Reassessing Race and Ethnicity through a Migration Lens

Published 16 December 2020 / By Michael Jones-Correa, University of Pennsylvania

Back to Articles

This blog is based on a presentation given during the COMPAS Seminar Series, Race, Ethnicity and Migration in Autumn 2020.

It is now taken as a given in the social sciences that race and ethnicity are constructed social categories, with situational elements that vary by individual's contexts. However, because of the role that state institutions play in assigning ethnicity, and reifying these categories from everything from the date they collect to their assignment of public goods, in practice social scientists still treat ethnic and racial categories as fixed. So, despite an underlying acknowledgment that race and ethnicity are of course socially constructed and situational (reflected in linguists' interest in code-switching and the like), we actually know quite little about the malleability of these identities in the course of people's lives—about the practice of situational ethnicity. But if there is any group of individuals for whom that should not be true, it should be migrants.

Migrants, as they leave their communities of origin, are constantly adapting to new social contexts as they move within and across national borders, and consequently play up (or downplay) aspects of their ethnic identities in favor of others, at times in order to minimize threat or friction, or to generally fit in. For instance, a migrant who leaves Niger, from a Djerma village traveling north en route to Europe may likely find themselves emphasizing their national identity as Nigerien, as Muslim, or as African as they travel to Libya or Algeria, and then, say, to Italy. In the course of their journey, some aspects of their ethnic identities will become salient; others will be submerged. Similarly, a migrant who leaves from a K'iche community in the western part of Guatemala making their way to the United States might have to negotiate their identity as indigenous K'iche in Guatemala, as Guatemalan or Central American while in Mexico, as Central American or Latino on arrival to the United States. Each of these identities, at different times and places, can trigger racial/ethnic discrimination, and therefore which identity becomes a migrant's "public" identity requires finessing, and at times, dissembling. Each of migrant's identities is therefore both "true" and "false" depending on the context, and the state's role in shaping that context.

This brings us to the role of states, and particularly the role of receiving states in the categorization of race and ethnicity, and the effect these have on migrant identification. The work, for instance of Nobles (2000) and Mora (2014), illuminates the role the US government agencies, such as the Census, have played, through ethnic categorization, in creating ethnic groupings. In the 1980s, for instance, as the Census was conducting trials of possible question wordings for the next decennial census. One of the possible categories the Bureau was pre-testing was "Central American" as a possible ethnic identifier. The Bureau had selected different sites around the US as test sites, including one in the state of Kansas. The results they received back from Kansas puzzled analysts—a higher than expected percentage of respondents were responding positively to this new identifier. Looking more closely at the data, it became apparent that a good number of US born whites were choosing this identity, which was even more of a puzzle. In probing further, it became clear, however, that these respondents were choosing that identifier because, of course, if Kansas is in the middle of the United States, of course they were "central Americans." The Census quickly dropped that category from consideration for use in its decennial census, but it illustrates, if a bit tongue in cheek, how easily the state's categorization of ethnicity can shape ethnic identification.

The point here is that if one examines the kind of information that governments collect, and in particular the racial and ethnic categories they use, and by adopting, impose more widely, then it is easy to see, first, how government imposed categories matter, and second, how the with the use of these categories mask more nuanced understandings of race, ethnicity and religion. The US Census, for instance, collects information on the country of origin of migrants from India, but nothing on their state of origin, their caste, or their religion. Once in the US, our hypothetical migrant from Kerala is simply "Indian," subsumed within a broader category of "Asian American." These state-assigned categories are picked up and replicated by highly respected policy and academic researchers. To give one example, the Migration Policy Institute, perhaps the highest profile immigration policy think tank in Washington DC, regularly puts out briefs on immigrant populations in the US. But because of its reliance on US Census data, its brief on Indian immigrants in the US makes no mention of characteristics by region of origin, religion, language or caste—information which is arguably hugely important in understanding these immigrants' outcomes (Hanna and Batalova 2020). Academic research like the National Asian American Survey (NAAS), which tries to survey a representative sample of all "Asian Americans"—which describes individuals with origins in Asia, itself a contested Census category—similarly pays no attention at all to key subnational identities. The work that I and collaborators have done on Indian migrants in the Atlanta and Philadelphia metropolitan areas also falls short: we collected information on respondents' religion and region of origin, but only imperfectly recorded migrants' language(s), and asked nothing at all about caste. There is careful research being conducted on the South Asian immigrant community in the US—see for example, work by Mishra (2018) and Kapur (2016)—but it is still little and far between.

Similarly, there is very good work on indigenous migrants to the US from Mexico and Central America (see for instance, research published by Fox, Rivera-Salgado, Menjivar, Abrego and others). But the fact is that in the social sciences the distinctiveness of indigenous identity and language gets lost as these indigenous migrants become subsumed under the "Hispanic" or "Latino" pan-ethnic categories in use in the United States. Other ethnic categories, notably Muslim or Arab, (which are quite salient in Europe) are played down in the United States, in part because of the oddities of how the US government categorizes individuals as "White," and the absence of any public data on religion. Immigrants from the Middle East are, for the purposes of race in the US, considered "White" rather than being considered as either Asian or African, and no public data is collected on religious identification and practices, traceable to the Constitution's emphasis on the separation of church and state. Existing estimates of the Muslim population in the US are simply that—estimates—and what we know of the demographics and public opinions of Muslims comes from private surveys, like those from the Public Religion Research Institute, or the Pew Research Center.

In the US context, then, caste disappears, subsumed under national origin; indigeneity disappears, subsumed under pan-ethnic labels; religion disappears, subsumed under racial or national origin categories. Perhaps more importantly, these state-sanctioned categories are eventually adopted by the people they cover. Research I and others have conducted on Latino/a populations in the US suggest, for example, that on first arrival Latin American migrants identify with their national origin over identification with any larger pan-ethnic grouping such as "Hispanic," or with national identities like "American," but these preferences shift over time in the US and over generations. (Fraga et al. 2012). In the end, people often adopt state-sponsored ethnic and racial categories.

Are state-sponsored ethnic categories detrimental? This is by no means clear. It could be, for example, that not having any data on caste is actually beneficial to low-caste Indian migrants in the United States. To return to our hypothetical lower-caste migrant from Kerala state in India: they may find, after migrating to the US, that their caste, language and religion simply don't carry the same valence in their new context, with their identities being subsumed under the label "Indian" or "Asian," allowing for kinds of social mobility that might not have been possible for them in India itself. It could be the case as well that the relatively lower salience of Islam in the United States contributes to Muslim's socio-economic mobility. However, a lack of attention may also mean the disappearance of language and culture for indigenous migrants, or the conflation of groups that might otherwise remain distinct, like Sikhs in the United States, who are lumped together with Indians or South Asians.

The argument in this essay is that while identities are understood to be situational and contextual, in practice social scientists and policy makers do not treat them as such, eliding key aspects of how identities function in social contexts. Migrants, however, provide particularly salient examples of individuals' adaptation of identities as they move from context to another, and their conscious (or unconscious) adaptation as they move should be informative to social scientists interested in the malleability of ethnic identities (and its limits), and the ways identities respond to social and institutional categorizations, even as these highlight some aspects of identity and subsume others.

Work Cited

Fraga, Luis, John Garcia, Rodney Hero, Michael Jones-Correa, Valerie Martinez and Gary Segura. 2012. Latinos in the New Millennium: An Almanac of Opinion, Behavior and Policy Preferences (New York: Cambridge University Press).

Jones-Correa, Michael, Helen Marrow, Dina Okamoto and Linda Tropp. Immigrant Native Relations in 21st Century America.

Kapur, Devesh. 2016. The Other One Percent: Indians in America (New York: Oxford University Press).

Hanna, Mary and Jeanne Batalova. 2020. Indian Immigrants in the United States. Migration Policy Institute.

Mishra, Sangay. 2018. Desis Divided: The Political Lives of South Asian Americans (Minneapolis: University of Minnesota Press).

Mora, Cristina. 2014. Making Hispanics: How Activists, Bureaucrats and the Media Constructed a New American (Chicago: University of Chicago Press).

National Asian American Survey.

Nobles, Melissa. 2000. Shades of Citizenship: Race and the Census in Modern Politics (Palo Alto: Stanford University Press).


Download as PDF

If you do not have Adobe Acrobat Reader, which is required to read this document, you can download it for free from the Adobe website.