An open access publication of the American Academy of Arts & Sciences
Winter 2005

One drop & one hate

David A. Hollinger
View PDF

David A. Hollinger, a Fellow of the American Academy since 1997, is Preston Hotchkis Professor of American History at the University of California, Berkeley. He has written extensively on the history of American science, religion in America, and the changing role of racial and ethnic factors in the country’s history. His two most recent books are Postethnic America (1995) and Science, Jews, and Secular Culture (1996).

Two portentous practices within the public discussion of ‘race’ in the United States since the late 1960s are rarely analyzed together. One is the method by which we decide which individuals are ‘black.’ The other is our habit of conflating the mistreatment of blacks with that of nonblack minorities. Both practices compress a great range of phenomena into ostensibly manageable containers. Both function to keep the concept of race current amid mounting pressures that threaten to render it anachronistic. Both invite reassessment at the start of the twenty-first century.

The prevailing criterion for deciding who is black is of course the principle of hypodescent. This ‘one drop rule’ has meant that anyone with a visually discernable trace of African, or what used to be called ‘Negro,’ ancestry is, simply, black. Comparativists have long noted the peculiar ordinance this mixture-denying principle has exercised over the history of the United States. Although it no longer has the legal status it held in many states during the Jim Crow era, this principle was reinforced in the civil rights era as a basis for antidiscrimination remedies. Today it remains in place as a formidable convention in many settings and dominates debates about the categories appropriate for the federal census. The movement for recognition of ‘mixed race’ identity has made some headway, including for people with a fraction of African ancestry, but most governments, private agencies, educational institutions, and advocacy organizations that classify and count people by ethnoracial categories at all continue to perpetuate hypodescent racialization when they talk about African Americans.1

This practice makes the most sense when antidiscrimination remedies are in view. If discrimination has proceeded on the basis of the one drop rule, so too should antidiscrimination remedies. But even when antidiscrimination remedies are not at issue, most Americans of all colors think about African American identity in either/or terms: you are black, or you are not. It is common for people to say, “I’m half Irish and half Jewish” without one’s listener translating the declaration into terms other than the speaker’s. One can even boast, “I’m one-eighth Cherokee” without causing the listener to quarrel with that fraction or to doubt that the speaker is basically a white person. But those who say things like “I’m half Irish and half black” are generally understood really to be black, and “I’m one-eighth African American” is not part of the genealogical boasting that infuses American popular culture.

The second portentous practice is the treating of all victims of white racism alike, regardless of how differently this racism has affected African Americans, Latinos, Indians, and Asian Americans, to say nothing of the subdivisions within each of these communities of descent. When federal agencies developed affirmative action programs in the late 1960s, they identified Asian Americans, Hispanics, and Indians along with African Americans as eligible groups. As John Skrentny has shown, entitlements for nonblack groups were predicated on the assumption that such groups were like blacks in their social experience.2 Other disadvantaged groups, including women, impoverished Anglo whites, impoverished European ethnics, and gays and lesbians, were less successful in gaining entitlements during the so-called minority rights revolution because they were not perceived as victims of white racism. Yet the officials who designed entitlement programs for the purposes of remedying white racism often homogenized those descent groups colloquially coded as black, brown, red, and yellow. There was a good reason for this. White racism was real, had expressed itself against every one of these color-coded groups, and was a problem in American life that demanded correction.

The notion that all descent groups whose ancestry could be located outside Europe were like blacks, however, had not been prominent previously in the proclaimed self-conception of these nonblack minority groups, nor in much of what public discussion there was of their history and circumstances. The histories of each of these communities were almost always presented to their own members as well as to the society at large in terms that took their differences into account, including the specific ways in which whites had abused them. These histories, moreover, were usually about particular descent groups, such as Chinese Americans or Mexican Americans, rather than about what came to be called ‘panethnic’ groups, such as Asian Americans and Latinos.3 Japanese Americans had been subject to property-owning restrictions and had been incarcerated without due process during World War II, and all but a few immigrants from Asia had been denied naturalization until 1952. Immigrants from Mexico had always been able to achieve citizenship and were not included in the miscegenation laws that prevented nonwhites from marrying whites, but these immigrants and their descendants had been subject to other abuses, including school segregation and exclusion from juries in many jurisdictions until courts eliminated these practices in the decade after World War II. Mexican Americans, moreover, despite their overwhelmingly immigrant origins, did come from a country that had lost territory to the United States, and sometimes defined themselves as a conquered people, like the Indians. The Indians themselves had their own story, featuring deaths on a horrendous scale through disease and genocide. But beyond emphasizing these and many other differences, spokespersons for these nonblack groups sometimes partook of the antiblack racism of the white majority. As late as the early 1960s, for example, spokespersons for Mexican Americans in Los Angeles made a point of saying that their community wanted little to do with blacks in the same city.

Utterances of this latter kind diminished rapidly in the late 1960s as political alliances were forged between black advocacy organizations and organizations speaking for other descent groups. The idea that Asian Americans, Latinos, and Indians were indeed like blacks gained ground and was marked vividly with a designation especially popular in the 1980s: ‘people of color.’ The downplaying of the differences between nonblack minorities and blacks was practiced first by officials and then by activists who came to understand that by applying ‘the black model’ to their own group they had a better chance of getting the sympathetic attention of officials and courts. White racism thus ironically came to be assigned the same capacity traditionally assigned to one drop of black blood: the capacity to define equally whatever it touched, no matter how the affected entity was constituted and what its life circumstances might have been. We have been living by a principle of white racist hypovictimization: we can call it the one hate rule, with the understanding that the colloquial use of ‘hate’ follows the language conventions of recent years, when we speak of ‘hate speech’ and ‘hate crimes.’

Both the one hate rule and the one drop rule have recently come under increasing pressure. But before I take up these pressures and suggest some of the potentially deep changes in American race discourse they might produce, I want to clarify the historical circumstances that have endowed these rules with such force.

The property interests of slaveholders and the social priorities of Jim Crow racism are central to the principle of hypodescent. Keeping the color line sharp facilitated the enslavement of children begotten upon slave women by white men. The offspring of these couplings would grow up as slaves in a race-specific slave system. The principle was sharpened under Jim Crow, when opposition to social equality for blacks was well served by a monolithic notion of blackness accompanied by legislation that outlawed as miscegenation blackwhite marriages but that left less strictly regulated any nonmarital sex in which white males might engage with black females. Some slave-era and Jim Crow governments did employ fractional classifications, providing distinctive rights and privileges for ‘octoroons,’ ‘quadroons,’ and ‘mulattoes,’ but this fractional approach was hard to administer, invited litigation, and blurred lines that many whites preferred to keep clear. ‘Mulatto’ was dropped from the federal census after 1920, and more and more state governments went the way of Virginia, whose miscegenation statute as revised in 1924 classified as white only a person “who has no trace whatsoever of blood other than Caucasian.”

The combination of these miscegenation laws with the principle of hypodescent consolidated and perpetuated the low-class positions of African Americans in much of the United States. By marking all offspring of white-black couplings as bastards, governments in many jurisdictions prevented these offspring from inheriting the property of a white father. Although the legendary Virginia statute, along with all other racial restrictions on marriage, was invalidated in 1967 by the U.S. Supreme Court in the case of Loving v. Virginia, the one drop rule classically formulated in the Virginia statute was not affected in its capacity as a convention operating throughout American society. Traditional white racism perpetuated this convention, but so, too, did the social solidarity of an African American community whose borders had been shaped by that racism. It is no wonder that the officials, courts, and advocacy organizations that designed and defended affirmative action measures showed no interest in mixture. Even if ‘light-skinned blacks’ had sometimes experienced a less consistently brutal style of discrimination than that experienced by the darkest of African Americans, there was no doubt that any person perceived as having any black ancestry whatsoever was rightly included in the antidiscrimination remedies being developed in the late 1960s and early 1970s.

But what about nonblack victims of white racism? Awareness of the reality of discrimination against nonblacks led to the conclusion that all ethnoracially defined victims of white racism might as well be made the beneficiaries of the same new set of entitlements being developed in the civil rights era, even in the absence of anyone’s having lobbied for that result. (Indians, to be sure, were always subject to an additional, separate set of programs following from the distinctive constitutional status of Indian tribes.) When the Equal Employment Opportunity Commission (EEOC) designed its precedent-setting employer reporting form (EEO-1) in 1965, the EEOC included Indians, Asian Americans, and Latinos along with African Americans as the groups to be counted in relation to its mission. In fact, the EEOC was almost entirely concerned with African Americans: what percentage of those employable were actually employed in a given labor market? At the public hearing designed to collect reactions to this reporting form, no voice mentioned even in passing the situation of the nonblack minorities.4

Virtually everyone in power at the time assumed the nonblack minorities to be so tiny a part of the picture as to require no discussion and to entail no policy dilemmas for the future. Support for the Civil Rights Act of 1964 and for the specific mission and methods of the EEOC established under its terms was deeply informed by a popular understanding of the history of the victimization of African Americans in particular, and not by any comparably deep understanding of the acknowledged mistreatment of Latinos and Asian Americans. To call attention to this truth about the civil rights era is not to downplay the reality of white racism against nonblacks in American history right up to the time officials and courts acted. Rather, the point is that remedying the abuse of nonblacks was almost an afterthought to remedying antiblack racism.

Nothing illustrates this fact more dramatically than the lack of sustained public debate on the eligibility of immigrants and their offspring for affirmative action. This silence resulted partly because the Latino and Asian American populations were still small (about 4.5 percent and 1 percent, respectively, in the census of 1970), and because the Immigration and Nationality Act of 1965 that eventually transformed the ethnoracial demography of the United States, and revolutionized the meaning of ethnoracially defined entitlements, was not expected to significantly increase immigration from Latin America and Asia.

Yet the numbers of Latin American and Asian immigrants mounted in the 1970s, yielding more and more nonblack Americans who were not the descendants of those Chinese American, Japanese American, and Mexican American families that had been abused in the United States, and who were thus less analogous than were nonimmigrant Latinos and Asian Americans to the descendants of enslaved Americans. Indeed, the number of new immigrants between 1970 and 2000 who were eligible for at least some affirmative action benefits came to about 26 million, the same number of eligible African Americans as measured by the census of 1980. More strikingly yet, many of the new immigrants and their children proved able, especially in the Asian American case, to make their way around racist barriers in education, business, and the workforce that continued to inhibit the progress of African Americans.

This emerging social reality might have triggered a rethinking of the one hate rule and stimulated a genuine effort to confront the distinctive history and needs of the several nonblack groups on each group’s own terms. But the system then in place created a huge disincentive for such a rethinking: the black model was working quite well. It helped get the attention of officials and courts, enabling them to recognize and understand the victimization of nonblack minorities. As early as 1968, the Chicano youth activists in Los Angeles were declaring “Brown and Black” to be one and the same. As the most careful scholar of that episode has observed, writers in the Chicano movement’s magazine La Raza, even while surrounded by older Mexican Americans whose group advocacy had been based on the affirmation of white identity, “asserted that Mexican identity, when measured in terms of history, geography, oppressions, and dreams, was functionally black.”5 Hence the one hate rule was quietly enacted by a variety of nonblack advocacy groups as well as by officials and courts.

Neither the EEOC nor anyone else designing and approving affirmative action programs predicated on the ideal of proportional representation seems to have anticipated what could have happened if one or another of the designated groups came to be overrepresented instead of underrepresented. In the late 1960s and very early 1970s, there were very few Asian Americans, Latinos, and Indians in most of the same employment and educational spaces in which African Americans were underrepresented in relation to their percentage in the total population. Instead of inquiring into the specific causes of the underrepresentation of the various groups, one could assume with some justice that behind all cases was white racism of one degree or another. The one hate rule was good enough. At least for a while.

But as the numbers of Asian Americans increased dramatically through chain migrations in the 1970s and 1980s, and began to affect the public face of American society especially in California, a striking challenge to the one hate rule appeared. It became hard to overlook that Asian Americans, even if subject to discrimination as ‘foreign’ and thus ‘not really American,’ were overrepresented rather than underrepresented in many universities and professions and among high-income householders. Well before the end of the 1980s, the Census Bureau reported that average family income for Asian Americans, even when the income for recently arrived immigrants from Southeast Asia was included, was higher than that for non-Hispanic whites. Asian Americans were quietly dropped from some private affirmative action programs (not from those operated by the federal government), but what public discussion there was of the success of Asian Americans was clouded by the problematic concept of ‘the model minority.’ The idea that African Americans, Latinos, and Indians had something wrong with them structurally – some genetic inferiority or deeply embedded cultural deficiency from which the wonderful Asians were free – was sometimes implied, and was of course vigorously contested.

Given the prior assumption that all ethnoracial minorities were more or less equally the victims of white racism, how could one talk about the success of Asian Americans without appearing to deny the power of white racism or to engage, however subtly, in a racist discourse against African Americans, Latinos, and Indians? That this pitfall could indeed be avoided was proved by a growing academic literature exploring with increasing rigor the different historical circumstances of the various American ethnoracial groups popularly called ‘minorities’ or ‘people of color.’ That literature recognized, for example, the unique legacy of slavery and Jim Crow for African Americans, and assessed the pre-immigration social position and commercial experience for many Asian Americans.6 Bengali engineers and Chihuahuan agricultural laborers really did bring different pre-immigration experiences and skills to the United States. Not innate ‘racial’ characteristics, but empirically warrantable social conditions could illuminate the contrasting destinies of different descent communities in the United States. Yet public policy discussions did not take much advantage of the invitation offered by Asian American success to rethink the one hate rule. Far from it.

A mark of the persistence of the one hate rule is its dominance of President Clinton’s Initiative on Race, as displayed in One America in the 21st Century: Forging a New Future, the 1998 report of the Initiative’s advisory board. Although the impeachment of Clinton distracted attention from this document at the time of its release, it is the only major president-sponsored assessment of race since the Kerner Commission’s report of thirty years before. The very banality of One America in the 21st Century renders that document all the more revealing a depository of publicly acceptable ‘race talk’ in the United States at the turn of the twenty-first century.

Central to that talk is the assertion that any differences between the particular varieties of ‘racial’ discrimination and abuse are incidental to what those varieties have in common, and the assumption that the same set of policies can deal with virtually all those varieties of disadvantage. The advisory board does point (with a series of “signposts of historical episodes,” which they distinguish from the “comprehensive” history they disclaim) to a handful of particular experiences: the conquest of the Indians, the enslavement and segregation of black people, “the conquest and legal oppression of Mexican American and other Hispanics,” the “forced labor of Chinese Americans,” and the “internment of Japanese Americans.” Even “new immigrants” from Southeast Asia “continue to feel the legacy of discriminatory laws against Asian Pacific Americans because they continue to be perceived and treated as foreigners.” In keeping with this last observation, which incorporates the most recent of voluntary immigrants into the same frame with the descendants of slaves and of the conquered and ruthlessly slaughtered indigenous population, the advisory board offers the following summary of the salient history: “Each of the minority groups discussed above share in common a history of legally mandated and socially and economically imposed subordination to white European Americans and their descendants.”7

This perspective informs the entire document, especially the advisory board’s recommendations. All but five of the more than fifty recommendations are general to all victims of racism. Four of the five exceptions deal with the special problems of Indians and Alaskan natives, and the fifth calls for better data-gathering on nonblack minority groups. Not a single one of the advisory board’s recommendations speaks to the specific claims of African Americans on the national conscience. Yet blacks, and blacks alone, inherit a multi-century legacy of group-specific enslavement and hypodescent racialization long carried out under constitutional authority in the United States.

The contrast between the Asian American experience in recent years and the African American experience during the same period is systematically deemphasized by One America in the 21st Century. Only in a footnote and in one easily missed chart does the advisory board acknowledge that by the end of the 1980s Asian Americans had achieved an average annual family income higher even than that of non-Hispanic whites, and almost twice that of blacks and Hispanics. Repeatedly, the advisory board tries to shoehorn the Asian American experience into the space prescribed for it by the one hate rule. In a single sentence, the advisory board praises law enforcement agencies for investigating both the decapitation of a black man in Texas and the death threats to sixty Asian American students at a campus in California. A statement in the text to the effect that “criminal victimization rates are significantly greater for minorities and people of color than for whites, especially with regard to violent crime,” makes no distinctions between the groups. But if one turns to the footnote documenting this statement, one learns that while the homicide rate is 58 per 100,000 for African Americans and 25 per 100,000 for Hispanics, it is only 8 per 100,000 for Asian Americans, which is close to the 5 per 100,000 for whites. Thus the proximity of Asian Americans to non-Hispanic whites in one statistical sector after another is downplayed, ignored, or concealed. Many of the charts in the report that show inequality by ethnoracial group omit Asian Americans altogether. This is true of charts showing rates of college enrollment, median weekly earnings of male workers, and employment – all of which contrast whites to blacks and Hispanics.

The advisory board is understandably determined to refute the myth that “the problem of racial intolerance in this country has been solved,” but in its reluctance to particularize and measure the dimensions of this problem and to deal directly with the reasons why some Americans mistakenly believe the problem to be solved, it ends up weakening its case.8 Asian American success in overcoming the worst consequences of white racism is the elephant in the advisory board’s room.

At stake is the more precise location of the barriers that inhibit Americans of various communities of descent from participating more fully in the life of the nation. The more confident we can be about the social location of those barriers, the more likely we are as a nation to develop policies that target remedy to wrong in the effort to achieve a more equal society. If economic and social conditions antecedent to immigration are significant factors in explaining the relative success many Asian American groups have achieved, that suggests that white racism does not always have the same effect on everything it touches, but rather affects those objects differently depending on how those objects are constituted.

Even One America in the 21st Century approaches this insight when it distinguishes between the different destinies of Asian American groups, noting in a footnote that while 88 percent of Japanese Americans between the ages of twenty-five and twenty-nine have a high school diploma, only 31 percent of Hmong Americans do.9 How recent the immigration and how strong or weak the class position of the group prior to immigration clearly make an enormous difference. This is true not only for Asian Americans but also for Hispanics. For instance, sociologists have explained repeatedly that recent illegal immigrants from Mexico encounter the United States and its white racism differently than do Cuban Americans whose families have been in the country for several decades, or than do descendants of earlier generations of migrants from Mexico who have more opportunities to learn English and to take advantage of whatever educational opportunities are at hand.

So great is the variety of experience among Hispanics that the Census Bureau would do well to think carefully about the basis for continuing to treat Hispanics as a single category at all. The census might drop this quasi-racial category and count instead those inhabitants who identify with descent communities from Mexico, Cuba, Puerto Rico, the Dominican Republic, Haiti, and other such defining points of origin. Instead of counting ‘Asians,’ the census might count people who trace their descent to China, Japan, Korea, Vietnam, India, Iran, the Philippines, Pakistan, Lebanon, Turkey, etc. Any public or private agency that wished for any reason – including the design and implementation of antidiscrimination remedies – to treat all Hispanics or Asians as a single group could easily reaggregate the groups counted separately by the census. Or a given agency might conclude, on the basis of what it learns about the social and economic circumstances of particular descent communities, and on the basis of its analysis of where responsibility for a given case of disadvantage lies, that some groups need affirmative action and others do not. Breaking down Hispanic into the actual descent groups that exist in the United States would facilitate this. So, too, with Americans of Asian descent. Neither Hispanics nor Asian Americans have an experience as unified as that of African Americans, and the Census Bureau needs a better justification than it has offered until now for the use of these panethnic, ‘racial’ categories. By rejecting racial and quasi-racial categories, the census can liberate itself from de facto responsibility for deciding who is eligible for this or that program.10

Analysis of different segments of the black population, too, yields more precise information about the location of the barriers to full participation in American life. Black immigrants from the Caribbean and their descendants are more likely than the American-born heirs of the Jim Crow system to advance in education and employment and to marry outside their natal community. So too are black immigrants from Africa, as the public has recently been reminded by the remarkable career of Illinois politician Barack Obama, elected to the U.S. Senate in 2004.11 Moreover, Dalton Conley has found that when blacks and whites with the same property holdings (as opposed merely to the same income, which is a less substantial indicator of economic position) are compared, the gap between black and white performance on Graduate Record Examinations and in several other arenas of achievement diminish to a point of statistical insignificance.12 Class position, when accurately measured, makes a formidable difference. What our social science is telling us today is not that white racism has disappeared, nor even that it is unimportant, but that it interacts with a variety of other realities to create the patterns of inequality that social policy must address.

It is in the context of these social scientific findings that the status of ‘underrepresented minorities’ invites reexamination with an eye toward better understanding those patterns of inequality. When the ideal of proportional representation entered affirmative action directives and jurisprudence in about 1970, a major objective was to get beyond ‘intentional’ discrimination in order to confront prior, structural conditions producing inequality. But by promoting the idea that the mere fact of underrepresentation constituted evidence of discrimination, however indirect, officials and courts deflected attention from any and all possible specific explanations for why a particular descent group might be underrepresented in a particular employment or educational sector. What was lost in the process was an ability to deal forthrightly with the appearance of Asian Americans as an overrepresented minority.

Underrepresentation and overrepresentation constitute a logical syndrome. Should we not expect the same principles of causation to apply to both sides of the phenomenon? Might what we learn about the overrepresentation of particular descent groups – Korean Americans and Jewish Americans, for example – help us to understand the underrepresentation of others, and vice versa? This might seem obvious, but the analysis of overrepresentation, and of the historical processes by which ethnoracial groups that were once underrepresented have become overrepresented, usually stops with the white color line. The Irish, the Italians, the Poles, and the Jews, we say, became white. But invoking whiteness does not carry us very far. Appalachian whites are not overrepresented in the medical profession and in the nation’s great universities, and some ‘people of color’ – Chinese Americans and South Asian Americans, for example – are.

Jewish experience since 1945 is the most dramatic single case in all American history of a stigmatized descent group that had been systematically discriminated against under the protection of the law suddenly becoming overrepresented many times over in social spaces where its progress had been previously inhibited. The experience since 1970 of several Asian American groups is a second such dramatic case. These cases of success invite emphasis and explanation in relation to explanations for the social destiny of other descent-defined groups. What explains the overrepresentation of Jewish Americans, South Asian Americans, and Japanese Americans in the domains of American life where African Americans and Latinos are underrepresented? The failure to pursue this question implicitly strengthens largely unexpressed speculations that Jews and Asians are, after all, superior genetically to African Americans, Latinos, and American Indians – the groups whose underrepresentation is constantly at issue.

Yet the grounds for avoiding talk about the overrepresentation of Jewish Americans and some groups of Asian Americans diminish, if not disappear, once the relevant statistics are explained by taking full account of the conditions under which the various descent communities have been shaped.13 Avoiding the forthright historical and social-scientific study of the question perpetuates the mystification of descent communities and subtly fuels the idea that the question’s answer is really biological, and if made public will serve to reinforce invidious distinctions between descent groups. The open discussion of overrepresentation will not be racist if it proceeds on nonracist assumptions. We will not understand patterns of inequality in the United States until overrepresentation and underrepresentation are studied together and with the same methods. The one hate rule is an obstacle to such inquiries. But if the overrepresentation of African American males in prisons can be explained, as it often is, with reference to slavery, Jim Crow, and the larger history of the institutionalized debasement of black people, so, too, can the overrepresentation of Jewish Americans and Korean Americans in other social spaces be explained by historical conditions.

So the one hate rule, however sensible it may have seemed when informally adopted in the 1960s and 1970s, is increasingly difficult to defend. And the less blinded we are by it the more able we are to see the unique invidiousness of the one drop rule, its ironic twin. The practice of hypodescent racialization has entailed an absolute denial of the reality of extensive white-black mixing. It has embodied a total rejection of blackness and it has implied a deep revulsion on the part of empowered whites. This variety of white racism was cast into bold relief in the 1980s and 1990s by the dramatic upsurge of immigration from Latin America and Asia. The first of these immigrations displayed from the start an acknowledged and often celebrated mixture of European and indigenous ancestry, and produced children who married Anglos at a rising rate and who were not subject to hypodescent racialization as Latinos. The new immigrants from Asia married Anglos at a considerably higher rate than Latinos did, and their offspring were not socially coerced to identify as 100 percent Asian.

Only a few years earlier, when affirmative action and the allied initiatives that eventually came to be called ‘multiculturalism’ got started, the assumption had been that all the standard minority groups were clearly bounded, durable entities, kept in place by the power of white racism and by the internal adhesives of their communities of descent. But the experience of nonblack minorities was sufficiently different from that of African Americans that the hypodescent racialization of the latter came to be more widely recognized as an index of the unique severity of antiblack racism in the United States. No wonder some frustrated African American activists campaigned for group-specific reparations. Hence the weakening of the one hate rule and the development of a critical perspective on the one drop rule proceeded dialectically. The more fully we understand the unique invidiousness of the principle of hypodescent as applied to ‘blacks,’ the weaker the hold of the one hate rule; and the weaker the hold of the one hate rule, the more able we are to confront at long last the exceptionally racist character of the one drop rule.14


1 For a more extensive account of the historic role of the principle of hypodescent, see my “Amalgamation and Hypodescent: The Question of Ethnoracial Mixture in the History of the United States,” American Historical Review 108 (5) (December 2003): 1363–1390, from which several paragraphs in this essay are drawn.

2 John D. Skrentny, The Minority Rights Revolution (Cambridge, Mass.: Belknap Press of Harvard University Press, 2002). For a vigorous critique of this book, see Victoria Hattam, “The 1964 Civil Rights Act: Narrating the Past, Authorizing the Future,” Studies in American Political Development 18 (Spring 2004): 60–69, followed by a generally convincing response by Skrentny, “Policy Making is Decision Making: A Response to Hattam,” 70–80.

3 For two overviews of the development of ‘panethnicity,’ see Jose Itzigsohn, “The Formation of Latino and Latina Panethnic Identities,” and Yen Le Espiritu, “Asian American Panethnicity: Contemporary National and Transnational Possibilities,” in Nancy Foner and George Fredrickson, eds., Not Just Black and White: Historical and Contemporary Perspectives on Immigration, Race, and Ethnicity in the United States (New York: Russell Sage Foundation, 2004), 197–216, 217–234.

4 I owe this information to John D. Skrentny.

5 Ian Haney López, Racism on Trial: The Chicano Fight for Justice (Cambridge, Mass.: Belknap Press of Harvard University Press, 2003), 167.

6 See, for example, Grace Kao, “Asian Americans as Model Minorities? A Look at Their Academic Performance,” American Journal of Education 103 (February 1995): 121–159.

7 Advisory Board to the President’s Initiative on Race, One America in the 21st Century: Forging a New Future (Washington, D.C.: Government Printing Office, 1998), 37–38.

8 Ibid., 46, 48, 65, 71–72, 75, 81, 128, 131.

9 Ibid., 126.

10 This suggestion about the census is a variation on proposals made during the 1990s by a number of demographers and social scientists. See, for example, Margo Anderson and Stephen E. Feinberg, “Black, White, and Shades of Gray (and Brown and Yellow),” Change 8 (1) (1995): 15–18, esp. 18. The substitution of the racial categories with more specific demographic categories would provide individuals greater opportunity to declare their cultural identity while also enabling public and private agencies to pursue antidiscrimination remedies on more empirically warranted grounds.

11 Obama, the keynote speaker at the Democratic National Convention in July of 2004, is the son of a black immigrant from Kenya and a white mother from Kansas. His life story and the questions that story raises about black identity and the dynamics of black progress in the United States today are explored in Scott Malcomson, “An Appeal Beyond Race,” The New York Times, August 1, 2004, IV, 5, and William Finnegan, “The Candidate,” The New Yorker, May 31, 2004, 32–38.

12 Dalton Conley, Being Black, Living in the Red: Race, Wealth, and Social Policy in America (Berkeley: University of California Press, 1999).

13 For an exploration of this terrain, see my “Rich, Powerful, and Smart: Jewish Overrepresentation Should Be Explained Rather Than Mystified or Avoided,” Jewish Quarterly Review 94 (Fall 2004): 596–602. The most ambitious and convincing study of Jewish success is Yuri Slezkine, The Jewish Century (Princeton: Princeton University Press, 2004).

14 For critical suggestions based on an earlier draft, I am indebted to Victoria Hattam, Jennifer Hochschild, Joan Heifetz Hollinger, Ian Haney López, Rachel Moran, Robert Post, Kenneth Prewitt, John Skrentny, Werner Sollors, Eric Sundquist, and Kim Williams. For other assistance I want to thank Jennifer Burns.