the decennial census will soon be upon us, and probably less assisted by technology than was originally planned. (as anyone who has used ancestry.com or any of the other genealogy services knows, paper and pen and big ledger books were a pretty good medium for the accurate recordation of data. however, with a much larger population and more questions, it would be nice to have a faster means of data input, cleaning, and analysis available. no matter how cool it is to see your grandfather’s census form from the 1920s.)

but i digress. one of the burning questions that may be settled – or skewed – by next year’s census is the prevalence of same-sex cohabiting couples in the united states. but it’s not just a matter of self-reporting, it’s a matter of how the government treats the information.

from the washington post:

The 2010 Census will not be the first in which same-sex couples have identified themselves as married. But it will be the first in which the raw numbers are publicly reported, reflecting an evolution in the way the Census Bureau keeps track of American lifestyles.

The issue of counting same-sex unions first arose in 1990, when the Census Bureau added the category of “unmarried partner,” primarily to count heterosexual couples living together. Since no state permitted same-sex marriage, the Census Bureau “edited” the sex of one person in each same-sex couple. For instance, if two women said they were spouses, the Census Bureau changed the sex of one to male.

this is troubling on a number of levels. i’m not sure how you get “spouses” from the “unmarried partner” category. if two women claim to be “unmarried partners,” they’re not “spouses,” are they? and if two women claimed to be “spouses” at a time where no state permitted same-sex marriage, it would have made sense to recode them as “unmarried partners,” no?

oh, wait. i bet politics were involved.

of course it was, because human numbers are never just about the numbers. they’re about social policy, and everyone’s political agendas:

Gay advocates say they plan to use “A Census that Reflects America’s Population,” as the Census Bureau calls its plan to report same-sex marriage statistics, to push for legislative and policy initiatives, while groups opposed to same-sex marriage weigh a counteroffensive.

Particularly at the state and local levels, gay advocacy groups say census data on income for same-sex couples will show the need for more protections against job discrimination. Statistics on households with children will help them challenge laws limiting gay adoptions and legal guardianship. With raw numbers to illustrate the need, it will be easier to demand services, they say.

(conventional wisdom would suggest that male same-sex couples probably have a higher standard of living than heterosexual couples – they’re more likely to be DINKS, and then there’s that whole men-earn-more phenomenon. but again i digress.)

“Why does the census ask if people are young or old, black or white, married or single?” said Joe Salmonese, president of the Human Rights Campaign, which promotes civil rights for gay, lesbian, bisexual and transgender people. “It’s because we want to understand if the country is meeting the public-policy needs of those Americans. That’s particularly so for LGBT Americans.”

that’s actually an excellent question. the constitution – the authority for conducting the headcount in the first place – doesn’t mandate any of those questions. article 1 section 2 clearly ties the census to political representation and taxation:

Representatives and direct Taxes shall be apportioned among the several States which may be included within this Union, according to their respective Numbers, which shall be determined by adding to the whole Number of free Persons, including those bound to Service for a Term of Years, and excluding Indians not taxed, three fifths of all other Persons. The actual Enumeration shall be made within three Years after the first Meeting of the Congress of the United States, and within every subsequent Term of ten Years, in such Manner as they shall by Law direct.

to satisfy the constitutional requirements, the only questions needed at the time were free vs. indentured vs. slaves, and indian taxed vs. indian non-taxed. of course, authority for the census now rests in title 13 of the u.s. code, and it’s slightly more specific. but the interesting bit is here:

(g) As used in this section, “census of population” means a census of population, housing, and matters relating to population and housing.

it’s difficult to imagine anything more political than “matters relating to population and housing” in a nation where restrictive property ownership covenants once flourished, where mixed-race marriages were routinely prohibited, and where local housing laws once discriminated against family units based on the specific nature of the family relationships. and there’s definitely the potential to construe the data collection requirements very broadly – what’s a “matter relating to population,” anyway?

judging from the draft short form (the long form has been ditched completely, according to BUCEN, and all housing questions not directly related to “who lives here” have been relegated to the yearly sample-based american community survey), “matters relating to population” include specifying asian and hispanic by country of origin (fascinating, as asian is considered a racial category, but hispanic is an ethnic category), and assuming all blacks have no country of origin other than the united states. it’s difficult to discern the original purpose behind including those categories. is it politically relevant today whether a hispanic person in the u.s. self-identifies as chicano rather than cuban? or if a third-generation asian american traces her family back to japan, rather than viet nam?

bah. too much thinking for a sunday morning. at any rate, with the increasing polarization of the political landscape, i’m expecting an enormous hue and cry surrounding the release of the 2010 census data. hold on for the ride!