Letter to the Census Bureau - Society for Linguistic Anthropology

On May 27, AAA President Virginia Dominguez sent a letter to the Census Bureau regarding its language questions and classifications, which you can find through AAA Challenges Questions on US Census and the AAA homepage.

The letter was written by the newly constituted SLA Committee on Language and Social Justice, which partners with the AAA Committee on Human Rights (CfHR). It suggests several alternatives for the classification of speakers who live in households where no one over the age of 14 speaks English “very well” as “linguistically isolated.” This is a follow up to the AAA’s 2007 AAA resolution (seeStatement on Language Questions in the US Census), and previous correspondence with the Census Bureau. Those letters, which can be viewed through Committee for Human Rights, reveal the unwillingness of the Census Bureau to change the language questions, but a willingness to re-consider their use of “Linguistically Isolated.” The Committee is hopeful that their efforts over the last four years will finally result in critical changes in the way the Census 2010 data are reported, and we will continue to press for changes in the language questions themselves.

Galey Modan July 5, 2010 at 10:29 am

The letter that Virginia Dominguez sent to the Census Bureau is really disturbing, as it reveals a fundamental lack of understanding of both the nature of the language data that the Census Bureau collects, and the system by which Census questions are developed. Although Dominguez signed the letter, it’s perhaps even more disturbing to know that that our own committee on Language and Social Justice wrote it. It’s sloppy and gives us a bad name.

The letter asserts that “The Census data prove that most of those who speak a language other than English at home are Proficient Bilinguals”. However, the Census data do no such thing. The Census only asks about the *English* proficiency level of people who speak languages other than English within the home. In order to be a proficient bilingual, one must speak *two* languages proficiently. But the Census gathers no information about how proficient such speakers are in the other language(s) that they speak. Given the amply documented language loss among heritage language speakers resulting in a shift to monolingual English speakers within 3-4 generations, it is highly likely that many people who speak a language other than English in the domestic sphere and who speak English “well” or “very well” are in fact *not* proficient bilinguals at all, but rather English-dominant speakers with limited competence in their heritage languages. Therefore, the category options proposed in Dominguez’s letter cannot possibly accurately reflect the data collected.

As for the Census Bureau’s alleged unwillingness to change the language questions: As Jennifer Leeman clearly explains in her article in the May issue of Anthropology News (which the writers of this letter unfortunately seem not to have read), the Census Bureau itself does not come up with census questions. Rather, Congress and Federal Agencies charge the Census Bureau with adding new questions. Therefore, if the AAA deems new language questions necessary (which I believe they are), it is incumbent upon AAA to work with other Federal agencies or through congressional committees to get *them* to charge the Census Bureau with adding new questions. (Leeman’s article details the problems with the current questions and suggests new questions that need to be asked, so I will not rehash that information here. But it makes me wonder: what is the point of publishing such articles if a committee working on the same issues doesn’t bother to read them?)

Given that the key undertaking of anthropologists is to observe and understand how social systems work, it’s an embarrassment that our professional organization would write a letter that makes clear that we have no idea how the Census Bureau works, or what kind of data it collects. If as a professional organization we hope to have an impact on the social institutions we study, we need to show that we understand those institutions, otherwise no one will take us seriously.

1 thought on “Letter to the Census Bureau”

Galey Modan July 5, 2010 at 10:29 am

The letter that Virginia Dominguez sent to the Census Bureau is really disturbing, as it reveals a fundamental lack of understanding of both the nature of the language data that the Census Bureau collects, and the system by which Census questions are developed. Although Dominguez signed the letter, it’s perhaps even more disturbing to know that that our own committee on Language and Social Justice wrote it. It’s sloppy and gives us a bad name.

The letter asserts that “The Census data prove that most of those who speak a language other than English at home are Proficient Bilinguals”. However, the Census data do no such thing. The Census only asks about the *English* proficiency level of people who speak languages other than English within the home. In order to be a proficient bilingual, one must speak *two* languages proficiently. But the Census gathers no information about how proficient such speakers are in the other language(s) that they speak. Given the amply documented language loss among heritage language speakers resulting in a shift to monolingual English speakers within 3-4 generations, it is highly likely that many people who speak a language other than English in the domestic sphere and who speak English “well” or “very well” are in fact *not* proficient bilinguals at all, but rather English-dominant speakers with limited competence in their heritage languages. Therefore, the category options proposed in Dominguez’s letter cannot possibly accurately reflect the data collected.

As for the Census Bureau’s alleged unwillingness to change the language questions: As Jennifer Leeman clearly explains in her article in the May issue of Anthropology News (which the writers of this letter unfortunately seem not to have read), the Census Bureau itself does not come up with census questions. Rather, Congress and Federal Agencies charge the Census Bureau with adding new questions. Therefore, if the AAA deems new language questions necessary (which I believe they are), it is incumbent upon AAA to work with other Federal agencies or through congressional committees to get *them* to charge the Census Bureau with adding new questions. (Leeman’s article details the problems with the current questions and suggests new questions that need to be asked, so I will not rehash that information here. But it makes me wonder: what is the point of publishing such articles if a committee working on the same issues doesn’t bother to read them?)

Given that the key undertaking of anthropologists is to observe and understand how social systems work, it’s an embarrassment that our professional organization would write a letter that makes clear that we have no idea how the Census Bureau works, or what kind of data it collects. If as a professional organization we hope to have an impact on the social institutions we study, we need to show that we understand those institutions, otherwise no one will take us seriously.

Comments are closed.