Social scientists have long considered high levels of racial and ethnic intermarriage — along with language acquisition, socioeconomic attainment, and residential patterns — a bellwether of social integration into the larger American society. Intermarriage requires individuals in different groups to form intimate attachments, which suggests that group boundaries are fading in importance and that preferences for marriage within the group are weak.
Moreover, the children of interracially married couples have complex racial backgrounds and often identify with two or more races, which further blurs the boundaries between groups and leads to more intermarriage in the next generation.
In the mid-20th century, Milton Gordon, a noted sociologist, presented the "straight-line" theory of racial and ethnic assimilation. He argued that levels of racial and ethnic intermarriage would increase steadily over generations as the social barriers between racial and ethnic groups diminished and preferences for in-group marriage faded.
Over the course of the 20th century, levels of intermarriage increased steadily among European immigrant descent groups. Levels of ethnic intermarriage were very low among European immigrants near the beginning of the 20th century, higher among their native-born children (the second generation) in the middle of the century, and very high among their grandchildren and great-grandchildren (the third and later generations) by the end of the 20th century.
The increases in levels of intermarriage across generations thus tracked and helped accelerate the integration of European groups, which originally were considered to be racially distinct, into American society.
During the last 40 years, however, most immigrants have come from Asia and Latin America. Are their children — the new second generation — showing higher levels of intermarriage than their immigrant parents?
Causes of Intermarriage
The "straight-line" theory of racial and ethnic assimilation suggests that levels of intermarriage increase across generations as social boundaries between groups diminish and elements of cultural distinctiveness, such as fluency in a non-English language, fade between the first and second, and between the second and third generations.
Levels of intermarriage between groups are, however, also affected by demographic factors in addition to preferences. For example, all else being equal, members of larger groups are less likely to intermarry than members of smaller groups because there have more potential partners.
Members of racial and ethnic groups that are more geographically clustered or segregated are more likely to marry within their own group due to the higher likelihood of interacting with one another within the shared space.
In addition, third parties can intervene. For example, some parents, especially immigrant parents, pressure their children to consider only prospective spouses of similar ethnic or racial descent.
The relative balancing of these forces varies according to generation. In some ways, the first generation is the most distinctive since many (although not all) marry before they arrive in the United States. Their marriages are thus not subject to the same demographic considerations as those of second and later generations.
Second-generation adults, who currently compose a relatively small number of people sandwiched between larger numbers of first- and third-generation adults, encounter relatively small numbers of prospective spouses who are also second generation.
Estimates from the March 2005 Current Population Survey (CPS) suggest that the first generation consists of about 35.3 million foreign-born Americans. The second generation, here defined as Americans with at least one foreign-born parent, consists of only about 21.1 million people, and the third generation, defined as Americans with native-born parents, contains over 221 million people.
If generation is a proxy for cultural distinctiveness, then many second-generation adults are pressured by demographic constraints to choose between first-generation immigrants who identify more strongly with their ethnic and racial origins and third-generation Americans for whom race and ethnicity are less important.
Because high rates of intermarriage between two or more groups are viewed by scholars as evidence that the social boundaries between the groups are fading, social science research on intermarriage typically relies on survey or census data. These data provide an overview of levels and patterns of intermarriage.
Unlike the U.S. census and most other major surveys, the U.S. Census Bureau's CPS contains information on the birthplaces of respondents and of respondents' parents. Consequently, it is possible to determine whether respondents are members of the first, second, or third (and later) generations.
To increase the sample size, data from the 2003, 2004, and 2005 Annual Social and Economic Supplements of CPS were merged. Because CPS does not explicitly ask for information on respondents' spouses, the analyses presented here only refer to married men and women who were living in the same household and whose information could be linked, not to all married couples.
Respondents were defined as foreign born or first generation if they were born abroad and did not have American-born parents. They were considered to be second generation if they were native born (i.e., born in the United States, Puerto Rico, and outlying areas, or born abroad to American parents) and reported having either one or two foreign-born parents.
Respondents were considered to be third (or later) generation if they were native born and both of their parents were also native born. Some "third-generation" respondents could therefore trace their ancestry on American soil back four or more generations.
The 2003-2005 CPS files contain information on respondents' race and Hispanic origin. Race is measured using the major categories of white, black, Asian, American Indian and Alaskan Native, and other, plus some complex categories such as black-Asian or black-white. Because the racial categories do not apply very well to Hispanic respondents, the category "Hispanic" is treated as a separate racial/ethnic category.
Very few second-generation respondents chose American Indian or any of the complex racial categories, and so the intermarriage statistics presented here focus on Asian, black, and white respondents, and Hispanic respondents of any race.
In 1970, just a few years after the U.S. Supreme Court struck down all antimiscenegation laws, less than one percent of marriages were interracial. Since then, levels of racial intermarriage have steadily increased. In 1980, about two percent of marriages were interracial; by 2000, about 5.4 percent were interracial.
The upward trend in levels of racial intermarriage has continued into the first part of the 21st century. CPS data gathered between 2003 and 2005 show that about 7.5 percent of all marriages are interracial.
Whatever the generational statuses of the husband and wife, most of the interracially married couples consist of a white spouse with an Asian, Hispanic, or black spouse; less than one percent of interracially married couples consist of two non-white spouses.
The combinations of races among the interracially married couples do differ by gender. For example, there are more Asian wives with white husbands than Asian husbands with white wives, and more black husbands with white wives than black wives with white husbands.
Figure 1 shows the percentages of married women by generation and race/ethnicity who have husbands of a different race or ethnicity than themselves.
The patterns of intermarriage across race and generation suggest that levels of racial intermarriage are strongly affected both by the differing opportunities and preferences for intermarriage within each generation and each ethnic/racial group. Levels of intermarriage are low among the first generation for all women. Except for Asian women, well over 90 percent of foreign-born women have husbands of the same racial or ethnic origins as themselves.
The low levels of intermarriage in the first generation are followed by higher levels of intermarriage in the second generation for all nonwhite women. Among Asians and Hispanics, the increase in levels of intermarriage continues into the third generation. For Asian and Hispanic women, then, the pattern fits the expectations generated by the "straight-line" assimilation theory, with steady increases in intermarriage across generations.
The picture differs for white women and black women. Levels of intermarriage among white women are relatively steady across generations, hovering around five percent. The steadiness can be attributed to the large numbers of whites in the American population — all else being equal, levels of intermarriage are always lowest among members of larger groups.
White foreign-born women are unlikely to be married to a man outside their ethnic/racial group because many arrive in the United States already married. Second- and third-generation white women are likely to meet and marry a native-born white man because the white population is still, by far, the largest racial group in the United States.
It is more difficult to explain why levels of intermarriage among blacks are substantially higher among the second generation than among the third-generation. Perhaps second-generation black women who grew up in a household with at least one foreign-born parent are less affected by the accumulation of racial discrimination in the American context and thus more open to marrying someone of another race. Second-generation blacks may also be more likely or be more able to emphasize their national or regional origins, e.g., Trinidadian, in lieu of identifying themselves as American-born black or African American.
Another possible reason for the gap is that second-generation black women are more likely than third-generation black women to live in major metropolitan areas, to have higher levels of education, and to have a racially complex ancestry — all attributes that lead to racial intermarriage.
And, finally, this result could be an anomaly generated by a small sample size: the level of intermarriage among second-generation black women is based on only 70 cases.
Since the results for men parallel those for women, those statistics are not presented here. Foreign-born men are very likely to have spouses of the same racial and ethnic origins as themselves. For Asian and Hispanic men, levels of intermarriage increase between the first and second, and the second and third generations. Levels of intermarriage among white men hover at low levels in every generation while levels of intermarriage among second-generation black men are higher than among first- or third-generation black men.
The next way to examine the data is by separating those with one foreign-born and one native-born parent from those with two foreign-born parents. The presumption is that those who grew up with two foreign-born parents identify more strongly with their racial and ethnic origins and so have stronger preferences for in-group marriage than those with one foreign-born and one native-born parent.
Second-generation Americans with one foreign-born and one native-born parent are also more highly educated and earn more than those with two foreign-born parents. This segment of the second generation is generally more integrated into American society.
In addition, many respondents with one foreign-born parent and one native-born parent are the children of racially intermarried parents, because marriages involving one immigrant and one native-born American spouse are likely to be interracial. Individuals with racially complex backgrounds are more open to the prospect of racial intermarriage than others.
Figure 2 shows that all racial/ethnic minority women with only one foreign-born parent and one native-born American parent are, in fact, more likely to be racially intermarried than second-generation women with two foreign-born parents. The same is true for men: racial/ethnic minority men with only one foreign-born parent are more likely to be racially intermarried than second-generation men with two foreign-born parents.
About half of Asian women with one foreign-born and one native-born parent, for example, are in interracial marriages versus a quarter of Asian women with two foreign-born parents. Asian women with one foreign-born and one native-born parent may be particularly likely to be intermarried because they are likely to be the daughters of Asian "war brides" and (white) American-born men who served in the military.
The difference in levels of intermarriage for second-generation women with two foreign-born parents versus those with one foreign-born and one native-born parent introduces complexities associated with "cross-generation" marriage.
The "straight-line" theory of assimilation presumes that marriage occurs within each generational cohort: immigrants marry other immigrants, members of the second generation marry other members of the second generation, and so on. This assumption is, however, simplistic.
Figure 3 shows the percentages of second-generation women of each racial/ethnic group with first-generation, second-generation, and third-generation husbands. Overall, relatively few second-generation women have second-generation husbands: the majority marry either first-generation or third-generation men.
However, there are striking differences across racial/ethnic groups. Over 75 percent of white second-generation women marry into the third generation. Black second-generation women are also very likely to marry into the third generation although some marry foreign-born men.
Hispanic second-generation women are fairly balanced with respect to marrying foreign-born, second-generation, or third-generation American men. Asian second-generation women are more apt than other women to marry someone of the same generational status as themselves.
The second generation thus appears poised between marrying either immigrants or third-generation Americans rather than other second-generation Americans. This pattern may be attributable to demographic constraints.
The adult second generation is still relatively small and so it is easier for white and black second-generation adults to marry into the very large third generation than to find prospective spouses among the smaller numbers of first- and second-generation white adults.
For second-generation Asians and Hispanics, the continuing high levels of immigration to the United States mean that the first generation is also fairly large.
The marriage behavior of the second generation lies betwixt and between that of the first and the third generations in two ways. Levels of racial and ethnic intermarriage increase substantially between the first and second generations for black, Asian, and Hispanic Americans, and increase again between the second and third generations for Asian and Hispanic Americans.
Even when the focus is narrowed to differences within the second generation, levels of intermarriage increase as the distance from the immigrant experience lengthens. Second-generation Americans with one foreign-born parent and one native-born parent (who were thus in a cross-generation marriage) are more likely to marry interracially than those who grew up with two immigrant parents.
The second generation is also poised between reaching back into the first generation or reaching forward into the third generation for spouses. Currently, only a minority of second-generation Americans are married to other second-generation Americans. This pattern probably reflects, in part, the relatively small size of the second generation, which is sandwiched between a very large number of third-and-later-generation Americans, and, for Hispanics and Asians, a growing number of immigrants.
The increases in rates of racial/ethnic intermarriage across generation, and the common pattern in which members of the second generation marry third-generation Americans, suggest that Asians and Hispanics are being quickly integrated into the larger American-born population.
Intermarriage is often considered to be one of the most important signs of assimilation and integration of immigrant-descent groups for several reasons. First, high levels of intermarriage demonstrate and accelerate the fading of cultural and social boundaries between immigrant descent groups and the larger American population. Second, high levels of intermarriage are also typically accompanied by growing similarities in the educational and labor force achievements of immigrant groups and the larger American population.
Gordon composed his theory of "straight-line" assimilation after observing marriage behavior among European-descent groups in the first part of the 20th century. The data presented here show that the two largest contemporary immigrant descent groups in the United States, Asians and Hispanics, are following the same generational patterns of intermarriage today.
Gordon, Milton M. 1964. Assimilation in American Life: The Role of Race, Religion, and National Origins. New York: Oxford University Press.