The Rodriguez case was decided in 1897, a year after the Supreme Court’s infamous Plessy v. Ferguson opinion.157 While Mexican Americans had experienced de facto segregation in the West (and would continue to do so until the 1960s in some places), blacks experienced a heightened level of de jure segregation after Plessy. The case helped entrench the hypo-descent rule for blacks, but, before Plessy, it was not at all inevitable that the one-drop rule would become dominant in the United States. In fact, for centuries before then, there had been many definitions of black status vying for supremacy. The American revolutionary generation defined someone as black only if one or more of the individual’s grandparents was black.158 Between 1850 and 1920, the census identified not only blacks in the United States but also mulattos, suggesting that the government viewed black/nonblack mixing as substantial enough to be counted.159 The 1890 census, the one that preceded Plessy, actually counted quadroons and octoroons, in addition to blacks and mulattos.160 Not until the 1920s and 1930s did the hypo-descent rule hastened by Plessy become consolidated as the American rule for defining black status.161
In part, it was the instability and variety of definitions of blackness that led to Mr. Homer Plessy’s case. Plessy was seven-eighths white and one-eighth black and was phenotypically white.162 Yet, under various Louisiana definitions of race in its Jim Crow laws, Plessy’s legal identity varied, despite the fact that he self-identified as black and participated fully in black community life (in other words, Plessy did not seek to “pass” for white). As legal scholar Cheryl Harris has observed, it was precisely Plessy’s appearance as white that made him an ideal candidate for the test case that would challenge Louisiana’s mandate that railway cars provide separate cars for whites and “colored persons.”163 It is telling that for Plessy’s case to receive the court’s review, a deal had to be struck between his civil rights attorneys and the Louisiana and Nashville Railroad to arrest him for violating state law. Without the deal, train personnel, like others in his day-to-day interactions, may well have assumed he was white and accordingly not challenged his selection of a seat in the white car.
Plessy’s lawyers made a number of arguments designed to expose the fallacies of Jim Crow laws. They argued that the laws were unenforceable—in the first instance, by those people who decided who was in the wrong railway car, for example, and also by appellate courts deciding whether the law had been correctly applied—because racial mixture in Louisiana’s population sometimes made it impossible to visibly ascertain an individual’s race and to definitively prove a person’s racial status in court.164 Since segregation laws could not be enforced consistently (and often were enforced in contradictory ways), Plessy’s attorneys argued, the Court should look for the underlying purpose of the law and, on that basis, find it unconstitutional. Plessy’s brief made the point this way: “Why not count everyone as white in whom is visible any trace of white blood? There is but one reason to wit, the domination of the white race.”165
The Supreme Court majority largely ignored these issues (though Justice Harlan, dissenting, did not). They did so because the mere fact of debating them would have loosened the boundaries of racial categories, thereby disrupting the myth of race as fixed and stable. The majority’s one concession to Plessy’s powerful arguments was, in the penultimate paragraph of its opinion, to acknowledge the variety of existing definitions that determined who was black:
It is true that the question of the proportion of colored blood necessary to constitute a colored person, as distinguished from a white person, is one upon which there is a difference of opinion in the different states; some holding that any visible admixture of black blood stamps the person as belonging to the colored race; others, that it depends upon the preponderance of blood; and still others, that the predominance of white blood must only be in the proportion of three-fourths. But these are questions to be determined under the laws of each state, and are not properly put in issue in this case.166
In the end, the Court’s ruling fueled the most extreme definition: the hypo-descent rule under which any visible proportion of African ancestry, however slight, marks a person as black.
Although cases like Plessy contributed to American ideas that race was immutable and fixed, the reality was that race was more fluid. A powerful example comes from the Senate’s 1902 debates on the Omnibus Statehood Bill (see Chapter 2), which sought to admit Oklahoma, New Mexico, and Arizona as new states. The racial composition of each of the proposed states was a major topic of floor debate. Senator Matthew S. Quay of Pennsylvania used official census numbers to estimate the non-Indian population of New Mexico at just under two hundred thousand.167 But Indiana senator Albert Beveridge, who as we know from earlier discussions strenuously opposed New Mexico statehood, said emphatically, “There is no ground for [the] assumption [that Mexicans are white].”168 Beveridge was supported by senators Knute Nelson of Mississippi and Eugene Hale of Maine, among others, who refused to countenance the census’s inclusion of Mexicans in the white category. Mexicans had been “white enough” to be collectively naturalized in 1848 and to be counted as white in 1897 under the naturalization laws, but, here, U.S. senators decided they were not white enough to warrant the full citizenship statehood would have provided.
These conversations in 1902 suggest that fluidity has been an enduring quality of American racial dynamics (especially in the construction of racial categories), rather than a feature unique to Mexican Americans. A similar conversation to the one involving New Mexico arose soon thereafter, when Congress took up the proposed state of Oklahoma, to be formed largely from the Indian Territory. The very senators who, by refusing to include Mexican Americans in the “white” category, argued that New Mexico’s white population was too small to warrant the territory’s admission as a state now argued that Oklahoma should be admitted because the number of “pure Indians” there was very small. Senator Nelson argued that, of Oklahoma’s 87,000 Indians enrolled in tribes, “a very small part are pure Indians and a very large portion of them are practically white men.” Senator Hale supported admission of Oklahoma because it possessed “just as pure and clean a white population as exists in Indiana or Pennsylvania or Maine.” Senator Beveridge made a parallel argument about the Indian Territory, saying, “Only a very small number, to wit 87,000 [of 400,000 persons] at most, are classed as Indians, and of those who are classed as Indians a very large number are pure white men, so far as blood is concerned, quite as white as we are.”169
By this point, the direction of the conversation had begun to seriously bother Senator Benjamin Tillman of South Carolina. He asked, indignantly,
How [have] white men as pure blooded as he or I . . . gotten the right to all that Indian land, and how [is it] that the 87,000 Indians, so called, are nearly all white people? Some stealing has been done somewhere, or something, and I should like to know how it happened. I know a bleaching process is going on over there; but still that does not make them as pure white men as he and I, because I do not think either of us could be suspected of having anything else but Caucasian blood in our veins. (emphasis added)
Tillman’s annoyance reveals the American racial order as, at once, fluid and rigid. Like the Mexican and Latin American process of “whitening” described earlier, he acknowledged “a bleaching process” in the United States. In the same breath, he embraced the primacy of blood purity, which invokes a more inflexible racial hierarchy and categorization. Senator Beveridge’s response to Tillman was to say that white men had become Indians by marriage and by adoption into tribes. But a more accurate and honest approach would have acknowledged that, despite the presence of Indians in the region, it should be admitted as a state because the Indians there would be dominated by whites.170 The Indians in the proposed state of Oklahoma did not then pose a threat to political domination by Euro-American men; in stark contrast, New Mexico’s Mexican men did, as the majority of those to be enfranchised in the proposed state.
This conversation contradict
s the claim that American racial categories were perceived by social actors in the past as rigid and immutable. American racial dynamics were and are often seen as the mirror opposite of Latin American race relations on this point. While Spanish-Mexican categories are viewed as flexible, American categories are viewed as historically fixed. In reality, the Spanish-Mexican system regularly produced racial dynamics that were harder than they appeared to be, and the Anglo-American system regularly produced dynamics that were more malleable than they appeared to be.171 Of course, the United States actually includes both models of race relations. Yet it is only by virtue of putting anti-Mexican racism in dialogue with antiblack racism that we can observe the full spectrum of American racial categories and formations.
Epilogue
One morning over breakfast in 2006 my ten-year-old son asked, “Mom, are we white?” From the habit formed of reading too many parenting books, I took a deep breath and then probed for more information (thinking to myself, is he asking what I think he is asking?). After acknowledging that his was a big question, I asked him to tell me more about what he was interested in finding out. “Well, back when there was slavery, were Mexicans white or black?” I began my answer by saying the question was complicated and one that people disagreed about. I told him that Mexican Americans had ancestors who were Indian, African, and Spanish, but they were primarily indigenous. I said that I did not believe we were white. He responded with another question—“Then why don’t we say whites, blacks, and browns?”—and then was quickly off to get ready for school. Thinking more about my son’s question, at least part of his dilemma comes from the continuing tendency in popular culture (including elementary school social studies) to see American history, from slavery to the civil rights movement, in black and white terms. My son was struggling with how to reconcile that standard story with the complex racial reality he knew.
Manifest Destinies has focused on how race evolved in the larger historical context, as well as how race and racial ideology organized social life and, particularly, how they shaped relations among racial groups. In this epilogue, I consider the contemporary relevance of Mexican Americans’ history. At least a small slice of that relevance concerns the twenty-first-century legacy of Mexican Americans’ history as off-white—sometimes defined as legally white, almost always defined as socially non-white. Unlike other off-white groups that moved definitively into the white category within decades after their initial arrival in the United States (for example, Jews, the Irish, Italians), Mexican Americans, as a group, have continued to be off-white, neither definitively white or definitively non-white.1
Remarkably, after more than 160 years, Mexican Americans still are insecure in their collective claim to white racial identity, and, in the early twenty-first century, may be further than ever from becoming white, or, to put it another way, they may be permanently off-white. At least by some indicia, there has been a growing rejection of white status among Mexican Americans, even as some of them continue to embrace whiteness as a route to social equality in the United States. In a sense, this is predictable given the history explored in this book. One of the effects of Mexican Americans’ existence as socially non-white has been the adoption of a non-white racial identity that has been both assigned by the dominant society and asserted by group members themselves.2 At the same time, Mexican Americans occasionally, and particularly in legal contexts, have been treated as white. It is not so surprising, then, that Mexican Americans have at times embraced both a white and a non-white racial identity, both collectively and individually.
To explore the contemporary manifestations of this dynamic I examine how Mexican Americans and other Latinos have responded to census questionnaires about racial identity. Census questions and responses to them are a fascinating way to explore racial identity. The Census Bureau stands in for the nation-state and for the government’s conception of racial and ethnic categories. The historical evolution of census categories powerfully illustrates how race is socially constructed.3 Moreover, the decadal enumeration of the American population serves symbolic purposes. The very act of counting the nation and dividing that count into particular categories has been an act of defining who we are and who we aspire to be as a nation.4
The process of defining the nation in this way also is deeply implicated in interest group and electoral politics. For example, President Lyndon B. Johnson created the Inter-Agency Committee on Mexican American Affairs, which early in the Nixon administration morphed into the Cabinet Committee on Opportunities for Spanish-Speaking People.5 These events at the national level stemmed from the larger context of the civil rights movement, as well as partisan estimations about the potential value of the Mexican American electorate.6 At around the same time, and partly in response to the criticism that Mexican Americans had been undercounted in 1970, the Census Bureau formed an advisory committee to explore how to better identify Spanish-speaking people in the 1980 census.7 The majority of the nineteen committee members were Mexican Americans (including leaders of the major civil rights organizations), and they recommended use of “Hispanic” as a term to include Mexican Americans, Puerto Ricans, Cuban Americans, and other Americans with origins in Latin American countries.8
Not long before this book appeared in 2007, Hispanics surpassed non-Hispanic blacks as the country’s most numerous minority group.9 Both the 2000 and 2010 census results showed Latinos to be the fastest growing ethno-racial group in the United States, with most of the growth stemming from births rather than immigration.10 Hispanics will continue to grow at a high rate because they are relatively young: whereas 15 percent of whites are sixty-five years old or older, only 5 percent of Hispanics are in that age group; 34 percent of Hispanics are under eighteen years old, compared to 20 percent of whites.11 Of the nation’s 41.3 million Latinos, nearly 60 percent are Mexican American, nearly 10 percent are Puerto Rican, and less than 4 percent are Cuban American; almost all of the remaining 26 percent come from six additional countries in the Caribbean and Central and South America.12 As might be expected given these points of origin, the population is regionally concentrated, with 75 percent of Cubans living in the South, 60 percent of Puerto Ricans living in the Northeast, and just over half of all Mexican Americans living in the West.13 More than 160 years after its invasion by the United States, New Mexico continues to have the highest proportion of Hispanics of any state (43 percent), followed by California and Texas (both at 35 percent).14
How the Census Bureau has counted Mexican Americans has varied considerably since the end of the war with Mexico in 1848.15 Between 1850 and 1920, Mexicans were not distinguished as a separate group in the census; they were counted in the white category. Inconsistently asked and inconsistently phrased questions pertaining to foreign birth, foreign parentage, language usage, and Spanish surname were used in the mid-1900s as proxies for counting Mexican Americans. Consistently unreliable estimates of the Mexican American population resulted. Foreign birth and foreign parentage, for example, failed to include the descendants of those Mexican Americans who lived in the Mexican Cession. Spanish surname was especially problematic, as it was both over- and underinclusive of Mexican Americans. One 1970s study found that reliance on Spanish surname resulted in overcounting Mexican Americans by about one-third and undercounting them by about one-third; in other words, Hispanics who did not have Spanish surnames (by virtue of intermarriage) were undercounted, while non-Hispanics whose names were mistaken as Spanish were included in the total.16 If this study was accurate, ironically, the proxy of Spanish surname may well have resulted in the correct aggregate number of Hispanics because overestimates and underestimates due to surname canceled each other out.
The pattern of counting Mexican Americans as “white” changed in 1930 when the census used “Mexican” appeared as a separate racial category.17 The designation of Mexicans as a racial group coincided with the Great Depression and with increased economic competition between whites and Mexicans, both native-born and immigrant.18
The economic climate fomented anti-Mexican racism, violence, and government hostility, including mass deportation (to Mexico). As many as one million Mexican-origin persons, including many American citizens, were rounded up by police and deported during this period.19 In Los Angeles, fully one-third of the Mexican American population returned to Mexico in the 1930s, either forcibly or voluntarily (in anticipation of the roundups).20 In northern New Mexico, where economic opportunities did not lure substantial numbers of Mexican immigrants until recent decades, Mexican Americans “remember being loaded in trucks and taken by train and truckload and dumped as ‘Mexican citizens’” across the U.S.–Mexico border in the 1930s.21
The racist, violent events of the 1930s were a sharp reminder to Mexican Americans of their marginal status. Although they were formally U.S. citizens, their Mexican American racial status kept them in a second-class position. Mexican American elites responded to the outrages of the 1930s by pressuring the Census Bureau to reclassify Mexican Americans as white.22 In 1940 and 1950, the census did so, but it also sought alternative ways to count Mexican Americans, as noted above. The 1960 census experimented with the category “white persons of Spanish surname” but only in the states of Arizona, California, Colorado, New Mexico, and Texas, where 90 percent of Latinos in the United States (virtually all Mexican Americans) then lived.23
Manifest Destinies, Second Edition Page 22