Abstract
The color line is still a central problem in the United States, as Du Bois declared more than a century ago. But economic, demographic, and social trends have subdivided it in ways that Du Bois could not have foreseen, creating tremendous intra-ethnoracial group diversity. A challenge for twenty-first-century scholarship is to make sense of the implications of growing intra-group diversity for the boundaries and meaning of group identity. Meeting this challenge requires treating intra-group diversity not merely as an outcome of various social processes. Intra-group diversity must also be seen as the origin of processes shaping the boundaries and meanings of group identities, as well as intergroup attitudes and relations. Meeting the challenge also necessitates adopting ethnographic and survey research practices that better capture the dynamism of the multiple color lines defining the American ethnoracial landscape and the implication of this dynamism for identity.
Keywords
Over a century ago, W. E. B Du Bois presciently proclaimed, “the problem of the Twentieth Century is the problem of the color-line” (1903:9) He was writing about the stark boundary between blacks and whites at the time. The color line of the twenty-first century, in contrast, is not so clear-cut. Economic, demographic, and social trends subdivide it in ways that Du Bois could not have foreseen. Now more than ever, economic inequality marks significant and varied divisions in the fortunes of individuals who share an ethnoracial origin. The Gini ratio, a standard measure of income dispersion within a population, for instance, has shot up for black, white, and Hispanic households since the late-1960s (Fields, Chavez, and Coddou 2013; see Figure 1), alongside differential levels of segregation and patterns of increasing residential integration (Logan and Zhang 2011). Among blacks, the most historically segregated group, the middle and upper classes have grown more residentially distant from their poorer counterparts (Sharkey 2014).

Gini ratios for households in the United States by race/ethnicity, 1967 to 2010.
Mass immigration has pushed the color line even further, challenging existing ethnoracial and panethnic boundaries beyond black and white (Lee and Bean 2010; Okamoto and Mora 2014). Assimilation across the U.S.-born generations of these groups adds linguistic, legal-status, and socioeconomic variation to all groups, but especially to Asians and Latinos (Alba, Jiménez, and Marrow 2014; Waters 1999; Wong et al. 2011). And partnering across ethnoracial lines has given rise to a population that increasingly recognizes itself and is recognized by others for having multiple ethnoracial ancestries (DaCosta 2007; Lee and Bean 2010; Wang 2012; Williams 2005). In sum, although the color line is still a central problem in the United States, that line has splintered a great deal since Du Bois’ time.
A challenge for twenty-first-century scholarship is to make sense of the implications of growing intra-group diversity for boundaries and meanings of group identity. Meeting this challenge requires treating intra-group diversity not merely as an outcome of various social processes. Intra-group diversity must also be treated as the origin of processes shaping the boundaries and meanings of group identities, as well as intergroup attitudes and relations. Meeting the challenge also necessitates adopting ethnographic and survey research practices that more effectively capture the dynamism of varied color lines defining the American ethnoracial landscape and the implication of this dynamism for identity.
Why Intra-group Diversity Matters
Some scholars see growing intra-group diversity as an indicator of both progress and stagnation when it comes to ethnoracial equality. Treating these trends as outcomes, research has generally come to a “two-handed” conclusion: on one hand, minorities have made tremendous economic and social progress, especially since the end of the Civil Rights Movement; on the other hand, there is sobering evidence of stagnated or even reversed progress (Bobo 2011; Telles and Ortiz 2008). Scholars and pundits use both sides when commenting on the status of U.S. ethnoracial minorities. Yet emphasizing intra-group diversity as an outcome of various social processes misses the fact that intra-group diversity also functions as a driving mechanism in the formation, activation, and expression of group identities. Indeed, intra-group diversity acts as both a dependent and an independent variable.
Research on class differences among African Americans provides a clear illustration. For African Americans, class mobility in relation to residential segregation is particularly central owing to historically high levels of spatial isolation. William Julius Wilson was an early observer of the broad economic and social forces that created a diverse experience of blackness, setting up one set of experiences for middle-class blacks and another for poor and working-class blacks (Wilson 1987). 1 Wilson treats class diversity among blacks primarily as an outcome: the black middle-class exodus from black communities produced a vacuum of middle-class sensibilities thought to be beneficial to segregated neighborhoods.
The next generation of research, however, began to treat this within-group heterogeneity as a cause, turning to the implications of black socioeconomic and residential mobility for identity. Pattillo (1999, 2007), for instance, showed that when the black middle class exited these neighborhoods, they did not entirely distance themselves from their poor ethnoracial brethren. Middle-class blacks navigate between their middle-class status and sensibilities, and a black identity that connects them to poorer African Americans, keeping them socially distant from middle-class whites. These class divisions come into fuller focus when middle-class and poor blacks work out what it means to be “black” in a neighborhood where both live side-by-side, but navigate different class contexts shaping their ideas about black authenticity (Pattillo 2007). But there is also a black middle class that is residentially distant from poorer blacks (Sharkey 2014). As Lacy (2007) shows, black middle-class suburbanites work to assert a sense of black identity consistent with contemporary conceptions of blackness. Some even exhibit the trappings of the “symbolic ethnicity” (Gans 1979) found among white ethnics. Residential distance from blacks in general, and poor blacks in particular, then, holds consequences and offers middle-class blacks latitude in how they fashion an ethnoracial identity.
Similarly, the intra-group diversity among Asians and Latinos is ripe for treatment as both outcome and cause. For Asians and Latinos, intra-group inequality often stems from immigration-driven compositional population change and assimilation. Even as an immigrant generation is prominent, assimilation makes second- and third-generation individuals distinctive from the first generation with respect to socioeconomic status (Park and Myers 2010), linguistic repertiore (Rumbaut, Massey, and Bean 2006), legal status (Bean et al. 2011; Yoshikawa 2011), and, increasingly, region of residence (Marrow 2011). Intra-group diversity also plays a key causal role in the formation and activation of group identities. Scholarship on panethnicity demonstrates that the categories of “Asian” and “Latino” are based on constructed—and contested—boundaries meant to include members of multiple ethnoracial and national-origin groups (Okamoto and Mora 2014). The meaning and importance of ethnoracial identity is, in part, driven by the different experiences of members in the group. For example, while assimilation generally leads to the development of panethnic identities, Schachter (2014a) found that immigrants from India—who often experience marginalization within the Asian American community—are less likely to identify themselves as Asian or Asian American when they are integrated into communities with large non-Indian, Asian populations.
Differences in national and ethnoracial origin are just one component of intra-group diversity. Even when individuals at different levels of assimilation recognize a similar ancestry, there is potential for more (or less) assimilated members of the group to shape the outlines of identity for all group members. 2 This dynamic is apparent in the experiences of later-generation members of groups that continue to be replenished by contemporary immigration. Ample ethnographic research shows how later-generation individuals of Polish, Mexican, Chinese, and Japanese descent construct identities in a context where immigrants of the same ethnoracial origin define the culture and boundaries experienced by people several generations removed from an immigrant generation (Erdmans 1998; Jiménez 2010; Tuan 1998; Vasquez 2011). And, level of assimilation—which varies among Asians and Latinos—shapes the sense of obligation to less well off members of the group that the more upwardly mobile members of the group exhibit. For instance, among second-generation children of poor immigrant groups who “make it,” economic striving can pull them in one direction whereas obligations of familial and coethnoracial networks made up of poorer individuals pull them in another (Agius Vallejo 2012). 3
Perhaps no factor has been more important to ethnoracial identity in the last three decades than romantic partnerships across ethnoracial lines, and the children these unions produce. These processes have historically been viewed as an outcome related to low social distance between groups (Gordon 1964; Waters and Jiménez 2005). 4 But tremendous growth in intermarriage (Wang 2012) and the institutionalized recognition of multiethnoraciality (Humes, Jones, and Ramirez 2011; Khanna 2012) are also important causes of contestation over the boundaries and content of ethnoracial groupings. Among individuals with multiple ethnoracial ancestries, there is simultaneously a desire for recognition of membership to a distinctive multiracial identity and an effort to be seen as belonging to the constituent groups to which they trace their various ancestries (Jiménez 2004; Williams 2005). With connections to multiple ethnoracial strands, these individuals often see the components of their ancestry as additive to an identity that recognizes more than one background. This recognition stretches group boundaries and their meaning such that organization and individuals committed to traditional outlines of ethnoracial groups are forced to contend with the membership of people with multiple ancestries (Lee and Bean 2010; Williams 2005).
Intra-group Variation and Out-group Attitudes
Intra-group diversity clearly drives the identity formation and expression within ethnoracial groups. But growing intra-group diversity is equally important to intergroup relations. Scholars have long recognized intra-group diversity to be a feature of identity construction from the inside, a research focus that is needed now more than ever. But how individuals perceive intra-group diversity among other groups, and the implications of these perceptions, is a topic warranting much more attention. New ethnographic research shows the perceptions of diversity within other ethnoracial groups matter for how definitions of group cohesion are constructed in diverse settings. For example, Jiménez’s (forthcoming) research in a black-majority-turned-Latino-majority city shows that the boundaries African Americans perceive amid dramatic immigration-driven change around them are not merely ethnoracial. Black residents treat speaking English and neighborhood tenure as important group boundaries that cut across ethnoracial lines. Thus, what some might describe as “black/brown” relations might be better described as “English-speaking/non-English-speaking,” “native/foreign-born,” or “longtime-residents/newcomer” relations (also see Watson and Saha 2013; Wimmer 2004; Woldoff 2011).
Survey research on out-group attitudes has been even slower to pay attention to the effects of intra-group differences. Survey questions about ethnoracial groups have come a long way from a time when intergroup relations were synonymous with black–white relations. But surveys still ask about “Latinos,” “Asians,” “whites,” and “blacks” as if respondents recognize none of the intra-group diversity with which the group members themselves increasingly contend (Bobo et al. 2000). New survey research offers broader support for ethnographic research showing the importance of intraethnoracial group diversity for intergroup relations. Schachter (2014b), for instance, developed a survey experiment that tests whether Americans hold different stereotypes about native and foreign-born whites, blacks, Latinos, and Asians. Preliminary results for a national sample of native-born-white respondents collected by YouGov (n = 1,397) in January of 2014 demonstrate that native-born white Americans are strongly aware of within-group diversity.
According to Schachter’s data, summarized in Figure 2, native-born whites hold distinct attitudes about native- and foreign-born members of racial groups, including their own. Native-born whites view Latino and Asian immigrants significantly more negatively than their native-born counterparts, although this disadvantage is most pronounced for Latino immigrants. While native-born whites do not as strongly distinguish between U.S.- and foreign-born blacks, the results trend in the opposite direction, suggesting that they hold somewhat lower levels of prejudice toward black immigrants than the native-born. These within-group differences have significant implications for between-ethnoracial group comparisons: The ethnoracial hierarchy shifts dramatically depending on examination of attitudes toward the native-born or immigrants. Blacks are the most negatively rated native-born group, while Latinos are the most negatively rated immigrant group, and Latino immigrants are the most negatively rated group overall. Yet, when only examining attitudes toward ethnoracial groups without specifying nativity (indicated by the circles on the graph), this nuance is lost; Latinos and blacks are rated equally negatively when no nativity information is given. These data offer broad-based evidence that intra-group differences resonate with ethnoracial group outsiders and are important determinants of intergroup attitudes.

Native-born whites’ views of race groups by nativity categories.
Changing Research Practices and Intra-group Diversity
The research we cite above admittedly only scratches the surface of what ought to be a more robust research enterprise examining the implications of intra-group diversity. Launching that enterprise requires research practices to better reflect intra-ethnoracial group differences that now pervade. A guiding heuristic of such an approach heeds Brubaker’s (2004) call to treat ethnoracial origin as a category to be investigated rather than assumed. In his “ethnicity-without-groups” approach, Brubaker notes that too often scholars assume the groups studied to be cohesive units, when the way group members and group outsiders conceive of the boundaries and content of group identity should be the very question under consideration. The relevance of that heuristic is clear in ethnographic research, which is more amendable to the inductive requirement of studying ethnoraciality without groups. Examples include Jiménez’s (2010) study of how recent Mexican immigrants shape the identities of later-generation Mexican Americans in Kansas and California, Fields’ (2011) examination of how competing definitions of black interests structure black political behavior, Pattillo’s (2007) examination of interactions between poor and middle-class blacks in a context of gentrification, McDermott’s (2006) and Hartigan’s (1999) study of poor whites, and Brubaker et al.’s (2007) study of ethnoracial and national identity in Transylvania, a region that has changed hands between Hungary and Romania multiple times.
As a corollary to Brubaker’s heuristic, making sense of ethnoraciality would also be well served by research strategies that distinguish how ethnoracial origin operates at both individual and collective levels. Growing Intra-group diversity in some ways individualizes the experience of ethnoracial group membership, making it important to specify how the “effects” of ethnoracial origin (as well as the outcomes of racialized treatment) might vary depending on its operationalization. For instance, Kalev, Dobbin, and Kelly (2006) found that corporate affirmative action and diversity programs have little effectiveness in increasing the share of ethnoracial minorities in management. That is, minorities do not benefit much from diversity efforts that simply embed them, compositionally, in majority white workplaces. However, Fields et al. (2013) found that being an ethnoracial minority in majority white workplaces is associated with a wage premium. Nonwhites who work in white contexts make more money than their coethnoracials, who do similar work in contexts characterized by members of the same group. 5 So while diversity programs may not help minorities in the aggregate, the individuals who are hired through them might see benefits. 6
Another example of scholarship that operationalizes intra-group diversity as a cause at the collective level is Okamoto’s (2003, 2006) focus on the conditions under which panethnic Asian American organizations and other forms of collective action are more or less likely to occur. And yet, as Schachter’s (2014a) research on Indian immigrants shows, such identities may not form at the individual level.
These heuristics are just as relevant, if more challenging to apply, for survey research. Capturing within-group variation among whites, in both how they perceive white identity and how they view differentiation in other groups, is possible in nationally representative surveys, such as the one described above that Schachter (2014b) used to study white attitudes toward native- and foreign-born members of various ethnoracial groups. Oversamples of minority populations, like the 2008 and 2012 National Election Surveys, or separate surveys, like the 2008 National Asian American Survey (Ramakrishnan et al. 2008), and 2006 National Latino Survey (Fraga et al. 2006), can generate the statistical power necessary to test for within-group differences (Segura and Rodrigues 2006). The increasing availability and declining costs of fielding original surveys allow social scientists to design and field surveys to better capture identity and attitudes when survey respondents are prompted to report their views of out-group members with different socioeconomic, nativity, legal status, and linguistic attributes. With this intra-group diversity in mind, survey instruments will better capture the intra-group diversity that people in the United States increasingly encounter.
Conclusion
History teaches that ethnoracial origin is an enduring feature of individual and collective life, even if it changes in form and meaning. We argue that the twin engines of socioeconomic inequality and assimilation have added to intra-group diversity in ways that complicate ethnoracial identity. Making sense of ethnoracial identity in view of significant intra-group diversity requires more attention to the ways the widening distribution of attributes within groups shapes how both out-groups and in-groups construct identity. An increased focus on intra-group diversity need not take away from tracking trends and theorizing about how ethnoracial origin shapes different outcomes and life chances between groups. Indeed, there is little doubt that ethnoracial origin matters in ways all too consistent with the past. Understanding how it matters for identity now and in the future requires maintaining a focus on the multiple consequences of within- and between-group variations.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
