whose ethnic backgrounds do not overlap with their own at all. . . . Only one fifth have spouses with identical backgrounds."
Under such conditions of great intermarriage, ethnic identity is increasingly a matter of choice for whites in the United States. An American of Italian, Irish, and Scottish ancestry, for example, can "choose" to identify with one or more of his or her ethnic ancestries and discard or "forget" others (Waters, 1990; Alba, 1990). For example, over 40 million Americans reported Irish ancestry in the 1980 census, a figure far in excess of any reasonable rate of natural increase from the 4.5 million Irish who were immigrants to the United States. Hout and Goldstein (1994) show that the number of Americans who identify as Irish Americans is possible only because of the high rate of intermarriage of persons with Irish ancestry and a very selective identification of Irish identity among offspring with multiple ancestries.
This fluidity of white ethnic categories stands in contrast to the seeming essentiality of race. But this fluidity is partially the result of the primacy of racial issues in American history, which necessitated unambiguous classifications, first to identify discrimination and now to implement affirmative action. But the social and legal forces of racial identity in the United States are also idiosyncratic, the product of complex and contingent processes, and subject to change over the coming decades.
Rates of intermarriage have been growing since 1960 for all groups, even for those defined as "racial" groups (Sandefur and McKinnell, 1986; Kikumura and Kitano, 1973; Kitano et al., 1984; Gurak and Fitzpatrick, 1982; Lieberson and Waters, 1988). Although it is still the case that only a small proportion of marriages by whites are to nonwhites and Hispanics (2 percent), the rate of increase in recent decades has been dramatic: "In 1960 there were about 150,000 interracial couples in the United States. This number grew rapidly to more than 1.0 million in 1990. When marriages with Hispanics are added the intergroup marriages totaled about 1.6 million in 1990" (Harrison and Bennett, 1995:165).
As we noted in Chapter 3, although 97 percent of whites and 94 percent of blacks married within their own groups in 1990, 70 percent of Asians and 73 percent of Hispanics did so. The percentage of intermarriages increased between 1980 and 1990, as younger people married outside their group to a greater extent. Intermarriage rates were much higher for native-born Asians and Hispanics in 1990. Among younger married persons in 1990, aged 25 to 34 years, 65 percent of native-born Hispanics had a Hispanic spouse (see Table 8.2). Among younger Asians, 53 percent of men had an Asian wife and 46 percent of women had an Asian husband.
Intermarriage rates vary regionally, with lower intermarriage rates for Asians and Hispanics in areas where there is a heavier concentration of immigrants. Outside the South and the Southwest, younger native-born Hispanics had a non-Hispanic spouse more than half the time (Farley, 1996). Younger native-born Asians living outside California and the other Pacific states had a non-Asian