Reporters are enamored of Census Bureau stats, but when it comes to America's complicated racial picture, the numbers are misleading
By TODD GITLIN
Numbers games are the national pastime, but whereas baseball fans content themselves with reports of homers already hit and pitches already pitched, so-called experts in business and politics lavish attention on speculation. We are a civilization more indebted to prognostication than the most caricatured of "savages." Oh, do we love precision! "The economy should grow at a rate of 2.1 percent for the next year" satisfies our primitive impulse to put the future in order.
Recently, the Census came out with a population projection and the media predictably swung into compliance. The New York Times put the story on the first page of its national section under the headline "Census Sees a Profound Ethnic Shift in U. S.," with the subhead, "By 2050, Non-Hispanic Whites Will Decline to a Slim Majority." Quickly, we began to hear the new projections cited as hard and fast evidence that race-based politics make sense. If whites are dwindling, presumably a little more push from race-based identity activists ought to be enough to push people of color into power. The numbers are brandished so firmly as if to say, It's going to rain tonight, take an umbrella.
First of all, what's the news? Already in 1992, the Census anticipated that the non-Hispanic white population would decline by 2050 to 52.7 percent. The current report estimates that the non-Hispanic whites will end up at 52.8 percent. No huge news here. The big shift from the 1992 projection to the 1996 projection lies elsewhere: in the higher anticipated numbers of Hispanics and the lower anticipated numbers of African-Americans and Asians. But what kind of headline would that make -- "Census Sees More Hispanics Than It Previously Saw but Fewer Blacks, Asians"?
So the first point is that projections come and go, but rarely is the current round juxtaposed to the previous one. Institutional amnesia applies. Were readers reminded of the last round, all of them would have to be seen as evanescent. Moral: These projections ought to be taken with enough salt to melt the snow on a wintry New York street (or heart).
Consider further that such projections assume that people will continue to be identified as they are now. But racial and ethnic categories don't grow on trees. There is one race, and it's called human. Our labels are matters of convention, often racist in origin, and conventions change. The category "Hispanic" did not even appear on the Census until 1970. The number of people identifying as Native Americans shot up as the stigma dwindled and the benefits increased. On census forms, Americans are told to pick a racial and ethnic identification from among a select list of categories. Our answers depend, in part, on the alternatives offered.
The Census Bureau projections assume a constant rate of immigration. They also ignore intermarriage. As they play out assumptions about the next three generations, they assume either that people from one group are not going to have children with people from other groups, or that if they do, the children will be identified as people are identified today. But this is statistical hubris. What we know about recent decades is that mixing is ever more common. Yet the Census continues to prohibit people from identifying themselves as members of two or more races. They can say "white" or "black," or, if they insist, "other," but not "white" and "black." Mixed-race activists have been demanding a change to permit a child of mixed race to identify with both parents, not one of the above or the residual "other." The Census hasn't yet decided what to do about 2000. If they open up, we can expect yet more blurring of statistical lines.
The point is that nobody is entitled to be quite so certain what people will mean by race or ethnicity fully half a century on. The Harvard sociologists William Alonso and Mary Waters have made the point: "Had there been population projections by groups during the period of peak European immigration (1880-1920), trying to guess what the nation would 'look like' in the year 1950 or 1980 or 2000, those predictions would almost certainly have been wrong in more ways than one. They would have seriously overestimated the proportions of some European origin groups and underestimated others." In the words of the Florida demographer Paul R. Spickard, "Almost no White American extended family exists today without at least one member who has married across what two generations ago would have been thought an unbridgeable gap."
Moreover, the labels shift their meanings. Technically, "Hispanic" does not mean Spanish-speaking, but rather, says the Census Bureau, "a person of Mexican, Puerto Rican, Cuban, Central or South American or other Spanish culture or origin, regardless of race." The label was an administrative convenience manufactured for the Census. Its usefulness for describing reality is something else again. It yokes together second-generation South Texas Chicanos and first-generation Los Angeles Chicanos, New York Puerto Ricans and Florida Cubans. The people thereby roped together do not share common music or food, and frequently, not even language, for by the third generation, many Hispanics don't speak Spanish. "Hispanic" is not a culture, and not nearly so hard and fast as an "identity."
In the future, Hispanics will choose how ethnic they want to be -- which makes their ethnicity far from the primordial destiny that the term "ethnos" implies. Many "non-Hispanic whites" and "Hispanics" will agree on political and cultural and economic matters more than they disagree, certainly more than the Dominican immigrant in Washington Heights with the Cuban immigrant banker in Miami. If they disagree over welfare stipends or foreign aid, their class position will be speaking louder than the language of their parents.
Numbers are stupid things, as Ronald Reagan might have said. They don't speak for themselves. They don't add up to the national fate. Reporters should stop deferring to the false god of precision. How about a new box at the top of the daily Times, opposite "All the News That's Fit to Print" -- a prophecy from the past that didn't pan out?
Todd Gitlin is the author of "The 60s: Years of Hope, Days of Rage" and, most recently, "The Twilight of Common Dreams: Why America is Wracked by Culture Wars." He is Professor of Culture, Journalism and Sociology at New York University. He can be reached at gitlint@is.nyu.edu