Anthropology essay final

[Pp. 15-36 in : How the U.S. Racializes Latinos: White Hegemony and Its Consequences, edited by José A. Cobas, Jorge Duany and Joe R. Feagin. Paradigm Publishers (2009)]

Pigments of Our Imagination: On the Racialization and Racial Identities of

“Hispanics” and “Latinos”

Rubén G. Rumbaut

“Why should Pennsylvania, founded by the English, become a Colo ny of Aliens, who will shortly be so numerous as to Germanize us instead of our Anglifying them, and will never

adopt our Language or Customs, any more than they can acquire our Complexion?” —Benjamin Franklin (1751)

“‘Race’ is a trope of ultimate, irreducible difference.” —Henry Louis Gates, Jr. (1986)

I have been telling my students since the 1970s that “race is a pigment of our imagination.” The play on words of the definition is meant as a double entendre, both to debunk baseless biological pretensions and to focus attention on the social, legal, and political construction of categories meant to put people “in their place” in hierarchies of power and privilege. “Race” is a social status, not a zoological one; a product of history, not of nature; a contextual variable, not a given. It is a historically contingent, relational, intersubjective phenomenon—yet it is typically misbegotten as a natural, fixed marker of phenotypic difference inherent in human bodies, independent of human will or intention. What is called “race” is largely the sociopolitical accretion of past intergroup contacts and struggles, which establish the boundaries and thus the identities of victors and vanquished, of dominant and subordinate groups, of “us” and “them,” with their attendant conceits of superiority and inferiority and invidious taxonomies of social worth or stigma. As such “race” is an ideological construct linking supposedly innate traits of individuals to their rank and fate in the social order. Racial statuses and categories (and the putative differences that they connote) are imposed and infused with stereotypical moral meaning, all the more when they become master statuses affecting all aspects of social life. The dominant “racial frame” (Feagin 2006) that evolved in what became the United States, during the long colonial and national era of slavery and after it, was that of white supremacy. Benjamin Franklin’s words in the epigraph above are illustrative; they were written in 1751, a quarter of a century before he signed the Declaration of Independence with no hint of irony, back when not even Germans were imagined to be “white,” mixing nativism and racism in what would become a familiar, habitual American blend.

Rubén G. Rumbaut 8

CREATING A “HISPANIC” AND “LATINO” CATEGORY IN OFFICIAL STATISTICS

Beginning in 1850, the U.S. Census relied on objective indicators, such as country of birth (or decades later, parent’s birthplace, mother tongue or “Spanish surname”), to identify persons of Mexican origin in its decennial counts. Mexicans were coded as “white” for census purposes from 1850 to 1920. They were then classified as a separate “race” in the 1930 census, amid the Great Depression. During that tumultuous decade perhaps a million or more were forcibly “repatriated” to Mexico, including many U.S. citizens (see Johnson, 2005; Kanstroom, 2007; Ngai, 2004); but Mexican American civil rights groups, with the support of the Mexican government, demanded not to be so designated. That racial usage was subsequently eliminated and Mexicans were again classified as “white” in the 1940 census and thereafter.

It was only in the 1950s, a decade in which more Puerto Ricans came to the U.S. mainland than did immigrants from any other country, that the Census Bureau first published information on persons of Puerto Rican birth or parentage. Tabulations on people of Cuban birth or parentage were first published in 1970, following the large flows that came to the U.S. after the 1959 Cuban Revolution. Efforts to demarcate and enumerate the “Hispanic ” population as a whole, using subjective indicators of Spanish origin or descent, date back to the late 1960s. At that time—in the context of surging civil rights activism, new federal legislation which required accurate statistical documentation of minority groups’ disadvantages, and growing concerns over differential census undercounts—Mexican American organizations in particular pressed for better data about their group (Choldin, 1986). The White House ordered the addition of a Spanish-origin self- identifier on the 1970 census (in the “long form” sent to a 5 percent sample); to test it, the same question was inserted in the November 1969 Current Population Survey (the first time that subjective item was used). Later analyses by the Census Bureau, comparing the results nationally of the (subjective) Hispanic self- identification in the CPS vs. the (objective) use of Spanish surnames, found wide ranging differences between the two measures, raising questions of validity and reliability. For example, in the Southwest, only 74 percent of those who identified themselves as Hispanic had Spanish surnames, while 81 percent of those with Spanish surnames identified themselves as Hispanic; in the rest of the U.S., only 61 percent of those who identified as Hispanic had Spanish surnames, and a mere 46 percent of those with Spanish surnames identified as Hispanic (U.S. Census Bureau, 1975).

In 1976, the United States Congress passed a remarkable bill—Public Law 94- 311—a joint resolution “Relating to the publication of economic and social statistics for Americans of Spanish origin or descent.” Signed by President Ford in June 1976, it remains the only law in the country’s history that mandates the collection, analysis and publication of data for a specific ethnic group, and goes on to define the population to be enumerated. The law, building on information gathered from the 1970 census, asserted that “more than twelve million Americans identify themselves as being of Spanish- speaking background and trace their origin or descent from Mexico, Puerto Rico, Cuba, Central and South America, and other Spanish-speaking countries;” that a “large number” of them “suffer from racial, social, economic, and political discrimination and are denied

Rubén G. Rumbaut 9

the basic opportunities that they deserve as American citizens;” and that an “accurate determination of the urgent and special needs of Americans of Spanish origin and descent” was needed to improve their economic and social status. Accordingly, the law mandated a series of data collection initiatives within the Federal Departments of Commerce, Labor, Agriculture, and Health, Education and Welfare, specifying among other things that the Spanish-origin population be given “full recognition” by the Census Bureau’s data-collection activities via the use of Spanish language questionnaires and bilingual enumerators, as needed; and that the Office of Management and Budget (OMB) “develop a Government-wide program for the collection, analysis, and publication of data with respect to Americans of Spanish origin or descent ” (Rumbaut, 2006).

In 1977, as required by Congress, OMB’s Statistical Policy Division, Office of Information and Regulatory Affairs, issued “Directive 15: Race and Ethnic Standards for Federal Statistics and Administrative Reporting” to standardize the collection and reporting of “racial” and “ethnic” statistics and to include data on persons of “Hispanic origin.” Directive 15 specified a minimal classification of four “races” (“American Indian or Alaskan Native,” “Asian or Pacific Islander,” “Black,” and “White”) and two “ethnic” backgrounds (“of Hispanic origin” and “not of Hispanic origin”), and allowed the collection of more detailed information as long as it could be aggregated within those categories. Since that time, in keeping with the logic of this classification, census data on Hispanics have been officially reported with a footnote indicating that “Hispanics may be of any race” (for usage rules, see U.S. Census Bureau, 2003).

Tellingly, however, the term led to the development of another category, “non- Hispanic white” (a catchall for persons who identify as white but whose ancestry does not include a Spanish-speaking nation), which has been typically set against “Hispanics” and the other racial minority categories, conflating the distinction. In the news media, academic studies, government reports, and popular usage the “ethnic” constructs “Hispanic” or “Latino” have already come to be used routinely and equivalently alongside “racial” categories such as “Asian,” “Black” and “non-Hispanic White,” effecting a de facto racialization of the former. It is now commonplace to see media summaries of exit polls tallying the “Latino vote” alongside “white” and “black” rates, or similar tallies “by race” of school dropout, poverty and crime rates; or for local TV news crime-beat reporters to quote police sources that “the suspect is a Hispanic male,” as if that were a self-evident physical description; or to read newspaper articles which report matter-of-factly that the country’s first “Hispanic” astronaut was Franklin Chang-Díaz, who is a Chinese Costa Rican; or that the first “Latina” chancellor of a University of California campus is France A. Córdova, a French-born physicist who majored in English at Stanford, whose mother is an Irish-American native New Yorker and whose father came to the U.S. as an 8-year-old from Tampico.

Later critic ism of the categories led to a formal review of Directive 15, beginning in 1993 with Congressional hearings and culminating in revised standards which were adopted in 1997 (U.S. Bureau of the Census, 1997; see also Snipp, 2003; Fears, 2003). The changes now stipulated five minimum categories for data on “race” (“American Indian or Alaska Native,” “Native Hawaiian or Other Pacific Islander,” “Asian,” “Black or African American,” and “White”); offered respondents the option of selecting one or

Rubén G. Rumbaut 10

more racial designations (an option used for the first time in the 2000 census); and reworded the two “ethnic” categories into “Hispanic or Latino” and “not Hispanic or Latino.” “Hispanic or Latino” was defined as “a person of Cuban, Mexican, Puerto Rican, South or Central American, or other Spanish culture or origin, regardless of race. The term, ‘Spanish origin,’ can be used in addition to ‘Hispanic or Latino.’” The notice in the Federal Register of these revisions to OMB Directive 15 (as adopted on October 30, 1997) pointedly added that “The categories in this classification are social-political constructs and should not be interpreted as being scientific or anthropological in nature… The standards have been developed to provide a common language for uniformity and comparability in the collection and use of data on race and ethnicity by Federal agencies.” Nonetheless, Directive 15’s definitions of "racial" and "ethnic" populations are used not only by federal agencies, but also by researchers, schools, hospitals, business and industry, state and local governments—and are conflated, abridged and diffused through the mass media, entering thereby into the popular culture and shaping the national self- image.

ASSERTING IDENTITIES: NATION, “RACE” AND PLACE IN THE 2000 CENSUS

Much has been made in the media and even in academic discourse about “the browning of America,” a misnomer based on stereotypes of phenotypes presumed to characterize peoples of Latin American origin. Does the Hispanic population differ significantly from non-Hispanics by “race,” as it does by place, socioeconomic status, and national origins? The American system of racial classification, employed variously since the first census of 1790, has been the sine qua non of externally imposed, state-sanctioned measures of group difference, distinguishing principally the majority “white” population from “black” and American Indian minority groups, and later from Asian-origin populations (Snipp, 2003). Yet as seen earlier, “Hispanics” were incorporated in official statistics as an “ethnic” category, and conceived as being “of any race.” Moreover, prior to 1970 Mexicans were almost always coded as “white” for census purposes, and were deemed “white” by law (if not by custom) since the nineteenth century. In addition, neither “Hispanic” nor “Latino” is a term of preference used by Latin American newcomers in the United States to define themselves; rather, the research literature has consistently shown that they self- identify preponderantly by their national origin. How then are racial categories internalized by Hispanics? Are there intergroup and intragroup differences in their patterns of racial self- identification? The 2000 census asked separate questions for “Hispanic” or “Latino” origin and for “race,” permitting a cross-tabulation of the two— and thus an examination of how “Hispanics” or “Latinos” self- report by “race” as well as by national origin.

Despite growing diversification and accelerating immigration from a wider range of Latin American countries over the past few decades, persons of Mexican, Puerto Rican, and Cuban origin still comprised 77 percent of the 35.2 million Hispanics counted by the 2000 census. Those of Mexican origin alone numbered 22.3 million—63 percent of the U.S. total at the time. Puerto Ricans on the mainland accounted for another 10 percent, and Cubans for 4 percent. [If Puerto Ricans living on the island (who are U.S. citizens by birthright) were added to the calculation, those three groups would comprise

Rubén G. Rumbaut 11

80 percent of the total.] Muc h of the remainder was accounted for by six nationalities of relatively recent immigrant origin: Dominicans, Salvadorans and Guatemalans make up 7 percent of the Hispanic total, while Colombians, Peruvians and Ecuadorians combine for nearly 4 percent more. Hence, nine ethnic groups accounted for nine out of ten (88 percent) Hispanics in the U.S. mainland. Their size and evolution reflect both the varied history of their incorporation in the United States and the relative geographical proximity of their source countries to the United States: Mexico, El Salvador and Guatemala from Meso-America; Puerto Rico, Cuba, and the Dominican Republic from the Caribbean; Colombia, Perú and Ecuador from South America. Persons who trace their ethnic identities to the ten other Spanish-speaking countries of Central and South America, plus Spain, comprised only 4 percent of the Hispanic total. And only 8 percent self-reported as “other Spanish, Hispanic or Latino” in the 2000 census, without indicating a specific national origin.

Hispanics as a whole are much more likely than non-Hispanics to consist of relatively recent newcomers to the United States: 45 percent of Hispanics are foreign- born, compared to less than 8 percent of non-Hispanics. Only the “other Spanish, Hispanic, Latino” is overwhelmingly a native-born population (94 percent)—some with ancestries that can be traced back many generations. Aside from that special case, the Mexicans and Puerto Ricans—the two populations of longest residence in the United States, and the largest by far—are the only ethnic groups that consist mainly of natives (58 percent of the Mexicans and 60 percent of the Puerto Ricans were born in the U.S. mainland). All others are primarily foreign-born populations—from two thirds of the Cubans and Dominicans to more than three- fourths of all the other groups.

Intergroup Differences: Self-Reported Race among Latinos in the 2000 Census

Table 1 compares Hispanics and non-Hispanics, as well as the largest Hispanic ethnic groups, by the main racial categories employed in the 2000 census. Of the 246.2 million non-Hispanics counted by that census, 97 percent reported their race either as white (79 percent), black (14 percent), or Asian (4 percent). In sharp contrast, among the 35.2 million Hispanics, only half reported their race as white (48 percent), black (1.8 percent), or Asian (0.3 percent). Most notably, there was a huge difference in the proportion of these two populations that chose “other race:” while scarcely any non- Hispanics (a mere 0.2 percent) reported being of some “other race,” among the Latin Americans that figure was 43 percent—a reflection of more than four centuries of mestizaje in Latin America and the Caribbean, as well as differing histories and conceptions of “race.” In addition, Hispanics in the 2000 census were more than three times as likely to report an admixture of “two or more races”—6.4 percent of Hispanics vs. only 2 percent of non-Hispanics—although among Hispanics who listed “two or more races,” the overwhelming majority (85 percent) specified “white” plus another race. Still, the main divide among Hispanics was between the 48 percent who racially self- identified as “white,” and the 43 percent who rejected all the official categories and reported “other race” instead. [The Census Bureau is considering eliminating the “other race” option in the 2010 census in order to force respondents to choose from among the five standard racial options mandated by the OMB directive.]

Rubén G. Rumbaut 12

Ethnic Identity % "Other" % Two or % Total N race % White more races % Black % Asian Indigenous*

Total U.S. Population 281,421,906 5.5 75.1 2.6 12.2 3.6 1.0

Not Hispanic/Latino 246,217,426 0.2 79.0 2.0 13.7 4.1 1.0 Hispanic/Latino 35,204,480 42.6 47.8 6.4 1.8 0.3 1.1

Dominican 994,313 58.8 22.4 9.4 8.2 0.2 0.9 Salvadoran, Guatemalan 1,532,512 55.2 35.8 7.2 0.6 0.1 1.1 Mexican 22,293,812 45.8 46.8 5.2 0.7 0.2 1.2 Peruvian, Ecuadorian 697,798 41.7 47.9 8.5 0.6 0.4 0.8 Puerto Rican 3,537,351 38.1 46.9 8.1 5.8 0.4 0.7 Other Central American 903,574 37.7 44.7 9.5 7.1 0.2 0.9 Colombian 648,731 28.2 62.0 8.2 1.1 0.2 0.4 Other South American 494,186 20.6 70.0 8.0 0.8 0.3 0.4 Cuban 1,311,994 7.6 84.4 4.1 3.6 0.2 0.2 Other Spanish, Hispanic, Latino 2,790,209 34.7 49.2 10.7 2.5 1.0 1.9

Source: 2000 U.S. Census, 5% IPUMS.

Table 1 Hispanic/Latino Ethnic Identity by Self-Reported "Race," 2000 Census,

Race (Self-Reported)

* Includes American Indians, Alaskan and Hawaiian natives and other indigenous Pacific Islanders.

Ranked by Proportion Identifying as "Other Race"

Examining these results for each of the main Hispanic ethnic groups, the proportions who identified racially as “white” ranged from a low of 22 percent among Dominicans to a high of 84 percent among Cubans. More than half of the Dominicans (59 percent) and the Salvadorans and Guatemalans (55 percent) reported “another race,” as did 46 percent of the Mexicans, 42 percent of the Peruvians and Ecuadorians, 38 percent of the Puerto Ricans, 28 percent of the Colombians, and less than 8 percent of the Cubans. The most likely to identify as “black” were the Dominicans (8.2 percent), while the “other Spanish, Hispanic or Latino” were the most likely to identify as multiracial (10.7 percent). The meaning of “race,” however, is problematic for a number of reasons. Consider, for example, the importance of geographic context in the determination and variability of self-reported racial identities in the census.

Location, Location, Location: Intragroup Differences by Race and Place

Self-reported “race” varies not only between Latin American-origin groups, but also within them—and over time and place as well. Table 2 presents 2000 census data on self-reported “race” for the largest Hispanic groups, now broken down by the largest states: California and Texas in the Southwest (where Mexicans, Salvadorans and Guatemalans are most concentrated), and New York-New Jersey and Florida along the east coast (where the Caribbean groups are concentrated). The differences are striking: In California 40 percent of the Mexican-origin population reported as “white,” but in Texas 60 percent were “white;” and 53 percent reported as “other race” in California, compared to only 36 percent in Texas. (Indeed, a 1998-2002 longitudinal surve y in Los Angeles and San Antonio found that Mexican Americans in San Antonio were five times more likely to identify as “white” than those in Los Angeles [Telles and Ortiz, 2008: 272, 312]). Similar if less pronounced patterns were observed for Salvadorans and Guatemalans in

Rubén G. Rumbaut 13

those two states: they were significantly more likely to be “white” in Texas and “other” in California.

Ethnic Identity Total N % Two or % "Other" % White more races

Hispanic/Latino (U.S. total): 35,204,480 In California 10,928,470 51.6 39.7 6.4 In Texas 6,653,338 36.7 58.0 4.1 In New York-New Jersey 3,972,595 43.3 42.1 7.6 In Florida 2,673,654 16.6 75.0 5.4

Mexican 22,293,812 In California 9,025,952 52.7 39.7 5.6 In Texas 5,706,532 35.8 59.6 3.6

Salvadoran, Guatemalan 1,532,512 In California 667,273 61.1 30.2 7.2 In Texas 146,781 54.1 40.0 5.2

Puerto Rican 3,537,351 In New York-New Jersey 1,462,393 40.5 45.4 6.5 In Florida 496,122 22.9 66.9 6.2

Cuban 1,311,994 In New York-New Jersey 151,744 13.3 72.6 5.5 In Florida 878,289 3.9 91.6 2.5

Dominican 994,313 In New York-New Jersey 709,755 62.4 19.8 8.9 In Florida 92,785 33.9 45.5 9.2

Colombian 648,731 In New York-New Jersey 224,391 34.0 56.4 8.1 In Florida 192,397 14.5 77.8 6.4

Peruvian, Ecuadorian 697,798 In New York-New Jersey 336,769 47.2 43.4 7.8 In Florida 96,754 18.5 73.6 6.9

Source: 2000 U.S. Census, 5% IPUMS.

Table 2 "Race" Self-Reported by Largest Hispanic Groups in Selected States, 2000 Census

Race (Self-Reported)

Even more striking are the differentials in the geography of “race” among the Caribbean groups: all were far more likely to be “white” in Florida than in New York- New Jersey. In Florida 67 percent of the Puerto Ricans reported that they were “white,”

Rubén G. Rumbaut 14

compared to only 45 percent in New York-New Jersey; the respective percentages for Cubans were 92 to 73 percent; for Dominicans, 46 to 20 percent; for Colombians, 78 to 46 percent. The gap was wider still for the Peruvians and Ecuadorians, 74 to 43 percent. In all cases, as Table 2 shows, the reverse obtained for self- reports of “other race.” Those systematic patterns across so many different nationalities are unlikely to be explained by selective migrations, but rather invite a contextual, counterintuitive explanation: the more rigid racial boundaries and “racial frame” developed in the former Confederate states of Texas and Florida, and the severe stigma historically attached to those marked as non- white there, may shape defensive assertions of whiteness when racial status is ambiguous; in states like California and New York, the social dynamics have been more open to ethnic options and a rejection of rigid U.S. racial categories. If “race” was an innate, permanent trait of individuals, no such variability would obtain. Instead, these data exemplify how “race” is constructed socially and historically—and spatially as well.

These striking contextual differences are supported by other relevant data. For example, the 2000 census conducted in Puerto Rico found that 81 percent of the population on the island self-reported as “white”—notably higher than the 67 percent of Puerto Ricans who self- reported as “white” in Florida, vs. 45 percent in the New York region. A census conducted by the U.S. when it occupied the island in 1899 found that 62 percent of the inhabitants were “white,” as were 65 percent of those counted in the 1910 island census; but that proportion grew to 73 percent in 1920, and 80 percent by 1950—an increase attributed by Loveman and Muñiz (2007) to changes in the social definition of whiteness and the influence of “whitening” ideology on the island, since they could not be accounted for by demographic processes, institutional biases, or other explanations. Similarly, a study of racial self- identification of Puerto Ricans surveyed in the U.S. and Puerto Rico (Landale and Oropesa, 2002) found that mainland Puerto Ricans more strongly reject the conventional U.S. notion of race than do their island counterparts. Contexts shape the meanings of identity assignments and assertions (Rumbaut, 2005).

ASSERTING IDENTITIES: THE MALLEABLE MEANINGS OF “RACE”

Varieties of Racial Identification among Dominicans

The meaning of “race” also varies depending on the history of the group, on the way questions are asked, and even on the response format provided in conventional surveys. In a survey of more than 400 Dominican immigrants in New York City and Providence, Rhode Island, the adult respondents were asked a series of three questions about their racial self- identification (Itzigsohn, 2004). First they were asked, in an open- ended format, how they defined themselves racially. Next they were given a close-ended question, asking if they were white, black, or other (and if other, to specify). Finally they were asked how they thought that “mainstream Americans” classified them racially. The results are summarized in Table 3. In response to the first open-ended question, 28 percent gave “Hispanic” as their “race,” another 4 percent said “Latino,” and still others offered a variety of mixed “Hispanic” or “Latino” answers; 13 percent said “Indio,” and another 13 percent gave their Dominican nationality as their race. Only 6.6 percent chose “black,” and 3.8 percent “white.”

Rubén G. Rumbaut 15

How to you Are you: white, How do you think define yourself black, or other? most Americans

racially? (if other, specify) classify you Responses (open-ended Q) (close-ended Q) racially?

% % %

Black 6.6 16.8 36.9 White 3.8 11.6 6.4

Hispano/a (Hispanic) 27.5 21.1 30.4 Latino/a 4.1 2.8 3.2

Indio/a 13.1 18.8 4.0 Dominicano/a 12.8 2.0 0.2

Mestizo/a 4.7 8.0 1.0 Trigueño/a 4.1 4.5 2.0 Moreno/a 1.9 2.0 2.2 Mulato/a 0.3 1.5 0.0

Indio Hispano/a 4.1 1.0 0.2 Black Hispano/a 0.6 1.0 2.0 White Hispano/a 0.6 0.3 0.5 Mixed Hispano/a 0.6 1.3 0.2 Latino-Americano/a 0.6 0.5 0.0 Latino-Hispano/a 0.3 0.5 0.5

Java-India claro/a 0.3 1.3 0.2 Amarillo/a (yellow) 0.3 1.0 0.2 Oscuro, prieto, de color 0.3 0.8 1.0

American 0.6 0 0.5 Puerto Rican 0 0 0.2 Human race, other 6.9 1.5 0.7 Does not know 5.0 1.3 6.9

Table 3

Source: Adapted from Itzigsohn, 2004.

(Survey of Dominican immigrants in New York City and Providence, N=418) Dominican Immigrants' Answers to Three Racial Self-identification Questions

The rest of their responses showed the extraordinary range of racial categories and labels common in the Spanish Caribbean—as well as the very significant responses obtained depending on the question asked, even though all three were ostensibly getting at the same thing: the respondent’s racial identity. When asked to choose from the

Rubén G. Rumbaut 16

closed-ended format of the second question, the largest response remained “Hispanic” (written in by 21 percent of the sample, in addition to 3 percent who chose “Latino”), though the catego ries “black” and “white” now more than doubled to 16.8 and 11.6 percent, respectively. And when asked how they thought that others classified them racially, the category “black” dramatically increased to 37 percent—reflecting the reverse way in which the “one drop rule” functions in the United States vs. the Dominican Republic—while “white” decreased to 6.4 percent. “Hispanic” was still given by almost a third of the sample (30.4 percent) as the “racial” category that they perceived others used to classify them. Indeed, “Hispanic” was the label most consistently given by the respondents to characterize their own racial identity, whether asserted by themselves or imposed upon them by others.

Intergenerational Differences: The “Race” of Immigrant Parents and their Children

Another recent study found that, in addition to significant change in their ethnic self- identities over time and generation in the United States (as measured by open-ended questions), the offspring of Latin American immigrants were by far the most likely to define their racial identities in sharp contrast to their own parents (Portes and Rumbaut, 2001; Rumbaut, 2005). During the 1990s in South Florida and Southern California, the Children of Immigrants Longitudinal Study (CILS) surveyed a sample of more than 5,200 1.5- and second-generation youths, representing 77 different nationalities, including all of the main Spanish-speaking countries of Latin America. Their immigrant parents were also interviewed separately. In one survey (conducted when the youths were 17 to 18 years old), the respondents were asked to answer a semi- structured question about their “race,” and were given the option to check one of five categories: “white,” “black,” “Asian,” “multiracial,” or “other;” if the latter was checked, they had to specify what that “other race” was. The results are presented in Table 4.

Among the Latin American-origin youths, less than a fourth of the total sample checked the conventional categories of white, black, or Asian; 12 percent reported being multiracial; and over 65 percent checked “other.” When those “other” self-reports were coded, it turned out that two- fifths of the sample (41 percent) wrote down “Hispanic” or “Latino” as their “race,” and another fifth (19.6 percent) gave their nationality as their “race.” The explicit racialization of the “Hispanic- Latino” category, as well as the substantial proportion of youths who conceived of their nationality of origin as a racial category, are noteworthy both for their potential long-term implications in hardening minority group boundaries, and for their illustration of the arbitrariness of racial constructions—indeed, of the ease with which an “ethnic” category developed for administrative purposes becomes externalized, diffused, objectified, and finally internalized and imagined as a marker of essentialized social difference.

The latter point is made particularly salient by directly comparing the youths’ notions of their “race” with that reported by their own parents. The closest match in racial self-perceptions between parents and children were observed among the Haitians, Jamaicans and other West Indians (most of whom self-reported as black), among the Europeans and Canadians (most of whom labeled themselves white), and among most of the Asian-origin groups (except for the Filipinos). The widest mismatches by far (and

Rubén G. Rumbaut 17

hence the most ambiguity in self-definitions of “race”) occurred among all of the Latin- American-origin groups without exception: about three fifths of Latin parents defined themselves as “white,” compared to only one fifth of their own children. More specifically, 93 percent of Cuban parents identified as white, compared to only 41 percent of their children; 85 percent of Colombian parents defined themselves as white, but only 24 percent of their children did so—proportions that were similar for other South Americans; two thirds of the Salvadoran, Guatemalan, and Nicaraguan parents saw themselves as white, but only one fifth of their children agreed; and about a third of the Dominican parents reported as white, more than twice the proportion of their children who did so.

National Origin Respondent White Black Asian Multiracial Hispanic, Latino

Nationality as race

Other

(Parent/Child) % % % % % % %

Latin American: Parent 58.1 1.5 1.1 14.7 6.4 8.3 9.8 Child 21.9 0.8 0.0 12.1 41.0 19.6 4.6

Mexico Parent 5.7 0.0 2.1 21.6 15.9 26.1 28.5 Child 1.5 0.3 0.0 12.0 25.5 56.2 4.5

Cuba Parent 93.1 1.1 0.3 2.5 1.1 0.5 1.4 Child 41.2 0.8 0.0 11.5 36.0 5.5 4.9

Dominican Republic Parent 30.6 11.1 0.0 44.4 0.0 5.6 8.3 Child 13.9 2.8 0.0 13.9 55.6 8.3 5.6

El Salvador, Guatemala Parent 66.7 4.2 4.2 16.7 8.3 0.0 0.0 Child 20.8 0.0 0.0 12.5 58.3 4.2 4.2

Nicaragua Parent 67.7 0.5 1.6 22.0 5.4 0.5 2.2 Child 19.4 0.0 0.0 9.7 61.8 2.7 6.5

Other Central America Parent 48.0 24.0 4.0 20.0 0.0 4.0 0.0 Child 8.0 8.0 0.0 40.0 44.0 0.0 0.0

Colombia Parent 84.6 1.1 0.0 9.9 2.2 0.0 2.2 Child 24.2 1.1 0.0 9.9 58.2 1.1 5.5

Perú, Ecuador Parent 61.8 0.0 0.0 26.5 2.9 2.9 5.9 Child 32.4 0.0 0.0 11.8 55.9 0.0 0.0

Other South America Parent 87.8 0.0 0.0 6.1 2.0 2.0 2.0 Child 28.6 2.0 0.0 14.3 40.8 14.3 0.0

question on racial identity. "White," "Black," "Asian" and "Multiracial" were fixed responses; all "Others" were open-ended entries. * Figures are row percentages. Interviews with immigrant parents and their teenage children were done separately, using the same Source: Children of Immigrants Longitudinal Study (CILS); Portes and Rumbaut 2001; Rumbaut 2005.

Table 4 Self-Reported “Race” of Children of Immigrants and their Parents, by National Origin Groups

Self-Reported Race*

The children, instead, largely adopted “Hispanic” or “Latino” as a racial label (41 percent—the largest single response), whereas scarcely any of their parents did so (6 percent); or they gave their nationality as their race (20 percent of the children vs. 6 percent of their parents). Indeed, well over half of the Dominican, Salvadoran,

Rubén G. Rumbaut 18

Guatemalan, Nicaraguan, Colombian, Peruvian and Ecuadorian youth reported their race as “Hispanic” or “Latino.” Among the Mexicans, whose pattern differed from all of the others, the children preponderantly racialized the national label, whereas Mexican parents were more likely to use “other” and “multiracial” as descriptors. These results point to the force of the acculturation process and its impact on children’s self- identities in the U.S.—indeed, they provide another striking instance of the malleability of racial constructions, even between parents and children in the same family, residing in the same place. More fully exposed than their parents to American culture and its ingrained racial notions, and being incessantly categorized and treated as Hispanic or Latino, the children of immigrants learn to see themselves in these terms—as members of a racial minority— and even to racialize their national origins. If these intergenerational differences between Latin American immigrants and their U.S.-raised children can be projected to the third generation, the process of racialization could become more entrenched still. It is indeed ironic that in a nation born in white supremacy, where citizenship was restricted on racial grounds till 1952 and immigration until 1965, and where it took a civil rights revolution to overthrow the legal underpinnings of racial apartheid, the children of Latin American immigrants, historically “white by law,” should learn to become “non-white” in the post- Civil Rights era.

Rubén G. Rumbaut 19

REFERENCES

Alba, Richard D., Rubén G. Rumbaut, and Karen Marotz. 2005. “A Distorted Nation: Perceptions of Racial/Ethnic Group Sizes and Attitudes toward Immigrants and Other Minorities.” Social Forces 84, 2: 899-917.

Aleinikoff, T. Alexander, and Rubén G. Rumbaut. 1998. “Terms of Belonging: Are Models of Membership Self-Fufilling Prophecies?” Georgetown Immigration Law Journal, 13, 1: 1-24.

Almaguer, Tomás. 1994. Racial Fault Lines: The Historical Origins of White Supremacy in California . Berkeley: University of California Press.

Choldin, Harvey M. 1986. “Statistics and Politics: The ‘Hispanic Issue’ in the 1980 Census.” Demography 23, 3: 403-418.

Duany, Jorge. 2002. The Puerto Rican Nation on the Move: Identities on the Island and in the United States. Chapel Hill: University of North Carolina Press.

Fears, Darryl. 2003. “The Roots of ‘Hispanic:’ 1975 Committee of Bureaucrats Produced Designation,” Washington Post, October 15.

Feagin, Joe R. 2006. Systemic Racism: A Theory of Oppression. New York: Routledge.

Foley, Neil. 2004. “Straddling the Color Line: The Legal Construction of Hispanic Identity in Texas.” Pp. 341-357 in Nancy Foner and George Frederickson, eds., Not Just Black and White: Historical and Contemporary Perspectives on Immigration, Race, and Ethnicity in the United States. New York: Russell Sage Foundation.

Franklin, Benjamin. 1751. “Observations Concerning the Increase of Mankind, Peopling of Countries, &c.” At: http://bc.barnard.columbia.edu/~lgordis/early AC/documents/observations.html

Gates, Jr., Henry Louis. 1986. “Race,” Writing, and Difference. Chicago: University of Chicago Press.

Gómez, Laura E. 2007. Manifest Destinies: The Making of the Mexican American Race. New York: New York University Press.

Haney López, Ian. 2006. White by Law: The Legal Construction of Race. Revised and updated edition. New York: New York University Press.

Haub, Carl. 2006. “Hispanics Account for Almost One-Half of U.S. Population Growth.” PRB Report (February). Washington DC: Population Reference Bureau.

Itzigsohn, José. 2004. “The Formation of Latino and Latina Panethnic Identities.” Pp. 197-216 in Nancy Foner and George Frederickson, eds., Not Just Black and White: Historical and Contemporary Perspectives on Immigration, Race, and Ethnicity in the United States. New York: Russell Sage Foundation.

Johnson, Kevin. 2005. “The Forgotten ‘Repatriation’ of Persons of Mexican Ancestry and Lessons for the ‘War on Terror.” Pace Law Review, 26, 1: 1-26.

Rubén G. Rumbaut 20

Kanstroom, Danie l. 2007. Deportation Nation: Outsiders in American History. Cambridge, MA: Harvard University Press.

Landale , Nancy S., and R.S. Oropesa. 2002. “White, Black, or Puerto Rican? Racial Self- Identification among Mainland and Island Puerto Ricans.” Social Forces 81:231-254.

Langley, Lester D. 2003. The Americas in the Modern Age. New Haven: Yale University Press.

Loveman, Mara, and Jerónimo O. Muñiz. 2007. “How Puerto Rico Became White: Boundary Dynamics and Inter-Census Racial Reclassification.” American Sociological Review 72, 6: 915- 939.

Maltby, William S. 1968. The Black Legend in England: The Development of Anti-Spanish Sentiment, 1558-1560. Durham, NC: Duke University Press.

Menjívar, Cecilia, and Rubén G. Rumbaut, “Rights of Racial and Ethnic Minorities and Migrants: Between Rhetoric and Reality.” Pp. 60-73 in Judith Blau, David Brunsma, Alberto Moncada, and Catherine Zimmer, eds., The LeadingRogue State: The United States and Human Rights. Boulder, CO: Paradigm Publishers, 2008

Montejano, David. 1987. Anglos and Mexicans in the Making of Texas, 1836-1986. Austin: University of Texas Press.

Ngai, Mae N. 2004. Impossible Subjects: Illegal Aliens and the Making of Modern America. Princeton: Princeton University Press.

Olivas, Michael A., ed. 2006. “Colored Men” and “Hombres Aquí:” Hernández v. Texas and the Rise of Mexican American Lawyering. Houston: Arte Público Press.

Passel, Jeffrey S., and D’Vera Cohn. 2008. “U.S. Population Projections: 2005-2050.” Washington, DC: Pew Hispanic Center.

Portes, Alejandro, and Rubén G. Rumbaut. 2001. Legacies: The Story of the Immigrant Second Generation. Berkeley and New York: University of California Press; Russell Sage Foundation.

Rumbaut, Luis E. and Rubén G. Rumbaut. 2007. “‘If That Is Heaven, We Would Rather Go to Hell:’ Contextualizing U.S.-Cuba Relations.” Societies Without Borders 2, 1: 131-152.

Rumbaut, Rubén G. 2005. “Sites of Belonging: Acculturation, Discrimination, and Ethnic Identity among Children of Immigrants.” Pp. 111-163 in Thomas S. Weisner, ed., Discovering Successful Pathways in Children’s Development: New Methods in the Study of Childhood and Family Life. Chicago: University of Chicago Press.

Rumbaut, Rubén G. 2006. “The Making of a People.” Pp. 16-65 in Marta Tienda and Faith Mitchell (eds.), Hispanics and the Future of America. Washington, D.C.: National Academy Press.

Snipp, C. Matthew. 2003. “Racial Measurement in the American Census: Past Practices and Implications for the Future.” Annual Review of Sociology, 29: 563-88.

Rubén G. Rumbaut 21

Telles, Edward E. and Vilma Ortiz. 2008. Generations of Exclusion: Mexican Americans, Assimilation, and Race. New York: Russell Sage Foundation.

Tienda, Marta, et al. 2006. Multiple Origins, Uncertain Destinies: Hispanics and the American Future. Washington DC: National Academies Press.

Tocqueville, Alexis de. 1969 [1832]. Democracy in America. Translated by George Lawrence; edited by J.P. Mayer. Garden City, NY: Doubleday Anchor.

U.S. Census Bureau. 1975. Comparison of Persons of Spanish Surname and Persons of Spanish Origin in the United States. Technical Paper No. 38 (Washington, DC: U.S. Government Printing Office).

U.S. Census Bureau. 1997. Revisions to the Standards for the Classification of Federal Data on Race and Ethnicity. At: http://www.census.gov/population/www/socdemo/race/Ombdir15.html

U.S. Census Bureau. 2003. “Guidance on the Presentation and Comparison of Race and Hispanic Origin Data.” At: http://www.census.gov/population/www/socdemo/compraceho.html

U.S. Census Bureau. 2008. “U.S. Hispanic Population Surpasses 45 Million, Now 15 Percent of Total.” At: http://www.census.gov/Press-Release/www/releases/archives/population/011910.html

Walton, John. 2001. Storied Land: Community and Memory in Monterey. Berkeley: University of California Press.

Weber, David J., ed. 1973. Foreigners in their Native Land: Historical Roots of the Mexican Americans. Albuquerque: University of New Mexico Press.

Weber, David J. 1992. The Spanish Frontier in North America. New Haven: Yale University Press.

Zolberg, Aristide. 2006. A Nation by Design: Immigration Policy in the Fashioning of America. New York: Russell Sage Foundation; Cambridge, MA: Harvard University Press.