ACRC Article

THE DECENNIAL census required by the U.S. Constitution has always been entangled with questions of race. The constitutional provision that, until passage of the Fourteenth Amendment, counted a black man as only three-fifths of a person raised problems from the beginning. But race remained a part of the government’s population tally even after the Civil War. Until 1970, the Census Bureau decided people’s race for them: the enumerator who knocked on the door determined it, sometimes by inquiring, sometimes with a quick look. Since then, respondents have been allowed to answer the question for themselves, picking from an ever-expanding list of racial, ethnic, and national possibilities.

Still, for all these changes, the census continued to conform in one key respect to most people’s understanding of what race meant. As the instructions on the form emphasized, “Fill ONE circle for the race that the person considers himself/herself to be.” Race was an exclusive category—if you were one thing, you were necessarily not something else, just as you were either male or female.

But the census forms that arrived in the nation’s mailboxes in March 2000 were different. It was not that the federal government had decided race and ethnicity were no longer important. Far from it: on both the short and long versions of the questionnaire, one of the first substantive items inquired, “Is this person Spanish/Hispanic/Latino?” A second question probed more directly for race but added, by way of instruction, something truly novel: “Mark one or more” (emphasis added). Still more surprising, some seven million Americans did just that, refusing to describe themselves as only white, black, Asian, Latino, Korean, Samoan, or one of the other categories listed.

Though the number of those who indicated more than one race was seemingly insignificant—less than 3 percent of the population—their action was nothing short of momentous, and may well herald the beginning of the end of racial classifications as we know them.

THE CHANGE in the 2000 census form was no bureaucratic accident, but rather the product of a long, bitter political battle. In the early 1990’s, a small group of interracial families began to lobby the government on behalf of their children, who, they maintained, should be able to describe themselves as “multiracial” on the census. A grassroots parents’ movement made its case on websites and in special-interest magazines, arguing that young people should not be forced to choose between the identities of their mothers and their fathers. In 1993, this group convinced the Office of Management and Budget (OMB), which determines the racial categories used by the federal government, to open an inquiry into the idea of putting a mixed-race box on the census form.

The civil-rights establishment immediately mobilized to fight back. Organizations like the NAACP, the National Urban League, and the National Council of La Raza recognized the threat such a mixed-race box would pose to the classifications that justify their existence: after all, every dark-skinned child registered in this category would shrink the government’s official count of blacks or Latinos or American Indians, eventually reducing the political influence of the organizations claiming to represent these groups. For the better part of the 90’s, minority activists and their allies in the federal government mounted an unstinting effort to block any change in the census form.

In 1997, the racial advocates appeared to prevail. After four years of study and angry public debate, OMB decided against allowing the Census Bureau to include the category of “multiracial” or “mixed race.” Only as an afterthought did OMB announce that the census would permit respondents to check more than one racial box. And the agency eventually decided that, for official purposes, those who checked both white and a minority category would be counted simply as members of that minority.

Together, these decisions seemed to spell certain defeat for the mixed-race families. Just to make sure, in the months before the census, the civil-rights establishment spent millions of dollars on advertising urging blacks, Latinos, and American Indians to check only one box, counting themselves as one and only one race.

Nevertheless, Americans of mixed heritage were not deterred. People of every color and from every region and age group checked more than one box: 5 percent of blacks, 6 percent of Hispanics, 14 percent of Asians, 40 percent of American Indians. In New York City and other areas heavily populated by immigrants, the proportion ran as high as one in four. Most tellingly for the future, people under the age of eighteen were twice as likely as their elders to identify themselves as multiracial.

The significance of these figures was apparent from the moment they were released—if only, at first, in the chaos they created. The Census Bureau’s racial percentages now added up to well over 100. Journalists and the social scientists they looked to for elucidation could not compare the 2000 census to the 1990 count or any that came before it. In short order, Americans found they no longer had clear-cut answers to the simplest questions about who we are as a nation.

How many blacks are there in the United States today? If one counts only the people who checked black alone, the answer is 34.7 million. But if one includes those who checked both black and something else, the number rises to 36.4 million. Are blacks still the country’s largest minority group, or do Latinos now outnumber them? It depends whether one includes those extra 1.7 million: yes if one does, no if one does not.

Even more confusing is what happened to the tally of American Indians. If one counts only the people who checked American Indian alone, there are 2.5 million. But add in those whites and blacks who think they have a little Indian blood—and who, prompted by pride or whim, took the trouble to say so on their census forms—and the number jumps to an astonishing 4.1 million, an increase of nearly 65 percent.

WHAT ACCOUNTS for this shift in how Americans are choosing to identify themselves racially? At the center of it, plainly, are mixed-race couples and their children—fruit of the unprecedented rise in intermarriage in recent decades. Even excluding Latinos, who are among the most likely to marry outside their group—but whom the government does not recognize as a race—marriages between whites and non-whites grew tenfold between 1960 and 1990. Among third-generation Latinos and Asians, intermarriage rates now exceed 50 percent, and in California, more mixed-race children than black children are already being born every year.

But intermarriage is not the only factor at work. Latino immigrants, most of whom come from heavily mixed-race countries, are often baffled by the rigid racial categories they discover in the U.S., and are among the Americans most likely to identify themselves as belonging to more than one group. The Census Bureau’s illogical insistence on asking a separate question about Hispanic identity—another concession to civil-rights advocates anxious to maintain their group’s official numbers—reveals even more starkly than the form otherwise would just how many Latinos consider themselves multiracial. Last year, 48 percent of those who identified themselves as Hispanic also checked white, 2 percent also checked black, 6 percent checked two or more additional boxes, and 42 percent marked “other” (probably a way of suggesting that they equate their race with their ethnicity). Altogether, that makes well over half who identified themselves both as Hispanic and as members of some other group conventionally recognized as a race.

What is more, as demonstrated most dramatically by the tally of American Indians, many of those declaring two or more backgrounds were not the children of recent mixed marriages at all. Some had simply decided that their blended ancestry, however remote, was important enough to them to be mentioned. Still others undoubtedly saw the new options on the census in a more radical light: a chance to stage a small, private revolt against the yoke of race by declaring at last that the monolithic categories long used to define them did not capture the reality of who they were.

The response to the 2000 census is hardly the first indication in the history of America or the modern world that human reality is too varied and fluid to be shoehorned into racial categories. In the 19th century, the U.S. census classified people with both black and white ancestors as mulattos, quadroons, and octoroons; but even at a time when interracial marriage was strictly prohibited, these labels soon became odious and absurd. South African apartheid considered a man black if he could not pull a comb through his tightly curled hair. The Nazis had their tests; white Southerners in the Jim Crow era had other tests. The classifications that resulted were always arbitrary and artificial, and in almost every context—when they did not lead to genocide—they eventually collapsed of their own weight.

Nor are the results of the 2000 census the only indication that Americans are tiring of rigid racial categories. While both history and politics argue that race is an immutable characteristic, social scientists find that ordinary people often do not feel this is the case. A recent survey by researchers at the University of Michigan asked teenagers to fill out a form about race at school, then had interviewers put the same questions to them at home, usually in the presence of their parents. On the form, twice as many indicated they were multiracial as said so when they were asked in person.

Recent trends in popular culture—from white teenagers’ fascination with rap to the popularity of mixed-race celebrities like Tiger Woods and Mariah Carey—make a mockery of racial categories. So do soaring rates of interracial friendship and dating. This is no fringe phenomenon: according to one recent survey, more than 60 percent of American teenagers have dated someone of another color or ethnic group. And a USA Today/CNN/Gallup poll conducted in March found that 64 percent of the public—and 75 percent of those under eighteen—thought it was “good for the country” to have more Americans “think of themselves as multiracial rather than belonging to a single race.”

NONE OF this means that race and ethnicity aredead or dying in America. Blacks, whites, Latinos, and Asians are not, in the foreseeable future, going to “melt” into one large, uniform brown race. In practice, even the children of intermarriage make ethnic choices. More often than not, they choose one side of the family over another; and often, several generations later, their descendants are still celebrating the heritage they have chosen. Nevertheless, the solidity of these categories, and their significance for individuals, are plainly changing.

For some people, race remains the most important aspect of their identity. But for others, it is more like a hobby, what sociologists call “symbolic ethnicity.” For intermarried Asians and Latinos, and even for some blacks, it can be a voluntary thing. For many Americans, it is a matter not of essence but of choices—about belonging, loyalty, cultural affinity, and the relative weight they assign to ethnic identity. The syndicated columnist Clarence Page, who is black, has characterized this complexity with a vivid metaphor. Race is real enough, he suggests, but also like Jell-O: too slippery and elusive to grab by the handful.

Thanks to the 2000 census, it is likely to become only more so, causing ever-widening ripples of dismay among racial advocates and their allies in government. Consider the confusion spawned by the greatly increased number of people who identified themselves as American Indian. Officials are well aware that this count is an artifact of the new form: no one thinks the largely static Indian population really grew by 65 percent in just ten years. Nevertheless, for official purposes, even those who marked Indian and something else are to be counted as Indians. Does this mean that the federal government is now obligated to increase the services it provides for Native Americans by 65 percent—despite the fact that most of the 1.6 million “new” Indians live nowhere near a reservation and do not formally belong to any tribe? On reflection, even Washington must recognize that this would make no sense.

This may be an extreme example, but it is a taste of things to come. American government spends more than $185 billion a year on special provisions for minorities, and all sorts of public goods are handed out according to racial percentages. Legislative districts, municipal contracts, slots at prestigious universities, government jobs, antipoverty benefits—name the public benefit, and we divide it up by race.

But how can this continue if we can no longer count reliably by race—as race itself is revealed as a largely artificial or voluntary construct? The policy most clearly at risk is affirmative action. How is a city or state to set aside some proportionate number of jobs for minority contractors if it cannot determine the number of minority contractors in the population? So too with racial gerrymandering. How will a state legislature go about creating, say, a majority Latino district if it cannot specify who is Latino and who is not? And how can that be determined when some people say they are Latino in one context but not in another? What about those who are half-Latino, or one-quarter, or who merely happen to have a Spanish surname?

Nor will it be any easier to use racial accounting as a tool to detect discrimination. Though the concept of “disparate impact” has been much challenged, and rightly so, demonstrating that a particular policy has such an effect remains the most common way to allege discrimination and build a case for racial redress. But what will happen when prosecutors can no longer show disparate impact—when they can no longer compare, for example, the number of Latino drivers being stopped and arrested on the New Jersey Turnpike with the number of Latinos who use that highway? Surely we would not want the courts to resort to the same criterion used by offending state troopers—namely, whether a driver “looks” Hispanic.

IF A new multiracial or “mestizo” order is emerging in the United States—and, in the wake of the census, more and more people are acknowledging that such a transformation is under way—a large, troubling question remains: will blacks be part of it?

For those who fear the answer is no, exhibit A is history: the long and often ugly experience of blacks in America and the consequences, psychological and other, that it continues to spawn. Prejudice may also play some part: despite an encouraging abatement of bigotry, many whites continue to view blacks in a less positive light than they view new immigrants, whatever their color or nationality. Then there is black poverty—the persistence of a small but intractable black underclass—that no notion of voluntary ethnicity seems likely to dispel. But perhaps the greatest obstacle of all is the angry, separatist attitudes so prevalent in much of black America, both poor and middle-class. That doing well in school is “acting white,” that the only way to be manly is to be ruthlessly predatory, that to cooperate with “the system” is to betray one’s origins: it is not difficult to see how this oppositional culture might prevent any significant softening of the hard outlines of racial identity among many blacks.

Still, troubling as these concerns are, there are also signs pointing in a different direction. It is true, for example, that blacks marry outside their group far less frequently than Latinos and Asians; but the rate at which they are doing so has grown significantly over the last generation. In 1960, fewer than 1 percent of married black men had non-black wives; today the figure is more than fifteen times higher. Even more dramatic is the number of black-white interracial births, which, according to one estimate, more than tripled in the 1990’s. And given the way attitudes are shifting, particularly in places like California where the racial and immigrant mix is most pronounced, this trend can only accelerate.

Nor is this the only sign that blacks too may be on the cusp of a new and looser racial understanding. In fact, the 2000 census suggests that blacks may be among the Americans most eager for a change in that direction. Not only were blacks under eighteen far more inclined than their elders to mark more than one racial box, but young blacks were twice as likely to do so as other young people: astonishingly, nearly one in 10 of them chose this option. Equally telling in this regard is the USA Today/CNN/Gallup poll, which found that blacks approved of the new easing of racial lines no less heartily than whites did.

Certainly the black-white color line is going to be harder to cross or blur than other racial and ethnic boundaries. When it comes to race, what other people think you are often matters as much as, if not more than, what you think. And blacks who checked more than one box on the census may be ignoring the political reality of the world around them—a reality shaped as much these days by the civil-rights establishment as by lingering white prejudice. Still, by making their choices known, multiracial Americans have taken a big step toward changing that reality.

IT WILL not do—and it helps no one—to bePollyana-ish. Obscuring or blurring the lines among groups will hardly solve all problems of race and ethnicity. It will not tell Americans how to negotiate their differences with one another, or how far to go in developing their identities. Nor will it answer the increasingly pressing question of what ideals and purposes we share as a nation. But it might help us to do something almost as important: to separate the realm where race and ethnicity matter, and should matter, from the realm where they do not, if only because they cannot.

An age-old American formula that has long worked for white ethnics, if not for blacks, distinguishes the role race and ethnicity play in the public sphere from the part they can and should play in our private lives, at home, and in our neighborhoods. Our long-held ideals argue that the public realm must be race-neutral, while in private, race and ethnicity may be fostered, celebrated—or ignored, as individuals choose. This is hardly a foolproof formula, but in a nation composed of a multitude of groups it is a necessary one. And the beauty of it today, given the way Americans’ feelings about race are changing, is that it need not be arbitrarily imposed or enforced from above.

On the contrary, as the 2000 census suggests, the line has begun to draw itself naturally. The more Americans realize that ethnic and even racial identification can be a matter of personal choice, the fuzzier such categories will grow—and, one can only hope, the harder it will be for government to traffic in them.

TAMAR JACOBY is a senior fellow of the Manhattan Institute and the author of Someone Else’s House: America’s Unfinished Struggle for Integration. Her article, “In Asian America,” appeared in our July-August 2000 issue.