The Mission of the Manhattan Institute is
to develop and disseminate new ideas that
foster greater economic choice and
individual responsibility.


Census' Ethnic Emphasis Outrageous
Thursday, March 30, 2000

By Tamar Jacoby

Like most people, I'm a procrastinator, and when the inevitable envelope arrived from the Census Bureau I let it languish on my kitchen table for a week or so. Part of me was looking forward to answering: Like voting, it seemed a rite of citizenship, my chance to help the government tabulate who we are as a nation.

But I also expected that it would be a time-consuming chore. So, when I finally opened the envelope, I was stunned. Besides my age, gender and who owns the house I live in, all the government wants to know about me is race and ethnicity.

Am I married? Am I employed? Am I a citizen? Washington apparently is not interested in any of this—doesn't think it's relevant to a meaningful national portrait. Some people are asked to respond to these additional queries and others: Fifteen percent of the questionnaires mailed in the last few weeks are what's known as the "long form"—52 broad-ranging questions about who you are and what you do. But the rest of us, in the eyes of the U.S. government, have no characteristics other than skin color. Washington could hardly be more brazen in telegraphing its values—and Americans of every color ought to be outraged.

Of course, like it or not, most people are accustomed by now to the idea that there is no such thing as a plain, unqualified American, only an array of hyphenated varieties. But imagine how bizarre, if not downright frightening, the form must appear to many of the people of color it's supposed to benefit, particularly recent immigrants. One friend of mine—a third-generation Japanese-American, proud of his ancestry,

but dogged all his life by questions about "where he comes from" -is so angered by this single-mindedness that he doesn't plan to fill out the form, and he surely isn't alone.

In fact, it turns out, the offensive form is only the beginning of the Census Bureau's demeaning essentialism. Bowing to growing demand from people of mixed race, the Office of Management and Budget, which helps to oversee the census, had reluctantly conceded that there might be Americans who don't fit neatly into the racial categories listed on the forms.

Pressure from ethnic advocacy groups—particularly black groups such as the NAACP—prevented OMB from adding a box labeled "multiracial." (The fear was that too many

mixed-race people might check it, reducing the minority percentages used for calculate affirmative- action benefits and legislative redistricting.) After much debate, the government gave up the conceit that race is like gender—you are either one or the other and permitted people to

check as many boxes as they wanted. It seemed a small victory: the real-world complexity of the American population triumphing over the divisive, self-interested chauvinism of ethnic advocates.

But now the government has reversed itself again: Under a new OMB "guidance," people who list themselves as white and members of a minority group will be counted as minorities. In the Jim Crow South, this kind of odious accounting was called the "one-drop rule." Now it's official government policy.

In addition to its housekeeping function, the census is also a national ritual, and this year virtually all the trappings of that ritual celebrate racial and ethnic differences. The forms come in six different languages: not just English and Spanish, but also Chinese, Korean, Tagalog and Vietnamese. A hugely disproportionate share of the bureau's advertising budget is being spent in minority communities on ads that appeal shamelessly to ethnic chauvinism.

But isn't there a purpose to all this categorization? Don't social scientists need the racial and ethnic data provided by the census? Even those, with philosophical qualms about making a shibboleth of color can see the utility of comparing the percentage of blacks in the population with the percentage who are employed or own their homes or hold health insurance. It turns out experts disagree about how much of the census data is actually useful or necessary for social science.

The problem begins with the way the race questions are phrased. The form doesn't ask: "Are you Hispanic?" or "Do your ancestors come from Asia?" What the government wants to know is what race "this person considers himself/herself to be."

The census' racial tallies don't match those compiled from birth and death certificates (which don't match each other). Some people's self-identification is so flexible that it changes from week to week with passing fashions: The number of people self-identifying as American Indian, for example, rose noticeably in the wake of the movie "Pocahontas." And even people who are sure about their origins are often befuddled by the census questionnaire—justifiably so. For complicated political reasons, whether one is Hispanic is a separate question ' from what race you consider yourself to be And among the possible answers to the latter question are a number of still further confusing categories, including Chinese, Japanese and Korean—all presented as if they were 4craces" rather than nationalities.

The issue is not just that the census' approach is politically wrong-headed. Far more troubling is the gulf between the government's standardized categories and the fluid, rapidly changing racial and ethnic reality of America. By the second generation, between a third and a half of both Hispanics and Asian Americans marry outside their groups. The number of those who prefer the multiracial designation is expected to multiply exponentially in 'coming decades. It's hard to see what value it has for sociologists or anyone else to label such people by their ancestors' country of origin. Yet the Census Bureau goes on trying—and pretending it is able—to capture and codify this changeable, subjective ethnic landscape.

What's the alternative to the bureau's outmoded schema? One possibility would be to skip the race question entirely, although that is probably politically implausible. Sociologist Nathan Glazer sees some utility in continuing to ask if people are black or white, but beyond that he would drop all questions about ethnic self-identification. (Far more useful, he suggests, would be to ask respondents where they and their parents were born, and what language they speak at home.) This would bring the census back into the world of verifiable fact—although it, too, would certainly raise political hackles. Still, another possibility would be to make ethnicity questions optional and replace the word "race"—not what the form is eliciting in most cases, anyway—with "origin." Possible answers could include African-American, Asian and Hispanic, but also Irish, Italian, mixed ethnic-background white, multiracial and so forth.

Of course, none of this is going to happen soon. In the meantime, the only option may be small acts of rebellion. Some people, like my Japanese-American friend, will be so offended by the racial emphasis that they won't fill out the form at all. Others will check no racial or ethnic box. Still another option would be to check "some other race" and then write in "American"—or better yet, "human." If enough people did that, by the next census in the year 2010, maybe the bureau would finally be prepared to take the measure of America's truly diverse and gloriously multicultural population.

©2000 Newsday

About Tamar Jacoby: articles, bio, and photo



Home | About MI | Scholars | Publications | Books | Links | Contact MI
City Journal | CAU | CCI | CEPE | CLP | CMP | CRD | ECNY
Thank you for visiting us.
To receive a General Information Packet, please email
and include your name and address in your e-mail message.
Copyright © 2009 Manhattan Institute for Policy Research, Inc. All rights reserved.
52 Vanderbilt Avenue, New York, N.Y. 10017
phone (212) 599-7000 / fax (212) 599-3494