Map of the US Census Bureau Regions
This is the sixth post in a multi-part series dealing with the race questions on the census. It examines the lessons learned from examining the changes over all the censuses from 1790 to 2010.
Looking over the "race" questions on the 23 censuses from 1790 through 2010--variously labeled as (a) free white/other free/slave (b) color (c) color or race (d) is this person (e) race--and the widely varied options for answering them, it is easy to see that they have varied greatly over time. Several trends are evident in these variations. These trends include: (1) a shift from "slave versus free" as a classifier to "race"; (2) the separation of race from Latin American descent; (3) oscillation between allowing for self-description ("is this person") and insistence on a race label; (4) oscillation between allowing for multiple descriptors and insisting on a single descriptor; and (5) proliferation of categories over a series of censuses followed by regrouping with a smaller number of options. If asked to fill out all 23 forms, many individuals in our increasingly diverse country would have to change the way they label themselves multiple times.
Those responsible for preparing the 2010 census would probably argue for the desirability of continuing the 2000 practice of using the category "race" along with substantially the same options. They would say that this allows for comparability of data so that population trends can be more accurately monitored. There are several responses to this argument.
To begin with, the census has been notably inconsistent in its terminology over the decades, so that to aim for consistency with the use of the scientifically untenable "race" category is to knowingly propagate error. In the words of Oscar Wilde, "Consistency is the last refuge of the unimaginative." Or, in computer speak, GIGO (garbage in/garbage out).
Since race is a socially constructed category rather than a scientifically observable one, and because social concepts vary as the culture changes, the assumption that the same term maintains the same meaning over time is inaccurate. For example, while individuals continue to list the identical date of birth every ten years, they may change their racial self-designation over the same period of time, even when choosing from an identical list of options. One instance of this is the increase in the number of "American Indians" following the introduction of casinos on reservations.
The statement on the U. S. government website that the race question has been "Asked since 1790" is inaccurate and misleading. The statement is inaccurate because the term "race" didn't appear until 1900, and it is misleading because it implies that the same question has been asked on all 23 censuses. As this series of posts has shown in great detail, the questions, the categories within the questions, and the options within the categories have varied wildly over time, with no element in common over the 220 year time span. By using cultural terms masquerading as biological ones, the use of the term "race" has had the effect of legitimizing an unscientific and (to many) offensive question.
Let us hope that the 2020 census will acknowledge science and abandon the term "race"--perhaps by reintroducing some form of the "Is this person" question.
Image Source: A map of the US Census Bureau Regions
Check out my most recent book, The Myth of Race, which debunks common misconceptions, as well as my other books at http://amazon.com/Jefferson-M.-Fish/e/B001H6NFUI
The Myth of Race is available on Amazon http://amzn.to/10ykaRU and Barnes & Noble http://bit.ly/XPbB6E
Friend/Like me on Facebook:http://www.facebook.com/JeffersonFishAuthor
Follow me on Twitter: www.twitter.com/@jeffersonfish
Visit my website: www.jeffersonfish.com