A brief history of federal affirmative action action categories


From Harold Orlans, The politics of minority statistics, Society, May 1989, No. 4, at 24, 24.

In 1956, the Committee on Government Contracts, which monitored compliance with President Eisenhower's orders barring contractor discrimination, initiated what became a recurrent survey of contractor personnel. The survey form asked contractors for the number of their "total," "Negro," and "other minority" employees and, if they had many "other," how many were "Spanish Americans, Orientals, Indians, Jews, Puerto Ricans, etc." The Puerto Rican influx to mainland cities was then receiving much attention.

To count Hispanic employees accurately, Hispanic spokesmen pointed out, a separate line was needed. When some critics said that Jews no longer suffered significant discrimination, three leading Jewish organizations agreed to their deletion. After Hawaii became a state in 1959, members of its congressional delegation called for explicit attention to Oriental employees. Accordingly, the revised 1962 survey form asked for the number of all contractor employees and four separate minorities": Negro, Spanish-American, Oriental, and American Indian.

In time, the definition of each group was modified by an interagency committee representing statistical and civil rights agencies. "Negro" became "black." "SpanishAmerican" became "Spanish-surnamed" and then (since a Mexican American's Irish wife had such a surname but an Irishman's Mexican-American wife did not) "Hispanic origin." The difficulties of distinguishing "blacks," "black Hispanics," and "white Hispanics" and "whites" were addressed, if not eliminated, by a trifurcation into "black (not of Hispanic origin), " "Hispanic origin," and "white (not of Hispanic origin)." "Oriental" became "Asian or Pacific Islander," since Polynesian Hawaiians or the Guamanians in southern California were not "Oriental." Neither were immigrants from India, Pakistan, and Sri Lanka who in 1977 were reclassified from "white" to "Asian" after requesting inclusion in affirmative action programs.

In the Reagan era, bureaucratic resistance hardened, for Arab Americans who sought a similar re-classification are still "white," as are Asian Tatars, Kurds, and Iranians. "American Indian" became "American Indian or Alaskan Native" because Eskimos and Aleuts are not Indians. Thus, by 1977 a recurrent government survey of Negro employees had evolved into an odd five-fold classification of the world's peoples, or of their emigrants and descendants in this nation.

As survey instructions note, the classification reflects disparate factors: race ("Black racial groups of Africa"-but not of Cuba or Melanesia); language or culture ("Spanish culture.., regardless of race"); above all, geography or country of origin, which has the virtues of identifiability and stability. Although not defined, "origin" is central to the classification. "Blacks" and "Hispanics" are persons with "Black racial" or "Spanish" origin, respectively; "whites," "Asians," and "American Indians" are persons "having origins in any of the original peoples" of Europe, Asia, and North America, respectively. The idea of "origin" is clear only if it is not pursued, for the "original peoples" of countless lands (for example, Finland, the Ukraine, the Indus Valley, Palestine, Sicily) are lost in the mists of history.

The classification is full of contradictions. Thus, a descendant of the "original peoples" north of the Rio Grande is an "American Indian"; but south of the river, such a person is "Hispanic" although of Indian stock. Immigrants from Cuba are "Hispanic," but from Brazil they are "white," "black," or "Hispanic." A Sephardic Jew is "Hispanic"; a Hasidic Jew is "white"; an Ethiopian Jew is "black"; if they all enter the United States from Israel, what is their country of origin?

NEXT: Today in Supreme Court History: October 30, 1735

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. Affirmative action is just the modern perpetuation of racial prejudice. Thankfully, I work in one of the more objective fields, where contributors can be judged not by their phenotypes but by their work product.

    1. The Republican Part is the modern perpetuation of racial prejudice.

      1. Which part?

        Probably the right hand or something.

      2. The only minority is clingers

  2. David,

    I don’t think it’s correct to conflate an executive order barring discrimination by federal contractors with affirmative action (such as SBA Section 8(a) requirements).

    1. He’s not conflating them. He’s explaining the history behind the current classification categories.

      1. Fine, in the alternative, the headline is wrong. Nothing in the quoted material relates to classifications for affirmative action programs.

        1. No, the headline is not wrong. The current federal affirmative action categories* are directly descended from the choices allowed in the 1956 survey of government contract compliance.

          * To be more specific, the current seven “race and ethnicity categories” mandated to be reported on the EEO-1 can be traced back to that survey. Affirmative action on the basis of gender, disability and veteran status has different roots.

          1. Moreover, the original meaning of “affirmative action” wasn’t “racial and ethnic preferences.” The original contracting compliance programs were intended to get contractors to take “affirmative action” to ensure that they were recruiting and considering all candidates fairly, and not passively allowing the inertia of past discriminatory practices to dictate their hiring policies.

            1. And if it had remained so would be entirely laudable by everyone. Here, shockingly, the millennial “check your privilege” claim actually has some merit.

              1. Saying that affirmative actions means blacks (&c) are the privileged ones these days is a pretty hot take.

  3. What is even more remarkable is the fact that none of these classifications address the question of persons of mixed ancestry. Instead, they all seem to be premised on the so-called “one-drop rule” once popular in the South. Hence, you get people claiming various preferences based on only tiny and even imperceptible levels of a preferred ancestry. This issue is being deliberately ignored because those most directly involved know that the only solution to that issue is the promulgation of racial definitions much like those used during the Third Reich and in South Africa. The absurdity of all this is illustrated by the fact that the South African authorities once classified Japanese citizens as “honorary white people.”

    1. Modern ID pol considers all East Asians honorary whites.

  4. Of course, if the law is to apply to all, none of this is necessary; just expensive.

  5. In a country of 330 million, a perfectly drafted law which operates only as intended, free of anomalies and injustices, is a unicorn. Yes, that argues for smaller government and fewer laws. It also means that when the decision is made to create a national rule, the need to revisit it from time to time to optimize tradeoffs and minimize unintended harm should be taken for granted. And like most original legislation that takes X steps forward and Y steps back, the best we can hope for in the corrective updates may be that X>Y.

    But even that modest goal depends on parties negotiating in good faith, which in turn requires at least a modicum of faith by each side in its opponent. Nowadays the ascription of good faith to a political opponent makes unicorns look plentiful.

    The taxonomy of groups covered by affirmative action needs updating to correct the anomalies you’ve shown, and then some. Neither side trusts the other, so I won’t hold my breath.

    1. I ascribe *arbitrariness* (and racism) to just about every policy of racial preference (with a few exceptions like studying sickle-cell anemia, if that can be construed as race-specific).

      What would opponents and supporters of affirmative action negotiate about? If supporters say “help us fix the policy you don’t think should exist in the first place,” the best reply would be “you’ve had these policies for several decades and still haven’t solved the problem of arbitrariness? Maybe you have bigger problems than tweaking the definition of hispanic.”

      1. You make my point. Racism may not be exactly the same thing as bad faith, but ascribing either one to your opponent works to the same effect as ascribing the other.

        1. One of WWWebster’s definitions is “racial prejudice *or* discrimination” (emphasis added). So racial discrimination is racist. An alternative definition by which discrimination against some races is racist and discrimination against other races *isn’t* racist – such a definition would itself be racist.


          1. In your hypothetical negotiations with supporters of affirmative action, I suppose opponents should use soothing euphemisms like “your preferred policy” or “your social system,” just as the supporters would presumably tone down the rhetoric about how it’s the *opponents* of affirmative action who are the *real* racists.

            1. In my hypothetical, both sides insult each other, because that’s what politics is right now. Were it up to me, both sides would use soothing euphemisms. Shocking, I know, but I’ve found that works better than name-calling when the goal is a productive negotiation.

              1. Of course it’s nice to be important, but it’s also important to be nice.

                What kind of compromise, where each side gets something, would be viable?

                1. (Just to be clear, I was asking from curiosity, not as a rhetorical question)

                2. Other than narrowing the definitional anomalies? Beats me. IMO affirmative action would be much more coherent if it was means tested, but that’s probably beyond the scope of this hypothetical.

  6. This is a real question. I’m not trying to make a point, but to get information. Why are “Eskimos and Aleuts . . . not Indians”? If we use the (admitedly weird) nomenclature “Indian” to refer to the people who crossed over from Beringia to North America at the end of the last ice age, why doesn’t that include Eskimos and Aleuts? Did their ancestors in fact come into North America at a different time or by a different route or from a different population? I’ve never heard that, but it’s interesting if true. Any citations (for non-specialists) would be appreciated.

    1. Not an expert here, but speaking as someone who lived in Alaska. Yes, they crossed the Bering Strait when it was frozen and stayed while others moved south, so Native Alaskans are all descended from the same stock as groups as geographically diverse as the Commanche in Texas and the Aztecs in Mexico.

      That said, the Native Alaskans I interacted with liked that phrase rather than “Indian” because they wanted to empathize a different lifestyle pre-European settlement than the mounted Plains Indians that people think of when they say Indian (even though the Indians in the Pacific Northwest are almost identical). That, and that they wanted to emphasize that they have been in the same location for a long time.

      1. Thanks mad_kalak. Being of German heritage (among others) I understand wanting to distinguish myself from others of similar heritage. But the native Alaskans ought to know that the plains Indians didn’t mount horses until after the Spanish showed up with horses from Europe in the 1500’s, because the early Americans killed off the New World horses (and other megafauna) ASP after arriving from Beringia.

      2. Actually, the Arctic natives were part of a recent second migration. Paleo-Indians were here around 20000 ybp but Arctic natives came around 2500 ybp. While “Indians” share close genetics (except for some SA populations with more variety) and cultural attributes or language groups, Arctic natives are distinct and far more closely related to Siberian tribes, to where some are were only separated by European colonization around the Bering Strait.

        It also helps they’re nowhere near the West Indies.

      3. Nooooo!

        The plains Indians (and I do hate that naming convention) were assuredly not mounted pre-European settlement, because the horses we so associate with the pains Indians were introduced to the Americas by those same European settlers (though more often, conquistadors).

        For ten thousand years America was equine free, until they came back aboard the Europeans ships.

        1. yes, I know the plains Indians were not mounted until they obtained Spanish horses. That doesn’t change the point that Native Alaskans want to disassociate themselves from them. However, I did not know there were multiple waves of migration.

      4. And while we’re talking about the wishes of some ethnic groups to distinguish themselves from others, it’s worth pointing out that the word “Inuit,” which certain people-formerly-known-as Eskimos prefer to be called, is roundly rejected by the Yupik (a kind of “Eskimo” dwelling in Alaska and Eastern Siberia), who insist that they are not Inuit.

  7. Holding racial classifications to an unrealistic standard for fidelity doesn’t prove racial classifications are bad policy.

    By the standard set here, what’s a search or seizure is full of contradictions as well.

  8. Anyone with the slightest familiarity with Latin America knows that those in power are often white. Or at least they look white European. So how does a Mexican television star who looks like she grew up in Westchester but speaks English accent fit under a discriminated group? She is not a part of any racial group but Western European. And she would certainly suffer no discrimination – other than by tight-assed lily-white American feminists. Too much lipstick and lingerie, no doubt.

Please to post comments