A Brief History of Census “Race” (E4)

Essays on the Color Line and the One-Drop Rule
by Frank W Sweet
June 3, 2010

T

he U.S. federal census was founded to apportion congressional representation among the states. In order to achieve additional goals, it switched in 1850 from recording households in summary, to recording individuals in detail. It became self-administered in 1960 to reduce costs. It has always been a political instrument of the administration in power. Today, the census encourages identity politics and so wavers between the goal of capturing “race” as a form of ethnic self-identity, and the equally desired but conflicting goal of capturing “race” as involuntary physical trait.

This brief history covers three major topics: The Changes of 1850 and 1960, Politics and Confidentiality, andThe “Race” Question. The third topic, the history of the “race” question, is then presented in six sub-topics:Changes in “Racial” Terminology, Changes in “Racial” Categories, Changes in “Racial” Criteria, Changes in Stated “Racial” Goals, The Religion Question Controversy of 1956, and The Legality of Refusal.

The Changes of 1850 and 1960

Figure 1 shows a page fragment from the first census, that of 1790. Each line represented a household. The columns recorded how many each of: men, women, children, slaves, etc. lived within the household. No need was felt to record individuals by name, age, race, or role in the home (husband, wife, dependent child, servant, etc.). Only the totals for each household were recorded. Hence, for example, the data offered no way to measure the rate of intermarriage.

Figure 2 shows a page from the 1880 form. It was typical of forms used during the century from 1850 to 1950, when census-takers asked and recorded many data items for each individual. The specific items captured varied from one census to the next, sometimes dramatically, but a few items, such as “name” and “color” were always present.

Figure 3 shows the form used in 1960, the first self-administered census. The advantage of a self-administered census was cost. It was cheaper for Americans to record their own information than to pay census-takers. The disadvantage was that it encouraged respondents to treat the “race” question as voluntary ethnic self-identity, rather than as an involuntary biological or genetic classification. More about this later. The 1960 form employed the same columns-and-rows layout used by census-takers in prior years. It was not well designed for a resident to fill in easily.

Figure 4 shows the 2010 form. It was the most frugal self-administered census questionnaire yet.

Politics and Confidentiality

Since 1850, Americans have lamented that the census has fallen from being an objective scientific data-collection instrument to being a propaganda tool of the administration. In fact, the census has never been objective. From the very first census that took on expanded goals beyond congressional apportionment (that of 1850), the census has been used to further the administration’s political aims. Indeed, the records of debates surrounding the census’s 1850 expansion suggest that its potential use as propaganda tool was one of the reasons why Congress expanded its mission. (1)

In 1850, following administration directives, the census bureau published reports that Afro-Euro intermarriage produced insane and feeble-minded children. At the time, the nation’s elite believed that intermarriage would destroy U.S. society. (2)

During the 1930s, following administration directives, the census bureau published frequent false reports that unemployment was falling and the economy was recovering. At the time, the administration believed it vital to convince the electorate that the Great Depression was coming to an end. (3)

During World War II, The Second War Powers Act explicitly voided census confidentiality. Following administration directives, the census bureau provided the FBI and local law enforcement with the names and addresses of American citizens of Japanese ancestry, so that they could be rounded up and sent to concentration camps.(4) Nevertheless, the broken promise of confidentiality continued to be promulgated. The administration believed it vital to convince the electorate that census data was confidential while, in fact, using it to target certain citizens.

Over the past decade, the Patriot Act explicitly voided census confidentiality. From 2002 to 2004, following administration directives, the census bureau routinely handed census records on Arab-Americans over to Homeland Security, so that these U.S. citizens could be monitored for disloyalty.(5) Nevertheless, the broken promise of confidentiality continues to be promulgated. The administration believes it vital to convince the electorate that census data is confidential while, in fact, using it to target certain citizens.

The “Race” Question

The following discussion focuses on how the census handles the U.S. Black/White dichotomy. It does not address other so-called “races” or ethnicities.

Although some form of “color” or “race” has been recorded since the first census, the terminology applied, the categories imagined, the criteria of “racial” classification, and the justification given for forcibly categorizing American citizens into mutually exclusive “races” have all varied greatly over the centuries.

A problem faced by the census today is that many Americans no longer take the “race” question seriously. This is a consequence of the 1960 switch to self-reporting. Having opened the door to self-classification, the bureau has been unable to stop millions of Americans from checking off the “other” category and filling in “human,” or something equally frivolous or non-racial. To discourage such lack of cooperation, the bureau has threatened criminal prosecution of anyone who fails to choose a government-approved “race.” More about this later.

Changes in “Racial” Terminology (6)

From 1790 to 1880, the census recorded only “color,” not “race.” This was not mere semantics. During this period, skin complexion was the prime determinant, not heredity nor ideology. For example, Eston Hemings was recorded on the 1830 census as a “White” man whose “Colored” mother (Sally) lived with him. Millions of Americans of the time, in both the free states and the slave states, were accepted as White despite having known slight Black ancestry.

In 1890, the census dodged the question of whether the qualities (Black, white, etc.) reflected mere skin-tone (as previously), or whether they denoted some sort of invisible ancestry (as subsequently). Hence, the question asked was, “Whether White, Black, … etc.” and had no noun modified by those adjectives.

The word “race” first appeared in the census in 1900. From 1900 to 1940, the census asked each person’s “color or race.”

In 1950, the word “color” was dropped and the census asked only for the person’s “race”.

In 1960 (the first self-administered census), the census again dodged whether the qualities (Black, white, etc.) reflected skin-tone or invisible ancestry. The question asked was, “Is this person white, black, … etc.,” leaving the noun being modified to the reader’s imagination. Unfortunately for census goals, many Americans saw this as asking for their voluntary ethnic self-identity.

In 1970, “color” made a brief comeback. The census asked each person’s “color or race,” as in the 1900-1940 period. This was the last census to use the word “color.”

In 1980 (as in 1890 and 1960), the census again dodged whether the qualities reflected skin-tone or invisible ancestry. As in 1960, the question asked, “Is this person white, black, … etc.,” with no noun.

Finally, every census since 1990 has asked for the person’s “race” with no ambiguity, with no longer any mention of color, and with no mention of ethnicity.

Changes in “Racial” Categories (7)(8)

From 1850 to 1880, there were three “racial” categories: White, Black, and Mulatto. Census-takers’ instructions repeatedly stressed the importance of distinguishing Mulatto from Black. Of course, there was no way of ensuring that different enumerators saw the same thing. They often classified siblings differently, or the same person differently on different censuses.

The 1890 census instructed enumerators to classify people into one of five precise gradations. The five categories were Black, Mulatto, Quadroon, Octoroon, and White. Few census-takers tried seriously to obey this impossible directive.

In 1900, for the first time, the census employed an explicit one-drop rule. There were only two categories: White and Black. Having any trace of known African ancestry forced a person into the latter category.

From 1910 to 1930, the census went back to the three “racial” categories of 1850-1880 (White, Black, and Mulatto). Again, different enumerators often saw different things. They often classified siblings differently, or the same person differently on different censuses. An interesting item is that the 1930 census instructed census-takers to classify Mexican-Americans as non-White. This produced an uproar from Mexican-Americans, as well as an international protest to the State Department from the nation of Mexico. Consequently, in 1940 census-takers were ordered to classify Mexican-Americans as White.

In 1940 the census returned to the one-drop rule of 1900. With one year’s exception, subsequent census instructions to this day suggest that any trace of African ancestry makes you involuntarily Black. The exception was the 1950 census. It was the only census in U.S. history that instructed enumerators to record membership in a triracial community as a legitimate “race.”

Report persons of mixed white, Negro, and Indian ancestry living in certain communities in the Eastern United States in terms of the name by which they are locally known. The communities in question are of long standing and are locally recognized by special names, such as “Croatian,” [sic] “Jackson White,” “We-sort,” etc. Persons of mixed Indian and Negro ancestry and mulattoes not living in such communities should be returned as “Negro.” (8)

Because immigrants from the Balkan state are usually called “Croats,” not “Croatian,” it is likely that the above instruction actually meant “Croatan.” “Croatan” is an obsolete title for the Lumbees of Robeson County, North Carolina. (9) It is also interesting that 1950, the only year in which people could be  “racially” labeled as “Jackson White,” “Croatan,” “Melungeon,” etc. was also a year when, according to most researchers, any census-taker who did this would have taken his life in his hands, risking a beating or worse.(10) At that time, being labeled “non-White” was an insult.

Since 1960, self-classification has enabled people to use whatever criteria they wish. This led many Americans to interpret the question as relating to voluntary ethnopolitical self-identity, rather than the traditional concept relating to continent of ancestry.

The 1970 census was the first to allow respondents to choose “other” and fill in a blank line with a “race” other than one of the government-approved choices.

Increasing numbers of Americans have chosen “other” ever since, most filling in something frivolous or non-responsive, such as “human” or “American.” As of 1980, 7 million Americans did this. In 1990, the number of Americans not taking the “race” question seriously had grown to 10 million. By the 2000 census, “other” had become the third most popular “race” in the United States, chosen by about 15 million Americans.(11)

The census bureau saw this disrespect as a problem. Their mandate is to force all Americans to choose one of the government-approved “races.” To stop the trend, the Census Bureau resolved in 2004 to remove the “other” choice from the 2010 census form entirely. This led to a conflict between the Executive Branch under which the Census Bureau operates, which approved the removal, and the Legislative Branch that provides their funding, which disapproved. In the 2005 appropriations bill, Congress gave the Census Bureau a harsh choice: either restore the “other” choice to the 2010 form, or lose their funding.(12) The Bureau restored the “other” choice but also changed the instructions, defining “race” as something physical, no longer based on self-identity. More about this later.

The 2000 census was the first to allow respondents to check off multiple “race” boxes. Nevertheless, with one exception, negligible numbers of African Americans check off anything other than simply “Black.”(13) To most African Americans, including the current U.S. president, if you are a member of the Black ethnopolitical community, you are entirely Black, and it would be “racial treason” to check off White also, no matter much Euro ancestry you have.(14) The exception is that about half of intermarried parents check both boxes for their children. (The other half overwhelmingly check off “Black.”) (15)

Changes in “Racial” Criteria

Before 1960, when census takers assigned people to the approved “races,” the rules for classifying people were not always clear. Nevertheless, it appears that census takers probably followed “racial” classification rules in effect in their society at the time, as reflected in court cases. They seem to have employed four different methods of assigning “racial” membership: appearance, blood fraction, association, and the one-drop rule.

Until about 1820, most people in authority were guided by a person’s physical appearance. If you looked White to the census-taker, then you were White (for example, Sally Hemings’s son Eston). But if you looked Black then you were Black (like Eston’s brother James). From 1830 until about 1890, blood fraction or association decided the issue. If most of your ancestors were White and you were accepted as White in society, then the census-taker classified you as White. Finally, from around the turn of the twentieth century to today, the one-drop rule of invisible Blackness emerged and gradually replaced all other methods. If you had any known African ancestry, no matter how slight, then you were involuntarily classified as Black. (16)

Since 1960, as mentioned earlier, Americans have interpreted the census “race” question as asking about voluntary ethnic self-identity. The wording of the 2000 census instructions probably encouraged this interpretation. It ordered, “indicate what this person considers himself/herself to be” [emphasis mine].(17)

Forbidden by Congress from removing the “other” choice entirely, the Census Bureau sought a different way of forcing Americans to take the “race” question seriously. As mentioned, to meet their mission, they needed to discourage the millions who in 2000 checked off “other” and wrote in something frivolous, like “human.” The solution was to remove the “considers” instruction in 2010. The census no longer allows one to select the “race” that a “person considers himself/herself to be.” Instead, it now demands to know what “is the person’s race.” This, however, led to a dilemma. On the one hand, “race” is no longer a matter of choice but on the other hand, the instructions no longer say just what criteria are to be used.(18)

Changes in Stated “Racial” Goals

The earliest mention of the need to classify free Americans by “race” was in the late 17th century, long before the census (or the nation) was founded). At that time (1691-1723), the reason given was that: (1) Racial classification was necessary to prevent intermarriage, (2) Intermarriage had to be prevented to stop the birth of Americans of mixed Euro-Afro ancestry, (3) The existence of mixed-ancestry Americans would enable alliance between European and African involuntary forced laborers, and (4) Such alliance would overthrow English rule and turn colonies into free maroon communities.(19) In fact, Latin American colonies encouraged intermarriage (and consequent mixed-ancestry colonists) and yet were not overthrown.

In the 1830s, as a consequence of the Nat Turner incident, free African Americans came to be seen as a threat. The justification for “racially” classifying Americans on the census was to prevent African-American migration to the free states. Several Midwestern states’ constitutions forbade African-American immigration.(20) In fact, no evidence ever emerged that free African Americans were a threat.

In the 1850s, intermarriage was believed to produce mentally defective offspring. Census “scientists” published findings that mixed-ancestry children were usually born retarded or insane. And so, the justification for “racially” classifying Americans on the census was to prevent intermarriage and thus protect public health.(21) In fact, even cursory examination of the data shows it to be falsified, since many communities reported more insane mixed-race individuals than their entire population.

In the Jim Crow era (1900-1965), the elites in power argued the importance of preserving “White racial purity.” The justification for “racially” classifying Americans on the census was that it helped to prevent contamination of the White “race.”(22) In fact, liaisons between White males and Black females were encouraged while the reverse was often punished with death by public torture.(23)

Since 1970, U.S. society has tried to atone for slavery and the state-sponsored terrorism of the Jim Crow era by enforcing laws, regulations, entitlements, policies, and practices that favor African Americans. Since that time, “racially” classifying Americans on the census has been justified on the grounds that such classification is needed in order to fight “racism.”(24) In fact, no “anti-racist” or civil rights federal regulation or law enforcement over the past 30 years has ever used census data.(25)

The practice of forcibly classifying Americans by “race” has been, and continues to be justified by rationalizations ranging from the need to “keep Blacks in their place,” to the need to “atone for past racism,” to the need to “fight future racism.” That the reasons are often contradictory and have always been counterfactual shows that involuntary “racial” classification is important to U.S. society, although Americans cannot rationally articulate just why it is so important. Hence, the practice is unlikely to end within the next few centuries.

The Religion Question Controversy of 1956 (26)

In 1956 the Census Bureau wanted to demand every American’s religion in the 1960 census. Religious leaders had previously been asked about their size of their congregations, the seating capacity of their churches or synagogues, and the value of their real estate in 1830, I860, and 1870. But 1960 would be the first time that every American would be required to confess his or her religion on the federal census.

Two parallels with today’s “race” question are striking.

First, the Census Bureau’s wording of the question, “What is your religion?” assumed that religion was an innate part of self-identity. Sociologist William Petersen suggested that the form should first ask, “Have you a religion?” and only if the answer is “yes” should a follow-up ask “What is it?” But this advice met derision. To the Census Bureau, this was as foolish as asking Americans today, “Have you a race?” As with “race” today, what some people saw as an arbitrary and divisive label, the Bureau enforced as an essential hereditary trait.

Second, the advocates’ stated justifcation, then as now, was to “protect minorities from discrimination.” What the Bureau actually wanted to know was evident in the first draft of the question. It simply asked, “Are you a Jew?” Supporters insisted that only by forcing every Jew in America to confess their Jewishness, could Jews be protected from bigotry.

The divergence from today’s “race” question is also striking. The U.S. Jewish community rose up in horror at the proposal. Jewish organizations lobbied and aroused public sympathy to defeat the idea. They pointed out that forcing Jews to identify themselves was precisely the first step that the Nazis had used in their extermination program. In the end, their objections triumphed. Congress refused to authorize the plan, and the Census Bureau eventually backed down.

In contrast, all major African-American political organizations today strongly support African Americans’ being forced to identify themselves.

The Legality of Refusal

In order to encourage compliance, the Census Bureau promulgates the claim that refusal to answer is a crime. U.S. Census Bureaus spokeswoman Shelly Lowe told the Washington Times that anyone over the age of 18 must fill out all the questions and any such person who refuses to answer “any of the questions” faces a $5,000 fine.(27)

The $5000 fine is questionable. On the one hand, United States Code, Title 13 (Census), Chapter 7 (Offenses and Penalties), SubChapter II, says that the maximum penalty for refusing to answer the census is a $100 civil fine. The maximum penalty for deliberately lying on the census is a $500 civil fine. Themaximum penalty for interfering with a census-taker with the intent of causing an inaccurate count (virtually impossible to prove) is $1000 plus one year in prison. Here is the actual statute:

221. Refusal or neglect to answer questions; false answers

(a) Whoever, being over eighteen years of age, refuses or willfully neglects, when requested by the Secretary, or by any other authorized officer or employee of the Department of Commerce or bureau or agency thereof acting under the instructions of the Secretary or authorized officer, to answer, to the best of his knowledge, any of the questions on any schedule submitted to him in connection with any census or survey provided for by subchapters I, II, IV, and V of chapter 5 of this title, applying to himself or to the family to which he belongs or is related, or to the farm or farms of which he or his family is the occupant, shall be fined not more than $100.

(b) Whoever, when answering questions described in subsection (a) of this section, and under the conditions or circumstances described in such subsection, willfully gives any answer that is false, shall be fined not more than $500.

(c) Notwithstanding any other provision of this title, no person shall be compelled to disclose information relative to his religious beliefs or to membership in a religious body.

Sec. 222. Giving suggestions or information with intent to cause inaccurate enumeration of population

Whoever, either directly or indirectly, offers or renders to any officer or employee of the Department of Commerce or bureau or agency thereof engaged in making an enumeration of population under subchapter II, IV, or V of chapter 5 of this title, any suggestion, advice, information or assistance of any kind, with the intent or purpose of causing an inaccurate enumeration of population to be made, shall be fined not more than $1,000 or imprisoned not more than one year, or both.(28)

On the other hand, recent Census Bureau claims apparently rely on 18 USC 3571 for the $5,000.00 figure. The Sentencing Reform Act of 1984 changed the maximum fine for federal “infractions” (not “crimes”) to $5,000.00, superseding sentences specified in individual statutes unless the specific statute exempts itself, which Title 13, Sections 221-222 does not seem to do. Specifically, Title 18 USC, Part II, Chapter 227, Section 3571 states:

(e) Special Rule for Lower Fine Specified in Substantive Provision.— If a law setting forth an offense specifies no fine or a fine that is lower than the fine otherwise applicable under this section and such law, by specific reference, exempts the offense from the applicability of the fine otherwise applicable under this section, the defendant may not be fined more than the amount specified in the law setting forth the offense.

Whatever the maximum allowable fine, only two court cases have ever been argued over refusal to answer the census. In U.S. v. Sharrow, 1962 (309 F.2d 77) the U.S. Second Circuit Court of Appeals ruled that the reason given for refusal (that the census did not record disfranchisement) did not suffice to prove the census unconstitutional. In Morales v. Daley, 116 F. Supp. 2d 801, 820 (S.D. Tex. 2000), the U.S. Second District Court ruled that the reason given for refusal (unreasonable search and seizure prohibited by the 4thamendment) did not suffice to prove the census unconstitutional. In both cases, the person was fined the maximum of $100.

Note that the written statute only authorizes a fine for refusal to answer “to the best of your knowledge.” If the race question intends to capture voluntary self-identity, then any answer you give is, by definition, “the best of your knowledge.” If it intends to capture something objectively definable, then its lack of definition automatically makes any answer you give “the best of your knowledge.” The statute authorizes a heavier fine only for deliberately lying. It is inconceivable that anyone could be prosecuted for lying about what “race” they consider themselves to be.

Incidentally, Mary Lee and I are friends with an attorney, accredited to plead before the U.S. Supreme Court, who longs for the opportunity to argue against the “race” question. Given that, in 2000, 15 million Americans refused even to take the question seriously, and that they did so with impunity, the attorney’s wish is unlikely to be granted.

Click here for an animated YouTube version of this topic.


References

  1. Margo J. Anderson, The American Census: A Social History (New Haven: Yale University, 1988), 34-35.
  2. Ibid., 28-29; but for the definitive explanation, see Lawrence Raymond Tenzer, The Forgotten Cause of the Civil War: A New Look at the Slavery Issue (Manahawkin NJ: Scholars’ Pub. House, 1997), Chapter 3, “Racial Theory and the Pre-Civil War Census.”
  3. Anderson (1988), 161-67.
  4. “Confirmed: The U.S. Census Bureau Gave Up Names of Japanese-Americans in WW II,” Scientific American, March 30, 2007.http://www.scientificamerican.com/article.cfm?id=confirmed-the-us-census-b
  5. “Department of Homeland Security Obtained Data on Arab Americans From Census Bureau” Epic.org, July 23, 2004.http://epic.org/privacy/census/foia/
  6. For a collection of downloadable census forms from 1790 to the present, visithttp://backintyme.com/rawdata/censusforms
  7. For all enumerator instructions from 1850 to the present, downloadhttp://backintyme.com/rawdata/censusforms/instructions.doc
  8. Enumerator instructions were summarized fromhttp://usa.ipums.org/usa/voliii/tEnumForm.shtml
  9. Gerald M. Sider, Lumbee Indian Histories: Race, Ethnicity, and Indian Identity in the Southern United States (Cambridge UK: Cambridge University, 1993).
  10. Brewton Berry, Almost White (New York: Macmillan, 1963), 38.
  11. See “Racial Categorization in the 2010 Census” (April 7, 2006) minutes of meeting, downloadable from link. Specific remarks by Commissioner Michael Yaki on page 9, lines 10-19.http://backintyme.com/rawdata/censusforms/meetingminutes.doc
  12. Ibid., lines 20-21.
  13. Frank W. Sweet, Legal History of the Color Line: The Rise and Triumph of the One-Drop Rule (Palm Coast FL: Backintyme, 2005), 113-115.
  14. Randall Kennedy, Sellout: The Politics of Racial Betrayal (New York: Pantheon, 2008).
  15. Sweet (2005), 291-96.
  16. Ibid., 315-21.
  17. http://backintyme.com/rawdata/censusforms/2000.jpg
  18. http://backintyme.com/rawdata/censusforms/2010.jpg
  19. Sweet (2005), Chapter 8, “Why Did Virginia’s Rulers Invent a Color Line?”
  20. Sweet (2005), 331-33,
  21. Tenzer (1997), Chapter 3, “Racial Theory and the Pre-Civil War Census.”
  22. Joel Williamson, The Crucible of Race: Black/White Relations in the American South Since Emancipation (New York: Oxford University, 1984).
  23. Philip Dray, At the Hands of Persons Unknown: The Lynching of Black America (New York: Random House, 2002).
  24. See “Racial Categorization in the 2010 Census” (April 7, 2006) minutes of meeting, downloadable from the link. Specifically, see the remarks by Yaki, Lee, and Prewitt on pages 24-31.http://backintyme.com/rawdata/censusforms/meetingminutes.doc
  25. Ibid., See Kirsanow comments on pages 30-31.
  26. Kevin M. Schultz, “Religion as Identity in Postwar America: The Last Serious Attempt to Put a Question on Religion in the United States Census” Journal of American History, September 2006.http://www.britannica.com/bps/additionalcontent/18/22453332/Religion-as-Identity-in-Postwar-America-The-Last-Serious-Attempt-to-Put-a-Question-on-Religion-in-the-United-States-Census
  27. See the claim made by Census Bureau spokeswoman Shelly Lowe to the Washington Times, reported athttp://www.cbsnews.com/8301-503544_162-5095844-503544.html
  28. http://people.howstuffworks.com/question345.htm
If you liked this essay, leave a tip using bitcoins.
Deposit to address: 1GDGfpdvoP5xw5bCJzazCyJoCKbQdJd6jh

Frank W. Sweet is the author of Legal History of the Color Line (ISBN 9780939479238), an analysis of the nearly 300 appealed cases that determined Americans’ “racial” identity over the centuries. It is the most thorough study of the legal history of this topic yet published. He was accepted to Ph.D. candidacy in history with a minor in molecular anthropology at the University of Florida in 2003 and has completed all but his dissertation defense. He earned an M.A. in History from American Military University in 2001. He is also the author of several state park historical booklets and published historical essays. He was a member of the editorial board of the magazine Interracial Voice, and is a regular lecturer and panelist at historical and genealogical conferences. To send email, click here.


Other Frank W Sweet sites: Essays on the U.S. Color Line Armed Citizens and the Law
Backintyme Performances YouTube Channel --

Comments are closed.