Race Relations in a Diversifying Nation

Citation:

Hochschild JL, Rogers R. Race Relations in a Diversifying Nation. In: New Directions: African Americans in a Diversifying Nation. edited by James Jackson. Washington D.C. National Planning Association ; 2000.

Full Text

Chapter 3: Race Relations in a Diversifying Nation

Jennifer Hochschild

Reuel Rogers

Department of Politics,

Princeton University

June 22, 1999

For New Directions: African Americans in a Diversifying Nation,

edited by James Jackson  (Washington D.C.: National Planning Ass’n., 2000)

NOTE: not quite final version.

“Diversifying” is indeed the key term for making sense of racial and ethnic relations in the United States over the next few decades. The most basic form of diversification in contemporary politics is that of racial or ethnic background and identification of Americans.  That divide is sometimes deepened, but at other times submerged into, other divisions – of class, gender, region, urbanicity, religion, ideology, and simply personal history and idiosyncrasy.  At this point in American history, one can plausibly predict at least three, mutually exclusive, ways in which diversity might develop over the next few decades:

• Pluralism: Conventional racial and ethnic identities are dissolving, at least around the edges, as groups come into close and complicated contact with one another.  Intermarriage, residential and job integration, cultural blends, and intergroup political alliances are all increasing.  Class divisions within a given racial or ethnic group, generational or nationality differences among recent immigrant groups, and ideological disagreements on particular policy issues are also working to break down the basic divide among European Americans, African Americans, Latinos, and Asian Americans.  Groups will not dissolve into an anodyne melting pot, but individuals will have considerable leeway in deciding how and how much group identity matters to their lives.
• Separation: Americans are moving toward mutual racial and ethnic separation, whether because groups choose to distinguish themselves from others or because they are separated from others despite their intentions or even their awareness.  Multicultural curricula and political programs sometimes emphasize differences and pride in distinctive identities more than shared interests or values.  Disputes over redistribution through governmental programs sometimes turn into zero-sum contests among groups.  The rapid growth of gated communities and the   provision of formerly-public services allow some Americans to exclude others from their daily lives.  Groups may not be overtly or even covertly hostile, but individuals’ life chances will be largely shaped by the nature of the racial or ethnic group into which they were born.
• Black exceptionalism: Anglos, Asian Americans and Latinos are slowly becoming a single, intermingled, population that will generally not include blacks.  The exclusion could be voluntary, or forced onto African Americans, or the result of an interaction between exclusion and separatist preferences.  Intermarriage and residential integration are increasingly dramatically among the first three groups, but not in the fourth.  Second- and third-generation nonblack immigrants are moving into the middle class, while African Americans who have been in the United States for dozens of generations remain trapped, physically and economically, in inner city ghettos and rural destitution.  Conservative white politicians garner support among newly enfranchised immigrants in our largest cities, but seldom even try to appeal to traditional black voters.  Middle class blacks and young black adults  increasingly endorse cultural forms of racial nationalism, even as they move into the economic mainstream.   In short, life as an African American will be qualitatively different – although not always worse – than life as a member of any other racial or ethnic group.

In this chapter we will not propose an overall conclusion as to whether pluralistic intermingling, ethnic separation, black nationalism, or white defensiveness will predominate over the next few decades.  Nor do we offer an overall judgment about the long-term possibilities for political coalition-building among racial and ethnic groups.  The structure of American racial and ethnic interactions is simply too complicated for summary judgments; the evidence is still too mixed – or even flatly contradictory.  Most importantly, too many outcomes are contingent on choices not yet made and circumstances not yet faced.  We will, however, offer more focused conclusions and draw some prescriptive lessons.[1]

Two main conclusions emerge from our analyses of survey data and coalitional efforts among racial and ethnic groups in several cities.[2] First, it will be very difficult to develop and maintain stable coalitions across many groups and many issues if people persist in the preferences and perspectives they now express.  The survey data show why stable coalitions across groups will be hard to maintain, and the case study evidence shows how hard they have in fact been to maintain over the past few decades.  Second, there is nevertheless much more opportunity to develop at least short-term coalitions than many observers of American politics now perceive.  The survey data suggest where the coalitional possibilities lie; the case studies suggest how political activists might turn those possibilities into reality.

Our prescriptive lessons emerge from those conclusions.  We start from the premise that the first of the three pathways outlined above – that of pluralism and porous boundaries among groups – is the most desirable.  Based on the evidence outlined below, we will seek to persuade readers that the best way to achieve intergroup pluralism is to develop coalitions around issues other than racial and ethnic concerns.  Fighting against discrimination will and should remain among the goals of a multiracial coalition, but it should probably not be at the forefront of such a coalition’s political agenda.

To put the point more aphoristically, both survey and case study evidence suggests that the more a multiracial coalition focuses directly on issues of racial and ethnic equality, the less stable it will be and the more likely it will be to fragment into competitive factions.  Conversely, the same evidence shows that the more a multiracial coalition focuses on issues that are not ostensibly about race, the greater its chance of persistence and success.  These results obtain because most of the time, issues of racial and ethnic equality come to have a zero-sum quality among people of color at least as often as between people of color and European Americans.  In contrast, issues that focus on economic needs, community improvement, or family policies (to pick a few examples) have at least the potential to benefit a wide range of people of all racial and ethnic identities, thereby generating more positive-sum games and cutting across racial and ethnic divides.

Diversification

In the 1980s, the United States absorbed 7.3 million immigrants, mostly from Asia and Latin America (at an annual rate of about 0.3 percent of the total population).  Between 1991 and 1996, another 6.1 million immigrants entered the country (at an annual rate of about 0.4 of the population).  An additional 5 million people live in the United States as undocumented immigrants ([U.S. Bureau of the Census, 1998 #486], tables 5, 10).   Assuming for the moment that ethnic and racial categories will remain fixed into the foreseeable future, Hispanics will replace African Americans as the single largest non-Anglo group in the United States by the end of the next decade or soon thereafter. As chapter 2 showed, by 2020 about one-third of the residents of the U.S. will be Asian, black, Latino, or Native American, up from about 15 percent in 1960.

Within each of those panethnic groups lie deep differences of language, religion, history, culture, and economic status.  Puerto Ricans usually speak English before they immigrate and are fairly familiar with customs and politics in the United States; other immigrants typically face much more dramatic cultural changes. People from Japan and China have been emigrating to the U.S. for 150 years; Samoans and Vietnamese have come only very recently.  Protestant Northern Europeans can expect to be welcomed; Moslem Middle Easterners should anticipate suspicion.

Geography interacts with demography to magnify diversity.  Over the next few decades, cities and regions across the United States will have increasingly different racial and ethnic profiles. In ten of California’s 58 counties, between one-third and two-thirds of the residents are Latino (mostly Mexican); in other counties the Latino population is as low as 3 percent ((Tolbert and Hero 1996): 808).  Most of the nation’s fastest-growing cities are in the West and Southwest, and their growth is attributable to immigration (Holmes 1998).  Currently, more than half of the residents of New York City are immigrants or children of immigrants ((Moss, Townsend et al. 1997)).  By 2025, in twelve states, non-Hispanic whites will comprise fewer than 60 percent of the residents; in another twelve states, they will comprise more than 85 percent of the population ((Frey and Farley 1996): 758).

In short, the United States is experiencing an unprecedentedly widespread “demographic balkanization,” in which

areas where immigrants account for most of the demographic change will become increasingly multicultural, younger, and more bifurcated in their race and class structures. Other parts of the country, whose growth is more dependent on internal migration flows, will become far less multicultural in their demographic makeup and will differ as well in other social, demographic, and political dimensions. What is new about this scenario is its geographic scope. ((Frey and Farley 1996): 742).

To what degree will demography and geography be destiny? In a majoritarian democracy like the United States, numbers matter greatly in any political calculus.  As the largest racial minority group in the country and by virtue of sheer numbers, African Americans historically have made stronger claims on the attention of policymakers and elected officials than any other non-Anglo group.  More particularly, blacks’ residential concentration in central cities has facilitated their political incorporation nationally and in some states.  As the number of Asian Americans and Latinos grows overall and especially in particular cities and states, the political standing of African Americans and Anglos will decline at least in relative terms, if the political game is played out in zero-sum terms. But if blacks and other non-Anglo groups are able to generate an alliance among people of color, and perhaps add a few liberal whites, then the lists of winners and losers could look very different.  In that case, we will see very different political dynamics in areas that remain predominantly white compared with areas that are demographically varied.

Now add a third form of diversification – economic – to diversification by race or ethnicity and location.  By conventional measures of income, education, or occupation, over a third of African Americans can now be described as middle class, as compared with over half of whites. That is an astonishing--probably historically unprecedented--change from the early 1960s, when blacks enjoyed the “perverse egalitarianism” of almost uniform poverty in which even the best-off blacks could seldom pass on their status to their children ((Hogan and Featherman 1978): 101). Today, there is greater disparity between the top and bottom fifths of African Americans, with regard to income, education, victimization by violence, job status, and participation in electoral politics, than between the top and bottom fifths of white Americans.  Consider, for example, the Gini ratio, a measure of income disparity between wealthiest and poorest; the closer it is to 1, the greater the inequality.  Among whites it rose from 0.419 to 0.446 between 1990 and 1996; in the thirty years since 1967, the Gini ratio has risen 0.055 points. Among blacks in the 1990s, the Gini ratio rose from a higher base to a higher top – from 0.464 to 0.479.  Inequality among blacks has risen almost as much (0.047 points) as has inequality among whites in the past three decades ([U.S., 1998a #1478]: table B-3).

Latinos are undergoing a similar process of class dispersion.  The Gini ratio rose from 0.424 in 1990 to 0.457 in 1996; since 1972 (the first year for which data are available), it has risen a dramatic 0.084 points ([U.S., 1998a #1478]: table B-3; see also (Institute 1997)). Comparable Census data are not available for Asian Americans, but several surveys show clearly the variations within Asian nationalities. In Houston, for example, half of Filipinos enjoy household incomes over $50,000 (in 1994 dollars), compared with only one-fifth of Vietnamese ((Klineberg 1996): 13). Across the United States, the median family income in 1989 of major Asian American groups ranged from$30,500 among Vietnamese to $51,500 among Japanese.[3] Poverty rates vary across the four main racial and ethnic groups, as well as within them. Throughout the 1990s, roughly 11 percent of whites were poor according to the Census Bureau; comparable rates for blacks averaged 31 percent, for Latinos 29 percent, and for Asian Americans, 14 percent ([U.S., 1998b #1481]: table C-1). Again, however, one must disaggregate; in the 1990s, for example, 30 percent of Samoans and Vietnamese were poor ((Frey and Fielding 1995); (Lien 1998): table 1-2; see also (Brackman and Erie 1995); (Wyly 1997): table 3). So perhaps not demography and geography, but economics, is destiny. That is, perhaps economic interests will swamp racial/ethnic identification or regional variations as the driving force behind intergroup relations over the next few decades. Common economic interests could, however, generate several distinct dynamics. Poor people might compete among themselves, perhaps along racial or ethnic lines, for scarce public resources and jobs ((Johnson and Oliver 1989); (Johnson, Farrell et al. 1996); (Miles 1992); (Sonenshein 1996); (Chen and Espiritu 1989)). Alternatively, they might unite to pursue common interests ((Mollenkopf 1997); (Sonenshein 1997a); (Schuck 1993); (Wilson 1999 forthcoming)). Well-off members of various groups similarly might unite around a shared economic conservatism ((Parent and Stekler 1985); (Welch and Foster 1987); (Tate 1994): 38-45;(Klineberg 1996): table 12), or split in contention over affirmative action, business set-asides, and simple competition over wealth and status. Yet other forms of diversity can, and do, come into play in particular circumstances. Religion, gender, age, recency or timing of immigration, and individual political ideology are among the long list of factors that can overcome racial, geographic, or economic imperatives in determining a person’s policy views and willingness to enter into political coalitions. For example, third generation Hispanics are not only economically better off than are earlier generations, but also they are more culturally liberal and economically conservative than more recent immigrants ((Klineberg 1996): 15-21; (Aguirre, Saenz et al. 1989)). Among some Hispanic groups, later generations and those who have lived in the United States longer resemble American blacks in their views of racial discrimination more than do recent immigrants ((Uhlaner 1991; Garcia, Falcon et al. 1996); (de la Garza, DeSipio et al. 1992)). We noted earlier that the only safe generalization one can make about “race relations in a diversifying nation” is that one should be excessively modest about making predictions. But we also proposed three paths that racial and ethnic relations might take over the next few decades. Let us turn now to a consideration of the ways in which the various types of diversity just described map onto those possible pathways of pluralism, mutual separation, and African American exceptionalism. . Agreement on Policy Preferences – Pluralism Results of some surveys suggests that the pluralist model best describes citizens’ political perceptions and views on crucial policy issues. We see little racial or ethnic disparity, for example, in judgments about the recent trajectories of American families: Table 1: Are the Problems of People Like You Getting Worse? “During the past ten years, has XX gotten better, worse, or stayed the same (OR gotten easier or harder) for people like you (OR families like yours)?” Percent saying “worse” or “harder,” ordered by whites’ level of concern  Whites N=802 Asian Ams. N=353 Latinos N=252 African Ams. N=474 To get good jobs 56 56 50 60 Public schools 55 47 45 57 To find decent, affordable housing 55 48 55 49 For families like yours to stay together 45 34 40 48 Health care 44 30 30 39 Source: (Washington Post 1995): 75-76 ________________ All groups rank their concerns in the same order, with the exception of one inversion by Latinos. African Americans and Hispanics are not overwhelmingly more distressed than members of the other two groups. Arguably, there is more of a vertical than a horizontal pattern – that is, more disparity in concern among the issues (health care is relatively unproblematic; job attainment is the most difficult) than across the groups. We see the same pattern -- greater disparity across issues than across racial or ethnic groups – in the way citizens rank their desires for governmental action on particular policy issues: Table 2: Policy Preferences for Congressional Action “For each issue, please tell me if you think this is something Congress should do or should not do” Percent saying “strongly feel Congress should do,” ordered by whites’ level of support for each action  Whites N=802 Asian Ams. N=353 Latinos N=252 African Ams. N=474 Reform the welfare system 83 68 81 73 Balance the budget 82 75 75 79 Reform medicare 53 58 59 58 Cut personal income taxes 52 46 55 50 Limited tax breaks for business 39 30 41 41 Limit affirmative action 38 27 30 25 Put more limits on abortion 35 24 50 32  Source: (Washington Post 1995): 73-74 The issue of abortion generates a huge disparity between two racial groups (Asian Americans and Latinos). The issues of welfare and affirmative action also reveal substantial differences among two or three of the groups. But overall, we see larger differences across the policy domains themselves than among racial or ethnic groups. Most respondents endorse reformation of the welfare system and balancing the budget; respondents are split on reformation of medicare and income tax cuts, and only a minority of respondents endorse business tax cuts, limits on affirmative action, and limits on the right to abortion. Furthermore, the preference ordering across the four racial and ethnic groups is, with occasional exceptions, the same. A final array, again from the 1995 survey, shows similarly that even when they consider policies to eliminate rather than to promote, members of various racial and ethnic groups still do not differ sharply from one another:  1 Table 3: Support for Spending Reductions “A number of spending reductions have been proposed in order to balance the federal budget and avoid raising taxes. Would you favor or oppose making major spending reductions in¼” Percent saying favor major spending reductions, ordered by whites’ preferences Whites N=802 Asian Ams. N=353 Latinos N=252 African Ams. N=474 Public assistance/ welfare 63 55 53 45 Federal aid for cities 55 49 45 41 Food stamps 52 49 46 38 Defense 51 63 57 56 Tax credits for low-income families 49 41 52 40 Public housing 49 41 49 37 Federal programs to create jobs and for job training 48 42 43 41 Legal aid for poor people 48 38 40 35 Federal aid for college loans 46 35 34 34 Medicaid 43 38 44 35 Medicare 39 44 34 33 Head Start 38 35 30 34 Social security 37 42 34 35 Child health 34 35 30 31 Source: (Washington Post 1995): 8 We see greater disparities in this set of preferences than in the previous set, with differences of eight to fourteen percentage points across at least two groups on most issues. (The only issue that generates huge disparities is welfare.) Whites are more inclined to cut spending than at least two of the other groups on all issues except defense, and blacks are almost always less inclined to cut spending than all of the other groups. These results, of course, accord with the general finding that blacks are the most liberal group and whites the most conservative. But even in this case, we find more variation between the top and the bottom of the list within each group than between the extreme positions on any single issue (except welfare) across groups. Furthermore, with a few exceptions (notably defense), respondents in all groups give roughly the same preference ordering for policies that they would be willing to sacrifice. This is not a portrait of a citizenry radically divided along racial and ethnic lines. (see also [Joint Center, 1997a #39]: table 4a).[4] If we focus more narrowly on specific policy preferences for solving particular problems, we again see considerable convergence across racial and ethnic groups. In 1996, for example, all four groups agreed that “education” was the “most important issue facing your community today,” although varying degrees of urgency about other issues produced differences in the actual proportions which focused on education. Roughly speaking, all groups agreed that drugs, crime, and gang violence were the next most important set of issues.[5] Perhaps more surprisingly, the groups mostly concur on how to solve the problems of crime and schooling. Consider crime policy first: Between 80 and 90 percent of black, white, and “other” Americans agree that it is “extremely important” to spend tax dollars on “reducing crime” and “reducing illegal drug use” among youth. Since 1982, nonwhites have consistently been slightly more likely to agree that our country spends too little money on “halting the rising crime rate” and “dealing with drug addiction” ([U.S. Department of Justice, 1997 #82]: 141, 142-3, 144-45). Three quarters of both blacks and whites agree that the penalties for powdered and crack cocaine should be the same; 70 percent of blacks and 80 percent of whites support “3 strikes and you’re out” laws ([Joint Center, 1996 #38]: tables A-5 and A-6). Blacks, whites, and Hispanics all concur that “a juvenile charged with a serious property crime should be tried as an adult” (about 60 percent agreement), that “a juvenile charged with selling illegal drugs should be tried as an adult” (about two-thirds agreement), and that “a juvenile charged with a serious violent crime should be tried as an adult” (over 80 percent agreement) ([U.S. Department of Justice, 1997 #82]: 155). Among eight proposals offered them on one survey, blacks, whites, and Hispanics all identified 3 strikes, “money for youth programs,” and “adding 100,000 more police officers” as the most effective ways to reduce crime ([U.S. Department of Justice, 1996 #81]: 167). Only on two issues -- the hope that prison can rehabilitate prisoners, and especially on support for the death penalty -- do nonwhites differ dramatically from whites ([Joint Center, 1996 #38]: table A-7; [U.S. Department of Justice, 1997 #82]: 155, 159-165). Finally, consider more briefly attitudes toward educational reforms. Over 80 percent of blacks, whites, and Latinos concur that it is “extremely important” to spend tax dollars on “educational opportunities for children” ([U.S. Department of Justice, 1997 #82]: 141). Both African Americans and whites are split in their views of vouchers (Latinos evince the most enthusiasm for them) ([Joint Center, 1996 #38]: table B-3; [Joint Center, 1997b #40]: table 7). Blacks, whites, and Latinos are almost equally supportive of a constitutional amendment to allow prayer in schools ([Joint Center, 1996 #38]: table B-4; (Post-Modernity 1996): tables 10C, 47E). We could continue with other specific policy proposals but the point should by now be clear: on important groups of social and political problems, people of all racial and ethnic groups concur on what the problems are, how serious they are, and what at least some solutions to those problems should be. Where there is not concurrence across the population on the solutions – e.g. with school vouchers -- the groups frequently resemble one another in the proportion supporting or opposing the given proposal. Thus divisions over many policy proposals lie along fault lines other than race or ethnicity, such as political ideology, class, or urbanicity. Here are encouraging grounds for developing coalitions across racial and ethnic lines. We should expect Americans to continue to dispute vigorously over policy proposals. But if responses to an array of surveys can be trusted, we have evidence that many of those disputes will not fall along racial and ethnic lines. Thus the first of the three models described in the introduction – that of pluralistic intermingling -- has some support: conventional racial and ethnic identities do not determine the anxieties and policy preferences of many, perhaps most, American citizens on many issues of deep importance to them.[6] Disputes in Perceptions of Racial Discrimination – Racial/Ethnic Separation However, the pluralist model is only one potential path for “race relations in a diversifying nation.” Consider next the second plausible path – racial and ethnic separation and separatism. That possibility is revealed in responses to a query about relative degrees of discriminatory treatment: Table 4: Which group faces the most discrimination? “Among these groups, which one do you think faces the most discrimination in America today?” Percent choosing each group, ordered by whites’ responses  White R’s N=802 Asian Am. R’s N=252 Hispanic R’s N=252 African Am. R’s N=474 African Americans 51 62 42 82 Hispanic Americans 28 18 40 11 Asian Americans 14 12 11 5 Don’t know 6 8 7 2 Source: (Washington Post 1995): 93 _______________ African Americans think that members of their own group suffer from the most discrimination by a very wide margin; Latinos think members of their own group suffer just as much. All groups, including Asian Americans, agree that Asian Americans do not win that dubious honor. (See (National Conference 1994), pp. 87, 103, 119, 137; (Klineberg 1996): table 14; [Los Angeles Times Poll #319, 1993 #1488]: ques. 19 for similar results.) Here are the seeds of coalitional trouble, at least between blacks and Hispanics. More detailed probing of arenas in which discrimination might occur produce the same pattern, with a few crucial additions: ________________ Table 5: Which groups face discrimination in key arenas of life, compared with whites? Percent choosing each group  White R’s (N=1093) Asian-Am. R’s Latino R’s (N=77) Black R’s (N=102) Credit loans and mortgages Blacks 31 -- 31 69 Latinos 31 -- 28 60 Asian Americans 24 -- 31 33 Promotion to management positions Blacks 44 -- 40 72 Latinos 46 -- 48 66 Asian Americans 36 -- 33 39 White R’s (N=1093) Asian-Am. R’s (N=154) Latino R’s (N=502) Black R’s (N=1006) Treatment by police Blacks 50 60 61 81 Latinos 46 54 54 72 Asian Americans 35 53 50 48 Portrayal in the media Blacks 45 37 46 70 Latinos 45 50 44 65 Asian Americans 40 56 43 44 Treatment in the legal system Blacks 37 37 47 71 Latinos 36 35 43 68 Asian Americans 35 19 35 45 Source: (National Conference 1994): 73-75, 134, 136 ________________ Overall, African Americans perceive much higher levels of discrimination against non-Anglos than do the other groups, and they always believe that their own race faces the most discrimination. In addition, they always perceive more mistreatment of Hispanics than Hispanics do themselves; they are rather less concerned about Asian Americans. Latinos perceive essentially the same amount of discrimination against themselves and African Americans, and they see almost as much bias against Asian Americans. Asian Americans themselves perceive their own race to suffer the most discrimination in one arena, similar levels as the other two groups in a second, and least in a third. Whites generally see the least bias, and they make the smallest distinctions among the groups (see also (Zubrinsky and Bobo 1996): 353-354; [Los Angeles Times Poll #319, 1993 #1488]: ques. 24-27; [Los Angeles Times Poll #395, 1997 #1489]: ques. 29-33). In short, we see a clear racial/ethnic ranking of how much discrimination there is, but no clear racial/ethnic ranking of who suffers from it the most. In this context, any group’s claim to unusually bad treatment will be subject to contest by the others – either because they deny the fact of bad treatment, or because they deny that the first group suffers unusually from bad treatment. Results of another survey question seeking comparisons across groups underlines the tendency toward racial/ethnic separation or separatism. This item is less cognitive than affective; it speaks to self-image or aspirations for the future, rather than to particular policy or political preferences.[7] _______________  Table 6: Toward whom do you feel affinity, and aversion? “Of these groups, if you had to say, which one do you feel you have the most in common with/ least in common with?” Percent choosing each group Respondents: Feel most in common with: White N=1093 Black N=1006 Latino N=502 Asian American N=154 Whites -- 34 55 50 African Americans 38 -- 25 12 Latinos 28 45 -- 27 Asian Americans 19 19 6 -- DK/NA 15 2 14 11 Respondents: Feel least in common with: White Black Latino Asian American Whites -- 36 21 21 African Americans 24 -- 36 53 Latinos 24 19 -- 13 Asian Americans 36 36 32 -- DK/NA 16 9 11 13 Source: (National Conference 1994): 32 ________________ Whites feel most in common with blacks, who feel little in common with whites. Blacks feel most in common with Latinos, who feel least in common with them. Latinos feel most in common with whites, who feel little in common with them. Asian Americans feel most in common with whites, who feel least in common with them. Each group is chasing another which is running from it. What these results imply for political coalitions depends on how one conceives the dynamic of coalition-building. On the one hand, it is possible to generate, and may be possible to sustain, a coalition among people who feel little affinity for one another if their instrumental or ideological motivations are strong enough. That is the insight behind the cliché, “politics makes strange bedfellows,” and there are plenty of examples of the phenomenon (that, after all, is what makes the phrase a cliché). From that perspective, these results may be psychologically interesting but they say little about potential political alliances to elect new officials or pass new legislation. On the other hand, it is surely hard to generate, and especially hard to sustain, coalitions among people who feel little in common beyond a purely instrumental or ideological goal. As we demonstrate below, coalitions do not develop fully enough to have significant impact over time unless participants are willing to take risks, are able to develop trust over time, and sense that they are fighting for some larger and shared purpose. The results of this survey question suggest that those elusive but essential emotional ties will be, to put it mildly, hard to develop across racial and ethnic groups in the foreseeable future. Disputes in Perceptions of Racial Discrimination – Black Exceptionalism It is distressingly easy to demonstrate how far apart African Americans and whites are on issues that explicitly evoke racial identity. Consider disparities in what might appear to be a simple description of fact, such as “How much racial discrimination exists in the United States?” Seldom do more than a third of whites believe that blacks continue to experience racist treatment in jobs, housing, the media, or the criminal justice system ((Gallup Organization 1997); (National Conference 1994); (Washington Post 1995); (Zukin 1997): chaps. 3, 4; (Schuman, Steeh et al. 1997): 158-159, 166-169). Blacks disagree. Twice as many blacks as whites agree that there is “a lot” of “discrimination and prejudice¼ against blacks in the U.S. today,” and more than twice as many claim that blacks are treated “not well” or “badly” in their community ([Joint Center, 1997a #39]: tables 1, 4, 6; (Gallup Organization 1997): ques. 18; (National Conference 1994); (Washington Post 1995); (Zukin 1997): chaps. 3, 4; (Schuman, Steeh et al. 1997): 260-265 ). We see the same racial disparities when people consider change over time rather than the current level of racial discrimination. Whites are increasingly convinced that racial equality is growing in the United States. In the mid‑1960s, a third or more (depending on the year and the wording of the question) felt that the nation was making progress in solving its racial problems; by the 1970s, over half concurred, and by 1988, almost 90 percent of whites believed that “in the past 25 years, the country has moved closer to equal opportunity among the races.” By the mid-1990s, whites had again become cautious. On one survey, only a third (compared with 12 percent of African Americans) asserted that discrimination against blacks had decreased over the previous ten years ([Hochschild, 1995 #30]: esp. chap. 3; (Washington Post 1995): ques. 41a; (Post-Modernity 1996): table 38). Nevertheless, whites are consistently more likely to see progress than regression in racial equality. African Americans, meanwhile, are becoming increasingly discouraged on the issue of change over time. The proportion of blacks who see increasing racial equality declined from between 50 and 80 percent in the mid‑1960s to between 20 and 45 percent in the 1980s.[8] In 1995, over half (compared with less than one-quarter of whites) agreed that discrimination has worsened for their race over the previous decade; two years later, only one in five, compared with twice as many whites, agreed that “the situation of black people is better [than it was] five years ago.” ([Hochschild, 1995 #30]: chap. 3; (Washington Post 1995): ques. 41a; Joint Center 1997a: table 7; (Post-Modernity 1996): table 4J). One example demonstrates most clearly this striking divergence. From 1988 to 1993 (when David Dinkins was mayor), an increasing proportion of black New Yorkers felt that race relations in their city were good, while a decreasing proportion of white New Yorkers concurred. In 1994, however (when Rudolph Guiliani became mayor), a majority of whites who perceived any recent change thought New York's race relations had improved recently; a majority of blacks who perceived change thought relations had worsened. Thus no matter what direction opinions about race relations move, blacks and whites follow opposite trajectories. Racial divergence, not merely growing black pessimism, is the underlying phenomenon ([Hochschild, 1995 #30]: 61). Blacks and whites disagree just as much on perceptions of how well blacks are doing as on perceptions of discrimination. About three in five whites believe that African Americans are as well off or better off than whites with regard to their jobs, access to health care, and education; more than two in five whites say the same with regard to income and housing. Again, African Americans disagree; only 14 to 30 percent accept the claim that blacks are as well off as whites, depending on the particular arena in question. Barely three in ten whites, compared with seven in ten blacks, doubt that African Americans have as good a chance to live a middle class life as do whites ((Washington Post 1995); see also (Hochschild 1995): 61-64; (Zukin 1997): chaps. 2, 4).[9] Ironically, one of the few areas of perceptual agreement between the races shows both to be equally mistaken: both groups agree that blacks comprise one-fourth of the American population ((Washington Post 1995): sec. 1; blacks actually comprise about 12 percent of Americans). The model of black exceptionalism, however, includes the proposition that African Americans are distinct from Hispanics and Asian Americans as well as from whites. The evidence on this broader point is more mixed. Latinos typically fall between blacks and whites on questions of race relations, whereas Asian Americans frequently resemble whites (so far as we can tell from very scanty survey data).[10] For example, three in ten African Americans, over four in ten Latinos, and six in ten whites describe race relations in their community as excellent ([Joint Center, 1997a #39]: table 1; see also (Zukin 1997): chap. 3). Few European Americans claim to face discrimination; many African Americans do; Latinos and Asian Americans fall in between with Latinos closer to blacks and Asians closer to Anglos. ((Washington Post 1995): ques. 4; (Burns, Schlozman et al. 1998): fig. 13; (Klineberg 1996): table 14; [Los Angeles Times Poll #319, 1993 #1488]: ques. 22). Over twice as many black as white teenagers (46 percent versus 22 percent) claim to have been “a target of some racial or religious incident;” Latinos fall exactly in between (at 32 percent) ((Louis Harris and Associates 1990): table 47; see also (Post-Modernity 1996): table 83B). Latinos are sometimes less convinced than African Americans that their own group faces discrimination: in one survey, 80 percent of blacks, 70 percent of Latinos, and 40 percent of whites agree that “there is not enough attention paid to discrimination against Hispanics” ([Joint Center, 1997a #39]: table 5; see also (Washington Post 1995): ques. 11b; (Zukin 1997): chap. 3; (Uhlaner 1991); (Louis Harris and Associates 1990): 34). Perhaps the most important indicator of black exceptionalism is the fact that African Americans attach more political significance to their racial identity than do other groups. More precisely, perceived racial interests strongly shape blacks’ political evaluations and choices, whereas other groups either see fewer racial/ethnic interests, or do not use them to shape their policy values so strongly. (On blacks, see (Dawson 1994b); (Kinder and Sanders 1996); (Tate 1994); (Bobo and Smith 1994). On Hispanics, see (Jones-Correa and Leal 1996); (Uhlaner 1991)). Thus blacks may be especially prone to avoid coalitions with groups that seem not to take their distinctive perceptions, preferences, and circumstances fully into account. Shifting Agreements across Policy Domains So far we have evidence for pluralism, for group separation, and for African American ex exceptionalism. But these results reveal more than indeterminacy; they show that the less a survey question draws attention to race or ethnicity, and the less it asks people to measure their group against another, the less people split along group lines. That makes sense analytically, but it leaves open a wide array of political possibilities: under what circumstances is it possible --or desirable -- to reduce citizens’ attention to their racial or ethnic context in order to increase their attention to something else? A final pattern of possibilities for coalition, and contestation, will lead us one step further toward answers to that question. Some issues are likely to pit two or three of the four groups against the other(s). Here we examine two, but others could be added.[11] Affirmative Action: Latinos support strong forms of affirmative action more than do whites, but usually less than do blacks. The evidence on Asian Americans is more volatile, both because sample sizes are usually too small for reliability and because Asian Americans are deeply conflicted about affirmative action. ([Onishi, 1996 #59]). For example in 1995, whites were more likely to agree strongly than were blacks, Asian Americans, and Latinos that Congress should “limit affirmative action” (see table 2 above). But on other affirmative action questions in the same survey, Asian Americans and Latinos always located themselves between Anglos and blacks, with Asian Americans closer to whites and Latinos closer to blacks ((Washington Post 1995); see also [Joint Center, 1997a #39]: table 9; (Zukin 1997): tables 4-5, 4-6; (Post-Modernity 1996): table 10Q; (Klineberg 1996): table 14; (Bobo 1999 forthcoming)).[12] To summarize a huge array of data: blacks are much more likely to support affirmative action for blacks than are all of the other groups; blacks and Latinos concur on strong though lesser support for affirmative action for Latinos; and all groups concur on relatively weak support for affirmative action for Asian Americans ([National Conference, 1994 #56: 86, 102, 118, 136; (Burns, Schlozman et al. 1998): figure 12; (Citrin 1996): table 1). Latinos and whites favor affirmative action for the poor rather than for “people of a specific race or sex” slightly more than do the other two groups ([Washington Post, 1995 #86]: ques. 61). From a political actor’s perspective, the best evidence on support for affirmative action comes from exit polls after voting on Proposition 209 in November 1996. The San Francisco Examiner, the Los Angeles Times, and Voter News Service all conducted exit polls for Proposition 209, with similar results: roughly 60 percent of whites, 26 percent of blacks, 30 percent of Latinos, and 45 percent of Asian Americans favored the proposition to abolish the public use of affirmative action ([Ness, 1996 #58]: A1; [Los Angeles Times, 1996 #16]: A29; (Citrin, Cain et al. 1997a): table 1). That suggests greater support from Latino voters and less support from Asian voters for affirmative action than do surveys of the population as a whole. In an unsuccessful referendum to eliminate affirmative action in Houston, Texas a year later, about three-fifths of white voters supported abolition, compared with fewer than 10 percent of African American and about a quarter of Latino voters [Mason, 1997 #1492]. Thus a coalition among African Americans, Latinos and occasionally Asian Americans in favor of at least soft forms of affirmative action is plausible but not rock solid. Blacks care much more about the issue than do the other two groups, blacks support its strongest forms more than do the other two groups, and blacks are more committed to affirmative action for blacks than are the other two groups. Evidence from particular disputes over affirmative action shows both mutual support among two and sometimes all three of the non-Anglo groups, and fierce zero-sum competition, depending on the location and particular circumstances. Immigrants and Immigration: Preferences with regard to immigration policy and views of immigrants show a different pattern. Although the data are again too sparse to be definitive, especially for Asian Americans, Latinos and probably Asian Americans are much more sympathetic to immigrants and much more supportive of policies to encourage more immigration than are African Americans and whites. Within that conclusion lies great variation – across surveys, over time, in different regions and cities, across different nationality groups within a given ethnicity – which allows for various types of political mobilization. But the pattern suggests that an affirmative action coalition would break down rather quickly if its members shifted their attention to immigration. A survey in Los Angeles shows the pattern clearly. Latinos and Asian Americans are consistently more likely to give the response sympathetic to immigrants than are whites and African Americans _____________ Table 7: Residents of Los Angeles’ Preferred Immigration Policy, by Race/Ethnicity, 1992 High scores indicate attitudes more favorable to immigration  Anglos African Americans Latinos Asian Americans Amnesty for undocumented residents 42 44 80 55 Additional spending to prevent illegal immigration 19 29 26 24 Require recent immigrant business owners to hire Am. citizens (NOTE: scoring is reversed) 29 14 13 18 Reduce level of legal immigration 21 22 30 26 Policy Index Mean 2.62 2.53 2.88 2.74 SOURCE: (Bobo and Hutchings 1994): table 1 _______________ Latinos are the most liberal on immigration policy, followed by Asian Americans. African Americans and non-Hispanic whites are least supportive of policies to encourage immigration and help immigrants. Other national and regional surveys show roughly the same pattern: African Americans and whites are much less enthusiastic about immigrants and more restrictive in their preferences for immigration policy than are Latinos ([Los Angeles Times Poll #306, 1993 #1493]: ques. 136, 137, 142, 143; [Los Angeles Times Poll #319, 1993 #1488]: ques. 67; (Fetzer 1995): 12-14; [Joint Center, 1997a #39]: table 6; (Post-Modernity 1996): table 40A); GSS 1994: ques. 514, 518A.[13] Surveys that include Asian Americans show a split on most, but not all, questions between Latinos and Asian Americans on the one hand and blacks and whites on the other, with regard to immigrants and immigration policy ((Washington Post 1995); (Klineberg 1996): 32-33; (Uhlaner 1991): 360-65). A recent national survey suggests why African Americans and Anglos resemble each other, and do not resemble Hispanics and Asian Americans, in their preferences for immigration policy. On every one of the seven opportunities they were offered, Hispanics found the effects of immigration on their community to be much more favorable than did the other two groups: Table 8: Effects of Recent Immigrants on Your Community, by Race or Ethnicity, 1997 percent responding “better”  Whites N = 800 African Americans N=229 Hispanics N= 220 overall quality of life in your city or town 11 15 35 quality of local public schools 8 9 34 local crime situation 4 2 32 local job opportunities for you and family 7 9 31 local food, music, arts, culture 30 28 52 local economy in general 20 21 39 local politics and government 8 11 34 Response alternatives were “worse,” “not much effect,” and “don’t know.” SOURCE: (Princeton Survey Research Associates 1997) In 1995, over half of the black residents of Houston, Texas declared the impact of immigrants in their community to be “bad” or “very bad,” and only a third judged it to be “good” or “very good” ((Rodriguez 1996): 118). In a survey at about the same time in Los Angeles, over half of African Americans and whites agreed that if immigration into the United States continues at the present rate, members of their group would lose political influence and economic opportunity. Not surprisingly, only a fifth of Latinos and a tenth of Asian Americans expressed the same fear. Blacks were especially fearful of a direct zero-sum loss to immigrants ((Johnson, Farrell et al. 1996): tables 1-3; see also [Los Angeles Times Poll #306, 1993 #1493]: ques. 138-140). Concern about the health of the economy, alienation from the society and polity, economic and political isolationism, fear of economic competition and symbolic racism all contribute additionally to anxiety about immigrants and immigration ((Espenshade and Hempstead 1996); (Vidanage and Sears 1995); (Citrin, Green et al. 1997b)). ­­­­­­­­­­­­­­­­­­­­­ As with affirmative action, support for a ballot proposition is an illuminating supplement to survey data. In 1994, Californians voted on Proposition 187, which proposed to prohibit the use of public services such as schooling and health care by illegal immigrants. The proposition passed by a three to two margin. Averaging across exit poll data, we find that about 64 percent of Anglos, 52 percent of blacks and Asian Americans, and only 27 percent of Latinos supported it ((Tolbert and Hero 1996)817; [Los Angeles Times, 1994 #46]; [Migration, 1994 #50]; (Campbell and Wong 1998)).[14] Comparing survey responses with these voting results shows that Asian Americans voters are less sympathetic to illegal immigrants than are Asian Americans overall, and that voters in the other three groups roughly reflect the views of all members of those groups. Finally, support for policies that indirectly address immigration shows the same pattern as support for immigration policies directly, thus widening the arena in which the “natural” alliance of African Americans and Latinos, or of people of color more broadly, is likely to be strained or broken. African Americans and whites are much more likely than Latinos to agree that “being able to speak and understand English [is] an absolutely essential obligation of Americans,” and twice as many blacks and whites as Latinos strongly favor making English the official language of the Unites States ((Post-Modernity 1996): tables 18D, 10S; (Fetzer 1995): 23); see also GSS 1994, ques. 512, 513. Surveys show that Latinos favor bilingual education programs in schools much more than do whites and somewhat more than do African Americans ([Los Angeles Times Poll #306, 1993 #1493]: ques. 134; (Yankelovich Partners Inc. 1995): ques. 22; [General Social Survey, 1994 #1600]: ques. 524). Proposition 227, a measure to almost eliminate bilingual education in public schools, was approved in June 1998 by over three-fifths of California’s voters. The results show once again that Asian American voters are more conservative, or more assimilated, than are Asian Americans in general. But the rest of the votes follow roughly the pattern we would expect from the survey data: two-thirds of whites, 48 percent of blacks, and 37 percent of Latinos supported the measure (as did 57 percent of Asian Americans) ([Los Angeles Times, 1998 #1496). In short, we should not expect a “rainbow coalition” of support for immigration policy any more than we should expect one for affirmative action policy. Surveys suggest that such a coalition is within the realm of possibility and might in some circumstances be joined by well-educated and well-off whites. But it would be difficult to sustain, especially if issues get framed in a way that entails losses for some proportional to gains for others. Once again, we see dramatic fluidity in coalitional possibilities. Thus judging by survey data, a clear prediction of victory by one of our three initial models is not warranted, for two reasons. First, the data themselves yield too many complicated or simply inconsistent stories, for methodological or substantive reasons or both. Second, it is up to political actors who have not yet acted to determine which of the various possible groupings will in fact coalesce, and whether issues that evoke antagonism or agreement will predominate. Race matters -- but politics matters too. Coalitional Politics on the Ground Surveys suggest possible coalitions; political activity creates or destroys real ones. Coalitions can be built on different bases, and probably need to combine several in order to thrive. Those bases include economic interests, a commitment to shared values, leaders willing to take risks to reach beyond their core supporters, interpersonal trust, and a conviction that the long-term benefits of a coalition will outweigh its short-term costs ((Sonenshein 1993; Sonenshein 1997a)). Consider interests first. African Americans are the most liberal group on social and economic policy issues, seeking more governmental involvement and higher public expenditures on employment, health care, aid to the poor, education, and aid for urban infrastructures. That is an uncontested, familiar finding (see Table 3, and (Dawson 1994b); (Tate 1994); (Hamilton and Hamilton 1997)). Latinos similarly favor “increased government spending on health and crime and drug control; education; the environment; child services; and bilingual education. They also overwhelmingly look to government to solve the problems that most concern them . . . [and they] are notably more liberal than [whites]” ((de la Garza, DeSipio et al. 1992): 14). Some whites and Asian Americans also support redistributive policies. Can this policy consensus be translated into a sustained coalition, so that common interests override conflict over scarce resources? To interests, add values: African Americans have historically been the group most committed to vigorous public efforts to attain racial equality and to show respect for distinctive group identities. That too is an uncontested and familiar finding. Most Latinos and some whites and Asian Americans share these values. So the same question arises: can this ideological commitment be translated into a sustained coalition, so that shared values override conflict over scarce status and differing views on how to put these values into practice? Over the past few decades, leaders skilled at bringing together disparate and mistrustful groups have been in short supply. Mayor Tom Bradley of Los Angeles succeeded within his city; the Reverend Jesse Jackson has made valiant efforts to develop a “rainbow coalition” across the nation; and Mayor David Dinkins spoke eloquently of the “gorgeous mosaic” of New Yorkers. The Congressional Black Caucus and the Congressional Hispanic Caucus have overcome deep disagreements to work closely together, even on issues that mattered more to one than to the other of their constituencies. Nevertheless, moderate or conservative white politicians have recently proven far more adept at appealing to nonwhite immigrants. In New York, Los Angeles, Chicago, Gary, Philadelphia and other cities, white conservatives have won enough support from Latino (and other) voters to replace or defeat black liberal mayors ([Sleeper, 1993 #69]; (Mollenkopf 1997); (Sonenshein 1997a); (Pinderhughes 1997)). This pattern of defeat despite shared interests and values is partly explicable in terms of the other essential bases of coalition-building – the willingness to take risks, the need for interpersonal trust, and sufficient political strength to accept imminent losses for the sake of eventual gains. Less abstractly, the current eclipse of black-led coalitions results partly from the distinctive history of earlier black political successes. Blacks began to win urban elections, even where they remained a minority, when they both mobilized internally and cultivated linkages with liberal whites. Latinos and Asian Americans were included in these coalitions only in subsidiary positions ((Browning et al. 1990a)). But the standard biracial formula – implying a black-white dichotomy and an axis of inclusion/exclusion -- produces four problems for would-be black leaders in the new environment. First, it “posits the racial collectivity as its central agency and racial discrimination as its main analytical category” ((Reed 1988)). While still useful in many political contexts, that approach can deflect attention from new problems and policy concerns that African Americans share with newcomers. Second, it obscures intraracial stratification and thereby assumes allegiances within the black population – or within other populations -- that may not exist. Third and conversely, the standard formula focuses too heavily on interracial differences and thereby misses the opportunity to build new alliances with new political actors ((Skerry 1993): e.g. 10, 370; (Jennings 1992): 44). Finally, the standard formula puts too much weight on winning positions from which blacks have historically been excluded. This insistence on descriptive representation comes at the expense of other political and policy goals that can be shared across groups. The biracial formula worked well in the 1970s and 1980s. But now, as whites are leaving central cities and Hispanic and Asian immigrants are taking their place, blacks who remain find that the terms of the political equation have changed. Whites still comprise a plurality if not a majority of active voters and will remain disproportionately influential. Traditional white redistributive liberalism is on the decline, however, giving way to concerns – such as environmentalism, anxieties about welfare dependency, and crime control -- far less sympathetic to minority group claims. Thus even the whites who do remain in cities are less available for traditional black-liberal white alliances ((Kaus 1992); (Sleeper 1997)). And nonwhite constituencies in central cities will increasingly insist on more complex, multiracial coalitions.[15] In short, “the traditional way of conceptualizing racial-ethnic issues and strategies along a Black-White dichotomy will not be effective in mobilizing Puerto Ricans and other Latinos,” as well as Asian Americans and sympathetic whites. However, all is not lost: “while Puerto Ricans [and other nonwhites] do not respond neatly to this bipolar racial classification, the notion of being a nonwhite group is a strong one and, on this basis, there appear to be many areas of public policy consensus between large majorities of Puerto Ricans and Blacks” ((Falcon 1995): 204). The question is whether African Americans, other people of color, and like-minded whites can surmount their partly outdated assumptions, mutual mistrust, and substantive disagreements enough to pursue their shared material and philosophical goals. New York and Los Angeles are good cases for considering this question. Both cities have a substantial African American population that achieved considerable political incorporation over the past few decades. Both cities boast increasingly heterogeneous populations, due mostly to immigration from Asia and Latin America. In both cities, African American political leaders have recently lost power to relatively conservative whites. If we can explain these cases, we should have a good handle on the issue of when multiracial coalitions are -- and are not – feasible. Multiracial Political Dynamics in New York Large numbers of immigrants have always flocked to New York and continue to do so. Until the mid-1960s, most migrants came from Europe, and most had arrived by the 1920s. Since 1965, however, the city has absorbed unprecedented waves of Caribbean, Asian, and Latino immigrants, so that, in combination with African Americans, nonwhites comprise 57 percent of the population.[16] By sheer numbers alone then, blacks, Latinos, and Asian Americans in New York have the makings of a strong minority coalition. But numbers do not suffice to accomplish such an enterprise. To begin with, voters differ from residents; there are twice as many white as black registered voters, and almost four times as many white as Latino voters. In addition, the panethnic categories such as “Latino” and “Asian” obscure a politically salient fragmentation. New York’s Hispanics divide into Puerto Ricans, Dominicans, Columbians, Cubans, and others. Black New Yorkers include native-born African Americans and a growing foreign-born population of Afro-Caribbean immigrants. Asian Americans include at least half a dozen nationalities – Chinese, Filipino, Korean and Vietnamese immigrants from East Asia, and Indian, Bangladeshi, and Pakistani immigrants from South Asia -- whose native languages are mutually unintelligible. Nevertheless, New York’s political culture and institutions display features which should encourage multiracial alliances. First, the city’s blacks and Latinos (specifically Puerto Ricans) have a long history of political mobilization, at least since the Depression.[17] Second, after years of underrepresentation in city government, the two groups have begun to approach legislative proportionality. “By 1992, blacks and Latinos had come to hold 21 of 53 seats on the city council, six of the city’s 14 congressional seats, six of 25 state senate seats, and 24 of the 60 assembly seats” ((Mollenkopf 1997): 99). Third, although whites far outnumber blacks and Latinos in the general electorate, the disparity narrows considerably among registered Democrats. “[B]lacks are actually better represented among registered Democrats than in the voting-age population as a whole; Latinos are less underrepresented among Democrats than among registered voters as a whole” ((Mollenkopf 1994): 86). African Americans have attained significant leadership positions within the Democratic party (which still controls local patronage and nomination slates), and the two groups have formed a joint caucus in the state Assembly. The relatively low level of neighborhood-based conflict between blacks and new immigrants should also enhance the prospect for minority coalitions. New York has its share of interminority tensions, but they seldom reach the level of hostility seen in cities such as Los Angeles and Miami. (On Los Angeles, see (Baldassare 1994); (Gooding-Williams 1993); [Sonenshein, 1996 #72; (Davis 1998)]. On Miami, see (Grenier and Stepick 1992); (Portes and Stepick 1993). On New York, see (Mollenkopf 1995); (Mollenkopf 1994); (Waldinger 1996)). Conflicts between blacks and Korean merchants, for example, have spurred protests, vitriolic rhetoric, and scattered, individual cases of violence – but not large-scale rioting or mass violence. Labor market competition between blacks and new immigrants in New York has also been relatively subdued. With the important exception of the government sector and social services, African Americans and nonwhite immigrants largely concentrate in different sectors of the city’s postindustrial labor market ((Waldinger 1996); (Bailey and Waldinger 1991)). This division of labor may help to reduce tensions between African Americans and new immigrants if, as some argue, newcomers are taking jobs that native-born blacks do not want.[18] Finally, many black New Yorkers and new immigrants share material interests that should provide the basis for an issue-oriented coalition. Until recently, they were largely excluded from the economic benefits of the post-industrial transformation that the city’s economy has undergone. Thus poverty rates among New York’s blacks and Latinos increased sharply in the 1980s despite the city’s growing prosperity ((Falcon 1988): 174). In addition, the most rapidly growing and remunerative sectors of the post-industrial economy have mostly excluded blacks and other nonwhite groups. Increased marginality in a thriving economy brings other shared ills: higher rents in a tighter housing market, “very high unemployment rates, ¼ extremely high dropout rates from the schools, and intense residential displacement due to factors such as disinvestment and gentrification” ((Falcon 1988): 174).[19] How have New Yorkers fared in building multiracial coalitions, given this mix of shared interests, relatively sustained political involvement, and demographic disparities? In a word, poorly. Electoral coalitions between blacks and other nonwhite groups have been infrequent, and those that have formed have proven ephemeral and susceptible to disruption. Between 1965 and 1973, blacks joined with Hispanics and liberal whites to support Mayor John Lindsay’s progressive administration. During his two terms, Lindsay set spending and service distribution priorities that favored the city’s black and Latino populations. He also experimented with new forms of minority political incorporation such as offices of neighborhood government ((Mollenkopf 1997)). The experiment, however, failed to take institutional root and lost ground in the subsequent conservative mayoral administrations headed by Abraham Beame and Edward Koch. In 1985, the effort to create a progressive coalition took a different form. Nonwhite politicians sought to “close ranks” and mount an independent minority-based political initiative. About forty black political activists and elected officials convened the Coalition for a Just New York to select a candidate for mayor. Although Hispanic congressman Herman Badillo aggressively courted the group and advocated issue positions in keeping with the group’s agenda, several leading black politicians were openly averse to endorsing a non-black candidate and the Coalition passed on the opportunity. Not surprisingly, this open denial exacerbated tensions between the black and Latino communities. The Coalition ultimately settled on Assemblyman Denny Farrell, who, by most accounts, was a weak choice and ran a deplorably ineffective campaign. In this case, not only did the black rejection of Badillo’s overtures throw away an opportunity to “go along to get along”–essential to all coalitions ((Stone 1989)) -- but it also substituted a weaker for a stronger candidate. Ironically, Badillo’s policy positions would have benefited African Americans as well as Hispanics, certainly more than Ed Koch’s, the eventual winner, did. But black leaders were unable to relinquish the implicit claim that blacks historically have made to the most visible positions of leadership and to the right to define the policy agenda in minority-based coalitions. In our analytic terms, they lacked sufficient trust and sufficient political strength to absorb short-term costs in order to take advantage of the interests and values they shared with New York’s Hispanics. In 1989 blacks joined with Hispanics to elect David Dinkins as the city’s first African American mayor. This venture seemed even more conducive to a stable coalition among people of color than the Lindsay experiment since Dinkins’ electoral and governing coalitions were clearly organized and guided by minorities (along with some liberal whites). Dinkins campaigned not only by seeking, successfully, to win virtually all African American votes, but also by making direct, purposeful appeals to Latinos, Afro-Caribbeans, and other nonwhite constituencies. This strategy marked an important departure from the conventional black-white electoral logic. Under Dinkins’ administration, the number of black and Latino government officials increased, and relations between the police and the city’s nonwhite residents improved modestly. But as in the 1970s, the coalition proved fragile. Many Hispanic leaders charged that they were denied an equitable share of political rewards, and insisted that the Mayor was far more interested in courting and solidifying white support than rewarding his nonwhite supporters ((Thompson 1996)). Several black politicians echoed the latter complaint. Once again, the most severe conflict centered on group visibility and descriptive representation. Blacks in New York have achieved relatively greater access to political office than have other nonwhite groups, so they could afford to shift to “post-access” concerns for themselves and allow access concessions to others. After all, “the significantly less access [sic] that Latinos have than blacks or whites to political and public bureaucracy representation . . . would indicate that this would be a more salient issue to Latinos . . . [They] seem to give greater emphasis to questions of basic group visibility, including symbolic recognition” ((Falcon 1988): 180). Nonetheless, Latino political leaders and the Latino press charged that Dinkins’ African American appointees to the 1990 city council districting commission favored blacks over Latinos in allocating council seats. As they saw it, “although Latinos and African Americans are roughly the same proportion of the city population, the commission created twelve so-called ‘winnable’ seats for African American candidates but only eight Latino winnable seats” ((Thompson 1996): 77). Dinkins pointed out plausibly that this lopsided distribution was due largely to the geographic dispersion, low voter registration, and comparative youth of the city’s Latino population. Nevertheless, the Mayor’s black appointees would have been well advised to yield a few more seats to Latinos to help build reciprocity and solidify their loyalty to the coalition. Instead, “the split with Latinos revealed that parts of the Latino leadership were so displeased that they might try to break the African American/Latino alliance that helped elect Dinkins [in 1989].” ((Thompson 1996)). That is just what happened in 1993, as conservative Republican candidate Rudolph Giuliani defeated Dinkins after one term. Latino voters, who had provided a critical bloc of support in Dinkins’ 1989 victory, turned out in far fewer numbers in 1993 and almost half shifted their allegiance to Giuliani (Mollenkopf 1994: Afterword). Their demobilization and defection, along with the loss of some liberal white supporters, helped to produce Dinkins’ defeat. Mayor Giuliani has subsequently proven adept at preserving his new coalition by addressing, if only symbolically, issues of great interest to the city’s new residents such as immigration reform. He also appointed Hispanics to his administration (including, ironically, Herman Badillo, selected to run for city comptroller on Giuliani’s ticket). He has simultaneously pursued fiscal and social policies that have had a disproportionately adverse impact on the poor – that is, predominantly black, Latino, and other nonwhite populations. Room remains, therefore, for New York’s liberal and/or nonwhite leaders to fashion alliances to pursue a progressive agenda. But several changes must occur for such an alliance to have a hope of success. African American political leaders must speak out on issues such as immigration, deportation, and access to welfare services that particularly concern the new nonwhite residents of the city.[20] Coalitions must be carefully focused on discrete economic and social interests that concern all of these groups – such as expanded economic opportunity, improvements in public education, and better relations with the police -- rather than on the single, short-term goal of winning elections. All groups must work hard to avoid the zero-sum competition that arises in debates over who faces the most discrimination, who should win the most offices, or who deserves the most governmental aid. There are plenty of common interests and many shared values among poor and working class New Yorkers of all backgrounds; if leaders can keep the focus on them and take a few personal and political risks, a coalition with greater stability could emerge. But that outcome is far from certain. Multiracial Political Dynamics in Los Angeles African American political elites in other cities with high levels of immigration such as Los Angeles and Chicago face the same challenges as in New York. Circumstances and contexts differ, however. In Los Angeles, Latinos (predominantly Mexican Americans) outnumber blacks by almost three to one. The proportion of Asian Americans is small but rapidly growing, and Asian Americans are making significant political inroads into old coalitions.[21] In Chicago, blacks still outnumber Latinos (mostly Mexicans, Dominicans, and other Central and South Americans) two to one, and Asian Americans are relatively unimportant politically. However, in both cities the proportions of African Americans and whites are declining, and will continue to do so at a dramatic pace. Forging coalitions with nonwhite immigrant-based groups is thus increasingly urgent for traditional black and white political leaders. Until recently, African Americans in Chicago and Los Angeles enjoyed much higher levels of political incorporation than did blacks in New York. They also enjoyed greater success in generating alliances with other nonwhites, particularly Hispanics. In Los Angeles, former Mayor Tom Bradley, an African American, led an effective if sometimes uneasy coalition of blacks, Latinos, Asian Americans, and liberal whites from 1973 to 1993. Chicago’s blacks joined with Latinos and liberal whites to elect Harold Washington, also an African American, to the office of mayor in 1983 and again in 1987. But in recent years, these political coalitions have stumbled on many of the same obstacles that have blocked alliances among people of color in New York. Blacks and Latinos in Los Angeles have clashed over redistricting and descriptive representation, while in Chicago, relations deteriorated over the same issues after Harold Washington’s death. Neighborhood-level conflicts among racial or ethnic groups have convulsed both cities. Nevertheless, previous coalitional successes in both cities offer important lessons on how to forge constructive political alliances. The Los Angeles case is arguably the most illuminating because the most successful, and will be our focus here. Consider first the successes: During the 1980 redistricting debates in the state Assembly, the African American Speaker, Willie Brown, wisely appointed Hispanic assemblyman Richard Alatorre from East Los Angeles to chair the committee. After much debate, the committee selected a plan that gave the city’s Hispanic constituency one assembly and two congressional seats, and cost African Americans none ((Skerry 1993): 85). During the 1985 and 1986 councilmanic reapportionment debates, Mayor Bradley similarly created a Latino district and vetoed a plan that would have eliminated the city’s only Asian American district ((Sonenshein 1997a): 46). Finally, in the 1995, African American and Latino political leaders resolved months of rancorous debate by ensuring that neither constituency would lose legislative seats through redistricting.[22] In each case, Los Angeles’ black political leaders were able to move beyond the zero-sum perspective and were even able to use the reapportionment decisions to build reciprocity with Asian and Latino constituencies. But these successes could not prevent tears in the coalitional fabric. By the late 1980s, Latino and Asian American political leaders had begun to chafe at their subordinate position in the Bradley coalition. Hispanics in particular have vacillated between remaining a part of the multiracial Bradley coalition and seeking to “go it alone” by relying on their own growing numbers in the general population ((Sonenshein 1997a): 60). Organizing Latinos into an independent unified political bloc is no easy task in light of their nationality, citizenship, and socioeconomic differences.[23] But for many Latino political leaders, the prospect of remaining junior partners in a multiracial coalition led by blacks or whites is even less palatable. This strategic uncertainty makes for an unpredictable Latino constituency perhaps increasingly prone to conflict with other groups. In the 1993 mayoral election, Latinos split their votes between the two candidates – one of whom was the clear heir to the progressive, multiracial Bradley dynasty -- more than did either blacks or Asian Americans ((Sonenshein 1993)). Partly as a consequence, a conservative, white-led coalition won. The rest of the Bradley coalition largely held together in the 1993 election. Michael Woo, an Asian American city council member, won decisively among liberal whites and attained almost 70 and 86 percent of Asian American and African American votes respectively (although turnout among blacks was somewhat depressed compared with the Bradley years). Woo’s showing is all the more impressive because it came only one year after the South Central battles between Asian Americans and blacks. African American support for Woo contrasts starkly with the refusal of New York’s black political leaders to support Herman Badillo in his 1985 mayoral bid.[24] Mayor Richard Riordon won reelection decisively in 1997, however; it remains to be seen whether what remains of Bradley’s progressive, multiracial coalition can revive enough to defeat Riordan’s coalition of moderates and conservatives.[25] Paralleling successes and failures in the politics of redistricting and voting have been a set of alliances and conflicts revolving around substantive policy concerns. Coalition members have real interests in common; poor and working class Angelenos have all been affected by structural phenomena such as the movement of jobs to the suburbs and overseas, a repressive police force, and discriminatory practices of banks and realtors ((Sonenshein 1993); (Davis 1990)). Middle class Angelenos shared a concern about racial and ethnic discrimination, both for its own sake and because discriminatory practices would inhibit economic development – a goal they also shared. These shared interests were crucial in creating, and maintaining the Bradley coalition over many years and elections. But New Yorkers also shared problems and goals and have been able to coalesce only intermittently. What was different about Los Angeles? The key point, especially in comparison with New York and Chicago, is that a history of compromise on policy issues of mutual concern – which started more or less accidentally -- has made the next compromise a bit easier to create. In the late 1960s, all of the eventual coalition members were shut out from policymaking influence and political incorporation by the conservative and aggressively racialistic regime of Mayor Sam Yorty. Yorty was “a relentless foe of minorities and progressives, [often] using racist appeals to defeat . . . challenges” to his regime (Sonenshein 1997a): 44). That situation provided the impetus for African Americans, liberal whites, and to a lesser extent Asian Americans and Latinos to unite around their shared interests in political incorporation and antidiscrimination.[26] Once the coalition took root, its own prior existence helped to keep it going. To assert that longevity is a critical dimension of most successful governing coalitions is more than tautological; historical memory, practice in negotiation, personal loyalties, a habit of reciprocity, and trust among leaders are all linchpins in the maintenance of political coalitions ((Stone 1989); (Hinckley 1981)). The interests of the various coalitional partners evolved over time, and middle class African Americans and liberal whites enjoyed a greater share of the rewards of incorporation than others; nevertheless, the alliance was sustained for twenty years largely because the groups avoided zero-sum conflicts and focused on sites of interest convergence wherever they could be found. Thus the well-founded reputation that “black and [Hispanic] elected officials enjoy extremely good relations” ((Skerry 1993): 84-85); even the damage done to Korean businesses during the South Central civil unrest precipitated a new wave of dialogue between black and Asian leaders ((Regalado 1994); (Sonenshein 1996)). Michael Woo’s 1993 campaign showed the legacy of this search for convergence. His policy platform was highly congenial to the interests of poor and working class blacks, and he included traditional antidiscrimination planks of great concern to middle class blacks and white liberals. He prominently opposed the city’s controversial police chief Darryl Gates and worked tirelessly to bring about rapprochement between African Americans and Korean Americans in the aftermath of the South Central disaster. Woo “deserves major credit for brokering some agreements between these highly opposed groups . . . . [He] helped push through city hall funding for the conversion of some [Korean-owned] liquor stores to laundromats.” In fact, he “nearly recreated a successful version of the Bradley coalition in very rough times” ((Sonenshein 1993): 302; see also (Park 1996): 161-62). But history does not always proceed in a straight line, and time can exacerbate tensions as well as provide resources for overcoming them. Even before the 1992 unrest, one could sense that the coalition was in danger of dissolving, largely because key coalition members’ interests and ideologies were changing. After two decades, middle class whites began to object to Mayor Bradley’s aggressive pro-growth policies and stances on crime and police accountability. Poor and working class blacks and Hispanics also became increasingly frustrated with a pro-development agenda from which very few benefits trickled down to them. However, these disaffected constituents were split among themselves over immigration and class polarization, and could not unite either within or outside of the old Bradley coalition. Relations between African Americans and immigrants became especially tense, playing out in neighborhood conflicts over jobs,[27], housing,[28] and immigrant entrepreneurial activity ((Johnson and Oliver 1989); (Johnson, Farrell et al. 1996)). For example, in the wake of the 1992 disturbance, a few African American community organizers called for building contractors to replace Latino laborers with blacks. ((Sonenshein 1993): 298; (Skerry 1993): XX). This substitution of zero-sum proposals for efforts to find common solutions to the common problem of underemployment provides little encouragement to those seeking a new progressive coalition. If the Bradley coalition is not yet in complete disarray, its future is doubtful. It will take an unlikely mix of interests, ideology, and political entrepreneurialism to put it back together. Despite the growth of a new array of multiracial progressive organizations, “the Establishment Coalition and its policies have come to dominate the agenda of the city” ((Park 1996)). Conclusion Los Angeles holds important lessons for African Americans seeking to make common cause with people of color and liberal whites in other cities. The Bradley coalition mobilized around two unifying interests with deep ideological resonance: antidiscrimination reform and political inclusion. Antiracist political reforms carried unimpeachable moral force and ringing social relevance, and political incorporation was an urgent goal for a variety of Angelenos. Today, political actors must identify equally compelling sites of convergence in order to build new political capital and more complex coalitions. But the sites have shifted and strategies must be updated. As the survey data show, antidiscrimination goals are no longer the most stable arena for making common cause across racial and ethnic groups. Affirmative action and business set-asides are weighed down with symbolic meaning, some of which is negative to potential allies ((Hochschild 1998)). In addition, the groups disagree on the degree to which they and others suffer from discrimination and therefore disagree on the urgency of measures to end it. The pursuit of inclusion among decision-makers may be equally unpromising as an arena for building alliances, unless blacks can refrain from guarding their political gains against encroachment. And both the survey data and the experience in New York suggest how difficult it will be for African Americans to set aside their (arguably correct) conviction of uniquely harmful circumstances in order to make room for new immigrants. At this writing, conservatives have shown more skill in finding an alternative unifying formulation than have liberals. “In the Establishment Coalition, the aggressive recruitment and the inclusion of racial minorities can be seen as the defining difference between the conservative coalitions forged by Riordan and the previous conservative coalitions in Los Angeles. This might be precisely why Riordan’s coalition will prove to be more successful and durable than its predecessors” ((Park 1996): 165). We see no reason to modify this comment much for New York, Chicago, Philadelphia, or elsewhere. Nevertheless, most blacks do share common interests and values with many Hispanics, some Asian Americans, and some whites that could provide the basis for new coalitions of the left. For one thing, political incorporation remains incomplete; the history of Los Angeles shows that it is possible to redistrict in a way that does not tear a progressive coalition apart, and the survey data show at least some concern on the part of each group for the political status of the others. In addition, many blacks remain poor, and their interests coincide with those of many nonwhite immigrants and poor whites. And for once, the survey data reinforce the more analytic argument about interests since they show considerable agreement on the need for redistributive policies. So the traditional grounds for interracial alliances have not disappeared, even if they have not shown much vigor recently. If economic interests or other shared goals are to be the basis of new progressive coalitions, our evidence indicates that at least four conditions must obtain: • Where possible, racial issues should not be the center of discussion and action; the focus should instead be on shared substantive policy and political goals such as jobs in the primary sector, better schooling, nonbrutal crime control, neighborhood development, immigrant incorporation, and decent housing. • When attention to race is deemed desirable or essential, everything possible should be done to avoid zero-sum conflicts over processes (such as redistricting and the selection of candidates) and outcomes (such as affirmative action or the funding of particular programs). • African Americans need to pay more attention to civil rights issues of concern to immigrants such as welfare rights, deportation, and immigration restrictions. They also need to recognize that other groups have the same intense desire for descriptive representation that they themselves have evinced. • Latinos, Asian Americans, and sympathetic whites, in return, need to accept that blacks’ history of enslavement and their continued suffering from poverty and racial discrimination are qualitatively different from the history of all voluntary immigrant groups, and perhaps require distinct treatment as a consequence. We are not calling for African Americans to suppress concerns about racial issues. Nor should they abandon their pursuit of remedies for discrimination, especially for poor blacks who have few resources of their own with which to fight on any issue. We are arguing that a focus on ending racial discrimination per se is not the best staging ground for seeking alliances with other peoples of color or potentially sympathetic whites. Instead, African Americans would do well to consider that engaging other groups in support of antidiscrimination goals will most likely occur in the pursuit of policy goals that are not ostensibly about race. Consider, for instance, African Americans’ concerns about the potentially adverse impact of recent welfare reform policies on poor black communities. These concerns resemble immigrants’ anxieties about the broadly harmful effects of California’s Proposition 187 on all immigrants, not just on illegal entrants to the United States.[29] Similarly, some provisions of the 1996 welfare reforms directly targeted poor immigrants, and the new policy as a whole has left the poorest Hispanics in the same dangerous condition as the poorest blacks.[30] Immigrant advocacy groups thus might see as much racial or ethnic bias in the new welfare law as do black advocacy groups ((Lieberman 1998)). Shared anxieties about bias in public service provision provide a good site for generating shared demands to reduce discrimination – but only by first emphasizing the mutual desire to maintain social services and supports for the poor of whatever race. One can make a similar case for the issue of police accountability. African Americans have complained of police brutality and harassment for decades, and immigrants from Latin America and the Caribbean have recently joined them. The cases in New York involving the torture of Abner Louima, a Haitian immigrant, and the killing of Ahmed Diallo, a newcomer from Guinea, are the most dramatic, but not the only, incidents generating this concern. How to control crime without violating individuals’ rights is a question that could bring together people of various racial and ethnic groups, including whites, in support of antidiscrimination measures. But as with welfare, the initial policy focus should not be race, per se, but rather a shared interest in the delivery of a crucial and delicate public service in central cities. Improving public education, ensuring health care, finding homes for children without stable families, providing jobs once the economy slows– all of these issues have an obvious racial dimension, but need not be approached through a framework of fighting racial discrimination. To the degree that African American leaders can find ways to bring others into a coalition that focuses on the problems, rather than on the identities of the coalition members, to that degree they need not choose between fighting bias and finding allies. All of these things are easier said than done, and the prescription we just gave has historically proven almost impossible to sustain for long periods of time. Analytically, we are proposing that politicians and policy actors seek to create the first of our three models of diversity – that is, to promote interactive pluralism rather than group separation or black exceptionalism. Pluralism need not, and probably should not, imply colorblindness or unrooted individualism, as it has done in the past. It does imply that groups will come together around particular issues and perhaps separate into new groupings for other issues; the key point is that the content of issue disputes, not ethnicity or race, will determine the coalitional patterns. We retain some optimism that, despite our history, it is possible to develop class-based coalitions around particular policy issues because American society is developing several unprecedented features. One is the growth of a black middle class that is moving away from poorer blacks, physically and politically. That phenomenon so far has produced declining intraracial solidarity on issues of economic redistribution, and greater intraracial solidarity on issues of discrimination and bias. Whether the growing African American middle class will be more swayed by its increasing economic conservatism or its increasing racial nationalism – or both, or neither – is a question for coalition builders of the left and the right to ponder. In our analytic terms, black exceptionalism seems to be growing culturally and politically, while declining economically and socially ((Hochschild 1999 forthcoming)). A second new phenomenon is the growing Latino population that is neither “white” nor “black” by conventional understandings, and that therefore can act as an intermediary between the two traditional antagonists. That population is becoming politically important as more Latinos become naturalized citizens and begin to engage in conventional politics.[31] It is also becoming more internally diverse, as some Latinos join the middle class and others emigrate from desperately poor communities in Mexico and Central or South America into the poorest ranks of our cities. Most Latinos seek to become upwardly-mobile joiners of mainstream society (Saad 1995), but some see their group as an oppressed racial minority. Some reject conventional means to if not conventional goals of success, while other “Latino politicians are becoming uncomfortable with old-style minority politics”[Rodriguez, 1999 #1601]. How that complexity will translate into mobilization into left or right coalitions remains uncertain.[32] Asian Americans show some of the same political mobilization and ideological diversity as Latinos.[33] They are on balance more like whites in their economic trajectory and political views, but their commitment to descriptive political representation and anxiety about racial bias make them potential members of at least some progressive coalitions. They are the most libertarian of the four groups, which makes them sometimes available for mobilization on the left (e.g. for the retention of the right to abortions), and sometimes on the right (e.g. for opposition to governmental regulation of business). Thus, in our terms, groups may not be moving toward political or economic separation, but they retain distinctive profiles that must be considered in any effort to create alliances. Perhaps even the white population is becoming more internally varied; the 1997 Joint Center surveys show greater disparity across age groups than across races on many controversial policy issues. Some whites, too, suffer when jobs leave the city, schools deteriorate, and neighborhoods crumble. They have, of course, less need to enhance descriptive representation, but whites have shown themselves to be staunch members of progressive coalitions when their interests are sufficiently guarded and their ideological commitments engaged. Hence, the possibilities are ripe for multiracial coalition building among across races, ethnicities, classes, and locations. African American political leaders will fully awaken to this potential, however, only with some reorientation of black political discourse. They must eschew the standard biracial formulation that has traditionally marked black politics in American cities and that has predisposed blacks to ally primarily, if not exclusively, with liberal whites. Other groups in the electorate can no longer be ignored. Nor can the fact that there are other bases for coalition building besides antiracism and inclusion. If progressives do not take heed, conservatives will continue to do so. Bibliography Aguirre, B. E., R. Saenz, et al. (1989). “Discrmination and the Assimilation and Ethnic Competition Perspectives.” Social Science Quarterly 70(3): 594-606. Bailey, T. and R. Waldinger (1991). The Changing Ethnic/Racial Division of Labor, in Dual City: The Restructuring of New York. J. Mollenkopf and M. Castells, eds. New York, Russell Sage Foundation: 43-78. Baldassare, M., Ed. (1994). The Los Angeles Riots: Lessons for the Urban Future. Boulder, Westview Press. Bell, D. (1992b). Race, Racism, and American Law. Boston, Little, Brown and Company. Bobo, L. (1999 forthcoming). Race and Beliefs about Affirmative Action, in Racialized Politics: Values, Ideology, and Prejudice in American Public Opinion, D. Sears, J. Sidanius and L. Bobo, eds. Chicago IL, University of Chicago Press. Bobo, L. and V. Hutchings (1994). Race, the Sense of Group Position, and Immigration Politics. New York, Annual Meeting, American Political Science Association. Bobo, L. and R. A. Smith (1994). Antipoverty Policy, Affirmative Action, and Racial Attitudes, in Confronting Poverty: Prescriptions for Change. S. Danziger, G. Sandefur and D. Weinberg, eds. New York, Russell Sage Foundation and Harvard University Press: 365-395. Brackman, H. and S. Erie (1995). Beyond 'Politics by Other Means'?: Empowerment Strategies for Los Angeles Asian Pacific Community, in The Bubbling Cauldron: Race, Ethnicity, and the Urban Crisis. M. P. Smith and J. Feagin, eds. Minneapolis, University of Minnesota Press. Browning, R., D. R. Marshall, et al. (1990a). Minority Mobilization in Ten Cities: Failures and Successes, in Racial Politics in American Cities. R. Browning, D. R. Marshall and D. Tabb, eds. New York, Longman: 8-30. Burns, N., K. L. Schlozman, et al. (1998). Active Intersection: Gender, Race or Ethnicity, and Participation. Chicago IL, Annual meeting of the Midwest Political Science Association. Campbell, A. and C. Wong (1998). 'Racial Threat' and Direct Democracy: Contextual Effects in Two California Initiatives. Chicago IL, Annual meeting of the Midwest Political Science Association. Chen, L. and Y. Espiritu (1989). “Korean Businesses in Black and Hispanic Neighborhoods: A Study of Intergroup Relations.” Sociological Review 32: 521-534. Citrin, J. (1996). “The Angry White Male is a Straw Man.” Public Affairs Report (Univ. of California, Berkeley) 1: 10-13. Citrin, J., B. Cain, et al. (1997a). Ethnic Context, Group Conflict, and Racial Attitudes: California Voters on Affirmative Action and Immigration. Washington, DC, Annual Meeting, American Political Science Association. Citrin, J., D. Green, et al. (1997b). “Public Opinion Toward Immigration Reform: The Role of Economic Motivations.” Journal of Politics 59(3): 858-851. Davis, M. (1990). City of Quartz: Excavating the Future in Los Angeles. London, Verso. Davis, M. (1998). Ecology of Fear: Los Angeles and the Imagination of Disaster. XXX, Metropolitan Books. Dawson, M. (1994b). Behind the Mule: Race and Class in African-American Politics. Princeton, Princeton University Press. de la Garza, R., L. DeSipio, et al. (1992). Latino Voices: Mexican, Puerto Rican, and Cuban Perspectives. Boulder, Westview Press. deParle, J. (1998). "Shrinking Welfare Rolls Leave Record High Share of Minorities". New York Times. New York: A1, A12. Espenshade, T. and K. Hempstead (1996). “Contemporary American Attitudes Toward U.S. Immigration.” International Migration Review 30(2): 535-570. Falcon, A. (1988). Black and Latino Politics in New York City, in Latinos and the Political System. F. C. Garcia, ed. Notre Dame IN, University of Notre Dame Press. Falcon, A. (1995). Puerto Ricans and the Politics of Racial Identity. Racial and Ethnic Identity: Psychological Development and Creative Expression. H. Harris, H. Blue and E. Griffith. New York, Routledge. Fetzer, J. (1995). The Causes of Immigration-Related Attitudes in the United States. Yale University, Political Science Department, unpublished paper. Frey, W. and R. Farley (1996). “Latino, Asian, and Black Segregation in U.S. Metropolitan Areas: Are Multiethnic Metros Different?” Demography 33: 35-50. Frey, W. and E. Fielding (1995). “Changing Urban Populations: Regional Restructuring, Racial Polarization, and Poverty Concentration.” Cityscapes: A Journal of Policy Development and Research 1: 1-65. Gallup Organization (1997). Black/White Relations in the United States. Princeton NJ, Gallup Organization. Gallup Organization (1997). The Gallup Poll Social Audit on Black/White Relations in the United States. Princeton, NJ, Gallup Organization. Garcia, F. C., A. Falcon, et al. (1996). “Ethnicity and Politics: Evidence from the Latino National Political Survey.” Hispanic Journal of Behavioral Sciences 18: 91-103. General Social Survey (1994). . p. i. James Davis and Thomas Smith. Chicago IL, National Opinion Research Corporation. Gooding-Williams, R., Ed. (1993). Reading Rodney King/Reading Urban Uprising. New York, Routledge. Grenier, G. and A. Stepick, Eds. (1992). Miami Now!: Immigration, Ethnicity, and Social Change. Gainesville, University Press of Florida. Hamilton, D. and C. Hamilton (1997). The Dual Agenda: Race and Social Welfare Policies of Civil Rights Organizations. New York, Columbia University Press. Hinckley, B. (1981). Coalitions and Politics. New York, Harcourt Brace Jovanovich. Hochschild, J. (1995). Facing Up to the American Dream: Race, Class, and the Soul of the Nation. Princeton, Princeton University Press. Hochschild, J. (1998). Affirmative Action as Culture War. The Cultural Territories of Race: White and Black Boundaries. M. Lamont. Chicago, Universitiy of Chicago Press. Hogan, D. and D. Featherman (1978). “Racial Stratification and Socioeconomic Change in the American North and South.” American Journal of Sociology 83(1): 100-126. Holmes, S. (1998). Immigration Is Fueling Cities' Strong Growth, Data Show. New York Times. New York: A10. Institute, o. P. P. (1997). The Emerging Latino Middle Class. Los Angeles, CA, Pepperdine University. Jennings, J. (1992). The Politics of Black Empowerment: The Transformation of Black Activism in Urban America. Detroit, Wayne State University Press. Johnson, J. J., W. J. Farrell, et al. (1996). Immigration Reform and the Browning of America: Tensions, Conflict, and Community Instability. Social Science Research Council Committee on International Migration, "America Becoming/Becoming American", Sanibel Island, Florida. Johnson, J. J. and M. Oliver (1989). “Interethnic Conflict in Urban America: The Effects of Economic and Social Dislocations.” Urban Geography 10: 449-463. Joint Center, f. P. a. E. S. (1996). 1996 National Opinion Poll - Social Attitudes. Washington, DC, Joint Center for Political and Economic Studies. Joint Center, f. P. a. E. S. (1997a). 1997 National Opinion Poll - Politics. Washington, DC, Joint Center for Political and Economic Studies. Joint Center, f. P. a. E. S. (1997b). 1996 National Opinion Poll - Children's Issues. Washington, DC, Joint Center for Political and Economic Studies. Jones-Correa, m. (1998). Between Two Nations: The Political Predicament of Latinos in New York City. Ithaca, NY, Cornell University Press. Jones-Correa, M. and D. Leal (1996). “Becoming 'Hispanic': Secondary Panethnic Identification Among Latin American-Origin Populations in the United States.” Hispanic Journal of Behavioral Sciences 18(2): 214-254. Kaus, M. (1992). The End of Equality. New York, Basic Books. Kinder, D. and L. Sanders (1996). Divided by Color: Racial Politics and Democratic Ideals. Chicago, University of Chicago Press. Kirschenman, J. and K. Neckerman (1991). We'd Love to Hire Them, But...: The Meaning of Race for Employers. The Urban Underclass. C. Jencks and P. Peterson. Washington, DC, Brookings Institution: 203-232. Klineberg, S. (1996). Houston's Ethnic Communities. Houston TX, Rice University. Lieberman, R. (1998). Shifting the Color Line: Race and the American Welfare State. Cambridge MA, Harvard University Press. Lien, P.-t. (1997). “Does Under-Participation Matter? An Examination of Policy Opinions Among Asian Americans.” Asian American Policy Review 7: 38-54. Lien, P.-t. (1998). Asian Americans and Political Participation: A Book Proposal. Philadelphia, PA, Temple University Press. Los Angeles Times (1996). Elections '96: State Propositions: A Snapshot of Voters. Los Angeles Times. Los Angeles CA: A29. Louis Harris and Associates (1990). The Reebok-Northeastern Study of Youth Attitudes on Racism. New York NY, Louis Harris and Associates. Lutrin, C. and A. Settle (1996). California's Immigration Initiative in the 1994 General Election. Annual Meeting, American Political Science Association, San Francisco, CA. Mason, J. (1997). Voters Keep Affirmative Action Program Alive. Houston Chronicle. Houston TX: A1ff. Migrant News (1998). Elections, Population, Education, migrant@primal.ucdavis.edu. Miles, J. (1992). Blacks vs. Browns: The Struggles for the Bottom Rung. Atlantic Monthly: 41-64. Mollenkopf, J. (1994). A Phoenix in the Ashes: The Rise and Fall of the Koch Coalition in New York City Politics. Princeton NJ, Princeton University Press. Mollenkopf, J. (1995). Urban Political Conflicts: The Case of New York. Social Science Research Council Committee on International Migration, "America Becoming/Becoming American". Mollenkopf, J. (1997). New York: The Great Anomaly. Racial Politics in American Cities. R. Browning, D. R. Marshall and D. Tabb. New York, Longman: 97-15. Moss, M., A. Townsend, et al. (1997). Immigration is Transforming New York city. New York, New York University, Taub Urban Research Center. National Conference, o. C. a. J. (1994). Taking America's Pulse: The Full Report of the National Conference Survey on Inter-Group Relations. New York, National Conference of Christians and Jews. Ness, C. and A. Nakao (1996). Opponents File Suit to Block Prop. 209; Voters Back Ban on Affirmative Action by 55-45 Percent. San Francisco Examiner. San Francisco, CA: A1ff. Onishi, N. (1996). Affirmative Action: Choosing Sides. New York Times, Education Life. New York: 26-35. Parent, T. W. and P. Stekler (1985). “The Political Implications for Economic Stratification in the Black Community.” Western Political Quarterly 38(4): 521-538. Park, E. (1996). Our L.A.? Korean Americans in Los Angeles After the Civil Unrest. Rethinking Los Angeles. M. Dear, H. E. Schockman and G. Hise. Thousand Oaks, Sage: 153-168. Patterson, O. (1989). “Toward a Study of Black America.” Dissent(fall): 476-486. Pinderhughes, D. (1997). An Examination of Chicago Politics for Evidence of Political Incorporation and Representation. Racial Politics in American Cities. R. Browning, D. R. Marshall and D. Tabb. New York, Longman: 117-135. Politico, -.-t. f. f. L. p. (1998). Democrat and Republican parties look to Latino vote in 2000, politico1@aol.com. Portes, A. and A. Stepick (1993). City on the Edge: The Transformation of Miami. Berkeley CA, University of California Press. Post-Modernity, P. (1996). The State of Disunion, Vol. 2: Summary Tables. Ivy, VA, In Medias Research Educational Foundation. Power at the Polls (1999). . Parade Magazine: 7. Princeton Survey Research Associates (1997). “Immigration.” The Polling Report(August 11): 7. Pyle, A., P. McDonnell, et al. (1998). Latino Voter Participation Doubled Since '94 Primary. Los Angeles Times. Los Angeles CA: A1. Reed, A. (1988). “The Black Urban Regime: Structural Origins and Constraints.” Comparative Urban and Community Research 1. Regalado, J. (1994). Community Coalition-Building. The Los Angeles Riots: Lessons for the Urban Future. M. Baldassare. Boulder, Westview Press: 205-235. Rodriguez, G. (1998). "Minority Leader. New Republic: 21-24. Rodriguez, G. (1999). "From Minority to Mainstream, Latinos Find Their Voice". Washington Post. Washington D.C.: B1, B4. Rodriguez, N. (1996). “U.S. Immigration and Intergroup Relations in the Late Twentieth Century: African Americans and Latinos.” Social Justice 23(3): 111-124. Schuck, P. (1993). “The New Immigration and the Old Civil Rights.” The American Prospect 15 (Fall): 102-111. Schuman, H., C. Steeh, et al. (1997). Racial Attitudes in America: Trends and Interpretations,. Cambridge MA, Harvard University Press. Skerry, P. (1993). Mexican Americans: The Ambivalent Minority. New York, Free Press. Sleeper, J. (1993). The End of the Rainbow? America's Changing Urban Politics. New Republic: 23-. Sleeper, J. (1997). Liberal Racism. New York, Viking. Sonenshein, R. (1993). Politics in Black and White: Race and Power in Los Angeles. Princeton, Princeton University Press. Sonenshein, R. (1996). “The Battle Over Liquor Stores in South Central Los Angeles: The Management of Interminority Conflict.” Urban Affairs Review 31: 710-737. Sonenshein, R. (1997a). The Prospects for Multiracial Coalitions: Lessons from America's Three Largest Cities. Racial Politics in American Cities. R. Browning, D. R. Marshall and D. Tabb. New York, Longman: 41-63. Stone, C. (1989). Regime Politics: Governing Atlanta, 1948-1988. Lawrence, University Press of Kansas. Swarns, R. (1998). Hispanic Mothers Lagging As Others Leave Welfare. New York Times. New York: A1, B8. Tate, K. (1994). From Protest to Politics: The New Black Voters in American Elections. Cambridge, Harvard University Press. Thompson, J. P. (1996). "The Election and Governance of David Dinkins as Mayor of New York". Race, Politics, and Governance in the United States. H. Perry, ed. Gainesville, University Press of Florida. Tolbert, C. and R. Hero (1996). “Race/Ethnicity and Direct Democracy: An Analysis of Claifornia's Illegal Immigration Initiative.” Journal of Politics 58(3): 806-818. U.S. Bureau of the Census (1998). Statistical Abstracts of the United States, 1998. Washington, DC, U.S. Government Printing Office. U.S. Department of Justice (1996). Sourcebook of Criminal Justice Statistics 1995. Washington, U.S. Government Printing Office. U.S. Department of Justice (1997). Sourcebook of Criminal Justice Statistics 1996. Washington, U.S. Government Printing Office. Uhlaner, C. (1991). Perceived Discrimination and Prejudice and the Coalition Prospects of Blacks, Latinos, and Asian Americans. Racial and Ethnic Politics in California. B. Jackson and M. Preston. Berkeley, IGS Press: 339-371. Vidanage, S. and D. Sears (1995). The Foundations of Public Opinion Toward Immigration Policy: Group Conflict or Symbolic Politics? Annual meeting, Midwest Political Science Association, Chicago. Waldinger, R. (1996). Still the Promised City? Afro-Americans and New Immigrants in Postindustrial New York. Cambridge MA, Harvard University Press. Washington Post, Kaiser Foundation, et al. (1995). The Four Americas: Government and Social Policy Through the Eyes of America's Multi-racial and Multi-ethnic Society. Washington, Washington Post. Waters, M. (1996). Ethnic and Racial Groups in the USA: Conflict and Cooperation. Ethnicity and Power in the Contemporary World. K. Rupesinghe and V. Tishkov. London, U.N. University. Welch, Susan and Lorn Foster. 1987. "Class and Conservatism in the Black Community." American Politics Quarterly. 15 (4): 445-70. Wilson, W. J. (1999 forthcoming). Interracial Solidarity: Rising Inequality and the Quest for Coalition Politics. Berkeley CA, University of California Press. Wyly, E. (1997). Social Trends and the State of the Nation's Cities. New Brunswick, NJ, Rutgers University, Center for Urban Policy Research. Yankelovich Partners Inc. (1995). Time, C.N.N., YYankelovich Partners Poll. Roper Center, University of Connecticut, POLL. Zubrinsky, C. and L. Bobo (1996). “Prismatic Metropolis: Race and Residential Segregation in the City of the Angels.” Social Science Research 25: 335-374. Zukin, c. (1997). Taking New Jersey's Pulse. New Brunswick, NJ, Rutgers University, Bloustein School of Planning and Public Policy. [1] Because the topic of “race relations in a diversifying nation” is so huge and sprawling, we limit our focus to politics, that is to conventionally-defined political viewpoints, policy preferences, political coalition-building (or the lack thereof), and strategies for political success. We reserve for another day (or other authors) sociological questions about intermarriage and residential integration, cultural questions about self-expression and artistic blending, and economic questions about class formation and occupational attainment. [2] Had we enough time, we would expand the discussion to consider ideologically-based racial and ethnic coalitions in Congress and among interest or advocacy groups. The very short version of those stories is that the coalitions work surprisingly well in the House of Representatives, despite very difficult moments, and work rather poorly among advocacy groups. [3] By comparison, median family incomes for whites were$37,600, for blacks $22,400 for Latinos$25,000 and for all Asian Americans combined, \$41,000 ((Lien 1998): tables 1-1, 1-2).

[4] As (Bobo and Smith 1994) and (Kinder and Sanders 1996) show, policy issues that are not ostensibly about race are often racially inflected in public discourse and in citizens’ responses to survey questions.  We saw that phenomenon above on the survey questions that yielded the strongest racial divisions, such as the desirability of cutting welfare.  Whether an issue is given a racial inflection, or whether an issue that is commonly thought of in racial terms can be made to seem more racially neutral, is a central concern for anyone seeking to build coalitions across races and ethnic groups.  Racial connotations are at least somewhat malleable, and much political contestation in the foreseeable future will presumably revolve around whether welfare, or crime control, or public education, and so on is “really” about blacks and Latinos, or “really” about families and neighborhoods.  See our discussion of this crucial issue in the conclusion of the paper.

[5] We say “roughly speaking” because whites were more concerned about health care and care for the elderly than about gang violence, and Latinos were more concerned about child abuse than about crime ([U.S. Department of Justice, 1997 #82]: 115). In addition, more careful probing or qualitative interviews would probably reveal racial inflections in how various groups perceive these issues, which would expose at least some points of deep disagreement.

[6] These results apparently contradict findings in other analyses, such as those of (Kinder and Sanders 1996): 30 and (Tate 1994): 29-39.  In part, those differences simply reflect house effects of the surveys – which is to say, variations in results across different survey organizations that no one knows how to explain.  In addition, both books focus on surveys conducted about a decade before the one we have been citing in the text; it is possible that attitudinal differences between blacks and whites have softened since the mid-1980s. Both books also rely heavily on American National Election Studies surveys, in which the black sample sizes are sometimes quite small and Latinos and Asian Americans are not identified separately.  Even so, some of the results in both books resemble the patterns we described above: blacks are consistently more liberal than are whites, but on some issues, there is essentially no difference in policy views across the races. And on some clusters of issues one sees greater disparity between the issues that are most and least favored than between blacks’ and whites’ views on the issues within that cluster.

[7] This interpretation probably does not fit white respondents (unless we were to engage in a deep Freudian analysis of attraction to the disfavored other).  We interpret whites’ sense of commonality with blacks as an identification with other Americans rather than with foreigners or immigrants – an interpretation that is supported by survey data on immigration issues that we report below.

[8] Perhaps they concur with Orlando Patterson’s enunciation of the “homeostatic principle” of racial domination (Patterson 1989) or Derrick Bell’s similar conviction that American racism is permanent and essentially unchanging (Bell 1992b).

[9] On all of these issues, well-off and well-educated African Americans diverge more sharply from the modal white view than do poor and poorly educated African Americans [Hochschild, 1995 #30]; [Joint Center, 1997a #39]; (Gallup Organization 1997).

[10] New immigrants may not interpret racially-inflected encounters as African Americans typically would, so their subsequent political response (or lack thereof) may seem weak to American eyes. For example, Mary Waters (Waters 1996) describes the discordant reactions of Afro-Caribbean immigrants and African Americans in New York to an incident in which a young black immigrant was attacked by a mob of white teenagers.  African American leaders were more inclined to politicize the incident, and were much more attuned to the racial overtones of the attack than were immigrant leaders.

[11] For example, whites, blacks and Latinos all agree – but Asian Americans do not – that Asian Americans are “getting more economic power than is good for Southern California” ([Los Angeles Times Poll #319, 1993 #1488]: ques. 45).

[12] However, Asian Americans are no more immune to self-interest than are other groups.  A survey of southern Californians pointed out that there are almost three times as many Asian Americans in California’s universities than would be warranted by racial proportionality among state residents.  Respondents were then asked if they preferred an admissions policy that would mirror the state’s racial makeup, or if “more ¼[Asian Americans] should be admitted to college than others,¼ if [they] are better qualified.”  Three-quarters of Asian respondents supported “merit” as specified in this question, compared with six in ten whites, and roughly four in ten Latinos and African Americans ((Lien 1997): table 4).

[13] One more illustration, from a national sample and perhaps the best survey: on a four-item index of views of immigrants, Hispanics average 63 percent favorability, Anglos 42.7 percent,  and African Americans only 34.7 percent. There were too few Asian Americans to tabulate (GSS 1996: ques. 960A-D).

[14]  Exit polls are poor indicators of all Americans’ views because the discrepancy between legal residency in the United States and voting varies across racial and ethnic groups.  For example, 55 percent of the residents of California are non-Hispanic whites, but they represent over 80 percent of the state’s voters ((Lutrin and Settle 1996): 4).  Conversely, 30 percent of Californians but only 12 percent of California’s voters are Latino ([Pyle, 1998 #1491]).   Nevertheless, exit polls provide significant evidence both because voting has vastly more external validity than do responses to survey questions, and because exit polls have a larger sample of Asian Americans and Latinos than do surveys .

In March 1995, almost three in five black and white voters favored immediate implementation of the proposition, whereas Latinos “were the only major ethnic group strongly favoring delay” ((Lutrin and Settle 1996): 10).

[15] The proportion of voting-age nonwhite immigrants is growing steadily.  In New York, for example, there are 1.5 million post-1965 immigrants eligible for naturalization.  Most are from Asia, Latin America, and the Caribbean.  Many of these immigrants (or their children) will become voters, and thus critical elements in any urban coalition in the foreseeable future.

[16] Non-Hispanic whites comprise 43 percent of the population, blacks 25 percent,  Hispanics 24 percent, and Asian and others the remaining 7 percent.

[17] Asian American groups have been much less visible in New York City politics and have been slow to mobilize.  Several Asian American candidates did, however, launch campaigns in the most recent round of city council elections.

[18].This point is debatable; if employers prefer to hire immigrants rather than African Americans, employment niche patterns result from employer discrimination, rather than group self-sorting  ((Kirschenman and Neckerman 1991); (Smith and Edmonston 1997)).

[19] However, economic inequality within the black population has increased, as has “the social and political distance between the black middle class and the black poor” (Mollenkopf, 1992: 68).  Similarly, Afro-Caribbeans, Koreans, and Columbians have fared much better than native-born blacks, Dominicans, and Puerto Ricans.

[20] Congressman Charles Rangel is a notable exception to the black silence on these issues.  See (Jones-Correa 1998) on the political concerns of the new immigrants, and the lack of efforts to integrate them into political organizations.

[21] Latinos now comprise almost 40 percent of Los Angeles’ population and outnumber all other groups.  The Asian American population has more than doubled in the last twenty years and currently constitutes ten percent of the population.  After peaking at 17 percent some years ago, the African American population now hovers around 14 percent; the white population, at 48 percent of the population in 1980, now comprises just over a third.

[22] Reuel Rogers’ conversation (July 15, 1997) with James H. Johnson, who served as a consultant to the

reapportionment debates.

[23] Although Hispanics comprise 40  percent of the Los Angeles population, they are only 11 percent of registered voters. (Whites and blacks account respectively for 65 and 15 percent of the city’s registered voters.)  The disparity is due mostly to the comparative youth and low

naturalization and voter registration rates of the city’s Latino residents.  Hence, Latino leaders

hoping to create an independent political bloc must undertake a massive mobilization effort

focused on both naturalization and voter registration.

[24] Several leading African American politicians, such as Congresswoman Maxine Waters and former primary candidate J. Stanley Sanders did not endorse Woo.  But he won the backing of other key black political figures, and the city’s black leadership expressed no concerted resistance to his candidacy.

[25] Park ((Park 1996)) has a very useful discussion of the details of Los Angeles’ politics

over the past few years, expressing even less optimism than we do here about the

resurgence of a coalition of the left.

[26] The 1983 alliance among blacks, Hispanics, and liberal whites in Chicago was forged

around similar sites of convergence. This coalition never took institutional root; rather, it

was held together by Mayor Harold Washington’s charisma and strong leadership.  When

Washington died suddenly in 1987, the alliance disintegrated almost immediately.  The

group of black political leaders whom Washington had unified became riven by factionalism

among reformers, party loyalists, and racial nationalists.  The latter drew a disproportionate

amount of public attention and quickly alienated the white liberals and Latinos who had

been subsidiary, but crucial, partners in Washington’s coalition.  Both groups shifted their

allegiance to Mayor Richard Daley Jr.’s conservative coalition, which has dominated

Chicago politics since 1989.  Mayor Daley now enjoys strong support from all of the city’s

racial and ethnic groups, including African Americans.

[27] As Los Angeles’ “branch-plant economy toward which working class [b]lacks and Chicanos had always looked for decent jobs” has “collapsed,” the two groups increasingly must compete for what remains. Even those blacks “willing to compete for . . . menial service jobs find themselves in a losing competition with new immigrants” (Davis 1990): 304-05).

[28] Formerly all-black neighborhoods have undergone uneasy transition as large numbers of Hispanics moved in and Asian immigrants opened fledgling family businesses there.

[29] Thus Hispanics rushed to become naturalized citizens, to register, and to vote after Proposition 187.  California’s Hispanics are now much more likely to vote Democratic than are Hispanics in Texas and Florida, even though they had been moving in the direction of the Republican Party in the early 1990s ((Pinkus 1999); (America at the Polls 1999): 100)

[30] “As the welfare rolls continue to plunge, white recipients are leaving the system much faster than black and Hispanic recipients, pushing the minority share of the caseload to the highest level on record….  The disproportionately large exodus of whites has altered the racial balance in a program long rife with racial conflict and stereotypes…. The legacy of those stereotypes makes the discussion of race and welfare an unusually sensitive one” ([deParle, 1998 #1597]: A1; see also [Swarns, 1998 #1599]).

[31] A few representative quotations after the 1998 elections: “The backlash from the passage of Propositions 187 and 209 in California have made Latinos the state’s most powerful voting bloc” ((Migrant News 1998): 4).  “With the mid-term elections over, Republicans and Democrats are now plotting their strategies to enlist the nation’s fastest growing minority in their bid to win the White House in 2000” ((Politico 1998): 1) “In the last election, Hispanics were the swing vote in several key races, including those in Texas, California, Florida, New York, Arizona, Nevada, and California” [Power at the Polls, 1999 #1506].

[32] The most systematic analysis of the political implications of a growing number of Hispanic voters finds that political and (especially) socioeconomic gains for African Americans are positively correlated with gains for Hispanics.  That is, increases for one group are often concomitant with the other – and both gains come at the expense of whites ((McClain and Tauber 1998). If that pattern persists, it permits greater optimism about coalitions among peoples of color, but less optimism about coalitions between blacks and whites or between Latinos and whites.

[33] “Voter turnout among Asian Americans reached an all-time high in the last election…. “Their  most lasting contribution may be their unique ability to transcend traditional ethnic fault lines; indeed, the political maturation of Asian Americans may signal a new era in racial politics” [Rodriguez, 1998 #1503]: 21, 22.