That politics undergirds censuses is a truism. At least since Benedict Anderson wrote Imagined Communities: Reflections on the Origin and Spread of Nationalism in 1983[i] scholars have accepted that censuses are both political and scientific enterprises. Census racial classifications are a case in point because they have historically become instituted through political efforts. For example, “Mulatto” became a census classification in 1850 after politicians, alarmed by racial miscegenation, demanded that the Census Bureau enumerate those of black/white parentage.[ii] More recent ethnoracial categories have arisen as a result of the political efforts championed by community stakeholders. To wit, the Hispanic/Latino classification emerged as Mexican, Puerto Rican, and other community leaders pressured the Census Bureau for official recognition during the 1970s.[iii] And if a Middle Eastern/North African category is added to the next census in 2020, as is predicted, it will be because activists, academics, and others have lobbied over two decades for its inclusion. In effect, rather than reflecting an existing reality, all census racial categories emerge, or are negotiated, in such a political fashion—none exists in nature.
Despite the political origins of our official racial and ethnic categories, lay and academic prognostications about the country’s demo- graphic future rarely take politics seriously.
Take, for example, sociologist Richard Alba’s provocative commentary published in The American Prospect, “The Likely Persistence of a White Majority.”[iv] Alba argues that recent reports about the looming “browning of America” and the development of a “white minority” are over- blown. His claim rests on two points, the first of which involves how the Census Bureau presents racial statistics to the public. Specifically, because it classifies Latinos and mixed-race (i.e., white/ racial minority) Americans as non-white by default on census reports, the Bureau, according to Alba, “produces the smallest possible estimate” of the size of the white population. Alba contends that at least some Latinos and some persons of mixed racial heritage would be more accurately classified as white. Doing so would increase the size of the white population and thus generate a less “biased” portrait of America’s racial future.
Alba’s second point concerns the Bureau’s racial forecasts, specifically its recent contention that America is moving toward a minority majority future. He argues that the Bureau’s (and other analysts’) prognosis ignores the extent to which Latinos, immigrants, and mixed-race persons are assimilating into the white-dominant mainstream. He points to increases in mixed-race marriage (white/non-white) and higher levels of individual social mobility among minorities to make his case. The “likely result” of assimilation, Alba contends, “will be to alter the circumstances under which individuals are seen as belonging to marginalized minorities.” As a result, the “longstanding processes of assimilation could produce a white-dominated mainstream at the national level.”
Neither of Alba’s points, however, deals squarely with politics. His claim about census reporting conventions ignores the implications that racial classifications have for racial justice. After all, various activists still fight for these categories to include whom they do for specific political reasons relating to the needs of their communities. Moreover, his argument about assimilation and the enlargement of the white mainstream overlooks the sociopolitical complexity of race, especially the role of privilege and hierarchy. This failure to grapple with politics hinders broader conversations about census race data because it narrows debates to the issue of “methodological accuracy,” a concept that itself is not devoid of politics[v]. As a result, conclusions about who racial minorities are and what census race data represent become decontextualized and incomplete.
In response to Alba, we bring politics to the fore, specifically as it relates to knowledge construction and the issue of racial incorporation. Our response focuses primarily on Latinos, as this population is a key category in Alba’s analysis and is the subject of our sociological expertise. We begin by considering the politics of census reporting conventions, highlighting the much-neglected issue of racial justice. We then move on to a discussion of census data and assimilation. We conclude by addressing racial forecasts, demonstrating how politics, rather than simply demographic inevitability or the dynamics of social mobility, will also ultimately determine the country’s future and the status of its ethnic and racial populations.
Census Statistics and Racial Justice Projects
Alba critiques the Census Bureau in the interest of producing more “accurate data.” He writes,
“Not only do its rigid and illogical classifications distort important new realities, the bureau is also not forthcoming about the errors and uncertainties involved.” This language of accuracy and validity—language the Census Bureau itself embraces—treats census classifications and reporting practices simply as technical and methodological matters. Accordingly, it narrowly assesses census data in relation to statistical procedures and presentation. But there are other ways to evaluate census statistics, such as looking at them “in terms of their practical utility for social projects.”[vi]
What practical utility do census race statistics have today? The answers vary, as different stake-holders use these data for different ends. Government agencies use race statistics to assess the social landscape and inform policymaking. Corporations rely on race data to market their products to different demographic communities. Social scientists draw on the statistics in their analyses. Census categories and statistics even foster narratives of community and belonging, not to mention a sense of moral worth, for those seeking an identity or recognition.
In addition, activists and social movements, particularly in the wake of the civil rights era, have turned to racial statistics to expose social inequality. In a political context that values numerical forms of knowledge, statistical evidence of underrepresentation and inequality has been indispensable to racial justice campaigns. For example, although Mexican-American and Puerto Rican activists in the 1960s argued continuously that their communities suffered from poverty and low levels of education, it was not until they possessed official census figures showing disparities that their arguments gained traction in many government sectors.[vii] Census race data have also been—and continue to be— vital for monitoring voting districts and curtailing the gerrymandering practices that often disenfranchise people of color. If there is any question about the importance of racial statistics for these efforts, we implore readers to follow what has happened in Texas and North Carolina—states where the revamping of districts effectively diluted the political clout of racial minorities.
Racial statistics, as the preceding examples illustrate, are employed to make visible the systems of domination and punishment that affect communities of color at material and cultural levels.[viii] Indeed, black and Latino classifications help to provide evidence of how these communities suffer from limited access to quality education, grapple with mass incarceration and racial profiling, and remain underrepresented in major centers of power, from the academy to elected office. In many ways, our census race data show that the cultural scripts and institutional barriers that treat minorities as morally and professionally inferior to whites are still very much operative.
Given this racialized system, census statistics and issues of classification cannot from our perspective be assessed narrowly in terms of statistical procedure. Questions of accuracy and validity are important, but exclusively focusing on these issues ignores larger matters. We argue that proposals for making changes to census categories or reporting practices must address the political utility of race statistics for racial justice. In other words, questions and analysis of racial classification must also grapple with the issue of how changes in reporting, classifying, and collecting race data will affect racial justice projects.
Evaluating census statistics with these overt political criteria reveals the limitations of Alba’s suggestion that some Latinos and mixed-race individuals would be more accurately classified as white. We believe that adopting this practice could make it more difficult to uncover certain forms of ethnoracial inequality, given that the distinct patterns of these communities may be lost when subsumed within the broader white data. Indeed, this lack of identifying information was precisely the problem prior to the adoption of the “Hispanic” category. Only when Hispanic data were disaggregated from the white category did it become possible to statistically track the discrimination and disparities facing the Latino population. In other words, classifying Latinos as white imperiled rather than aided their fight for racial justice. Undergirding Alba’s suggestion, we believe, is an unserviceable conception of whiteness. This conception ignores the institutionalized complexity of race, including patterns of hierarchy, privilege, and domination. Thus, we now turn to the issue of precisely how to conceptualize race and whiteness within census analyses.
Assimilation and the Meaning of Whiteness
In government and marketing reports, whiteness is just another variable, a racial category among many: a self-selected identity with certain correlated attributes (e.g., education level, income). Whiteness, in other words, is a statistical artifact that distinguishes a certain population from others.
However, as conceived in racial justice projects, and for race scholars more generally, whiteness is about privilege and hierarchy.[ix] Being white in America is not simply about identities or individual attributes that can be gained or lost, but rather about a privilege that has been reified within most institutions in America—from schools to workplaces to jury rooms to police forces. Moreover, it is about a hierarchy that attaches narratives about moral worth and legitimacy to images of whiteness. As such, those deemed “white” can lose income, status, or other individual attributes but not necessarily become divested of the privilege that whiteness affords, benefits that the pre-eminent sociologist W. E. B. Du Bois described as the “wages of whiteness.”
These two competing conceptions of whiteness have different implications for how we theorize the present and potential future position of Latinos and mixed-raced individuals. Viewing whiteness as linked to power and privilege complicates claims about assimilation into the white mainstream. We do not deny that people of color can become more upwardly mobile and gain access to spaces that have historically excluded them. Alba emphasizes this in his essay, noting that many Asians and other minorities now constitute a larger part of the upper class. Nor do we wish to trivialize growing rates of inter-marriage among some Latinos and mixed-raced persons. However, we are critical of those analyses that are quick to link mobility and marriage patterns to blanket statements inferring white boundary blurring.
Our criticism of these assimilation conclusions is based on the fact that America’s racial scripts and hierarchies are still heavily institutionalized in ways that validate whiteness.[x] The recent incorporation of white nationalist (sometimes euphemistically called “alt-right”) groups within the present White House administration is but one clear example. Moreover, these racial scripts persist despite the fact that people of color have experienced improvements in social mobility. As such, even those folks of color who attain high professional status are often devalued because they are not white. Professors and other professionals of color, for example, are often presumed less competent than their white counterparts[xi] President Obama, the first black president in U.S. history, and the first family were subjected to overtly racist tropes splashed across the covers of high-profile publications. And while we agree that a certain strata of minorities may become (or has already become) what sociologist Eduardo Bonilla-Silva calls “honorary whites,”[xii] we insist that the broader racial scripts and systems of domination remain by and large intact.
With respect to Latinos in particular, the suggestion that on the whole they are becoming white underestimates the ongoing racial stigmatization and exclusion faced by many in this community. Such arguments do not account for present-day anti-Latino and anti-immigrant policies and rhetoric. Should we assume that mass deportations, which, too, often lead to the separation of Latino families, or inflammatory electoral campaigns will have no effect on how Latinos identify and understand their place in U.S. society for years to come? And what of the widespread presumptions of illegality that Latino citizens must contend with?
Moreover, scholars who see the growing number of Latinos selecting “white” on the census as straightforward evidence of Latinos becoming white rarely contextualize such data within political history or current social dynamics. To be sure, some Latinos understand themselves as racially white. This identification may be interpreted as aspirational—a kind of racial passing—but it does not necessarily provide blanket evidence for the inevitable social inclusion of Latinos in the white-dominant mainstream. In fact, Professor Julie Dowling of the University of Illinois shows that Mexican Americans identify as “white” on the census not because they are accepted as white or even because they see themselves as white. Rather, by reframing the borders of whiteness to include them, Mexican Americans resist racial “othering,” in an effort to be accepted as fully American.[xiii]
Political sociologist Mara Loveman uncovers a similar dynamic in her research on early twentieth-century census enumeration in the U.S. colony of Puerto Rico.[xiv] These examples suggest that white self-identification—the very evidence some scholars have used to make claims about the whitening or assimilation of Latinos—may in fact register conditions of ethnoracial exclusion. In other words, choosing whiteness is also related to the perils of living as a minority for many.
Our discussions about assimilation and racial identification, then, must entail more than inter-marriage and social mobility patterns, however important these might be. They should also recognize that in the U.S. ethnoracial order, entrenched privileges and opportunities are afforded to some and not to others. Only once this is understood can scholars begin to interpret the relationship between social mobility and racial identification, for example. These realities of power and inclusion must also be accounted for when forecasting trends, as Alba does. Such analyses, however, also need to attend to the ways that forecasts themselves shape the political context in which assimilation or integration unfolds.
Racial Forecasts in National Politics
It is one thing to classify and report on current trends, another to make projections and forecasts about the nation’s future. We hesitate to join the chorus of commentators who proclaim one demographic future or another, as Alba and others do. Our reluctance stems from the fact that demographic projection typically ignores political context. Politics are, quite literally, not factored into the predictive statistical models. But this is not our only source of hesitation. Demographic prognosticators also tend to ignore or be less concerned with the political impact of their projections.
Indeed, forecasts have real consequences for national politics, both on the left and right. On the liberal left, the “browning of America” prognosis has quickly fueled a race to secure the Latino and Asian vote, because these communities are believed to be the impetus behind the nation’s demographic changes. Therefore, the Democratic Party has steadily built a ground game in states such as Nevada, Virginia, Arizona, and Florida with a focus on securing Latino and Asian support. These moves seem promising, but far too often, such efforts seek to simply capitalize on these populations rather than invest in or support them. Rarely, if ever, do electoral campaigns genuinely take on issues of racial domination and privilege, instead holding communities of color captive to the slow, almost glacial, progress of electoral politics.
On the right, arguments about the browning of America can fuel paranoia, backlash, and hate campaigns.[xv] The forecasts become twisted into narratives that minorities, especially immi- grants, are “taking over” and thus diminishing the values and morals upon which the nation was founded.[xvi] Alba himself makes the point that “the anxieties about the end of white majority status have fueled a conservative backlash against the growing diversity of the country.” This script has historically helped to shore up Republican Party support: President Donald Trump’s campaign rhetoric that Mexican immigration brings in “criminals” and “bad hombres” is a case in point. Such nativist resentment is fueled in part by a sense that the country is no longer majority Anglo.
Given this context surrounding demographic projections, our analyses must begin to factor in the real political consequences for how we understand race and for how Americans understand their position within the nation. Racial forecasts are not neutral; they are politically charged, particularly as they travel beyond scientific contexts. When we forget this, focusing instead on which forecast is more unbiased or more accurate, we reinforce the tendency to take projections as an inevitable truth rather than a contingent prediction. Moreover, in concerning ourselves with matters of accuracy, we may inadvertently contribute to the nativist backlash and fuel the very political trends that reinforce racial inequality. Of course, we do not contend that racial forecasts are the only, or most important, influence on national racial politics. However, we do believe that ignoring the political context and consequences of such forecasts leaves us blind to one of the factors that may be shaping the political conditions of incorporation for non-white populations.
For these reasons, demographic projections must be handled with care. No future scenario, including the white minority thesis, should be viewed as inevitable. Taking a stance against one forecast by offering another, as Alba does, simply perpetuates the faulty logic of assuming that there is a “true,” scientifically valid mode of interpretation. In addition, such an approach ignores the very ways that census forecasts play a role in reifying America’s pernicious racial scripts.
Let us reiterate the centrality of politics in demographic analyses. Racial classifications and race census data are political constructs that nonetheless help to code past and present forms of inequality and discrimination. They can provide vital measures of the exclusionary power structure, data that can bolster racial justice projects. However, when analyses and discussions about race data fail to acknowledge the central role of politics, arguments about accuracy, bias, and measurement will fall dangerously short. It is only when scholars begin to take politics seriously that our analyses of census data will more comprehensively reflect how race is lived in America.
[i] Benedict Anderson, Imagined Communities: Reflections on the Origin and Spread of Nationalism (London: Verso Press, 1983).
[ii] Melissa Nobles, Shades of Citizenship: Race and the Census in Modern Politics (Palo Alto: Stanford University Press, 2004).
[iii] G. Cristina Mora, Making Hispanics: How Activists, Bureaucrats, and Media Constructed a New American (Chicago: University of Chicago Press, 2014).
[iv] Richard Alba, “The Likely Persistence of White Majority,” American Prospect, January 19, 2016.
[v] Theodore Porter, Trust in Numbers: The Pursuit of Objectivity in Science and Public Life (Princeton: Princeton University Press, 1995).
[vi] Bruce Curtis, The Politics of Population: State Formation, Statistics and the Census of Canada, 1840-1975 (Toronto: University of Toronto Press, 2001), 34.
[vii] Mora, Making Hispanics.
[viii] Michael Omi and Howard Winant, Racial Formation in the United States (New York: Routledge, 2015).
[ix] Ian Haney Lopez, White by Law: The Legal Construction of Race (New York: New York University Press, 2006).
[x] Natalia Molina, How Race Is Made in America: Immigration, Citizenship and the Historical Power of Racial Scripts (Berkeley: University of California Press, 2014).
[xi] Gabriela Gutierrez y Muhs, Yolanda Flores Nieman, Carmen Gonzalez, and Angela Harris, Presumed Incompetent: the Intersection of Race and Class for Women in Academia (Salt Lake City: Utah State University Press, 2012).
[xii] Eduardo Bonilla Silva, “From Bi-racial to Tri-racial: Towards a New System of Racial Stratification in the USA,” Ethnic and Racial Studies 27, no. 6 (2004): 931-50.
[xiii] Julie Dowling, Mexican Americans and the Question of Race (Austin: University of Texas Press, 2013), 7.
[xiv] Mara Loveman, “The U.S. Census and the Contested Rules of Racial Classification in Early Twentieth-Century Puerto Rico.” Caribbean Studies 35, no. 2 (2007): 79-114.
[xv] Melissa Abrajano and Zoltan Hajnal, White Backlash: Immigration, Race, and American Politics (Princeton: Princeton University Press, 2015); Michael Rodríguez-Muñiz, “Temporal Politics of the Future: National Latino Civil Rights Advocacy, Demographic Statistics, and the ‘Browning’ of America,” in Department of Sociology (Providence: Brown University, 2015).
[xvi] Samuel Huntington, Who Are We: The Challenges to America’s National Identity (New York: Simon & Schuster, 2005).