigms. It remains to be seen how much negotiation and strategy is involved in racial identity formation and maintenance among Black and W hite Biracials; however, it is still the case that racial identities are subject to a degree of constraint that ethnic identities are not. Specifically, racial identities are constrained by historical stratification that is directly tied to bodily characteristics. It is these characteristics 50 BRUNSMA AND ROCKQUEMORE that linger in our cultural symbols and traditional ways of thinking and being. The existence, experience and voice of Biracial individuals is challenging the current categorization of race, altering the meaning of racial identity, and, in the process, altering the cultural toolkit (Swidler, 1986) of race in the United States. For individuals who are caught in between existing cultural categories at this time of uneven but emerging changes in those categories, appearances remain a significant constraining factor in identity construction and maintenance. REFERENCES Alba, R. (1990). Ethnic identity: The transformation of White America. New Haven, CT: Yale University Press. Berzonsk y, M. (1989). The self as theorist: Individual differences in identity formation. International Journal of Personal Construct Psychology, 2, 363–376. Berzonsky, M. (1992). Identity style and coping strategies. Journal of Personality, 60, 771–788. Berzonsk y, M. (1994). Individual differences in self-construction: The role of constructivist epistemological assumptions. Journal of Constructivist Psychology, 7, 263–281. Berzonsk y, M. (1999a). Identity processing style and cognitive attributional strategies: Similarities and difference across different contexts. European Journal of Personality, 13, 105–120. Berzonsk y, M. (1999b). Identity styles and hypothesis-testing strategies. Journal of Social Psychology, 139, 784–789. Berzonsk y, M. (1999c). Reevaluating the identity status paradigm: Still useful after 35 years. Development Review, 19, 557–590. Blanz, M., et al. (1998). Responding to negative socialidentity: A taxonomy of identity management strategies.European Journal of Social Psychology, 28, 697–729. Bonilla-Silva, E. (1997). Rethinking racism: Toward a structural interpretation. American Sociological Review, 62, 465–480. Bowles, D. D. (1993). Bi-racial identity: Children born to African-American and White couples. Clinical Social Work Journal, 21, 417–428. Boyd-Franklin, N. (1989). Black families in therapy: A multisystems approach. New York: Guilford. Brown, P. M. (1990). Biracial identity and social marginality. Child and Adolescent Social Work Journal, 7, 319–337. Cooley, C. H. (1902). Human nature and the social order. New York: Scribner’s. Côté, J. E. (1996) . Sociological p erspectives on id entity formatio n: The culture-iden tity link and identity capital. Journal of Adolescence, 19, 417–428. Côté, J. E. (1997). An empirical test of the identit y capital model. Journal of Adolescence, 20, 577–597. Cross, W. E. (1971). The Negro-to-Black conversion experience: Toward a psychology of Black liberation. Black World, 20, 13–27. Daniel, R. (1996). Black and W hite iden tity i n the new mi llenn ium: Unsev ering the ties that bind . In M. Root (Ed.), The multiracial experience: Racial borders as the new frontier. Thousand Oaks, CA: Sage. Davis, F. J. (1991). Who is Black? One nation’s definition. CITY: Pennsylvania State University Press. Deaux, K. (1993). Reconstructing social identity. Personality and Social Psychology Bulletin, 19, 4–12. SKIN COLOR AND BIRACIAL IDENTITY 51 Deaux, K., et al. (1995). Parameters of social identity. Journal of Personality and Social Psychology, 68, 280–291. Demo, D., & Hughes, M. (1990). Socialization an d racial identity among Black Americans. Social Psychology Quarterly, 53, 364–374. Ethier, K., & Deaux, K. (1994). Negotiating social identity when contexts change: Maintaining identification and responding to threat. Journal of Personality and Social Psychology, 67, 243–251. Field, L. D. (1996). Piecing together the puzzle: Self-concept and group identity in biracialBlack/White youth. In M. Root (Ed.), The multiracial experience: Racial borders as the new frontier. Thousand Oaks, CA: Sage. Gans, H. J. (197 9). Symbo lic ethn icity: The future of ethnic groups and cultures in America. Ethnic and Racial Studies, 2, 1–20. Gergen, K. J. (1991). The saturated self: Dilemmas of identity in contemporary life. New York: Basic. Giddens, A. (1991). Modernity and self-identity: Self and society in the late modern age. Stanford, CA: Stanford University Press. Gilliam, A., & Gilliam, O. (1999). Odyssey: Negotiating the subjectivity of Mulatta i dentity in Brazil. Latin American Perspectives, 26, 60–84. Goffman, E. (1959). The presentation of self in everyday life. New York: Anc hor Doub leday. Hall, C. (1980). The ethnic identity of racially mixed people: A study of Black-Japanese. Unpublished doctoral dissertation, University of California, Los Angeles. Harvey, A. R. (1995, January). The issue of skin color in psychotherapy with African Americans. Families in Society: The Journal of Contemporary Human Services, 3–10. Herring, R. D. (19 95). Devel oping bi racial et hnic id entity: A review of the increasing dilemma. Journal of Multicultural Counseling and Development, 23, 29–38 Hughes, M., & Hertel, B. R. (1990). The significance of color remains: A study of life chances, mate selection, and ethnic consciousness among Black Americans. Social Forces, 68, 1105–1120. Ichheiser, G. (1970). Appearances and realities: Misunderstanding in human relations. San Francisco: Jossey-Bass. Ignatiev, N. (1995). How the Irish became white. New York: Routledge. Johnson, D. J. (1992). Developmental pathways: Toward an ecological theoretical formulation of race identity in Black–White Biracial children . In M. Root (Ed.), Racially mixed people in America. Beverly Hills, CA: Sage. Kerwin, C., Ponterotto, J. G., Jackson, B. L., & Harris, A. (1993). Racial identity in Biracial children: A qualitative investigation. Journal of Counseling Psychology, 40, 221–231. Kich, G. K. (1992). The developmental process of asserting a Biracial, Bicultural identity. In M. Root (Ed.), Racially mixed people in America. Beverly Hills, CA: Sage. Lifton, R. (1993). The protean self: Human resilience in an age of fragmentation. New York: Basic. Loveman, M. (1999). Is “race” essential?: Comment on Bonilla-Silva 1997. American Sociological Review, 64, 891–898. Mead, G. H. (1934). Mind, self, and society. Chicago: University of Chicago Press. Miller, R., & Miller, B. (1990). Mothering the biracial child: Bridging the gaps between AfricanAmerican and White parenting styles. Women and Therapy, 10, 169–180. Morten, G., & Atkinson, D. (1983). Minority identity development and preference for counselor race. Journal of Negro Education, 52, 156–161. Neal, A. M., & Wilson, M. L. (1989). The role of skin color and feat ures in t he Black communi ty: Implications for Black women and therapy. Clinical Psychology Review, 9, 323–333. Porter, C. P. (1991). Social reasons for skin tone preferences of Black school-aged children. American Journal of Orthopsychiatry, 61, 149–154. Porter, J., & Washington, R. (1993). Minority identity and self-esteem. Annual Review of Sociology, 19, 139–161. 52 BRUNSMA AND ROCKQUEMORE Porterfield, E. (1978). Black and White marriages: An ethnographic study of Black–White families. Chicago: Nelson-Hall. Poston, W. C. (1990). The Biracial identity development model: A needed addition. Journal of Counseling and Development, 69, 152–155. Poussaint, A. (1984). Study of interracial children presents positive picture. Interracial Books for Children, 15, 9–10. Rockquemore, K. A. (1999). Between Black and White: Understanding the “Biracial” experience. Race and Society, 2, 197–212. Root, M. (1990). Resolving “other” status: Identity development of biracial individuals. Women and Therapy, 9, 185–205. Root, M. (Ed.). 1992. Racially mixed people in America. Newbury Park, CA: Sage. Root, M. (Ed.). 1996. The multiracial experience: Racial borders as the new frontier. Thousand Oaks, CA: Sage. Russel, K., Wilson, M., & Hall, R. (1992). The color complex: The politics of skin color among African Americans. New York: Anchor. Spickard, P. R. (1989). Mixed blood: Intermarriage and ethnic identity in twentieth-century America. Madison: University of Wisconsin Press. Stone, G. (1962). Appearance and the self. In A. M. Rose (Ed.), Human behavior and social processes. Boston: Houghton Mifflin. Storrs, D. (1999). Whiteness as stigma: Essentialist identity work by mixed-race women. Symbolic Interaction, 22, 187–212. Swann, W. B., Jr. (1987). Identity negotiation: Where two roads meet. Journal of Personality and Social Psychology, 53, 1038–1051. Swidler, A. (1986). Culture in action: Symbols and strategies. American Sociological Review, 51, 273–286. Taylor-Gibbs, J. (1997). Biracial adolescents. In J. T. Gibbs & L.-N. Huang (Eds.), Children of color: Psychological interventions with culturally diverse youth. New York: Jossey-Bass. Tizard, B., & Phoenix, A. (1995). The identity of mixed parentage adolescents. Journal of Child Psychology and Psychiatry, 36, 1399–1410. Turkle, S. (1997). Life on the screen: Identity in the age of the internet. New York: Touchston e. Wardle, F. (1987). Are we sensi tive to interrac ial children’s special ident ity needs? Young Children, 43, 53–59. Wardle, F. (1992). Biracial identity: An ecological and developmental model. CITY/STATE: Center for the Study of Biracial Children. Waters, M. (1990). Ethnic options: Choosing identities in America. Berkeley: University of California Press.