Leaving Race Behind

Our growing Hispanic population creates a golden opportunity

Detail from "Workers," Tarsila do Amaral, 1933 (Archive of Hispanic Culture)
Detail from "Workers," Tarsila do Amaral, 1933 (Archive of Hispanic Culture)


Some years ago the United States government asked me what my race was. I was reluctant to respond because my 50 years of practicing sociology—and some powerful personal experiences—have underscored for me what we all know to one degree or another, that racial divisions bedevil America, just as they do many other societies across the world. Not wanting to encourage these divisions, I refused to check off one of the specific racial options on the U.S. Census form and instead marked a box labeled “Other.” I later found out that the federal government did not accept such an attempt to de-emphasize race, by me or by some 6.75 million other Americans who tried it. Instead the government assigned me to a racial category, one it chose for me. Learning this made me conjure up what I admit is a far-fetched association. I was in this place once before. When I was a Jewish child in Nazi Germany in the early 1930s, many Jews who saw themselves as good Germans wanted to “pass” as Aryans. But the Nazi regime would have none of it. Never mind, they told these Jews, we determine who is Jewish and who is not. A similar practice prevailed in the Old South, where if you had one drop of African blood you were a Negro, disregarding all other facts and considerations, including how you saw yourself.

You might suppose that in the years since my little Census-form protest the growing enlightenment about race in our society would have been accompanied by a loosening of racial categories by our government. But in recent years the United States government has acted in a deliberate way to make it even more difficult for individuals to move beyond racial boxes and for American society as a whole to move beyond race.

Why the government perpetuates racialization and what might be done to diminish the role of race in our lives are topics that have become especially timely as Hispanics begin to take a more important role demographically, having displaced African-Americans as the largest American minority. How Hispanics view themselves and how they are viewed by others are among the most important factors affecting whether or not we can end race as a major social divide in America.

Treating people differently according to their race is as un-American as a hereditary aristocracy, and as American as slavery. The American ethos was formed by people who left the social stratification of the Old World to live in a freer, more fluid society. They sought to be defined by what they accomplished, not by what they were born with. As Arthur M. Schlesinger Jr. puts it in his book The Disuniting of America, one of the great virtues of America is that it defines individuals by where they are going rather than by where they have been. Achievement matters, not origin. The national ideal says that all Americans should be able to compete as equals, whatever their background. American society has been divided along racial lines since its earliest days.

Racial characterizations have trumped the achievement ideal; people born into a non-white race, whatever their accomplishments, have been unable to change their racial status. Worse, race has often been their most defining characteristic, affecting most, if not all, aspects of their being.

As a result, we have been caught, at least since the onset of the civil rights movement, in an ambivalence. On the one hand, we continue to dream of the day when all Americans will be treated equally, whatever their race; we rail against—and sometimes punish—those who discriminate according to race in hiring, housing, and social life. At the same time, we have ensconced in law many claims based on race: requirements that a given proportion of public subsidies, loans, job training, educational assistance, and admission slots at choice colleges be set aside for people of color. Many Americans, including African-Americans, are uneasy about what some people consider reverse discrimination. Courts have limited its scope; politicians have made hay by opposing it; and some of its beneficiaries feel that their successes are hollow because they are unsure whether their gains reflect hard-won achievements or special favors. There must be a better way to deal with past and current injustice. And the rapid changes in American demographics call for a reexamination of the place of race in America.


We have grown accustomed to thinking about America in black and white, and might well have continued to do so for decades to come except that Hispanics complicate this simplistic scheme: they do not fit into the old racial categories. Some Hispanics appear to many Americans to be black (for example, quite a few Cuban-Americans), others as white (especially immigrants from Argentina and Chile), and the appearance of still others is hard for many people to pigeonhole. Anyone seeing the lineup of baseball players honored as Major League Baseball’s “Latino Legends Team” would find that the players vary from those who are as fair-skinned as Roger Clemens to those who are as dark-skinned as Jackie Robinson. More important by far, survey after survey shows that most Hispanics object to being classified as either black or white. A national survey conducted in 2002 indicated that 76 percent of Hispanics say the standard racial categories used by the U.S. Census do not address their preferences. The last thing most of those surveyed desire is to be treated as yet another race—as “brown” Americans.

Hispanics would have forced the question of how we define one another even if they were just another group of immigrants among the many that have made America what it is. But Hispanics are not just one more group of immigrants. Not only have Hispanic numbers surpassed those of black Americans, who until 2003 made up America’s largest minority group, Hispanics have been reliably projected to grow much faster than African- Americans or any other American group. Thus, according to the Census, in 1990 blacks constituted 12 percent of the population and Hispanics 9 percent. By 2000, Hispanics caught up with blacks, amounting to 12.5 percent of the population compared to 12.3 percent for blacks. By 2050, Hispanics are projected to be 24.3 percent of the American population, compared to 14.7 percent for blacks. In many cities, from Miami to Los Angeles, in which African-Americans have been the largest minority group, Hispanics’ numbers are increasingly felt. While once Hispanics were concentrated in the areas bordering Mexico, their numbers are now growing in places like Denver, St. Paul, and even New England.

Immigration fuels the growth of Hispanics relative to the growth of African-Americans because Latin American immigration, legal and illegal, continues at an explosive pace, while immigration from Africa is minuscule. Hispanics also have more children than African-Americans. During the most recent year for which data is available, 2003–2004, one of every two people added to America’s population was Hispanic. And while black Americans have long been politically mobilized and active, Hispanics are just beginning to make their weight felt in American politics.

The rapid growth in the number, visibility, and power of Hispanics will largely determine the future of race in America, a point highlighted by Clara E. Rodriguez in her book Changing Race: Latinos, the Census, and the History of Ethnicity in the U.S. If Hispanics are to be viewed as brown or black (and some on the left aspire to color them), and above all if Hispanics develop the sense of disenfranchisement and alienation that many African- Americans have acquired (often for very good reasons), then America’s immutable racial categories will only deepen.

If, on the other hand, most Hispanics continue to see themselves as members of one or more ethnic groups, then race in America might be pushed to the margins. Racial categories have historically set us apart; ethnic categories are part of the mosaic that makes up America. It has been much easier for an individual to assimilate from an ethnic perspective than from a racial one. Race is considered a biological attribute, a part of your being that cannot be dropped or modified. Ethnic origin, in contrast, is where you came from. All Americans have one hyphen or another attached to their ethnic status: we’re Polish-, or German-, or Anglo-, or Italian-Americans. Adding Cuban-Americans or Mexican-Americans to this collage would create more comfortable categories of a comparable sort.


Many people take it for granted that genes determine race, just as genes determine gender. And we also tend to believe that racial categories are easy to discern (though we all know of exceptions).

One way to show how contrived racial divisions actually are is to recall that practically all of the DNA in all human beings is the same. Our differences are truly skin deep. Moreover, the notion that most of us are of one race or another has little basis in science. The Human Genome Project informs us not only that 99.9 percent of genetic material is shared by all humans, but also that variation in the remaining 0.1 percent is greater within racial groups than across them. That is, not only are 99.9 percent of the genes of a black person the same as those of a white person, but the genes of a particular black person may be more similar to the genes of a white person than they are to another black person.

This point was driven home to college students in a sociology class at Penn State in April 2005. Following their professor’s suggestion, the students took DNA tests that had surprising results. A student who identified himself as “a proud black man” found that only 52 percent of his ancestry traced back to Africa, while the other 48 percent was European. Another student who said she takes flak from black friends for having a white boyfriend found that her ancestry was 58 percent European and only 42 percent African. These two students are not alone: an estimated one-third of the African-American population has European ancestry.

Which people make up a distinct race and which are considered darkskinned constantly changes as social prejudices change. Jewish-, Slavic-, Irish-, and Polish-Americans were considered distinct races in the mid-19th and early 20th centuries—and dark races at that, as chronicled in great detail in Matthew Frye Jacobson’s book Whiteness of a Different Color: European Immigrants and the Alchemy of Race and in a well-documented book by Noel Ignatiev, How the Irish Became White. Ignatiev found that in the 1850s, Irish people were considered non-white in America and were frequently referred to as “niggers turned inside out.” (Blacks were sometimes called “smoked Irish.”)

The capriciousness of racial classifications is further highlighted by the way the U.S. Census, the most authoritative and widely used source of social classifications, divides Americans into races. When I ask my students how many races they think there are in America, they typically count four: white, black, Asian, and Native American. The Census says there are 15 racial categories: white, African-American, American Indian/Alaska Native, Asian Indian, Chinese, Filipino, Japanese, Korean, Vietnamese, “other Asian,” Native Hawaiian, Guamanian/Chamorro, Samoan, and “other Pacific Islander,” and as of 2000 one more for those who feel they are of some other race. (Hispanic is not on this list because the Census treats Hispanic as an ethnicity and asks about it on a separate question, but immediately following that question, the Census asks, “So what is your race, anyhow?”)

The arbitrary nature of these classifications is demonstrated by the Census Bureau itself, which can change the race of millions of Americans by the stroke of a pen. The Census changed the race of Indian- and Pakistani-Americans from white in 1970 to Asian in 1980. In 1930 the Census made Mexicans into a different race but then withdrew this category. Similarly, Hindu made a brief appearance as a race in the 1930 and 1940 Censuses but was subsequently withdrawn.

Anthropologists have found that some tribes do not see colors the way many of us do; for instance, they do not “see” a difference between brown and yellow. Members of these tribes are not colorblind, but some differences found in nature (in the color spectrum) simply don’t register with them, just as young American children are unaware of racial differences until someone introduces them to these distinctions. We draw a line between white and black, but people’s skin colors have many shades. It is our social prejudices that lead us to make sharp racial categories.

I am not one of those postmodernists who, influenced by Nietzsche and Foucault, claim that there are no epistemological truths, that all facts are a matter of social construction. I disagree with Nietzsche’s description of truth as “a mobile army of metaphors, metonyms, and anthropomorphisms—in short a sum of human relations, which have been enhanced, transposed, and embellished poetically and rhetorically and which after long use seem firm, canonical, and obligatory to a people.” However, there is no doubt that social construction plays a significant role in the way we “see” racial differences, although our views may in turn be affected by other factors that are less subject to construction, for example, historical differences.

Most important is the significance we attribute to race and the interpretations we impose on it. When we are told only that a person is, say, Asian-American, we often jump to a whole list of conclusions regarding that person’s looks, intelligence, work ethic, character; we make the same sort of jumps for Native Americans, blacks, and other races. Many things follow from these kneejerk characterizations: whether we will fear or like this person, whether we will wish to have him or her as a neighbor or as a spouse for one of our children—all on the basis of race. In short, we load on to race a great deal of social importance that is not a reflection of the “objective” biological differences that exist. To paraphrase the UNESCO Constitution, racial divisions are made in the minds of men and women, and that is where they will have to be ended.


If racial categories have long been settled, the social characterization of the Hispanic is up for grabs. We still don’t know whether Hispanics will be defined as a brown race and align themselves with those in the United States who are or who see themselves as marginalized or victimized—or if they will be viewed as a conglomerate of ethnic groups, of Mexican-Americans, Cuban-Americans, Dominican-Americans, and so forth, who will fit snuggly into the social mosaic.

The term Hispanic was first used in the Census in 1980. Before that, Mexican-Americans and Cuban-Americans were classified as white (except when a Census interviewer identified an individual as the member of a different racial group). Until 1980, Hispanics were part of the great American panorama of ethnic groups. Then the Census combined these groups into a distinct category unlike any other. It was as if the federal government were to one day lump together Spanish-, Italian-, and Greek-Americans into a group called “Southern European” and begin issuing statistics on how their income, educational achievements, number of offspring, and so on compare to those of Northern Europeans.

And as we’ve seen, those who define themselves as Hispanic are asked to declare a race. In the 1980 Census, the options included, aside from the usual menu of races, that ambiguous category “Other.” There were 6.75 million Americans, including me, who chose this option in 1980. Most revealing: 40 percent of Hispanics chose this option. (Note that they—and I—chose this category despite the nature of the word Other, which suggests the idea of “not being one of us.” Had the category been accorded a less loaded label, say “wish not to be identified with any one group,” it seems likely that many millions more would have chosen this box.)

To have millions of Americans choose to identify themselves as “Other” created a political backlash because Census statistics are used both to allocate public funds to benefit minority groups and to assess their political strength. Some African-American groups, especially, feared that if African-Americans chose “Other” instead of marking the “African-American” box, they would lose public allotments and political heft.

But never underestimate our government. The Census Bureau has used a statistical procedure to assign racial categories to those millions of us who sought to butt out of this divisive classification scheme. Federal regulations outlined by the Office of Management and Budget, a White House agency, ruled that the Census must “impute” a specific race to those who do not choose one. For several key public policy purposes, a good deal of social and economic data must be aggregated into five racial groups: white, black, Asian, American Indian or Alaska Native, and native Hawaiian or other Pacific Islander. How does the government pick a race for a person who checked the “Other” box? They turn to the answers for other Census questions: for example, income, neighborhood, education level, or last name. The resulting profiles of the U.S. population (referred to as the “age-race modified profile”) are then used by government agencies in allotting public funds and for other official and public purposes.

But the Census isn’t alone in oversimplifying the data. Increasingly, other entities, including the media, have treated Hispanics as a race rather than an ethnic group. This occurs implicitly when those who generate social data—such as government agencies or social scientists—break down the data into four categories: white, black, Asian, and Hispanic, which is comparable to listing apples, oranges, bananas, and yams. In their profile of jail inmates, the Bureau of Justice Statistics lists inmates’ origins as “white, black, Hispanic, American Indian/Alaska Native, Asian/Pacific Islander, and more than one race.” The New York Times ran a front-page story in September 2005 in which it compared the first names used by whites, blacks, Asians, and Hispanics. Replace the word Hispanics with the name of another ethnic group, say Jews, and the unwitting racial implication of this classification will stand out.

Still other studies include Hispanics when they explicitly refer to racial groups. For example, a 2001 paper by Sean Reardon and John T. Yun examines what they call “racial balkanization among suburban schools,” where there is increased segregation among black, Hispanic, and Asian students. A 2005 Seattle Times story uses racial terminology when it reports “Latinos have the fewest numbers among racial groups in master’s-of-business programs nationwide, with about 5,000 enrolling annually.” Similarly, The San Diego Union Tribune states: “A brawl between Latino and black students resulted in a lockdown of the school and revealed tensions between the two largest racial groups on campus.”

A handful of others go a step further and refer to Hispanics as a brown race. For example, following the recent Los Angeles mayoral election, The Houston Chronicle informed us that “Villaraigosa’s broad-based support has analysts wondering whether it is evidence of an emerging black-brown coalition.” And, National Public Radio reported: “There is no black and brown alliance at a South Central Los Angeles high school.”

One way or another, all of these references push us in the wrong direction—toward racializing Hispanics and deepening social divisions. America would be best served if we moved in the opposite direction.


Thus far, workers at the U.S. Census Bureau, following the White House’s instructions, seem determined to prevent any deemphasis of race. They are testing iterations of the wording for the relevant questions in the 2010 Census—but all of these possibilities continue to require people to identify themselves by race. Moreover, Census bureaucrats will continue to impute race to those who refuse to do so themselves, ignoring the ever-growing number of people, especially Hispanics, who do not fit into this scheme.

Imagine if instead the federal government classified people by their country (or countries) of origin. For some governmental purposes, it might suffice to use large categories, such as Africa (which would exclude other so-called black groups, such as Haitians and West Indians that are now included in references to “black” Americans), Asia, Europe, Central America, and South America (the last two categories would not, of course, include Spain). For other purposes, a more detailed breakdown might work better—using regions such as the Middle East and Southeast Asia, for example—and if still more detail was desired, specific countries could be used, as we do for identifying ethnic groups (Irish, Polish, Cuban, Mexican, Japanese, Ethiopian, and so on). Kenneth Prewitt, a former director of the U.S. Census Bureau, has suggested the use of ethnic categories. As we have seen, ethnic origins carry some implications for who we are, but these implications decline in importance over time. Above all, they do not define us in some immutable way, as racial categories do. A category called something like “wish not to be identified with any particular group” should be included for those who do not want to be characterized even by ethnicity or for others who view themselves as having a varied and combined heritage.

The classification of Americans who are second-generation, and beyond, highlights the importance of the no-particular-group category. Although a fourth-generation Italian-American might still wish to be identified as Italian, he might not, particularly if he has grandparents or parents who are, say, Greek, Korean, and Native American. Forcing such a person to classify himself as a member of one ethnic group conceals the significance of the most important American development in social matters: out-marriage. Out-marriage rates for all groups other than African-Americans are so high that most of us will soon be tied to Americans of a large variety of backgrounds by the closest possible social tie, the familial one. Approximately 30 percent of third-generation Hispanics and 40 percent of third-generation Asians marry people of a different racial or ethnic origin. Altogether, the proportion of marriages among people of different racial or ethnic origins has increased by 72 percent since 1970. The trend suggests more of this in the future. Even if your spouse is of the same background, chances are high that the spouse of a sibling or cousin will represent a different part of the American collage. At holidays and other family events, from birthdays to funerals, we will increasingly be in close connection with “Others.” Before too long most Americans will be “Tiger Woods” Americans, whose parental heritage is black, Native American, Chinese, Caucasian, and Thai. Now is the time for our social categories to reflect this trend—and its capacity for building a sense of one community—rather than conceal it.


Changing the way we divide up society will not magically resolve our differences or abolish racial prejudices. Nor does a movement toward a colorblind nation mean that we should stop working for a more just America. A combination of three major approaches that deal with economic and legal change could allow us to greatly downgrade the importance of race as a social criterion and still advance social justice. These approaches include reparations, class-based social programs, and fighting discrimination on an individual basis.

To make amends for the grave injustice that has been done to African-Americans by slavery and racial prejudice, as well as to bring to a close claims based on past injustices—and the sense of victimhood and entitlement that often accompanies these claims—major reparations are called for. One possible plan might allot a trillion dollars in education, training, and housing vouchers to African-Americans over a period of 20 years. (The same sort of plan might be devised for Native Americans.)

Such reparations cannot make full compensation for the sins of slavery, of course. But nothing can. Even so, if Jews could accept restitution from Germany and move on (Germany and Israel now have normal international relations, and the Jewish community in Germany is rapidly growing), could not a similar reconciliation between black and white Americans follow reparations? A precedent in our own history is the payment of reparations to Japanese-Americans because of their internment in World War II. In 1988, the U.S. government issued a formal apology in the Civil Liberties Act and awarded $20,000 to each living person who had been interned. About 80,000 claims were awarded, totaling $1.6 billion.

Part of the deal should be that once reparations are made for the sins against African-Americans in the past, black people could no longer claim special entitlements or privileges on the basis of their race. Reparations thus would end affirmative action and minority set-asides as we have known them.

At the same time, Americans who are disadvantaged for any reason not of their own doing—the handicapped; those who grew up in parts of the country, such as Appalachia, in which the economy has long been lagging; those whose jobs were sent overseas who are too old to be retrained—would be given extra aid in applying for college admissions and scholarships, housing allowances, small-business loans, and other social benefits. The basis for such aid would be socio-economic status, not race. The child of a black billionaire would no longer be entitled to special consideration in college admissions, for instance, but the child of a poor white worker who lost his job to outsourcing and could not find new employment would be.

Social scientists differ in their estimates of the extent to which differences in opportunity and upward mobility between blacks and whites are due to racial prejudice and the extent to which they are due to economic class differences. But most scholars who have studied the matter agree that economic factors are stronger than racial ones, possibly accounting for as much as 80 percent of the differences we observe. A vivid example: In recent years, Wake County in North Carolina made sure that its public school classes were composed of students of different economic backgrounds, disregarding racial and ethnic differences. The results of this economic integration overshadowed previous attempts to improve achievement via racial integration. While a decade ago, only 40 percent of blacks in grades three through eight scored at grade level, in the spring of 2005, 80 percent did so.

Class differences affect not only educational achievement, health, and job selection, but also how people are regarded or stereotyped. Fifty years ago, a study conducted at Howard University showed that although adjectives used to describe whites and blacks were quite different, that variance was greatly reduced when class was held constant. People described upper-class whites and upper-class blacks in a remarkably similar fashion, as intelligent and ambitious. People also described lower-class whites and lower-class blacks in a similar way, as dirty and ignorant. The author concluded that “stereotypes vary more as a function of class than of race.”

If race-based discrimination were a thing of the past, and black Americans were no longer subjected to it, then my argument that reparations can lead to closure would be easier to sustain. Strong evidence shows, however, that discrimination remains very much with us. A 1990 Urban Institute study found that when two people of different races applied for the same job, one in eight times the white was offered the job and an equally qualified African- American was not. Another Urban Institute study, released in 1999, found that racial minorities received less time and information from loan officers and were quoted higher interest rates than whites in most of the cities where tests were conducted.

The victims of current racial discrimination should be fully entitled to remedies in court and through such federal agencies as the Equal Employment Opportunity Commission. These cases should be dealt with on an individual basis or in a class-action suit where evidence exists to support one. Those who sense discrimination should be required to prove it. It shouldn’t be assumed that because a given workplace has more people of race x than race y, discrimination must exist.


In the end, it comes down to what Americans envision for our future together: either an open society, in which everyone is equally respected (an elusive goal but perhaps closer at hand than we realize), or an even more racialized nation, in which “people of color” are arrayed in perpetual conflict with white people. The first possibility is a vision of America as a community in which people work out their differences and make up for past injustices in a peaceful and fair manner; the other is one in which charges of prejudice and discrimination are mixed with real injustices, and in which a frustrated sense of victimhood and entitlement on the one hand is met with guilt and rejection on the other.

A good part of what is at stake is all too real: the distribution of assets, income, and power, which reparations, class-based reforms, and the courts should be able to sort out. But don’t overlook the importance of symbols, attitudes, and feelings, which can’t be changed legislatively. One place to start is with a debate over the official ways in which we classify ourselves and the ways we gather social data, because these classifications and data are used as a mirror in which we see ourselves reflected.

Let us begin with a fairly modest request of the powers that be: Give us a chance. Don’t make me define my children and myself in racial terms; don’t “impute” a race to me or to any of the millions of Americans who feel as I do. Allow us to describe ourselves simply as Americans. I bet my 50 years as a sociologist that we will all be better for it.

Permission required for reprinting, reproducing, or other uses.

Amitai Etzioni chairs The Communitarian Network, which he founded in 1993. He is University Professor and professor of international affairs at The George Washington University. Among his many books is The Active Society(1968).


Please enter a valid email address
That address is already in use
The security code entered was incorrect
Thanks for signing up