PART III—
DIALOGUES OF DOMINANCE AND SUBORDINATION

Chapter Five—
Intergroup Feelings and the Definition of Group Interests

It is expedient to hate the adversary with whom one fights (for any reason), just as it is expedient to love a person whom one is tied to and has to get along with. . . . The mutual behavior between people can only be understood by appreciating the inner adaptation which trains in us feelings most suitable to a given situation. . . . By means of psychological connections, these feelings produce the forces which are necessary to execute the given task and to paralyze inner countercurrents.
George Simmel, Conflict (1955, 34)

Simmel's reasoning exemplifies the logic that has prevailed in the literature on intergroup attitudes—that love for our own and hatred for our adversaries are the natural pivots in one group's orientations toward another, providing the necessary energy for the conduct of their unequal relationship. I discussed that reasoning in Part I and found it wanting. I argued that the ongoing, relational basis of the inequality between groups demands that one must get along with rather than fight with one's "adversaries," and, indeed, it is problematic whether the other group will even be defined as an adversary. Ironically, this means that the feelings that dominant groups find "most suitable" to the "given situation" of their relationship with subordinates may not be hatred, but instead affection or at least temperance. These pressures should mitigate against the expression of disparity and hostility and instead draw the dialogue between groups toward more amicable declarations.

In this part of the book, I appraise the part played by hostility and affection in the shaping of the dialogue that takes place between groups who are bound together in different kinds of unequal relationships. Over the next few chapters, I will be examining data on different elements of the intergroup attitudes that accompany race, gender, and class relations in the United States.

I begin, in this chapter, at the most fundamental level: do contending, unequal groups express negative or positive feelings toward one another, and do they view their interests as adversary or unified? The character of intergroup feelings is basic because it sets the tone of the relationship between groups: before we can interpret any other elements of people's

intergroup attitudes, we need to delineate their emotional packaging. The perception of group interests tells us about the cognitive understandings that accompany intergroup feelings: do unequal groups believe they are separated by a deep divide in their interests or do they see each other as having a shared community of interests?

I start by assessing the differing meanings attached to intergroup feelings and the perception of group interests in traditional conceptions of prejudice, tolerance, and group consciousness. I then discuss problems involved in the measurement of intergroup feelings and the perception of group interests, for despite their intrinsic conceptual significance, these two elements have rarely been directly measured. With those issues in hand, I outline my approach to the meaning of intergroup feelings and the perception of group interests. The second part of the chapter proceeds with an examination of my data on these two key elements of intergroup ideology.

Two broad sets of questions govern that analysis. First, to what extent do the feelings expressed by dominant and subordinate groups reflect basic currents of hostility or warmth? How sharply do people express an affective preference for their own group, and how much is hostility a factor in their feelings toward the other group? How does the element of hostility vary across the three intergroup relationships, and is there any consistent tendency for the expression of hostility to come more from one side of a relationship than the other?

Second, to what extent are group interests defined in conflictive or integrative terms? Do the members of unequal groups tend to perceive existing social arrangements as serving the interests of all the affected groups, as serving the interests of the dominant group at the expense of subordinates, or as detrimental to the interests of all groups? How does this vary according to the nature of the intergroup relationship and the position of the group within that relationship?

Answers to these questions will uncover the elementary thrusts of the ideologies that distinguish race, gender, and social class relations in the United States. Are those ideologies framed, emotionally and perceptually, by a sense of disparity or unity between groups?

The Issues

Despite their centrality, intergroup feelings and the perception of group interests have had different specific meanings attached to them from one body of literature to another. That variation reflects two main factors: primarily, whether intergroup attitudes are thought to have a rational or an irrational basis; and second, different ideas about the natural attitudinal expression of rational or irrational forces. I briefly consider the

different ways that intergroup feelings and the perception of group interests have been treated in traditional conceptions of prejudice, tolerance, and group consciousness: as we move from one concept to the next, rationality assumes a larger presence. I then turn to measurement issues that have hampered the assessment of intergroup feelings and the perception of group interests. Finally, I discuss my theoretical approach, and I draw out the primary empirical questions that confront us.

Prejudice

Prejudice has usually been conceived as an intrinsically psychological phenomenon, in which negative feelings constitute the hot core. Recall that Allport's widely accepted definition of prejudice is "an antipathy based upon a faulty and inflexible generalization" ([1954] 1979, 9). The centrality of negative feelings to the concept of prejudice has been restated many times (see, for example, Williams, Jr. 1947, 36–42; Harding et al. 1969, 3–6; Newcomb et al. 1965, 430–431; Simpson and Yinger 1972, 24; Berry and Tischler 1978, 235; Kinder and Sears 1981, 416; Pettigrew 1982, 1–5; Sniderman and Tetlock 1986, 186; Marger 1994, 74–75). The meaning attached to those feelings is somewhat different than in the literature on group consciousness. Most analysts of prejudice have paid almost exclusive attention to the attitudes of dominant-group members within a framework that seems to view all of the energy as coming from the members of the dominant group as they unleash their personality or cultural deficiencies on subordinate victims. Thus, the feelings that drive prejudice are not usually linked to the rational interests of a group within a relationship. Instead, things are seen as more of a one-way tirade: the members of the dominant group are the actors, and their feelings emanate from factors endogenous to their own ranks.

Within this perspective, negative intergroup feelings are seen as destructive, and theories have concentrated on how to change negative feelings into positive ones, in order to convert rejection into acceptance. In this endeavor, the locus of energy is clearly within the ranks of the dominant group: they initiate the hostility, and social change can only be accomplished by altering the emotional proclivities of dominant-group members toward subordinates.

And because intergroup feelings have not usually been linked with the tangible interests of the group, the way group members perceive their interests has not been regarded as critical in most analyses of prejudice. Of course, analysts in the realistic group-conflict school have argued that negative feelings are founded in a sense of threat that people feel vis-à-vis subordinates, and that sense of threat derives from their group's "realistic" interests (see, for example, LeVine and Campbell 1972; Bobo 1983; Giles and Evans 1986). Similarly, Sherif (1965) emphasized the

importance of establishing competitive or cooperative group interests in instigating or de-escalating intergroup hostility. The most common governing assumption, however, has been that it is not disparate interests that are ultimately responsible for dividing racial groups (see, for example, Allport [1954] 1979, 229–233), but rather that prejudice embroils an erroneous belief that the groups have incompatible interests. Thus, although analysts working on the effects of intergroup contact have repeatedly stipulated that contact situations must not contain competitive group interests if there is to be a positive effect on intergroup attitudes (see, for example, Williams, Jr. 1947; Wilner, Walker, and Cook 1955; Allport [1954] 1979; Amir 1969; Cook 1984; Miller and Brewer 1984), the import of that stipulation has been that policies designed to eradicate prejudice must be structured so as to expose the participants to the "common interests and common humanity" that exist between their own and the other group. In this way, policies may break down the artificial barriers that prejudiced people have erected between the groups (Allport [1954] 1979, 281; Wilner et al. 1955: 3–6).

Students of prejudice regard the antipathy that is the core of the concept as a complex puzzle. Some ambiguity hangs over the issue of whether negative feelings derive logically from "faulty and inflexible generalizations" that have been formulated in the absence of accurate information about a group or whether the faultiness and inflexibility of those generalizations are themselves conditioned by preexisting negative feelings (see, for example, the discussions of prejudice in Williams 1947, especially pages 40–41; and Allport [1954] 1979, chapters 1 and 2). Theories emphasizing personality or socialization factors (for example, Adorno et al.'s theory of the "authoritarian personality," 1950) have promoted the latter view. Theories focusing on personal contact with the object-group (for example, Wilner et al. 1955; Stephan and Stephan 1984) have put more emphasis on the idea that the initial negative beliefs derive primarily from ignorance but that once established they become the logical basis for negative feelings toward the object-group. Analyses done in the minimal-group paradigm (see, for example, Tajfel 1969, 1978; Brewer 1979) imply that antipathy toward other groups springs from cognitive and emotional biases that are elicited simultaneously by any intergroup situation, no matter what, or how flimsy, the basis of the division.

In any event, students of prejudice view negative feelings as a destructive force that is hurled upon subordinate groups and that must be eradicated in order to progress toward a more egalitarian society. Those feelings have sometimes been linked to the perception of disparate group interests, but since most students of prejudice regard such a perception as misplaced rather than rationally derived, the source of negative feelings

toward subordinates remains a puzzle—the solution to which holds the key to the problem.

Tolerance

Like other students of intergroup attitudes, analysts of tolerance have assumed that negative feelings toward other groups provide the motivational basis for negative behavioral dispositions toward those groups. However, those feelings are neither irrationally based, as most students of prejudice would have it, nor rationally escalatory, as students of group consciousness would have it. Instead of seeing the existence of negative intergroup feelings as a puzzle that needs to be solved (like students of prejudice) or as an ultimate state that needs to be attained (like students of group consciousness), analysts of tolerance have assumed that such feelings are a constant and unremarkable feature of relations among contending groups in democratic societies. This relegates both intergroup feelings and the perception of group interests to a latent, rather than an active, presence—they are neither the obstacle to intergroup harmony nor the key to constructive conflict.

Contending groups are assumed to represent different interests, and intergroup antipathy is regarded as a natural outgrowth of those differing interests. However, unlike students of group consciousness, students of tolerance do not seem to regard those contending interests as irreconcilable. Instead, they are committed to the utility of a continuing democratic choice among contending political alternatives. Thus, although they view intergroup antipathy as an inevitable outcome of social diversity, they also, like analysts of prejudice, view intergroup antipathy as destructive, or at least potentially so, rather than rationally constructive (see, for example, Stouffer 1955; Nunn, Crockett, and Williams 1978; McClosky 1964; Sullivan et al. 1982; McClosky and Brill 1983, chapter 1).

This line of thinking leads students of political tolerance away from any direct focus on the affective or perceptual boundaries that they assume exist between contending groups. Those boundaries are seen as potentially dangerous, but not subject to eradication. The pertinent empirical problem has been defined as how people learn to override their negative feelings toward contending groups and abide by "the rules of the game" that grant full civil and political rights to the groups they dislike. And because of a primary concern with the danger of a "tyranny of the majority," analysts have focused on the tolerance expressed by politically dominant groups toward "nonconformists" or other minorities.

The difficult attitudinal task of tolerance has been regarded as unattainable when intergroup antipathy is intense (Dahl 1956, especially 75–80; Sullivan, Piereson, and Marcus 1979, 1982). Mild feelings and

perceptions of intergroup disparity are regarded as optimal for the continuance of democratic procedures—enough to ensure a political choice but not enough to fragment society irreconcilably.

Group Consciousness

Analysts of group consciousness have viewed intergroup attitudes as rationally derived: groups are generally expected to espouse attitudes that are consistent with their interests. The class literature, in which the analysis of group consciousness is rooted, has implied this in drawing a clear link between the fundamental nature of socioeconomic inequality and the state of class consciousness. The primary concern has been the state of awareness of subordinate groups, with the question of dominant-group consciousness receiving less attention. The result has been a lack of clarity about what form a rational expression of dominant-group interests should take. This, in turn, has led to some confusion about subordinate attitudes: after all, the rational response of subordinates cannot be ascertained precisely without an understanding of the political reality that they confront. The literature on group consciousness offers several possible scenarios.

The picture that comes most immediately to mind is that of emotional and cognitive polarization between groups. This was the picture conveyed by Marx and Engels, as when they wrote in "The Manifesto of the Communist Party" ([1888] 1959, 8) that "the epoch of the bourgeoisie . . . has simplified the class antagonisms . Society is more and more splitting up into two great hostile camps , . . . bourgeoisie and proletariat" (emphasis added). Marx's central thesis about the inevitability of increasing class polarization under capitalism rests on the expectation that as capitalism develops, it reveals its workings more and more sharply: the mutually opposed interests of exploiting and exploited classes become manifested unmistakably to both sides, and this engenders feelings of solidarity within groups and disparity and hostility between them.

Within this framework, two elements have been regarded as basic in identifying subordinates' level of consciousness. First, do they identify emotionally with their group and distance their feelings from the dominant group? Indeed, do they hold negative feelings toward those against whom they must fight for social change? Second, do subordinates perceive their group as an entity that has distinct interests that lie opposed to those of the dominant group and that cannot be served within the status quo? These two elements have been considered essential to a sense of "shared fate" that binds subordinates together in opposition to the dominant group (see, for example, Centers 1949, chapter 2; Landecker 1963; Morris and Murphy 1966; Olsen 1970; Oberschall 1978; Tilly 1978; Fireman and Gamson 1979; Gurin, Miller and Gurin 1980; Miller

et al. 1981). Thus, unlike most students of prejudice, analysts of group consciousness have viewed intergroup disparity and hostility as rational expressions of a group's interests that provide the necessary energy for conflict and change.

Although this general expectation provides the backdrop for the analysis of group consciousness, it is not clear how such a state is reached. In particular, there is disagreement about which side is expected to initiate the path to conflict. Some analysts have seemed to assume that subordinates routinely encounter hostility and derogation from dominant groups. This assumption has been especially prominent among those studying ethnic identification (see, e.g., Brand et al. 1974). Many such scholars appear to have accepted the prevailing viewpoint of the prejudice literature that dominant groups are unable to contain their irrationally founded negativism toward minority ethnic groups. Other scholars follow the assumption employed in the realistic group-conflict school that the negativism of dominant groups is the rational expression of their privileged interests. In either case, the issue has been whether subordinates can distance themselves sufficiently to develop a positive sense of group identity in the face of that derogation, assert their interests, and counter with hostility for hostility. Whether or not the presumed negativism of dominant groups is thought to have a rational basis, signs of intergroup negativism among subordinates are interpreted as a constructive and rational response to the barrage that they receive from the dominant group.

Many analysts, however, have argued that it is in the interests of those in command to mask their privileged receipt of benefits. To that end, dominant groups are expected to represent their interests in universalistic terms as being at one with those of society as a whole (see, for example, Marx and Engels [1846] 1970; Dahrendorf 1959, 280–289; Genovese 1968; Parkin 1971, chapter 3; Gramsci 1971; Huber and Form 1973; Giddens 1979, 193–197). This changes the nature of the task for subordinates, putting the onus on them to raise the voice of disparity.

The manifest appeal of universalism to dominant groups, with the self-satisfaction and integrated view of life that it provides, is illustrated in the following excerpts from two essays by Stringfellow in Eliott's collection of proslavery arguments (1860). Tracts such as these show the seemingly boundless breadth of the claims that can be made by a system's beneficiaries about the universality of those benefits.

It is or ought to be known to all men, that African slavery in the United States originated in, and is perpetuated by a social and political necessity, and that its continuance is demanded equally by the highest interests of both races. . . . The guardianship and control of the black race, by the

white, in this Union, is an indispensable Christian duty, to which we must as yet look, if we would secure the well-being of both races. (Stringfellow 1860a , 521)

The facts which have been reviewed prove with equal clearness, that where slavery exists, the white race, and the black, have prospered more in their religious, social and moral condition, than either race has prospered, where slavery has been excluded. We see that an increased amount of poverty and wretchedness has to be borne in New England by both races. Ecclesiastical statistics will show an increased amount of prosperity in religion that is overwhelming.

Such is the prostration of moral restraint at the North, that, in their cities, standing armies are necessary to guard the persons and property of unoffending citizens, and to execute the laws upon reckless offenders. This state of things is unknown in the slave States.

The census shows that slavery has been a blessing to the white race in these slave States. They have prospered more in religion, they have more homes, are wealthier, multiply faster, and live longer than in New England, and they are exempt from the curse of organized infidelity and lawless violence.

A comparison of the slave's condition at the South, with that of his own race in freedom at the South, shows with equal clearness, that slavery, in these States, has been, and now is, a blessing to this race of people in all the essentials of human happiness and comfort. Our slaves all have homes, are bountifully provided for in health, cared for and kindly nursed in childhood, sickness, and old age; multiply faster, live longer, are free from all the corroding ills of poverty and anxious care, labor moderately, enjoy the blessings of the gospel, and let alone by wicked men, are contented and happy. (Stringfellow 1860b , 538–539)

With such arguments as these, dominant groups promote their own cause without stooping to the level of appearing self-interested. Their appeal is made in the name of moral righteousness, and at the same time, the potentially damaging issue of group interests is neutralized.

This implies that, if there is to be conflict, it is subordinates who initiate it. The way that subordinates respond to dominant claims of universalism, however, has been the subject of some disagreement. One argument has been that the political task of subordinates is to reveal the mutually opposed interests that divide the groups and to withdraw their emotional support from the modus vivendi. Dominant groups prefer to deny that there are opposed group interests, but they are eventually forced to accept that interpretation as a result of the concerted political efforts of subordinates. Indeed, Mannheim (1936, 229-235) has argued that dominant groups remain blissfully unaware of their special interests until the matter is rudely brought to their attention by subordinates. Appealing as it may be to view the world from which one benefits as one happy

family, this view may become increasingly difficult to maintain when confronted with repeated shrill denials of happiness from the other "family members."

Another well-known argument is that when subordinates mount a challenge, they are as liable as dominant groups to opt for the language of universalism. That was the argument made by Marx:

Each new class which puts itself in the place of one ruling before it, is compelled, merely in order to carry through its aim, to represent its interest as the common interest of all the members of society, that is, expressed in ideal form: it has to give its ideas the form of universality, and represent them as the only rational, universally valid ones. The class making a revolution appears from the very start, if only because it is opposed to a class , not as a class but as the representative of the whole of society. (Marx and Engels 1970, 65–66)

This suggests that both established and insurgent groups avoid characterizing society in terms of narrow, adversary interests, as they make competing claims to be the sole representatives of the common good. In this vein, Gamson (1968, 53–54) has argued that all contending solidary groups tend to represent their interests in universalistic terms. Indeed, if dominant groups eschew hostile assertions of their own interests and instead claim that current arrangements are to everyone's benefit, this may condition the moral framework so that any challenge is also couched in universalistic rather than "selfish" terms. In that spirit, socialist doctrine claims that redistributive measures would benefit the well-being and fulfillment of all elements of society, not just the working class, and the liberal doctrine of racial integration depicts segregation as a universal evil that works to the cultural detriment of whites as well as African Americans.

Other analysts have argued that subordinates' responses to the universalistic claims of dominant groups fall short of any outright challenge. Subordinates grapple with the reality of their day-to-day experience juxtaposed against the constant ideological onslaught of the dominant group. As they cope with these inconsistent stimuli, they may succumb to abstract representations of dominant ideological principles but formulate concrete micro-interpretations that are more in keeping with their personal experience (Mann 1970; Parkin 1971, chapter 3; Huber and Form 1973, chapter 9; Abercrombie et al. 1980, 140-155).

For scholars committed to a conflict perspective, subordinate attitudes that do not display some form of challenge are characterized as "false consciousness" (see, for example, Mills 1946; Mann 1970). As I discussed in chapter 1, this phenomenon poses an awkward puzzle for such analysts. Indeed, this is the only point at which students of group conscious-

ness have toyed with the idea of irrationally founded attitudes. A distinction is drawn between what people in a group "ought" to think and what they do think—a distinction that is rather tenuous, especially within a general framework that presupposes rationality in people's intergroup attitudes.^[1] Indeed, this line of explanation credits the dominant class with a rational pursuit of its interests, as it dupes the masses with a glossy picture of the status quo: it is only subordinates who, in allowing themselves to be duped, fail to demonstrate the rational awareness of their condition that is so urgently expected of them. Thus, in the attempt to define away the empirical problem of the lack of any sharp political challenge among subordinates, some conflict-oriented scholars have cut into the very building-block on which the expectation of conflict rests—that is, the assumption that individuals comprehend and respond rationally to the conditions that they experience. Analysts have attempted to resolve the contradiction by relegating false consciousness to the status of a temporary lapse over which rational political organization will eventually triumph, but the ideas that I have reviewed do not offer any sound basis for expecting the development of such an organization from grassroots that are so devoid of rational consciousness.

This is especially true of the literature on cultural hegemony, which represents the most sustained attempt to grapple with the issue of false consciousness. Marx and Engels' famous discussion in The German Ideology provides the basis for ideas on hegemony: "The ideas of the ruling class are in every epoch, the ruling ideas, i.e. the class which is the ruling material force in society is at the same time its ruling intellectual force" (Marx and Engels 1970, 64). Building on that argument, Gramsci (1971) posited a process whereby the dominant group gains the consent of subordinates by infiltrating their system of values. Williams gives a good description of the process:

[Hegemony is] an order in which a certain way of life and thought is dominant, in which one concept of reality is diffused throughout society in all its institutional and private manifestations, informing with its spirit all taste, morality, customs, religious and political principles, and all social relations, particularly in their intellectual and moral connotations. (Williams 1960, 587)

From this perspective, dominant groups succeed in imposing a single interpretation of social reality, in which all parties embrace the status quo as representing the common interest. Genovese has argued that

[1] See Mann (1970) for an interesting attempt to make an empirical determination of "false consciousness." Ultimately, however, the "falseness" of someone's consciousness is a definitional rather than an empirical problem, resting on prior assumptions about what someone's "true" interests are (see Connolly 1972) and about the rational course of action to pursue those interests.

hegemony depends on much more than consciousness of economic interests on the part of the ruling class and unconsciousness of such interests on the part of the submerged classes. The success of a ruling class in establishing its hegemony depends entirely on its ability to convince the lower classes that its interests are those of society at large—that it defends the common sensibility and stands for a natural and proper social order. (1968, 407)

The idea of cultural hegemony poses a contradiction within the conflict framework. In order for hegemony to work, subordinates must be relatively passive ideological consumers, which makes them improbable agents of ideological challenge. Subordinates are left wallowing in a state of false consciousness without any structural process being posited that might counter the effects of hegemony and instigate the path to emotional and perceptual estrangement from the status quo.^[2] At the same time, conflict scholars continue to uphold such estrangement as the catalyst for a political initiative to attain a more egalitarian society, and thus it is seen as the ultimate rational course for subordinates.

Measurement Issues

Despite their pivotal conceptual status, intergroup feelings and the perception of group interests have rarely been measured directly in analyses of prejudice, tolerance, or group consciousness. Direct measurement of intergroup feelings has largely been confined to the "thermometer scales" used in the National Election Studies (NES 1952–1990 Cumulative Data File Codebook), and those measures have been employed occasionally to make rudimentary assessments of feelings toward other social groups or one's own group in studies of prejudice or group consciousness (see, for example, Campbell 1971; Jackman 1978; Gurin, Miller, and Gurin 1980; Miller et al. 1981; Bobo 1983).^[3] Direct measurement of the perception of group interests is even more unusual; a rare example is found in Landecker (1963).

[2] Gramsci was writing as a political activist as much as an analyst, and his interest in the concept of hegemony appears to have been motivated in large part by the implications it held for the Communist party political agenda. His ideas about cultural hegemony led him to advocate that, in order to accomplish a complete overthrow of capitalism, in all its manifestations, it was essential to foster a working-class cultural hegemony (Williams 1960; Cammett 1967, 204–206; Anderson 1976). However, as Scott (1985, 316) has observed, it is difficult for the analyst to divine how, given one hegemony, there can simultaneously be the latitude to cultivate a counter-hegemony: the two ideas are logically inconsistent.

[3] The community studies conducted by Williams, Jr. (1964, 280–281) included a rare, early attempt to measure blacks' feelings toward whites directly. Two items were used: "Sometimes I hate white people [agree/disagree]," and "In general, how friendly or unfriendly are your feelings toward whites? Would you say you are very friendly, fairly friendly, rather unfriendly, or not friendly at all?" The first of these items was used again by Paige (1972) in a study of blacks' racial attitudes in the 1960s.

The minimal measurement of these concepts is paradoxical, but it is probably attributable to the compounding of two factors. First, ironically, assumptions about these phenomena have been so deeply embedded in various conceptual positions that there has not been much pressure to measure them directly: their importance has been so confidently assumed that empirical validation has rarely entered the research agenda. Students of prejudice and group consciousness have assumed that people's emotional dispositions toward other groups are artlessly manifested in every aspect of their intergroup attitudes, especially in stereotypical attributions of personality traits, social-distance dispositions, and policy views. Meanwhile, students of tolerance simply assumed that people's feelings toward minority groups were negative—and they then concentrated on measuring people's policy dispositions toward a group to assess whether they had attained a state of political tolerance (see, for example, Stouffer 1955). Inferences about people's perception of group interests have frequently been made in studies of class consciousness, but those inferences have generally been drawn from measures of people's policy views—it was presumed that the latter constituted a clear reflection of the perception of group interests.

Second, those proclivities have been compounded by the fact that intergroup feelings and the perception of group interests are relatively difficult to measure directly. Analysts have found other attitudinal elements, such as stereotypes, social-distance dispositions, and policy views, more susceptible to measurement. The inclination to measure indirect manifestations of the attitudinal core rather than the core itself was perhaps encouraged by the influence of Freudian thinking among some early empirical researchers. Some researchers felt that direct measures of people's feelings in particular would be invalid because respondents would be unable to retrieve or reflect clearly on such a deeply embedded and sensitive matter. This kind of concern is apparent in the widely influential California F-scale, whose creators (Adorno et al. 1950) deliberately relied on projective rather than direct questions. But one does not need to be a Freudian to acknowledge the difficulties that confront the survey researcher in measuring intergroup feelings and the perception of group interests.

Feelings are intrinsically nonverbal, and our culture gives us little experience in articulating our feelings verbally. In our everyday efforts, we usually find speech a clumsy instrument to convey the rich texture of felt emotions, and we often rely on nonverbal cues (posture, facial expression, tone of voice, body movements) to transmit the emotional coloration of our words. We may expect, then, that people would find it difficult to retrieve and convey their feelings in the somewhat stilted verbal back-and-forth of the survey interview.

The perception of group interests presents fewer intrinsic measurement difficulties, but because public political discourse rarely portrays political issues explicitly in those terms, the mass public is unschooled in the concrete attribution of group interests. Thus, they may have trouble understanding or responding to questionnaire items that ask about the specific policy interests of their own and other groups, because they are unused to the process of sizing up social life explicitly in that light.

These various difficulties are substantial, but not insurmountable. Indeed, it is essential to develop direct measures of these phenomena if we are to develop more specific expectations about their place in the mapping of intergroup attitudes. I return to these issues when I present my measures in the Data section below.

An additional consideration has probably discouraged researchers from measuring feelings toward social groups especially, and that is concern about social desirability pressures. On a topic as sensitive and personal as one's feelings toward social groups, some researchers may feel that respondents would be reluctant to reveal themselves openly, especially when asked about groups that have been the topic of heated public-policy debate. The potential for social desirability pressures must be respected in designing survey items on sensitive issues, but the significance of such pressures may have been overestimated.

Evidence suggests that people tend to assume that others hold similar views to their own (Schuman and Kalton, 1985, 655), a phenomenon that has been termed "looking-glass perceptions" (Fields and Schuman, 1976) or the "false consensus effect" (Ross, Greene, and House, 1977). Although that may seem surprising, most people are caught up in their own networks of information and experiences: because their own feelings and opinions have been formed within that context, they generally find that their own reactions are consistent with those of others in their social environment. This leaves them unself-conscious about their own reactions. Individuals whose feelings and opinions are out of step with the normative proclivities of their group might be expected to be especially susceptible to social-desirability pressures, but such people generally come from pockets that have been bypassed by the primary information channels that flow through the group, and hence they may also be oblivious to the "social undesirability" of their views. In other words, the same factors that shaped their unusual reactions also buffer them from an acute awareness of their unusual nature. Thus, for example, Fields and Schuman (1976) report that only 2 percent of Detroit whites believed that white and black children should be prevented from playing with one another even at school, but this tiny minority believed overwhelmingly that their own attitude was shared by the majority of Detroit whites. And even when people are aware that their feelings are in the minority

for their group, it is probably more compelling to them that others in their local network share their reactions than are the relatively distant reactions of others who lie outside their day-to-day lives.

On a more fundamental level, the concept of social desirability is less straightforward than it first seems. The problem is that if social desirability exerts pressure on how people reveal themselves, it may not be a narrow problem of measurement error confined to the survey interview, but a general phenomenon affecting the way people express their attitudes to their friends and associates, and even affecting the way they form those attitudes in the first place. If attitudes are seen primarily as the personal, private property of individuals, it is somewhat easier to conceive of pressures outside the individual as a contaminant. If, however, we see intergroup attitudes as being shaped by the exigencies and information that emanate from the group, the notion of the culture of the group exerting an extraneous pressure on its individual constituents' attitudinal expressions no longer makes sense. My argument is that people do not develop their attitudes independently and then lie about them when they believe they are out of step with the predominant cultural values of the group. Instead, the individual's attitudes are acquired without originality from the array provided by the group, and that array is itself shaped by the pressures to which the group is subject: in this unself-conscious process, the distinction between private dispositions and public ones becomes more difficult to maintain. Thus, although it is essential to construct measures that minimize any artifactual pressures in the interview situation itself, the notion of social desirability is not necessarily a simple issue of measurement alone, but entails broader questions about the forces that shape the attitudes we are trying to measure.

Dispensable Hostility

All the arguments I have outlined from past literature on prejudice, tolerance, and group consciousness pivot, for different reasons, on the notion of intergroup disparity and hostility. But I contend that dominant groups, far from having negative feelings toward subordinates, gravitate toward positive, or at least neutralized emotional dispositions. By the same token, I expect dominant groups to maintain an integrated view of the interests of their own and subordinate groups. If this is correct, it changes the nature of the game dramatically from any of the scenarios that have framed past discussions.

First, it means that subordinates do not have to distance themselves from the dominant group in order to have positive feelings about themselves. Second, it means that if subordinates are to take a hostile disposition toward the dominant group, they cannot simply reciprocate with venom for venom. Instead, it is incumbent upon them, from their weaker

political position, to renounce amicable feelings and inject hostility into the relationship—a dangerous game, since the dominant group has the resources to reciprocate more than generously. Third, if the amicability of the dominant group is packaged with an interpretation of social reality that binds the interests of the groups together, subordinates may have a difficult time prying those two items apart, and thus the acceptance of the amicability carries a price tag.

I assume that groups gravitate to an ideology that is rationally consistent with their interests. But I also assume that the rational actor does not naively pursue goals on the basis of their desirability alone, but is sensitive to constraints. Occasional mistakes and miscalculations are to be expected, but groups do learn from their mistakes. Most of the time, the articulated attitudes of unequal groups reflect responses that strive to maximize their control over resources within the conditions that they confront . This does not lead me to expect a great deal of hostility from either side of an intergroup relationship. Indeed, even the literature on group consciousness, for all its endorsement of intergroup hostility as the ultimate rational outcome, does not specify an avenue by which such hostility may be realized. None of the ideas articulated in that literature can be reconciled with the expectation of an eventual conflict, such as that envisioned by Marx and Engels, in which both sides attain an understanding of their mutual relationship and draw a sharp affective boundary between each other.

If, as many students of group consciousness have argued, dominant groups are engaged in establishing the common interest as the moral guideline, it would seem more in keeping for them to espouse the spirit of friendship toward subordinates than to adopt hostile feelings toward them. It is not in the interests of the dominant group to introduce or encourage feelings of hostility because that would (1) expose the inequality in the harshest possible light, and (2) remove their most potent weapon of social control—the offer of amicable relations, on their terms. On subordinates' part, it is hard to see how they can break completely free of such a sticky moral framework and push it to one of sharp mutual disparity and hostility.

At the same time, the totality of the concept of hegemony seems too unyielding and unidirectional to take account of the varying opportunities that different kinds of unequal intergroup relationships offer for dominant groups to invade the perceptions and feelings of subordinates. Subordinates are not passive recipients of dominant ideology, and their frame of mind is much affected by the opportunities that are afforded them to alienate their thinking from that of the dominant group. The dominant group, in turn, is sensitive to the kinds of communication channels that are available and attentive to the mood of the group over

which it seeks to maintain control. In other words, the thinking on both sides responds to the specific constraints that are presented.

In relations that are structured with frequent personal contact across group lines, dominant groups have more opportunity to infiltrate the lives of subordinates. Under these circumstances, there are escalated costs to individual subordinates of emotional and cognitive estrangement from the prevailing arrangements. When relations are structured more distally, warm intergroup feelings are more difficult for the dominant group to maintain and less personally costly for subordinates to abandon—but even here, there is good reason for the dominant group to strive to avoid disparity and for subordinates to approach it with circumspection.

To begin to address these issues, we need to delineate the emotional and perceptual currents that flow back and forth from one group to another in different intergroup relations. Three pieces of information are central. First, is there an emotional rift between one group and another, such that the members of each group feel emotionally attached to their own and separated from the other group(s) in the relationship? In other words, is there a sense of affective solidarity within groups and disparity between them? Second, to what extent is hostility present in people's feelings toward the other group(s) in the relationship? The degree of affective differentiation between groups and negative feeling directed toward the other group(s) represent the two elementary pieces of information that we need to assess one group's emotional disposition toward another. How those dispositions are exchanged between unequal groups throws light on their use and meaning in the interplay of dominance and subordination. Of particular importance is the question of which group—dominant or subordinate—seems more wedded to intergroup amicability or further advanced into the realms of hostility. Third, how do the contending groups define their interests? Do dominant groups tend to portray status quo arrangements as being in the common good, or do they identify distinct group interests? Do subordinates accede to the views of the dominant group? And if they mount a challenge, what form does it take? As we examine the admixture of feelings and perceptions that characterize race, gender, and social class relations, we can identify the place of hostility and amicability in the conduct of different kinds of unequal relations.

Data:
Intergroup Feelings

I begin by examining the intergroup feelings that characterize race, gender, and class relations. After describing my measures, I examine the

data for each intergroup relationship, in turn, and I then draw out the general patterns across the three cases.

Measures

I measured people's feelings toward their own and other groups with two sets of questions that were contained in a booklet that respondents filled out privately during the interview. Interviewers introduced the booklet as follows:

The questions in this booklet ask about various feelings people might have toward different groups. On the first page, you are asked generally how warm or cold you feel toward different groups. The warmer you feel toward a group, the higher the number you should select from the scale at the top of the page. The colder you feel, the lower the number you should select. Don't feel you have to spend a lot of time on this. Again, we're just interested in your first reactions.

Interviewers were instructed to hand the booklet and a pencil to the respondent, to busy themselves with other tasks while the respondent filled out the booklet, and to put the booklet in an envelope when the respondent handed it back.

On two separate pages in the booklet (back to back) were two sets of questions of the following form:

In general, how warm or cold do you feel toward ———?

In general, how close do you feel to ———?

The stimulus-objects were poor people, working-class people, middle-class people, upper-middle-class people, upper-class people, women, men, blacks , and whites , in that order for the warm/cold questions, and rearranged as gender, race, and then class groups for the closeness questions. For both sets of questions, subordinate groups preceded dominant groups as stimulus-objects within each relationship, so that respondents would use subordinate groups as the anchor. The scales that accompanied these questions at the top of the page were marked in single digits from 1 to 9 going across the page: 1 was identified as VERY COLD and NOT AT ALL CLOSE for the respective sets of questions, 5 was identified as NEITHER COLD NOR WARM or NEITHER ONE FEELING NOR THE OTHER, 9 was identified as VERY WARM or VERY CLOSE, and the other numbers were left unmarked.

These questions are modeled, with several modifications, on the thermometer questions used repeatedly in the National Election Studies conducted by the Inter-University Consortium for Political and Social Research at the University of Michigan (NES 1952–1990 Cumulative Data File Codebook). Like the thermometer items, these questions rely on

numbers rather than words as the vehicle to communicate feelings, thereby avoiding the difficulty that respondents might have in using words to articulate their feelings accurately. The use of numbers seems to require a less complex transliteration of a nonverbal phenomenon. The thermometer scale relies on the analogy with a thermometer to assist the respondent, but I used a shorter numerical scale because I feared that the use of 100 points might convey an expectation of precision that would lead the respondent to become more self-conscious and guarded. A relatively crude numerical scale, with higher numbers corresponding in the intuitively easiest way with warmer or closer feelings, should maximize respondents' ability to express their feelings with as little self-consciousness as possible.

The questions have four features to alleviate possible social-desirability pressures. First, and most important, the presentation of the questions in a self-administered booklet made people's responses private from the interviewer, thus removing the most immediate potential stimulus for social-desirability responses. It is hard to assess how much or what the respondent assumes about the attitudes of the female stranger who interviews him or whether the respondent cares, but whatever pressures exist to impress another should be more potent when face-to-face with a live person than vis-à-vis an anonymous researcher at the other end of a computer somewhere. Second, reliance on numbers rather than words for responses avoids the social overtones that adhere to many words associated with emotions. By intercepting responses without using the language of social discourse, any pressures stemming from that source are kept more at bay. Third, by asking respondents about their feelings toward each group separately, they are spared the invidious task of drawing explicit comparisons between different groups. Although comparisons are still implied in the sets of questions (and are of primary interest in our analysis), they are broached with more delicacy than would be the case with explicit comparisons. This, too, should make respondents more relaxed in answering the questions. Finally, I used balanced questions with graduated response-options in order to avoid loading the questions in one direction or the other.

In their content, my measures focus on the two affective responses that are most fundamental to discussions of intergroup attitudes—feelings of like or dislike, and feelings of closeness or distance. Although the former presents clear positive and negative poles (warm versus cold), the latter has no clearly negative pole but instead strikes at the degree of affinity someone feels for her own and other groups. These two kinds of feelings are very similar, but their particular nuances may elicit subtle differences. In particular, if people tend to avoid the expression of hostility, they may be more inclined to give responses that suggest a feeling of closer

affinity to their own group than an outright preference for their own group on a positive-negative dimension.

With both kinds of feelings, I am interested in two aspects of their expression. First, and most important, I gauge the extent to which affective boundaries are drawn between groups. Are groups solidified by emotional bonds within ranks and ruptured from other groups by affective disparity? This is measured by examining the difference between the feelings respondents express for their own group and for the other referent group. Each respondent's warmth- or closeness-score for the other group was subtracted from her score for her own group. With 9-point scales, difference-scores could range from +8 (strong preference for one's own group) through 0 (same score for both groups) to –8 (strong preference for the other group).^[4] These scores are collapsed into three categories in tables 5.1 (for race), 5.3 (for gender), and 5.5 (for class), to show the percentages of each group who (1) feel warmer or closer to their own group, (2) do not differentiate in their feelings toward the two referent groups, and (3) say they feel warmer or closer to the other group.

Second, I look behind the issue of affective differentiation at the absolute level of feeling respondents express for other referent groups. This gauges the way their emotional dispositions toward other groups are cast. The 9-point scales for warmth and closeness were each collapsed into three categories: negative feelings (scores of 1–4), neutral feelings (score of 5), and positive feelings (scores of 6–9). Those data are presented in tables 5.2 (for race), 5.4 (for gender), and 5.6 (for class).^[5]

Results

For each intergroup relationship, there are two tables presenting data on, first, the affective boundaries that lie between groups and, second, the degree to which hostility or affection is the emotional currency that is being tendered. I examine race, gender, and class, in turn, before drawing out general patterns.

Race

Race relations present our best evidence for reciprocated hostility between groups. Even here, however, the hostility is muted. Table

[4] The difference-scores for warmth and closeness are highly correlated. The Pearson's r's (with missing data excluded) are as follows: for race, .77; for gender, .59; and for class, they range from .42 to .67 across the ten possible class comparisons.

[5] In tables 5.1 through 5.6, respondents with missing data or who said "don't know" are excluded. Such respondents comprised about 3–5 percent of the total sample on any given measure.

5.1 presents the percentages of whites and blacks making different kinds of affective distinctions between the races, as well as the mean difference-scores for each group. Table 5.2 presents the percentages of whites and blacks expressing negative, neutral, or positive feelings toward each other, along with the mean warmth and closeness scores.

Whites and blacks almost mirror one another in the affective distinctions they draw along racial lines. No blacks and virtually no whites express greater warmth or closeness toward the other racial group: affective differentiation ranges from neutrality to preference for one's own group. In feelings of closeness, own-group preference predominates, with about 60 percent of both whites and blacks saying they feel closer to their own group.^[6] In terms of warmth, where the scale contains a clear negative pole, there is a little more reticence about expressing a preference for one's own group, especially among whites: 45 percent of whites and 56 percent of blacks feel warmer toward their own racial group.

Expressions of own-group preference are unlikely to be fulsome, however, on either side, and whites are especially inclined to confine their expressions of preference to modest levels. More detailed data for the uncollapsed scales (not presented here) indicate that about half of the whites who express a preference for their own group on either scale are drawing a distinction of only one or two points, and the percentages drawing larger distinctions drop off rapidly. Blacks are just as unlikely to draw gaping distinctions (that is, to have difference-scores of 5–8

[6] Middle-class blacks are slightly more likely than blacks in the poor and working class to say they feel closer to their own race. In the poor and working class, blacks express about the same degree of own-race preference as do whites, but in the middle class, 72 percent of blacks and 64 percent of whites say they feel closer to their own race.

points), but they are a bit more inclined to draw moderately large distinctions: about 80 percent of the blacks who prefer their own racial group have difference-scores of 1–4 points, with differences of 3–4 points occurring with about equal frequency to differences of 1–2 points.

When we look at the absolute feelings that are exchanged between groups (table 5.2), we find that outright negative feelings are expressed by a minority on either side. However, negativism is a little more likely to be expressed in terms of not feeling close than in terms of feeling cold toward the other group, and it is found more among whites than among blacks. Among whites, as many as 25 percent say they feel cold toward blacks and 39 percent say they do not feel close to blacks. Among blacks, only 14 percent say they feel cold toward whites and 20 percent say they do not feel close to whites.^[7] The balance of responses is titled slightly toward neutral expressions among whites and slightly toward positive expressions among blacks. The result is that among whites, professions of warmth still slightly outweigh professions of coldness toward blacks, but expressing a lack of closeness outweighs some degree of closeness by about two to one; among blacks, positive dispositions outweigh negative ones by more than three to one in terms of warm versus cold and by more than two to one on the close/not-close dimension.

The data in tables 5.1 and 5.2 indicate that whites and blacks roughly mirror each other in an exchange of muted hostility. Blacks are somewhat more likely than whites to express a preference for their own group, but a larger minority of whites invokes negative feelings toward the other

[7] Black women express a little more negativism than do black men in their feelings toward whites: 6.5 percent of black men and 17.2 percent of black women say they feel cold toward whites; 8.1 percent of black men and 25.6 percent of black women say they do not feel close to whites.

group. These small differences aside, the prevailing feelings on both sides range from neutrality to restrained affective disparity.

Gender

The affective character of gender relations contrasts sharply with that of race relations. Although men and women also tend to mirror each other's feelings, the tenor of the exchange is markedly more positive. The pertinent data are presented in tables 5.3 and 5.4.

The data in table 5.3 show that few men express an emotional preference for their own sex, in terms of either warmth (under 5 percent) or closeness (14 percent). Instead, most men either draw no affective distinction between the sexes (56–60 percent) or actually state a preference for women (30–35 percent). Thus, men's gender feelings tend to cluster around the neutral point, and deviations from affective neutrality are more likely to extend in the direction of preferring the other sex. In addition, men are likely to draw slightly larger affective distinctions when their feelings favor women: among men favoring their own sex, difference-scores are almost never larger than 2 points, but among men favoring the other sex, the differentiation generally falls in a range from 1–4 points.

Women more or less reciprocate men's feelings, although they are slightly more likely than men to feel warmer toward their own sex (20 percent) or closer to their own sex (32 percent), and they are a bit less inclined to express a preference for the other sex (15–16 percent). Nor do women share men's slight proclivity to draw larger affective distinctions in favor of the other sex than their own: among women, affective distinctions that favor either their own or the other sex tend to fall in the range from 1–4 points.

Not surprisingly, the data in table 5.4 indicate that in the affective

dispositions of each sex toward the other, positive feelings abound. About two-thirds of men and women have positive emotional dispositions toward each other, and negative feelings are expressed by only 3–8 percent of either group.

Thus, the overwhelming tenor of gender feelings is a lack of affective differentiation along group lines and a positive emotional disposition toward the other sex. Feelings of closeness are slightly less likely than feelings of warmth to evince such inclusive feelings toward the other sex, and women are a little more hesitant than men, but the general pattern stands in sharp contrast to race relations.^[8]

Class

Because class relations do not involve a simple dyad, assessment of the exchange of feelings among social classes is necessarily more complex. Table 5.5 presents data on the affective differentiation made by members of the upper-middle class, middle class, working class, and poor between their own class and each of the other classes. Table 5.6 presents data on the level of warmth/coldness and closeness expressed in people's feelings toward each of the other classes. Upper-class identifiers are excluded from these tables because there is an insufficient num-

[8] Black men and women are more likely than their white counterparts to express an affective preference for and positive feelings toward the other sex, but this tendency is especially pronounced among black men. Most notably, among black men 63 percent feel warmer toward women than toward their own sex, and 65 percent feel closer to women (as compared with 33 percent and 27 percent of white men expressing those feelings). Among black women, 25 percent feel warmer toward men than toward their own sex (as opposed to 13 percent of white women) and 33 percent feel closer to men (as compared with 14 percent of white women). Thus, black women do not display the slight tendency observed among white women to prefer their own sex over men; but, at the same time, the asymmetry between black men and black women in the degree to which they favor the other sex is considerably larger than it is among whites, with black men leading even more in the expression of inclusive feelings.

ber of them to allow for reliable inferences (see chapter 3), but the upper class is retained as a stimulus for those who identified with other classes.

The amount of affective disparity expressed between classes and the level of warmth/coldness and closeness expressed toward other classes are both highly variable, being dependent on two main factors. First, affective distinctions and negative feelings are more in evidence when people are contemplating classes higher than their own than when they regard those lower than their own. Second, such feelings come more to the fore as the status-distance of the referent class increases. These effects are compounded in the overall pattern of class feelings, to which

I now turn. Examples are drawn from the warm-cold data, but note that affective disparity and negative feelings are both more likely to be evinced in terms of degree of closeness than in terms of coldness versus warmth, especially when the referent class is lower than one's own.

The heightened sensitivity of people to the gap that separates them from higher classes than from the gap between their own and lower classes is illustrated neatly in the affective distinctions drawn by the middle class. When the upper class is the referent, about one-half of the middle class express warmer feelings toward their own class and virtually no one states a preference for the higher class; but in relation to the poor (an equally distant, lower class), only 28 percent of the middle class express a preference for their own class, and this is almost matched by 22 percent who express warmer feelings toward the lower class. The more generalized display of this phenomenon is in the marked asymmetry between the intergroup feelings expressed by lower and higher classes in relation to one another. Lower classes express greater affective disparity from higher classes than is reciprocated, and this asymmetry is more pronounced than that observed between unequal racial or gender groups. For example, only 27 percent of the upper-middle class feel warmer to their own class than to the poor, and this is matched by 28 percent of their class-compatriots who profess warmer feelings toward the poor. But among the poor making the same affective comparison, 55 percent feel warmer to their own class, and fewer than 8 percent express greater warmth toward the upper-middle class.^[9]

This asymmetry is also evident in the exchange of feelings between

[9] The asymmetry in the feelings exchanged between higher and lower classes is more sharply pronounced among blacks, even though their class identifications effectively span only the distance from the poor to the middle class. In the exchange of feelings among those three classes, blacks in the working and middle classes are less likely than their white class-peers to express own-class preference vis-à-vis lower classes, and are more likely to prefer the lower class; in addition, there is a (weaker) tendency for blacks in the poor and working class to be more likely than their white class-peers to feel warmer and closer to their own class than to higher classes. For example, in differentiation in feelings of closeness between the poor and the middle class, only 17 percent of the black middle class prefer their own class whereas 43 percent prefer the poor (as compared with 42 percent of the white middle class who prefer their own class and 12 percent who prefer the poor); among the poor, 58 percent of the blacks and 43 percent of the whites prefer their own class over the middle class. These distinctive class feelings among blacks are most readily interpreted in terms of the intensity of racial identity combined with the overrepresentation of blacks among lower classes and their marginal presence in the middle and upper-middle classes: blacks in all classes may feel more distant from the middle and upper-middle classes because these classes are predominantly white, whereas they feel closer to the poor and the working class because they have a stronger black contingent and because they identify their historical roots in those classes (these issues are discussed in greater detail in Jackman and Jackman 1983, 46–53).

proximate classes, where class feeling is weakest. Between any pair of proximate classes, own-class preference is generally expressed by about one third of the lower class, whereas affective neutrality prevails much more heavily in the higher class (only between the poor and the working class do feelings of warmth fall closer to the neutral point on both sides). Expressions of affective disparity increase steadily among both lower and higher classes as the status-distance of the referent class increases, with lower classes maintaining greater affective distinctions than are reciprocated.

Thus, affective differentiation between the classes encompasses a wide range. In some class comparisons, there is a virtual absence of affective disparity (as when higher classes regard proximate lower ones). This is akin to men's feelings vis-à-vis women, except that men's feelings show more of a tilt toward an actual preference for the subordinate group. At the other extreme are the class feelings expressed by the poor and working class vis-à-vis the upper and upper-middle classes: here, the extent of affective disparity exceeds that expressed by blacks vis-à-vis whites. The response of the upper-middle class, however, is more tempered than the response blacks get from whites. The size of the affective distinctions that are drawn between the classes also varies widely, in tandem with their frequency: the smallest distinctions are comparable to those expressed by men vis-à-vis women and the largest are about as large as or larger than the racial distinctions drawn by blacks.

The absolute level of people's affective dispositions toward other classes reflects the same mixture of elements (see table 5.6). In feelings toward the upper and upper-middle classes expressed by the working class and the poor, feelings of warmth generally outweigh feelings of coldness by a factor of two or three to one, with about one-third to one-half holding positive dispositions. This is similar to blacks' level of warmth toward whites. However, in the emotional dispositions of the upper-middle class toward the working class and poor, there is an even stronger positive bent: well over half of the upper-middle class profess feelings of warmth toward the working class and poor, and only a negligible percentage (5–9 percent) express negative feelings. Positive feelings are generally heightened between more proximate classes, although they usually fall short of the level of positive feeling expressed by men toward women.

General Patterns

Two general patterns are present to a greater or lesser degree across all three intergroup relationships. I discuss each pattern in turn and then consider the sources of their varied expression from one relationship to another. Figure 5.1 provides a quick reference for this discussion, by summarizing the differentiation in feelings of

warmth expressed between whites and blacks, men and women, and the poor and the upper-middle class.

First, and most important, there is a general avoidance of hostile feelings by both dominant and subordinate groups. Even in the staunchest examples of affective disparity (as in the feelings of the poor and working class vis-à-vis the upper-middle and upper classes, and in the feelings exchanged between blacks and whites), one-third to one-half of the members of those groups refrain from socially divisive feelings. In addition, the affective distinctions that are made are most likely to be small or moderate in size—gaping affective distinctions are rare. By the same token, most respondents stay away from the negative side of the two scales in expressing their affective dispositions toward other groups.

The general avoidance of inflammatory feelings is further demonstrated in the amplified expressions of own-group preference and slightly freer use of the negative side of the scale for other groups when respondents were asked about degree of closeness to each group rather than feelings of warmth versus coldness. The lack of a clear negative pole in the former scale gave respondents an opportunity to express a feeling of closer affinity to their own group without seeming to disparage the other group. The pattern of responses in all groups was affected at

Figure 5.1.
Summary of affective differentiation between groups: Percentages of whites and blacks, men and women, and the
upper-middle class and the poor expressing warmer feelings toward their own group, neutral feelings, and warmer
feelings toward the other group. Based on figures from tables 5.1, 5.3, and 5.5.

least slightly by that delicate distinction (even in the gender relationship), but dominant groups appear somewhat more sensitive to this than do subordinate groups.

The second general pattern that holds to some degree across all three intergroup relationships is a slight asymmetry between the feelings of dominant and subordinate groups. Subordinate groups run a little ahead of dominant groups in expressing a feeling of preference for their own, and the affective distinctions they make are likely to be a bit larger than those made by dominant groups. This asymmetry is never dramatic—the feelings that are exchanged between contending groups are generally in the same vein—but dominant-group members seem slightly more reticent than subordinates about drawing affective boundaries along group lines. Interestingly, the asymmetry is more noticeable on the warm-cold scale (with its clear negative pole) than it is on the closeness scale: in the safer context of affinity-without-disparagement, dominant-group members come nearer to matching subordinates in expressions of intergroup disparity.

The general restraint that is shown in the expression of intergroup feelings suggests that pressures on both sides of unequal relationships mitigate against flagrant emotional divisiveness. That dominant groups show particular restraint probably reflects a slightly different balance of pressures to which they are subject. To begin, dominant groups may be less sensitive than subordinates to the experiential disparity that exists between the groups. The experience of deprivations may be a more compelling nudge to awareness of disparity than is the enjoyment of benefits. It is easier to overlook or minimize the rude division between unequal groups when the arrangement delivers benefits rather than deprivations. By the same token, as dominant groups enjoy the sweet benefits of harmony within the status quo, divisiveness looms more unequivocally as a risk that could lead to a narrower circumscription of those benefits or even their loss. For subordinates, however, the costs of conflict (which are still considerable in view of their weaker position) are weighed against the daily price they pay for their integration in the status quo as incumbents of an inferior position.

This discrepancy in the pressures felt on either side of an unequal relationship is exacerbated in class relations, where the upper-middle class has been the tardiest in reciprocating the affective disparity expressed by the poor and working class. I suggest two main factors are at work here. First is the importance of self-identification, rather than imposed group definitions, for the existence of social classes as status groups. Dominant groups should be especially aversive to reciprocating affective disparity vis-à-vis groups whose very existence they would prefer to obscure. A second factor is the step-wise structure of class relations,

which has important implications for interclass contact and communication. People in the upper-middle class live far removed from the day-to-day existence of those in the poor and working class; this gives those distant subordinates the same opportunity as blacks to develop separate affective bonds, at least vis-à-vis classes located at the more distant extremes. But the feelings of the poor and working class toward higher classes are fragmented, with less disparity felt vis-à-vis more proximate classes, with which they are more likely to have interclass contact. For their part, those in the upper-middle class also have relatively frequent contact with more proximate subordinates in the middle class, which conditions that relationship more amicably. The limited reach of their downward exposure to those whose interests are only slightly estranged from their own may leave them less keenly aware of the depth of disparity felt by people in classes that are further removed. The buffering action of the middle class leaves the upper-middle class less directly exposed to the more divisive feelings that lie further below and thus it obscures their perception of those feelings. And at the same time, people in the upper-middle class are prevented from allying their feelings unequivocally against those beneath them because they still turn to face a class yet higher than their own, from whom they also harbor a feeling of disparity.

The exchange of intergroup feelings comes closest to reciprocated disparity in race relations, where physical separation of the groups is the sharpest, and where the dyadic structure leaves communication between groups uncomplicated. Because of the pervasive spatial segregation of racial groups, whites have forfeited the opportunity to penetrate and condition the emotional vulnerabilities of blacks. At the same time, the dyadic structure of the race relationship leaves the members of the dominant group with no buffer between them and restive subordinates. Thus, strict physical separation of the groups gives subordinates more opportunity to disentangle their emotions from the grasp of the dominant group; and because there are no buffering groups, the dominant group receives direct delivery of subordinates' feelings and may also feel beleaguered more quickly when confronted with feelings that are less than amicable. In such a context, whites have come closer to reciprocating the affective disparity expressed by subordinates and have slightly outstripped blacks in expressing negative feelings.

The most amicable exchange of feelings is found between men and women, where intimate intergroup contact is the norm, and where the dyadic structure of the relationship makes communication between groups straightforward. In this situation, men's and women's lives are emotionally intertwined, and without any other pertinent groups to create interference, each side can read the other's mood relatively easily.

Even here, however, there is a slight asymmetry between men's and women's feelings, with men's feelings being a bit more eagerly inclusive across group lines than are women's.

These patterns all suggest that no group in an unequal relationship seems predisposed to hostility, but subordinate groups are a little more inclined to gravitate toward feelings of intergroup disparity, albeit tentatively. As they retreat emotionally behind group lines, subordinates tend to stay away from expressions of outright hostility toward the dominant group and instead curb their feelings to a preference for their own group. Because the data are cross-sectional, they do not offer direct information about which side introduces affective disparity into the relationship, but it is subordinates who seem to be at the vanguard in every case, even in the prevailingly amicable gender relationship. The most plausible interpretation is that it is subordinates who introduce affective disparity and dominant groups who try to prevent or restrain such a movement.

The success of dominant groups in that regard depends on their ability to entangle subordinates emotionally within a social structure that binds the groups together in close physical proximity. That entanglement is severed when the groups are spatially separated. Without repeated physical contact, emotional ties across group lines are harder to maintain on both sides, but it is subordinates, as they experience the daily sting of deprivation, who are more inclined to emotional withdrawal behind group lines. The extent to which dominant groups feel the direct blast of subordinates' change in mood depends on whether their relationship is dyadic or whether there are intermediary groups who buffer communications between top and bottom and fragment the exchange of feelings. But once introduced, hostility changes the tenor of the relationship unavoidably, and dominant groups are shunted fitfully into the game of muted conflict.

Data:
The Definition of Group Interests

As before, I describe my measures and then proceed to an examination of the way group interests are perceived in race, gender, and class relations, in turn. To what extent do the groups on different sides of those relationships define their interests as mutually opposed? Or do they take a more inclusive view by seeing their interests as either communally served or communally hurt by the status quo? After delineating the exchange of perceived interests in each case, I turn to more general questions about the proclivities of dominant and subordinate groups to portray social arrangements in divisive or integrative terms as they seek to defend or challenge those arrangements.

Measures

To measure the perception of group interests, I used specific policies and practices as the stimuli, rather than asking about the global interests of groups. The latter is a more abstract concept and, in view of the particularistic quality of public political discussion, respondents are most likely to have contemplated the interests of their own and other groups in relation to specific policies and practices. By asking about the implications for relevant groups of well-known, specific policies or practices that have high symbolic value in the conduct of a relationship, it should be easier for the respondent to understand the question and to consider its meaning in concrete terms. In this way, the questions also avoid hazy generalizations by anchoring people's responses to a more tangible reality.

The items focused on two policies or practices for each intergroup relationship, presented as follows:

People have different opinions about who benefits and who is hurt by certain policies and practices in America today. I am going to read you a short list of such policies and practices and ask you for your opinion about who benefits and who is hurt by each of these.

If blacks generally live in black neighborhoods and whites live in white neighborhoods, who benefits from this—whites, blacks, everyone, or no one?

Who is hurt by this—whites, blacks, everyone, or no one?

How about private ownership—rather than government ownership—of business corporations? Who benefits from private ownership of business corporations—upper-class people, upper-middle-class people, middle-class people, working-class people, poor people, everyone, or no one?

Who is hurt by this? [same response-options presented on show-card]

Who benefits from the practice of women staying home to take care of children and men working to support the family financially—men, women, everyone, or no one?

Who is hurt by this—men, women, everyone, or no one?

If blacks generally go to some schools and whites go to other schools, who benefits from this—whites, blacks, everyone, or no one?

Who is hurt by this—whites, blacks, everyone, or no one?

Who benefits from the tax policy that reduces taxes for some types of business investments—upper-class people, upper-middle-class people, middle-class people, working-class people, poor people, everyone, or no one?

Who is hurt by this? [same response-options presented on show-card]

There are more women in certain kinds of jobs such as nurses and secretaries while there are more men in other kinds of jobs such as engineers and doctors. Who benefits from this—men, women, everyone, or no one?

Who is hurt by this—men, women, everyone, or no one?

All these policies and practices are highly salient to the conduct of the intergroup relationship in question. Five of them have also been the focus of considerable public discussion; the only one that has not is private ownership in industry, but since this is the core tenet of capitalism, it is implicated in much public discussion of subsidiary economic issues and is too central to our system of economic organization to be omitted from any inquiry into perceptions of class interests. At the same time, the wording of all the items was kept as neutral as possible, avoiding such value-loaded terms as "racial segregation" that might trigger defensive or pat responses. The purpose was to focus on the prime substance of specific arrangements without appearing to cast them in a negative or positive light.

For similar reasons, the question of who, if anyone, is hurt by each arrangement was separated from the question about beneficiaries. I wanted to leave respondents with as many degrees of freedom as possible in answering the questions. Responses to the items in each pair were combined afterwards to assess the degree to which people's perceptions of group interests fall into categories that have been discussed most often in the literature. Respondents who think the dominant group benefits and the subordinate group is hurt are taking the classic view of mutually opposed group interests in which the status quo serves the interests of the dominant group at the expense of subordinates. Those who think everyone benefits and no one is hurt are interpreting group interests as being communally served by the status quo . People who think no one benefits and everyone is hurt are of the logically contrary opinion that the status quo universally hurts all the groups involved. Many other combinations of responses are also possible, which may suggest a partially formed view of distinct group interests (such as "the dominant group benefits but no one is hurt"), a neutral view of the matter (such as "no one benefits and no one is hurt" or "everyone benefits and everyone is hurt"), or an idiosyncratic view that is less readily interpretable.

Results

The pertinent data are presented in tables 5.7 (for race), 5.8 (for gender), and 5.9 (for class). Responses to each pair of questions about who benefits and who is hurt by each policy or practice were combined to yield the categories listed down the left side of each table, with the "who-benefits" response given on the left of the pound-sign (#) and the "who-is-hurt"

response given on the right of the pound-sign. The list begins with the most divisive perception, that the dominant group benefits and the subordinate is hurt by a specific policy or practice—that is, that the interests of the groups are mutually opposed. For class relations, the existence of multiple groups means there is no single place to draw the line, and so three different kinds of responses are presented as alternative expressions of mutually opposed interests: (1) that the upper and/or the upper-middle class benefit and the poor and/or the working class are hurt; (2) that the upper and/or the upper-middle class benefit and the poor, the working class, and/or the middle class are hurt; and (3) that the upper class benefits and "everyone" is hurt (I assume that, since the upper class constitutes such a tiny minority of the population, "everyone" in this context was shorthand for "everyone else").

The next row presents a muted version of a zero-sum view, which is that only the dominant group benefits, but no one is hurt. The next three categories involve different ways of denying special group interests: in turn, that the policy is detrimental to everyone's interests ("no one benefits and everyone is hurt"), that it is beneficial to everyone's interests ("everyone benefits and no one is hurt"), or that it is neutral in its impact (almost all of these responses were that "no one benefits and no one is hurt"). The next category ("don't know who benefits and don't know who is hurt") involves a more passive kind of neutrality, stemming from uncertainty or ignorance about the issues.

The final category, labeled "Other," incorporates an amalgam of dispersed responses. Most of these responses represent slightly different twists of the main categories already listed (for example, "no one benefits and blacks are hurt" or "everyone benefits but women are hurt"), which make sense in their own terms but which do not fall neatly into any of the classic positions. Most of the specific combinations of responses incorporated under the "Other" label were articulated by only a handful of respondents, but a few of them (such as the examples given above) were expressed by a sufficient number of respondents to merit separate recognition: these will be mentioned in context.^[10]

Race

Table 5.7 presents whites' and blacks' views of how racial group interests are served by racially segregated neighborhoods and schools.

[10] The number of different "Other" response-combinations in each intergroup relationship was as follows. For race, there were 21 separate other combinations on each item-pair. For gender, the sex-role items elicited 27 other response-combinations, and the occupational-segregation items elicited 20. For class, where there are many more possible permutations, the private-ownership items elicited 86 separate other combinations of responses and the tax-policy items elicited 75. In each case, no more than 2–8 of the separate combinations attracted more than 15 respondents.

The two sides differ in the degree to which they take a critical view of racial segregation, but plainly divisive interpretations are not popular in either group.

The view that has the plurality among whites (34–38 percent) is that racial segregation in neighborhoods and schools benefits both racial groups and hurts no one. When one takes account of the additional 15 percent or so who think that racial segregation is neutral in its impact and the 6 percent who don't know who benefits or is hurt by it, well over one-half of whites espouse a view of racial segregation that denies that it harms anyone or that it serves any special group interests. Very few whites (7–9 percent) see racial segregation as something that benefits whites and hurts blacks, and even the muted view of special group interests that whites alone benefit but no one is hurt is espoused by only another 1 percent of whites. About 20 percent of whites view racial segregation as harmful to the interests of both groups, and another 5 percent (incorporated in the "Other" category) take the slightly different view that blacks alone are hurt by it, but that no one benefits from it.

Blacks are more likely to see the two racial groups as having mutually opposed interests, but still only about one-fifth to one-quarter take that view. Equally popular among blacks is the less divisive criticism of segregation as detrimental to both groups' interests: about one-quarter or more of blacks believe that racial segregation benefits no one and hurts everyone. Most blacks who gave responses in the "Other" category be-

lieve that segregation is harmful, either to blacks solely or to everyone, with the combination that no one benefits and blacks are hurt being given by about 10 percent. Blacks are about as likely as whites to see segregation as neutral in its impact on racial groups (15–16 percent) or to say "don't know" on the issue. But blacks are considerably less likely than whites to go as far as to believe that segregation bestows benefits in a color-blind way—only 14 percent of blacks think neighborhood segregation benefits both groups and hurts no one, and as few as 8 percent think this about school segregation.

Gender

The muted conflict of race relations stands in sharper relief against the amicable consensus that permeates gender relations. There are no important differences between men and women in the way they see, or rather do not see, gender-group interests.

Approximately one-half of both sexes believe that everyone benefits from and no one is hurt by traditional domestic sex-roles or the sex-typing of occupations. When one also takes account of the respondents who said these gender arrangements have no impact on the interests of either group or who said they did not know (11–12 percent on sex-roles and 20–26 percent on the sex-typing of occupations), a total of about two-thirds of men and women either view gender arrangements as a communally positive good or fail to see any involvement of group interests. Only a small minority see men's and women's interests as mutually

opposed (5–11 percent) or even view gender arrangements as harmful to both groups (4–10 percent).

The amalgamated "Other" responses display the same tendencies toward consensual inclusiveness. The items on traditional domestic sex-roles elicited a larger number of responses that fell into the "Other" category (about 23 percent), which in part reflects respondents who named children as beneficiaries of traditional domestic sex-role arrangements. The majority of the "Other" responses to this item-pair involved perceptions either that no one is hurt by traditional domestic sex-roles or that if women are hurt, they are also among the beneficiaries. A slightly larger portion of the "Other" responses to the occupational-sex-typing items involved allegations that it is harmful to either women or everyone, but only 8–10 percent of all responses fell into the "Other" category on this item-pair.

Class

There is more evidence of conflictive views of class interests than there is of either gender or race interests. The specific pattern of responses varies, however, across the two issues of private ownership of business corporations and tax reductions for investments.

Both class issues elicited the perception of mutually opposed group interests with considerably more frequency than did either gender or race issues. The first two rows in table 5.9 give the percentages of respondents with unambiguous views of class interests as mutually opposed, while the third row presents separately the relatively small percentage of respondents who offered the slightly less clear-cut zero-sum response that upper-class people are the sole beneficiaries and everyone [else] is hurt. When contemplating private ownership, respondents from

different social classes vary somewhat in their tendency to perceive mutually opposed class interests: just over 40 percent of the poor and the working class see private ownership as benefiting higher classes and hurting lower ones, whereas 36 percent of the middle class and 25 percent of the upper-middle class take that view. Correspondingly, the more integrative view that everyone benefits and no one is hurt by private ownership gains in popularity with ascending social class, with the percentages increasing monotonically from under 9 percent of the poor up to 37 percent of the upper-middle class. On the issue of tax-reductions for business investments, more than one-half of every social class see class interests as mutually opposed; and the more integrative perception that everyone benefits and no one is hurt is taken by only 5–6 percent of the poor, working class, and middle class, and by just 12 percent of the upper-middle class. Virtually no one in any social class espouses the view that either private ownership or investment tax-credits are harmful to the interests of all classes.^[11]

Consistent with the greater polarization that characterizes the perception of class interests, almost no one in any class sees either of these policies as neutral in its effects. However, as many as one-fifth of the poor and one-tenth of the working class take a more passively neutral position by saying "don't know." And between one-sixth and one-quarter of each class gave responses that were amalgamated in the "Other" category. A substantial minority of the latter (about one-sixth) involved the perception of mutually opposed class interests in which the middle class was named as a beneficiary rather than being aligned with the poor and working class, and there were many other responses that drew slightly different cut-points in identifying specific classes that benefit or are hurt by these economic practices. As with the patterns in interclass feelings, these data suggest the importance of the step-wise structure of class relations in fragmenting and scattering the ingredients for polarization.

General Patterns

Two general patterns mark the perception of group interests in race, gender, and class relations, and these patterns parallel

[11] On both of the class issues, blacks are somewhat more likely than their white class-peers to take the position that the classes have mutually opposed interests and less likely to take the benign view that everyone benefits and no one is hurt. In people's views of private ownership, it is primarily among the middle and working classes that blacks are more caustic than whites (with differences of about 14–22 percentage points). In people's views of investment tax-credits (where perceptions in general have a more divisive cast), it is among the poor that blacks differ most from whites in their adoption of a class-divisive interpretation (among the poor, 59 percent of blacks and 45 percent of whites take the view that class interests are mutually opposed, whereas no blacks and 9 percent of whites take the position that everyone benefits and no one is hurt). These patterns are discussed in greater detail in Jackman and Jackman (1983, 66).

those observed for intergroup feelings. First, sharply divisive perceptions of group interests are far from being the order of the day. This holds even in class relations, where they put in their strongest appearance. Second, subordinate groups tentatively lead the way in drawing divisive distinctions between the interests of contending groups. Dominant groups show more of a preference for the inclusive view that current arrangements serve the needs of all pertinent groups. As with intergroup feelings, both sides appear to tread cautiously, with subordinates showing slightly less reluctance to express divisiveness.

The presentation of these patterns varies from one intergroup relationship to another. In gender relations, which are marked by intimate contact between groups and permeated by role segregation, divisive perceptions are the farthest removed from everyone's mind. Both sides tend to embrace gender arrangements as serving communal interests.

In race relations, which are pervaded by physical separation of the groups, blacks show more indication of seeing status-quo arrangements in a critical light. But they are as likely to include whites along with their own group as being ill-served by racial segregation as they are to accuse whites of benefiting from segregation at black's expense. The prevailing response among whites is to persist with an inclusive view of racial segregation as being communally beneficial or at least neutral in its effects.

The perception of class interests is the most polarized, but the polarization is fragmented and dispersed by the step-wise structure of class relations and the ensuing pattern of communication among social classes—there is no single rift to clarify polarization. That aside, the pattern of responses in class relations suggests that economic interests have a tangibility that is harder to obscure than are interests based on cultural distinctions (such as in race). This is particularly noticeable on the concrete issue of tax reductions for business investments, with respect to which a majority of every class sees class interests as mutually opposed. On the more abstract issue of private ownership, ascending social class is associated with an increasing proclivity to the integrative view that it serves the interests of all classes, but those in the poor and working class do not see mutually opposed class interests quite as readily as they do on the more explicitly distributive issue of investment tax-credits.

Conclusions

In this chapter, I have addressed the rudimentary questions of how contending groups in the unequal relationships of race, gender, and social class define the boundaries between them, emotionally and perceptually. What part is played by hostility and affection in the definition of group boundaries, and do groups perceive their interests as opposed or shared?

Past research on racial prejudice, political tolerance, and group consciousness has overwhelmingly emphasized hostility and the perception of mutually opposed interests as the key germs in intergroup ideology. Contrary to that emphasis, my contention has been that hostility is dispensable—indeed, an encumbrance—in the dynamics of intergroup inequality. The members of dominant groups are inclined to strike an inclusive tone as they orient themselves toward those from whom they derive everyday benefits. And subordinates restrain themselves from a costly expression of negative affect.

The data in this chapter on intergroup feelings and the perception of group interests suggest there is a strain toward inclusiveness in the emotional and perceptual delineation of intergroup ideology. Most significantly, the pattern of results suggests that the expression of inclusive feelings toward subordinates cannot be construed as an abandonment of the unequal terms of the relationship. Indeed, it appears that dominant groups lean toward inclusive intergroup feelings as part of an effort to integrate subordinates within status-quo arrangements. Their success in molding the feelings and cognitions of subordinates is never complete, but their efforts do not meet with absolute failure in any of the three intergroup relationships. Whether one encounters the near-incorporation of subordinates or their contained estrangement depends on the opportunities for ideological incursion that are offered by the structure of the relationship.

When the normal conduct of the relationship provides frequent one-to-one contact between groups, conditions appear most conducive for luring subordinates into mutual affection and the perception of shared interests. In this paternalistic nest, dominant-group members express affection for subordinates and they define subordinates' needs as being met by current arrangements. Their efforts appear to meet with success. Unsullied by hostility, such a relationship depends on mutual good feeling and shared perceptions of communal interests to embrace the role segregation that sustains the inequality.

When the conduct of an intergroup relationship separates the groups physically, the dominant group forfeits the opportunity to infiltrate the feelings and awareness of subordinates. Under these circumstances, subordinates become more inclined to withdraw their good feeling and to separate their interests from those of the dominant group. However, through sporadic trial and error, they grasp that radical confrontation is not efficacious in changing the system in which they are the weaker party. They come to couch their withdrawal of support in terms that minimize its alienating effect on the more powerful party while providing themselves with an adaptive (rather than a frustrating) interpretation of the world in which they live. Confronted with recalcitrant subordinates,

the dominant group learns that the least costly response is to contain the damage by restraining the element of hostility and continuing to reach out to subordinates with an inclusive appeal. Each side opts to engage the other with persuasion rather than to resort to open hostilities.

However, even as subordinates tread wearily, they still step a little ahead of dominant groups on the path to divisiveness. For subordinates, there is a trade-off between their implicit understanding of the need for restraint and their discomfort with the costs they bare under present arrangements. This formula works in all three intergroup relationships to lure subordinates to the fragrance of challenge. Always with caution and with varying ardor, subordinate estrangement ranges from the tempered recalcitrance of women to reciprocate men's proffered affection to the stronger expressions of group bonds and group interests among the poor and working class and among blacks.

For dominant groups facing challenge, their satisfaction with present arrangements and their understanding of the costs of divisiveness both work to restrain confrontation in their response and to steer them toward an inclusive, persuasive appeal. The specific form their restraint takes depends on the exigencies of the relationship. For example, the tangibility of material differences in interests between social classes makes them more difficult to obscure, but the step-wise fragmentation of class relations makes it easier for the upper-middle class to stay away from negative feelings toward lower classes. Whites, caught in a dyadic relationship from which many members of the other group have withdrawn emotionally, find it more difficult to contain their emotional response, but the cultural basis of their unequal relationship makes it easier to cling to the inclusive perception that racial arrangements do not injure the interests of either group. In both instances, however, the dominant group takes reluctant leave of the comfortable ideological system that was built on an exchange of good feeling and shared perceptions of communal interests. It is prodded into the forum of muted conflict.

Chapter Six—
The Articulation of Policy Goals

Views about policies that affect the relationship between groups are a first-order expression of grassroots political alignments. The choice between egalitarian and expropriative policies defines the fundamental issue of contention between unequal groups. To what extent do groups align themselves on contending sides and pursue policy goals that promote their control over resources?

Policy goals are articulated as part of the dialogue that evolves between unequal groups. In that dialogue, the political posture of each group is not formulated freestyle. Groups maneuver within constraints. Those constraints arise from two sources. First, the day-to-day practice of the relationship with another group generates specific pragmatic and moral exigencies that mold the formulation of policy goals. The institutional means by which the expropriation is practiced and the communication channels that are thus created between groups generate varying opportunities and risks for each of the participating groups. Second, at the same time broader societal pressures bear down on the relationship and steer the moral parameters of discussion. The society in which the intergroup relationship resides is enveloped by a broad moral climate that is the product of the various intergroup relations and institutional arrangements that have shaped its history. Those pressures lend moral credibility to certain ideas at the expense of others and hence steer the direction of policy discussion in each intergroup relationship.

Intergroup policy goals in the contemporary United States are shaded by the normative climate of capitalist democracy, a climate in which egalitarianism has interfaced with individualism. This has had the dual effect of transmogrifying the popular understanding of egalitarianism and of lending a special moral significance to individualism. The specific pres-

sures generated by the practice of race, gender, and class relations are each nested within that broader ideological context.

The chapter opens with a discussion of the meaning and significance of people's intergroup policy views. Those views have been the subject of repeated deliberation in research on racial prejudice, tolerance, gender attitudes (especially "sex-role attitudes"), and class consciousness. And unlike the attitudinal elements under examination in the last chapter, they have also been subjected to extended empirical analysis. I start by considering, briefly, the various concerns that have guided past research and the measurement issues that have confounded that research. I then delineate my theoretical approach, in which I emphasize the pursuit of resources within pragmatic and moral constraints, and I consider the significance of individualism as an overarching moral principle that shapes contemporary policy debates. With those issues delineated, I outline the empirical questions that confront my analysis, before turning to an examination of my data.

The measurement of intergroup policy goals has been snared by a number of issues. In particular, many analysts have drawn a distinction between people's articulation of abstract policy principles and their positions on specific, concrete policy issues. Uncertainty about the resilience and meaning of that distinction obstructs our understanding of intergroup policy goals. Accordingly, the analysis of my data proceeds in two stages. I begin by asking about patterns of support for abstract policy principles. To what extent is there consensual endorsement of egalitarian (or individualistic) policy principles, and how does this vary across the three intergroup relationships of race, gender, and social class? Second, I assess views that are articulated about specific governmental policies. What are people's beliefs—both prescriptive and existential—about governmental intervention to promote affirmative change on behalf of subordinate groups? That is, how much affirmative activity do people think should be taking place on behalf of blacks, women, and lower social classes—and how much activity do they believe is already taking place? How much of a rift is there between dominant and subordinate groups in their beliefs about what should be, and what already is, the thrust of governmental policy?

The Issues

Analysts of prejudice, tolerance, gender attitudes, and class consciousness have each pursued and interpreted people's articulated policy views within their own distinctive framework of concerns. I briefly discuss the varying meaning and significance attached to people's policy views in past research, along with the measurement issues that have snagged that

research. My theoretical approach is then delineated, and I draw out the primary empirical questions that inform my analysis.

Prejudice

For students of prejudice, who have always been motivated by a strong social-problems orientation, whites' behavioral predispositions toward blacks have been a continual target of investigation. When physical segregation remained virtually unchallenged, researchers' policy concerns were primarily occupied by the question of how whites would react to intrusions by blacks in their personal lives—with acceptance or antagonism. But as the nation began to grapple legislatively and legally with racial segregation and the issue assumed increasing public visibility in broad policy terms, scholars enlarged their focus. Increasingly, whites' orientation toward racial integration and blacks' civil rights, in terms affecting blacks as a whole, became an absorbing policy question of immediate practical import. After the violent eruption of black urban discontent in the late 1960s, some scholars also began to ask about the racial-policy views of blacks (G. Marx 1967; Campbell and Schuman 1968; Aberbach and Walker 1972; Schuman and Hatchett 1974; Schwartz and Schwartz 1976; Turner and Wilson 1976; Schuman et al. 1985; Sigelman and Welch 1991). But the primary focus has been on the degree to which whites—the status-dominants—have resolved the "American dilemma" by retracting their support for established patterns of racial discrimination and endorsing affirmative change on behalf of blacks' civil rights.

At a deeper level, whites' responses to survey items about racial segregation have also been monitored as a barometer of the underlying level of racial antipathy. Because of the urgent significance attached to both the policy goals themselves and the deeper antipathies thought to lie beneath, there has been intense debate among scholars about the extent to which whites have abandoned support for segregation. Numerous studies have documented a marked decline in whites' support for racial segregation, in principle, over the past few decades (for example, Greeley and Sheatsley 1971, 1974; Jackman 1978; Schuman et al. 1985). And yet this has not been accompanied by a comparable change in whites' enthusiasm for the idea of specific governmental interventions to promote racial integration (for example, Jackman 1978; Fairchild and Gurin 1978; Sears et al. 1979; Kinder and Sears 1981; McConahay et al. 1981; Schuman et al. 1985; Taylor 1986; Sigelman and Welch 1991). This apparent gap between principle and implementation has left students of prejudice divided and uncertain about both the state of whites' dispositions toward blacks and the future of legislated racial integration.

Tolerance

Policy prescriptions have been the attitudinal element of prime interest to students of tolerance. Indeed, with the exception of Jackman (1977, 1978) and Sullivan, Piereson, and Marcus (1979, 1982), policy prescriptions have been the sole empirical target of research on tolerance. Questions address support for civil liberties and civil rights, and people's responses have been interpreted as indicating the degree to which they adhere to the democratic norm of tolerance. As with studies of racial prejudice, the problem has been couched as one involving dominant predispositions toward minorities, but the construction put on those predispositions is very different from that found in research on prejudice. Far from being a barometer of underlying feelings toward target groups, such feelings are instead presumed to be less than positive, and policy prescriptions are taken to reflect the extent to which people can override their affective dispositions in the way they orient themselves toward political minorities. Thus may people demonstrate their graduation to democratic citizenship.

The norm of tolerance occupies such a vital role in theories of democratic legitimacy and stability that observations of its weak presence have struck scholars as a daunting puzzle. Different assaults on that puzzle have roughly paralleled the lines of debate in the literature on prejudice. Prothro and Grigg (1960) found consensual endorsement of the guiding principles of majority rule and minority rights, when those principles were stated in their most general form, but they found sharply diminished levels of adherence when people were asked about specific policy-situations that embroiled those principles. Discouraged, they ironically concluded that the solvent that stabilizes the democratic system is mass apathy. They and other scholars offered the reassurance that those most actively engaged in the democratic endeavor—social or political elites—were the most likely to apply the norm of tolerance consistently to specific policy-contexts (Stouffer 1955; Prothro and Grigg 1960; McClosky 1964; Converse 1964). Some analysts have optimistically observed increasing levels of tolerance over the past few decades (Davis 1975; Nunn, Crockett, and Williams 1978), but Sullivan, Piereson, and Marcus (1979, 1982) found tolerance was still largely wanting, even among social elites, when they explicitly built affective reactions into their measure and asked people about their policy dispositions toward groups they personally disliked. They concluded that democratic stability is left undisturbed primarily because of the diversity of personal targets among the citizenry.

Gender Attitudes

The division of labor between the sexes is the most visible symptom of gender inequality. Like spatial segregation in race relations, the specifics

of role segregation between men and women have assumed high symbolic significance, and analysts of gender attitudes have had a primary interest in measuring support for that traditional division of labor, in its various manifestations (for example, Mason and Bumpass 1975; Mason, Czajka, and Arber 1976; Thornton and Freedman 1979; Spitze and Huber 1980; Cherlin and Walters 1981; Helmreich, Spence, and Gibson 1982; Thornton, Alwin, and Camburn 1983; Mason and Lu 1986; Simon and Landis 1989). Questions about endorsement or rejection of traditional sex-roles and equality of opportunity for women are often interspersed with, and not sharply distinguished from, questions on people's beliefs about the traits of women and the needs of children. All of these indicators are seen as parts of the psychological nexus that bolsters or opposes the traditional gender division of labor, and, indeed, they have commonly been termed sex-role attitudes.

As I discussed in chapter 1, the prime and sometimes exclusive interest has been in the views of women. This contrasts sharply with research on racial prejudice and tolerance, where it is the views of dominant-group members that are treated as the focal concern. Research on sex-role attitudes has not generally assumed any necessary power relationship between the sexes. Instead, the underlying presumption seems to have been that, since women have participated in the roles that differentiate them from men, those roles must be consistent with their preferences, in a power-neutral sense (for an extended discussion of this point, see Kane 1989). With women held as responsible at least in part for their own fate, the question becomes whether they indicate a ground swell for change or satisfaction with traditional gender arrangements. Empirical studies have suggested decreasing support for traditional sex-roles over the past few decades, and this has usually been treated as an important indicator of contemporary social currents.

Class Consciousness

Policy views are the most common empirical focus in analyses of class consciousness, where they are treated as an indicator of the rational perception and pursuit of class interests. The question has been whether different classes, and especially lower classes, take policy stances that suggest they understand and pursue their interests on distributive issues. This information is assumed to be vital to assessments of the fundamental organizational properties of class, including whether classes exist as social entities. Like women, the working class is asked to demonstrate an open sense of dissatisfaction with current arrangements, but, unlike women's views, the views of the working class are almost always assessed in comparison with those of higher classes. The goal has been to show evidence of either class conflict or benign consensus.

Because of the underlying concern with the existence of politically opposed classes based in distinctive social communities, analysts have attended not only to articulated policy goals but also to existential perceptions about the degree of equality and openness that already holds within the status quo (for example, Mann 1970; Parkin 1971; Huber and Form 1973; Abercrombie and Turner 1982). It is assumed that privileged classes will portray the status quo in benign terms as unbiased and open as they articulate conservative policy goals, whereas lower classes, if anything is amiss in the dominant portrayal, will indicate so with more jaundiced perceptions of current policies as they frame goals that call for affirmative change.

Studies have generally shown modest differences between classes in their perceptions of and prescriptions for socioeconomic policy, leaving analysts to argue about the implications of such differences for the viability of class conflict, and therefore of class itself. A repeated observation has been that when asked about policies as broad abstractions, socioeconomic policy views are more consensual and conservative than they are for specific, concrete issues (Mann 1970; Parkin 1971; Huber and Form 1973; Abercrombie and Turner 1982). Analysts have concluded that at an abstract level, there is a pervasive dominant ideology, but that subordinate classes are more inclined to pull away from the dominant interpretation of social reality when dealing with concrete issues that have definite experiential correlates. Whether the latter tendency heralds incipient class conflict or merely reflects minor quibbling on the fringes of a broad consensus about governing principles of distribution is a matter that hangs unresolved over the literature.

This failure of subordinates to deviate decisively from the dominant view has been interpreted by some as evidence of "false consciousness" (see, e.g., Mann 1970). But, as Mann's article demonstrates, the task of distinguishing empirically between false consciousness and spontaneous consensus is fraught with ambiguities. Besides, if the working class is judged to have fallen short of a rational awareness of its "true" interests, analysts are left with the theoretical problem of explaining how such a lapse in rationality could occur without at the same time undermining the long-term prediction of confrontational politics growing out of a rational processing of class experiences. The injection of false awareness among subordinates raises a theoretical inconsistency that weakens the long-term prognosis of rationally based class conflict.

Overview

Despite their urgent significance for research on race, gender, and class attitudes, people's articulated policy views have been interpreted quite variously from one body of research to another. At one extreme are

policy dispositions taken as a reflection of prejudice, where whites' views are examined to assess their readiness to relinquish irrational, parochial opposition to blacks' civil rights. At the other extreme are class policy views, where different classes (if they exist) are expected to reveal themselves in the articulated pursuit of rationally opposed goals: policy views are scrutinized to ascertain the extent to which classes manifest their Marxian-ordained destiny for head-on conflict. Between these extremes lie the analyses of sex-role attitudes and political tolerance, each of which incorporates idiosyncratic combinations of elements from either extreme. Sex-role attitudes have generally been examined as barometers of a benignly changing climate of opinion about women's roles in society—a climate in which women are presumed to take the lead by indicating freely any changing aspirations. And tolerance is measured to assess the degree to which society has elevated itself above narrow group interests to embrace the concept of universally equal political and civil citizenship—an elevation that is regarded as lying critically in the hands of dominant groups.

All analyses of articulated policy goals seem to share an understanding that what is at stake is the allocation of valued resources. But only analysts of class consciousness treat that allocation in clearly relational terms that recognize that if disadvantaged groups get more, then currently advantaged groups must give something valuable up. This perspective on policy goals casts them as a contest between the have's and the have-not's rather than as a test of the dominant group's readiness to advance itself to a higher moral state or of subordinates' benignly spontaneous election of change. What analysts of class consciousness have also assumed, however, is that if the have-not's are being exploited by the have's, the only rational course ultimately for the have-not's is to issue a head-on challenge to the unequal arrangement. In this framework, only one set of policy goals for subordinates is recognized as a rational expression of their interests: uncompromised confrontation.

Measurement Issues

Although people's policy views have probably been measured more frequently than any other aspect of intergroup attitudes over the last twenty years, doubts have beleaguered that research about what kinds of measures are most appropriate and how to interpret people's responses to various measures. Those doubts generally spin out from one or more of the following three issues.

Abstract Principles and Applied Policy Goals

The distinction between abstract policy principles and specific policy implementation has been a centerpiece of empirical debate in research on intergroup policy views.

The focus of disagreement is about which level of analysis provides the more valid indication of people's policy views. And as with many other measurement debates, the substantive implications are profound: indeed, the discussion turns more on conceptual and interpretive issues than on methodological issues per se.

Prothro and Grigg (1960) were the first to comment on the significance of the distinction between abstract principles and applied policy views. In their well-known article, they found widespread popular support for the abstract democratic principles of majority rule and minority rights, but the level of support dropped precipitously when respondents were asked to react to specific policy situations that involved an application of those principles. The nonequivalence of abstract and applied policy principles has continued to intrigue students of public opinion, especially in the contexts of racial policy attitudes and class consciousness, where scholars have struggled to explain the meaning of this distinction.

Students of racial policy views have repeatedly observed that whites exhibit higher levels of support for the general principle of racial integration than they do for specific governmental policies that would implement integration. Further, whites' opposition to the general principle of racial segregation has been steadily increasing over the past few decades to near-consensual levels while their support for specific governmental policies to implement integration has languished with virtually no change over the same time-period (Jackman 1977; Schuman, Steeh, and Bobo 1985). This persistent gap between abstract and applied policy principles among whites has been a central puzzle that has been the focus of considerable research, speculation, and debate (see, for example, Jackman 1977; Kinder and Sears 1981; McConahay, Hardee, and Batts 1981; Jackman and Muha 1984; Pettigrew 1985; Schuman, Steeh, and Bobo 1985; McConahay 1986; Schuman and Bobo 1988; Kuklinski et al. 1991; Sigelman and Welch 1991).^[1] The assessment of the current state of whites' racial attitudes and the prognosis for their trend-line and future direction have been seen as hinging critically on the interpretation of the discrepancies between abstract and applied views.

Various interpretations have been offered. Jackman (1977) argued that whites' commitment to racial integration is sincere but superficial: in specific contexts, the commitment to racial integration is not deep enough to override other values or interests that come into play. Several other scholars have argued that whites do have a serious commitment to racial integration but that in the applied context, the injection of the

[1] This discrepancy in whites' racial attitudes was the subject of an article in Research News , published for general dissemination by the University of Michigan (Katterman 1987). The article's title captures the state of the literature: "The Attitudes Gap: White Americans Endorse Racial Equality, Yet Show Little Support for Policies to Achieve It."

federal government as the proposed agent of change causes many whites to back off: many citizens lack confidence in the federal government and are unable to endorse policies that rely on that agency for implementation (Margolis and Haque 1981; Kuklinski and Parent 1981). Empirical analyses by Jackman (1981a , 1981b ) turned up no evidence to support that optimistic interpretation. Others have argued that it is the mechanics of some racial policies, such as school busing, that make them distasteful to many whites: whites' genuine desire to see racial integration implemented does not find a satisfactory outlet in existing policy options (Greeley and Sheatsley 1974). Yet another interpretation that puts a relatively good face on whites' racial attitudes is that many whites resist specific intervention by the federal government because they resent the coercive power of government in general or they resent any intrusions on individual liberties (for example, Lipset and Schneider 1978; Stinchcombe and Taylor 1980; Sniderman and Hagen 1985; Taylor 1986; Merriman and Carmines 1988).

Several more skeptical arguments have also been made. Some have argued that the positive sentiments elicited from whites on survey questions about abstract support for racial integration reflect only lip service to democratic platitudes: the underlying resistance to racial equality is manifested as soon as one starts talking about specific policies that would implement racial integration (Crosby, Bromley, and Saxe 1980; Dovidio and Gaertner 1986). Scholars in the "symbolic racism" school have argued that the old, more blatant expressions of racism (such as overt support for segregation) have become socially unacceptable, but the racism itself is deep-seated and finds expression in objections to specific policies that are designed to redress racial inequalities (Kinder and Sears 1981; McConahay et al. 1981; Kinder 1986; McConahay 1986). Finally, I have argued (Jackman and Muha 1984) that many whites oppose racial segregation as a general principle because it violates their reverence for individualism—but, by the same token, they are unable to support governmental policies that would enforce integration, since those too would trample on individual rights. And the commitment to individualism is an integral part of the defense of racial privilege, not an orthogonal value that merely collides accidentally with racial policy goals.

At the same time as students of racial attitudes have debated about whether the abstract or the specific level is the more valid reflection of people's policy views, students of class consciousness have used the discrepancy between abstract and applied policy views as a wedge to identify the limits of consent in the working class. Focusing on the degree to which different social classes show consensual or conflicting views of the social order, they have observed the dissipation of conservative consensus as one moves from the abstract to the specific level in both

perceptions of how the system operates and in economic policy goals. In this research, the dominant class reveals itself fairly constantly across abstract and specific contexts, but subordinates slip from an abstract endorsement of the regime to specific views indicating disaffection. Mann (1970) uses the discrepancy between abstract and specific views to infer that working class endorsement of the status quo reflects a "false consciousness," imposed from above, rather than an internalized set of values. He reasons that dominant values are absorbed by the working class at only the abstract level: in concrete applications, the divergent experiences of the working class lead them to take a more critical view of how the system operates and a more assertive position on economic policies. Without entangling himself in the question of what constitutes "true" or "false" consciousness, Parkin (1971) makes a similar attribution to a "subordinate value system" which he describes as accommodative: although working-class people unthinkingly soak in the broad moral precepts of the dominant value system, the milieu of the local working-class community provides the source for specific perceptions and normative judgments that are more jaundiced. The disjuncture between the two levels of abstraction is left unresolved within working-class culture. Other analysts too have emphasized the duality of working-class political beliefs, in which abstractions endorsing the status quo reside next to specific views and moral judgments that are far more cynical about how the system actually works (Huber and Form 1973; Abercrombie and Turner 1982).

Two questions are suggested. First, is the discrepancy between abstract principles and applied policy views a constant, or does it vary with different kinds of policies or in different intergroup contexts? Second, is there anything in the patterns of discrepancy that can throw light on which level of abstraction carries more significance? Is the specific application confounded with idiosyncratic factors that make it a messy test of policy commitments, or is the abstract level too vague to be meaningful? Perhaps it is only in the abstract that policy principles are unsullied by particulars that create interference. Or is it only when we get to specifics that the logical implications of conflicting interests are manifested?

Agents and Methods of Change

This issue intersects with the one above: what are the legitimate agents of change and the desirable methods of change? Some analysts have argued that, in asking people about their specific policy views, idiosyncracies about the proposed agent or method of change may determine people's responses, irrespective of their views about the general target of change. This issue has been discussed above, and I will not reiterate the various arguments here. The crux of the matter is whether people's disaffection from the agency of change or

fears about the disruption or inconvenience associated with the method of change are confounding our inferences about their preferred policy outcomes.

But what are the policies that can alter the patterns of discrimination that are built into our way of life without at the same time disrupting elements of that way of life? Inconvenience to the more privileged parties seems intrinsic to any effective policy, since the only way to eradicate socially patterned discrimination is to circumvent or counteract it with the introduction of socially "artificial" measures. Policies that are not socially disruptive will by the same token fail to alter the discrimination that is the "natural" product of established social practices.

By the same token, to what agency of change might we turn if we deprive the federal government of that role? No other governmental body has the same breadth of authority or the same powers of enforcement. For that reason it is hard to digest the plea that some people would be more comfortable if local governmental agencies had the responsibility for blacks' civil rights: a preference for local government connotes a preference not just for another governmental agency but for a weaker one. Indeed, in the context of the history of blacks' civil rights, the federal government has been indelibly associated with any moves for affirmative change (from the Emancipation Proclamation on), and any argument that a more local government agency could do the job better flatly contradicts the experience of history. I suggest that what most whites find objectionable in the proposition of federal government enforcement of racial integration is the prospect of enforcement . That flies in the face of individualism and free choice and implies an imposed uniformity of standards from which personal or local deviation is impossible.

Tangible Policies and the Framing of Policy Debate

This measurement issue is more elusive, buried in the depths of our conceptual approach. As empirical analysts, we prefer to focus on policies that have found some tangible expression in the extant political arena—and we ignore those that lie outside. This means that our empirical thinking about policy issues is directed by the way those issues are framed on the political agenda that is the site of our analyses. Examples of this slant can be found in research bearing on class, race, and gender policy views. In analyses of class policy views, there is a plentitude of questionnaire items on support for welfare policies but very few items asking about radical redistribution of income or about the elimination of private ownership. Items about racial policy focus almost exclusively on racial integration and equal opportunity, bypassing the question of racial equality in economic resources, status, or power. And in research on gender policy views, there has been a repeated focus on opinions about the particulars

of gender-role definition along with a dearth of items on male status or power prerogatives.

It is unlikely to be productive to ask people about their support for policies that are completely hypothetical and removed from the realm of everyday political reality. Respondents would have no real-world framework in which to place such questions and hence would have no reality-based cues to interpret their meaning. Under such circumstances, responses would tend to be unstable, the stimulus would fail to have a standardized referent, and for many respondents it would have no referent at all, leaving the researcher with a harvest of haphazard reactions and "don't know's" in lieu of valid information about their policy goals.

For these eminently sensible reasons, empirical analysts contain their attention to issues that have been given life in the extant political arena. And yet, as Bachrach and Baratz (1970) have pointed out (in the context of the empirical debate on how to analyze community power structures), this strategy creates a slippery path whereby we allow our conceptual definition of the range of policy options to be captured within the net of those that already exist. Polsby has rightly argued that we cannot observe that which does not exist (1980), but we must nonetheless apply the conceptual corrective that what exists is but a restricted subset of a much wider theoretical range, a subset created by non-neutral political pressures. The only practical empirical strategy is to devise measures of people's policy goals that draw on known policy options being aired on the public agenda. But as we allow the existing political agenda to determine the focus of our empirical inquiry, we must resist the pressure to define our theoretical parameters within the same narrow scope. As we interpret responses to the policy options that are the necessary focus of our measures, our inferential framework should also incorporate those unmeasured policy options that lie outside the bounds of public political discussion.

The Constrained Pursuit of Resources

The sustained, unequal distribution of resources creates groups that are bound to each other by their mutual relationship. Dominant groups depend on the compliance of subordinates for the continuation of the way of life that they enjoy, and subordinates are obliged to pay heed to their more powerful protagonists if they wish to hold on to whatever limited control over resources they have attained. What this implies is that groups on both sides are bound by the constraints imposed by their mutual relationship as they formulate their policy goals.

Within this framework, the members of both dominant and subordinate groups seek to maximize their control over resources. At the same time, neither side puts all or even part of its current holdings at risk in

the attempt to gain more. The central goal is to maximize: neither side will invest energy in quests for advancement that have few or no prospects of success, and there will be an even greater aversion to goals that may overstep the moral latitude of the relationship and thus incur the wrath of the other group. The malevolent interdependence created by the relational basis of the inequality forces the participating groups to be cognizant of their protagonists as they formulate their political goals.

I view the policy goals that are articulated by the members of unequal groups as the rational expression of their desire to control resources. And because I assume that the rational person avoids danger as much as he or she seeks to improve his or her condition, I cannot pronounce what goals are in the "true" interests of a group without attention to the constraints that are operative. Indeed, I assume that people seek to improve their lot only to the extent that it can be done without jeopardizing what they already have. Thus, I expect that over the long haul, barring occasional miscalculations and mistakes (which become part of the group's corporate history and knowledge), the articulated policy goals of group members will reflect their desire to improve their control over resources, limited by an implicit cognizance of the operative constraints. Those constraints emanate not only from pressures indigenous to their specific relationship with another group but also from broader societal pressures that bear down on the relationship from without. Those pressures affect both the formulation of policy goals themselves and the particular expression those goals take.

Within the confines of the debate between functionalists and conflict-oriented scholars, analysts were obliged to assert prima facie whether a group's "true" interests are served better by support for or opposition to inequality. Anything short of opposition to inequality among subordinates was in danger of being taken as evidence that inequality is benign rather than exploitative—with the only escape hatch being offered by the tenuous interpretation of "false consciousness." But by disentangling conflict from expropriation, we can avoid the bind of interpreting the absence of conflict as either the sign of benign consensus or as an incorrigible puzzle. We are freed from the imperative to make an assumption about whether the true interests of a group lie either in seeking complete equality or in embracing inequality. As Connolly (1972) has convincingly demonstrated, determining a group's ultimate, true interests remains a complicated philosophical issue that presents no ready solution. With the question of expropriation separated from the expectation of conflict, we can view the political goals of unequal groups in a less exalted light.

As people muddle through their daily lives in a reality that holds much uncertainty, they grope to maximize their control over resources that ease the business of living. In that unmeditated endeavor, the knowns

in life that are imposed by the institutional structures and the moral climate within which the person is born provide the guides as to what is possible and what is safe. Attitudes are not contrived actively through the independent thought of individuals, but instead evolve unself-consciously through the collective experience of a group that is bound to another group in an unequal relationship. The individual members of those groups are trying to maximize their control over resources while avoiding danger. It follows that the political goals of individuals will be molded by the pragmatic and moral reality that surrounds them.

In this light, the limits placed on subordinates' goals by the culture and morality that infuses their lives are not a block to their perception of their true interests (as Gramsci would have it). Instead, those limits are an affirmation of the link between pragmatic and moral constraints. The morality that is promoted by dominant groups is not an artificial phenomenon—it springs from the pragmatic reality in which the members of dominant groups find themselves and over which it casts a delicate veil. And, further, that morality itself dominates the political climate within which subordinates grapple for resources, setting bounds that are at once pragmatic and normative on what kinds of proposals have political credibility.

Martin Luther King, Jr.'s recollections of the Montgomery bus boycott provide an interesting example of how these pressures operate to constrain the substance and tone of subordinate challenge. In their initial meeting with the city fathers and bus officials in December 1955, the Montgomery Improvement Association limited itself to three fairly modest proposals:

(1) a guarantee of courteous treatment; (2) passengers to be seated on a first-come, first-served basis, the Negroes seating from the back; and (3) employment of Negro bus operators on predominantly Negro routes. (King, Jr. 1958, 109)

Each of these proposals was explained to the Montgomery city fathers and bus officials, drawing on arguments that might appeal to their specific moral sensibilities:

I made it clear, for instance, that our request for a first-come, first-served seating arrangement, with Negroes loading from the back and whites from the front, was not something totally new for the South; other Southern cities—such as Nashville, Atlanta, and even Mobile, Alabama—followed this pattern, and each of them adhered as rigorously to a pattern of segregation as did Montgomery . . . (King, Jr. 1958, 110)

Despite the modest nature of the proposals, the representatives of the white community remained intransigent:

The commissioners and the attorney for the bus company began raising questions. They challenged the legality of the seating arrangement that we were proposing. They contended that the Negroes were demanding something that would violate the law. We answered by reiterating our previous argument that a first-come first-served seating arrangement could exist entirely within the segregation law, as it did in many Southern cities. (King, Jr. 1958, 111)

Ultimately, segregated seating in buses was completely overthrown by a Supreme Court decision imposed on the community from outside. But within the context of locally based negotiations, the leaders of the African-American community in Montgomery were mindful of the mentality of their white opponents as they formulated their initial demands. Those are the pressures that operate, on a more diffuse scale, in the general formulation of subordinates' policy goals. The issues that are addressed in the public forum are a constrained subset of the full range of possibilities, and that subset then frames policy awareness at the grassroots level.

The morality that infuses an intergroup relationship derives from the specific exigencies imposed by the expropriative basis and structural form of that relationship. In addition, the broad moral climate of the society in which the intergroup relationship is nested is likely to wield some influence. As I discussed in chapter 3, within a particular society each intergroup relationship does not operate in isolation but is affected—sometimes tangentially, sometimes critically—by events in other intergroup relationships that cohabit in the same society. In this sense, there is the potential for cultural diffusion from one unequal relationship to another when they exist in the same society and embroil the same people. The diffusion could take place directly, as individuals apply lessons learned in one intergroup setting to their orientation toward another intergroup relationship. Or the diffusion might take place indirectly, as when a particular intergroup relationship plays an important part in the historical development of a society and thus shapes the general moral code of that society. A potent example of such a process is the unfurling of individualism as a governing principle of the American moral code.

Individualism and Egalitarianism

The principle of individualism is an undisputed bedrock of American morality and, indeed, of the Western capitalist-democracies in general. The centrality to Western ethics of the associated norms of individual liberty, achievement based on individual merit, and "free" competition has been observed by such disparate social analysts as Edmund Burke ([1775] 1954), Alexis de Tocqueville ([1850] 1969), Karl Marx ([1857–1858] 1971, 70–73, 128–131), and Max Weber (1946b ; [1930] 1958). Its pertinence has repeatedly been stressed in anal-

yses of class consciousness (for example, Parkin 1971; Huber and Form 1973; Abercrombie and Turner 1982; Jackman and Muha 1984), racial attitudes (for example, Kinder and Sears 1981; Jackman and Muha 1984; Sniderman and Hagen 1985; Bobo 1984; Kluegel and Smith 1986; Kluegel 1990), and contemporary American culture (for example, Bellah et al. 1985; Gans 1988).

The Western democracies also put a high stake on egalitarianism, or so it would seem, but the pervasive influence of individualistic values has worked to twist the popular meaning of equality to make it consistent with individualism. The "egalitarian" distribution of the vote in the Western democracies has been cast in terms of providing everyone with the same right to vote, not in terms of ensuring that everyone's voice is actually heard (Parkin 1971, 185). People do not have the equal distribution of public goods as their normative ideal—what is instead upheld, in the name of equality, is equality of opportunity. And as Schaar (1967) has deftly observed, equality of opportunity ironically presupposes a competitive world in which resources are allocated unequally. The morality of inequality hangs on the rules by which the race is conducted, on whether all competitors are given an equal chance in the race:

The doctrine of equality of opportunity is the product of a competitive and fragmented society, a divided society, a society in which individualism . . . is the reigning ethical principle. It is a precise symbolic expression of the liberal-bourgeois model of society, for it extends the marketplace mentality to all spheres of life. It views the whole of human relations as a contest in which each man competes with his fellows for scarce goods, a contest in which there is never enough for everybody and where one man's gain is usually another's loss. . . . The fundamental character of the social-economic system is unaltered. All that happens is that individuals are given the chance to struggle up the social ladder, change their position in it, and step on the fingers of those beneath them. (Schaar 1967, 237)

The same sentiment is expressed in a more glowing light by Morris Abram (appointed as vice-chairman of the United States Commission on Civil Rights during the Reagan presidency):

Civil rights have a unique meaning in this country. Elsewhere, in some of those societies where engineering a certain distribution of wealth and goods is part of the state's mission, people have economic rights—the right to housing, health care, and other goods. But civil rights have a different meaning in this country. We live in a constitutional democracy built not on the proposition that each [individual] has a fundamental entitlement to a particular piece of the economic pie, but rather on the concept that it is up to each individual to compete for economic goods, constitutionally protected from interference by guarantees of equal protection under the

law, due process, the Bill of Rights and, most fundamentally, the ballot. (Abram 1984, quoted in Taylor 1986, 203)

In this moral code, competition is hallowed. Indeed, the contractual, competitive work relations that mark capitalism have created a framework in which "freedom" takes on the individualistic meaning, "freedom to compete."

This shift in the meaning of equality to connote equality of opportunity is captured in the term "discrimination," which is the term that recurs most frequently in popular discussions (and in many scholarly discussions) of inequality. It is discrimination that collects all the feelings of moral disapprobation associated with the systematic disadvantages experienced by subordinate groups as a result of inequality. The term "discrimination" conceives of the problem as a violation of individual rights—that individual members of subordinate groups are not given the same opportunity to compete for public goods because they are treated categorically rather than on the basis of their individual attributes. With this normative emphasis, any ideological tendencies that encourage the categorical treatment of group members (most notably stereotypes) are seen as damaging, because they lower the probability that group members will be treated on an individual basis. By the same token, if discrimination is conceived as a violation of individual rights, so too is "reverse discrimination" (Gamson and Modigliani 1987). The speed with which this relatively new term was adopted in common usage reflects the moral precedence of individualism: it is regarded as morally equivalent to give a categorical preference to the members of either dominant or subordinate groups.

Scholars have made different arguments about the means by which individualism enters into the articulation of egalitarianism. Some have taken the more optimistic view that the prominence of individualism is just an unfortunate happenstance that gets in the way of support for specific egalitarian policies (for example, Merriman and Parent 1983; Sniderman, Brody, and Kuklinski 1984; Sniderman and Hagen 1985; Merriman and Carmines 1988). This view has been especially prominent in discussions of whites' lackluster support for interventionist policies on behalf of blacks' civil rights. It is argued that whites' objection to specific racial policies, such as quota systems, is based on the fact that these policies violate individualistic principles: the commitment to individualism is seen as a coincidental obstacle driven by factors independent of racial inequality. Indeed, Sniderman and Hagen (1985, 22) go so far as to say:

Individualism . . . undercuts efforts to realize racial equality not out of hostility to the idea of equality but for just the opposite reason: Individualism is itself a species of egalitarianism. . . . Even were this a world free of preju-

dice and meanness of spirit and ignorance, there would still be opposition to racial equality—an opposition rooted in egalitarianism in the form of individualism.

Other scholars have taken a less benign view of the role of individualism in popular responses to egalitarianism. Students of class have been especially likely to see individualism as an intrinsic part of the defense of inequality in the contemporary Western world (for example, Parkin 1971; Jackman and Muha 1984). Parkin (1971) has argued that the individualistic goals fostered by equality-of-opportunity issues have provided a convenient distraction from core redistributive issues, and that the political agendas of left-wing parties in Western democracies have shifted more toward equality-of-opportunity issues because they are less threatening to established groups and thus present a more fruitful avenue for contention. And some students of racism have also identified the morality of individualism as something that is actively invoked in the defense of racial inequality (for example, Kinder and Sears 1981; Jackman and Muha 1984; Gamson and Modigliani 1987). Kinder and Sears have argued that the traditional American values of individualism and self-reliance form the moral basis for racism and also provide a "safe" outlet for voicing objections to affirmative action programs. For these scholars, the popular whine of "reverse discrimination" reflects the moral envelopment of racism in individualism. The tone of principled moral outrage in the "reverse discrimination" argument is drawn out clearly by Gamson and Modigliani (1987), and they show how exponents of that argument sanctimoniously defended the rights of individuals against incursions on behalf of groups. For example, they quote Cohen (1979, 44) as saying, "Injuries are suffered . . . by individual persons. . . . The sacrifice of fundamental individual rights cannot be justified by the desire to advance the well-being of any ethnic group," and Glazer (1975, 220) is quoted as saying that affirmative action abandons the "first principle of a liberal society, that the individual and the individual's interest and good and welfare are the test of a good society" (quoted in Gamson and Modigliani 1987, 146). In this way, antiracist and "equality" symbolism were adroitly summoned to oppose affirmative action on behalf of blacks (Gamson and Modigliani 1987, 169–170).

I have argued that individualism has been the hallmark cry of dominant groups under challenge in the capitalist-democracies, precisely because it offers a principled way of denying the moral legitimacy of egalitarian demands made on behalf of groups (Jackman and Muha 1984). As Gamson and Modigliani (1987) point out, the symbolism that is thus invoked sounds conveniently like egalitarianism, and it also makes a trenchant appeal to "fairness" norms. Individualism has been especially

appropriate for the defense of class inequality in the "free-market" era because dominant classes have no incentive to emphasize the boundaries between classes. The practice of the expropriation does not require an emphasis on the "groupness" of the arrangements, and, indeed, as I argued in chapter 3, dominant classes prefer to encourage the notion of unsegmented socioeconomic differentiation where people's placement from one generation to the next is seen as fluid. Industrial economies require the recruitment of talent into skilled positions, and individualistic, meritocratic principles are consistent with that requirement.

The shift to equality of opportunity has had such overriding significance in the conduct of political life in the capitalist democracies that it should influence the articulation of policy issues even in relationships in which there is less indigenous pressure for individualism. Ideas that have worked in one context drift comfortably into another: they are the ideas that are the most readily available, and they have demonstrated credibility. However, it is in relationships marked by prolonged challenge from subordinates that individual rights become established as a bedrock moral tenet. This has a profound effect on the shaping of subordinate demands and the tenor of political debate between groups.

Empirical Questions

My argument is that unequal groups pursue resources under constraints. Those constraints emanate from the specific pragmatic and moral exigencies that are posed by the intergroup relationship of which they are a part, as well as from the broad moral code of the society in which the relationship resides. By examining the policy goals of unequal groups in relationships where different constraints are operative, we can assess the degree to which those goals do appear to reflect an ongoing endeavor to maximize, within constraints, control over resources. I have argued that the more distally relations are structured in the day-to-day exchange between unequal groups, the more opportunities there are for subordinates to disentangle themselves from the dominant group cognitively and emotionally and to frame policy goals that aim at affirmative change. However, one set of constraints is traded for another. When the dominant group is faced with institutional arrangements that deliver benefits to them in a cruder, more aggregated fashion and with subordinates who are more restive about those arrangements, dominant-group members move to contain the damage by redirecting people's moral sensitivities into more individualistic channels. The moral emphasis is shifted from the group to the individual, and subordinates are pressed to transform their group demands into individual aspirations. With the political credibility of groups undermined, those demands that continue to be issued by subordinates shift away from the redistributive issues that lie at the core of the intergroup relationship to

the safer moral ground of individual rights and equal opportunity. Indeed, those pressures have been such a central feature of American social history that we may find the morality of individualism is omnipresent in the articulation of policy views in all three intergroup relationships.

This argument directs us to the following empirical questions. What are the policy goals that are voiced by groups as they face each other under different kinds of constraints, and what refrains, if any, are found in all three intergroup relationships? To what extent do the members of each of the groups attempt to increase their group's control over resources or restrict their goals to more of a holding operation? And in their dialogue back and forth, what concessions, if any, do groups on either side appear to grant to their protagonists?

On the dominant side of the relationship, is there a tendency to pursue reactionary policies that would enlarge the inequality, to maintain a conservative stance that aims at keeping things as they are, or to make concessions to subordinates by acknowledging the need for some affirmative change? What kinds of concessions are most likely to be offered, and to what extent do dominant groups attempt to satisfy demands from subordinates with concessions that are symbolic or that involve only minor readjustments, enough to shave off the most offensive aspects of the inequality while leaving the main thrust of arrangements intact?

And on the subordinate side, how much of a tendency is there to demand affirmative changes or to hold on conservatively to current arrangements? How moderately or radically do subordinates compose their demands, and what direction is followed in the shaping of their policy goals? Is there any evidence of a readiness among subordinates to yield some ground to their protagonists by supporting reactionary changes that would further diminish their control over society's resources? Are there circumstances under which subordinates are prepared to suffer some additional symbolic indignity or to advocate some small tangible loss in order to stave off greedier assaults on their resources?

As we evaluate the policy goals of unequal groups, we should differentiate abstract principles from more specific policy views. Are there systematic differences in the way people think about general policy principles on the one hand and specific policy issues on the other? And it is important to assess people's specific policy goals against the existential perceptions of governmental policy that frame those goals. Without knowing the subjective existential framework through which people see ongoing public policy, it is impossible to assess the meaning of their prescriptions about how much activity they think the government ought to be promoting on behalf of various groups. It is also important to assess the extent to which group membership determines political perceptions.

Do the members of a group share the same perception of their fate at the hands of the government? And does the experiential quality of the intergroup relationship create a common estimate across group lines of the current level of governmental activity, or does it lead to sharply disparate existential views between the contending groups?

Data:
Abstract Policy Principles

I begin with the abstract principles that are espoused by contending groups in race, gender, and class relations, and then in the next section move to people's views—both existential and prescriptive—about specific policies. At each level, I start by assessing the opinions that are exchanged between groups in each intergroup relationship, and I then draw out the general patterns that are suggested by comparing the three cases. These data allow us to assess the way policy goals are shaped as unequal groups face each other under constraints.

Measures

Race

I use two items to reflect abstract racial-policy principles. Both items are modeled on those used repeatedly in the National Election Studies:

Generally speaking, are you in favor of racial integration, racial segregation, or something in between? (IF NECESSARY: By integration we mean when things are racially mixed, and by segregation we mean when races are separate.)^[2]

Which of these two statements do you agree with more?: Blacks have a right to live wherever they can afford to.
or
White people have a right to keep blacks out of their neighborhoods if they want to.

These items have been frequently used in the past and need little comment here. Both items clearly address the general principle of racial integration, which has had high visibility and symbolic importance over the past few decades. Note, however, that they do not address the principle of racial equality per se—that principle has been relatively distant from public discussion about racial policy and is thus absent from these items. The second item even idealizes the concept of economic inequality

[2] This item was worded a little differently in the National Election Studies: "Are you in favor of desegregation, strict segregation, or something in between?" (National Election Studies 1991, 274). I modified the wording to reflect the terms that had more currency in the 1970s.

with the qualifier, "Blacks have a right to live wherever they can afford to ": that phrase gives voice to the assumption that not everyone can or should be able to afford the same quality housing, and more obliquely, it also leaves unmolested the expectation that blacks lie considerably behind whites in what they can afford. As I have argued elsewhere (Jackman and Muha 1984), both items implicate the norm of individualism rather than equality. During the height of the civil rights movement, Southern-style segregation of facilities was repeatedly depicted by the media as a serious violation of individual rights; and the right of individuals to buy and sell freely on the housing market without neighborhood interference is another premise of individual freedom in an economy ruled by private ownership.

Gender

There are three items to reflect abstract gender-policy principles, and they focus on the issues that have been aired the most in public discussion of gender policy. The first two items address the classic division of domestic responsibilities between men and women, and the third item focuses on equality of job opportunity for both sexes:

Generally speaking, do you think that doing the housework and taking care of the children should be mainly the woman's responsibility or that it should be the man's responsibility as much as the woman's?

Generally speaking, do you think that the man should have the responsibility for providing financial support for the family, or that the woman should also have some responsibility for providing financial support?

Which of these two statements do you agree with more?:

A woman should be considered just as seriously as a man for any job that fits her interests.
or
Certain jobs should be open to men only.

As with the race items above, these items are straightforward and deal with familiar topics, but a few points are in order. First, the two items on domestic gender-roles necessarily deal with a more personal and privatized aspect of life than do the items for race or class. This is in keeping with the conduct of gender relations, in which home and family comprise the prime theater. But by the same token, this means that many of the central symptoms of gender inequality lie in a zone that has been exempted from the normal purview of public, governmental policy. Thus, they are not as susceptible to policy manipulation. This is intrinsic to the structure of gender relations and indicates yet again the advantages that accrue to dominant groups when they establish social control over subordinates through paternalistic arrangements that hinge on personal, one-to-one contacts.

Second, the two gender-role questions focus on the delineation of gender-appropriate responsibilities rather than whether certain tasks are beyond the ken of either sex. The purpose was to gauge opinions about basic principles of role definition rather than about whether one spouse may help another with gender-defined activities. For example, someone may consider it fine for the wife to take paid employment if she wishes and even to supplement the family pot with her earnings but at the same time consider the financial provision of the family to remain the husband's responsibility . Similarly, someone may think it appropriate for the husband to help his wife wash the dishes or the baby's bottom, but still believe that the cleanliness of house and family is the wife's worry. Public discussion of these issues usually presents them in terms of individual freedom of choice to deviate from prescribed norms if circumstances permit rather than in terms of how responsibilities should be assigned. Some respondents may not have been able to break out of that pattern of thinking, despite the stricter wording used on those items.

Third, the question on housework and childcare poses an egalitarian division of responsibility as one option, while the question on the family's financial support does not pose a completely egalitarian arrangement as an option, just some diffusion of responsibility. Pretesting indicated only minuscule support for an egalitarian division of financial responsibility between husbands and wives: the "egalitarian" response option was diluted to its present wording in order to capture whatever leanings away from traditional male role-definitions could be found.

Finally, the item on equality of opportunity is closer in spirit to the racial-policy items above: it deals with a public sphere more within the purview of government, and at the same time it implicates the principle of individual rights rather than equality. In this regard, note that the item only asks whether women should be considered for any job, not whether they should be hired. This distinction is important: the wording makes it plain that the focus is on procedures rather than outcomes, and endorsing the principle of an open procedure is of course not tantamount to expecting an egalitarian outcome. Indeed, such prominent theorists as Emile Durkheim ([1897] 1951, 385) and John Stuart Mill ([1869] 1970) were moved to advocate open procedures for women's employment opportunities while at the same time reassuring their readers that such procedures would not lead to any serious competition between the sexes in the job market—because, they argued, women's natural aptitudes would steer them into different occupations than men without any imposition of discriminatory rules.

Class

The general policy principle that is most central to class is economic inequality. I have two measures reflecting that principle, both phrased in terms of occupational inequality in earnings:

Thinking about the amount of money people in different occupations earn, do you think there should be a great deal of difference, some difference, or almost no difference in how much people in different occupations earn?

We're interested in getting people's opinions about how much money people in different occupations should make. Please look at these occupations and tell me how much money you think people in each one should make, regardless of how much they do make. Just tell me what you think would be best.

Now, how about a school teacher? How much money do you think an average school teacher should make in a year?

How much money do you think an average assembly-line factory worker should make in a year?

How much money do you think an average doctor should make in a year?

How much money do you think an average business executive should make in a year?

How much money do you think an average janitor should make in a year?

The first of these measures is the simpler test of support for income inequality as an abstract principle. Note that it does not include a response option for complete income equality: as with the item on the gender division of financial responsibility, pretesting indicated that complete income equality is outside the moral bounds of American society.

The second measure relies on a set of questions about desired incomes for a range of occupations to construct a general recommended ratio of incomes of low-level occupations to high-level occupations. Each respondent's concept of how much income inequality there should be was calculated by taking the mean of four recommended-income ratios: janitors to doctors, janitors to business executives, assembly-line factory workers to doctors, and assembly-line factory workers to business executives.^[3] The closer the ratio to 1.0, the smaller the desired income gap. Note that the occupations chosen as stimuli do not represent extremes in the occupational distribution (as, say, between migrant farmworkers and corporation presidents), but are intended to capture opposing poles in the mainstream of the labor force. This measure addresses general notions of the optimal income-gap by asking about concrete salary preferences: it thus allows us to validate the first measure, which relies on broader language to capture the same concept.

Results

Race

Racial segregation is largely repudiated by both blacks and whites—and yet there is a divergence between the two groups in their

[3] Respondents with missing data on two or more of the four individual ratios were excluded from analysis. (Application of a more stringent decision-rule excluding those with missing data on one or more of the individual ratios resulted in a slightly smaller N , but yielded virtually identical distributions.)

enthusiasm for what would seem to be the most obvious alternative, racial integration . That divergence reveals itself suggestively rather than starkly. Table 6.1 displays the percentage distributions of responses given by whites and blacks to the two items on racial integration.

Only about 15 percent of whites endorse either the right of whites to exclude blacks categorically from their neighborhoods or an overall policy of racial segregation; this is not very different from the near-complete lack of support for segregation among blacks. For many whites, however, rejection of segregation is not tantamount to endorsement of racial integration. When forced to choose between the right of whites to exclude blacks categorically from their neighborhoods and the right of blacks to live wherever they can afford, about eight out of ten whites endorse the latter right. But on the other item, which offers the option of "something in between" racial segregation and integration, fully 44 percent of whites are drawn to that response, leaving only 35 percent supporting unmodified racial integration. By contrast, only 15 percent of blacks find "something in between" racial segregation and integration an attractive proposition: 73 percent of blacks support racial integration and 97 percent support blacks' residential rights. Thus, overall, the principle of racial integration receives overwhelming support from blacks, whereas among whites it has frequent but guarded support.^[4]

[4] Data from the American National Election Studies reported in Schuman, Steeh, and Bobo (1985, 74–75, 95–96, 144–145) suggest even higher levels of attraction to the "something in between" response among whites (and also among blacks) than is indicatedin my data. In the NES data, whites' support for "something in between" increased from 46 percent in 1970 to 60 percent in 1978, and blacks' support for this option increased from 18 percent to 39 percent over the same time period. Schuman, Steeh, and Bobo provide evidence from a split-ballot experiment that the NES's placement of the item in the interview schedule immediately after an item on federal school intervention (a particularly sensitive issue) was contaminating responses to the "general principle" item and increasing the number of "something in between" responses. In my interview schedule, the item on general support for racial integration preceded the item on federal school intervention; the only racial items asked immediately before were on respondents' personal preferences for interracial contact in the workplace and neighborhood, which I judge to be relatively innocuous.

The figures in Schuman, Steeh, and Bobo (1985) indicate that the percentage of blacks to be drawn to the "something in between" response has also increased, although it still trails behind the percentage of whites giving that response. This pattern suggests that blacks may have backed away from the goal of racial integration in the face of unrelenting white opposition to racial integration in specific policy contexts, especially in public schools. Blauner (1989) argues that blacks have become disillusioned with the goal of racial integration, and the growing interest in Afrocentrism among blacks in the late 1980s and early 1990s lends support to that view. Confronted with unyielding white opposition, the goal of seeking more contact with whites may have lost its appeal as a means of attaining racial equality. This may be especially true in the context of public schools, where the prospect of sending one's children to learn in a hostile white environment may seem an unsatisfactory outcome.

One might well ask what "something in between" racial segregation and integration could be. What it probably conjures up for most people is the idea of racial integration or segregation being decided on the basis of individual free will, without any external interference—so-called "voluntary integration." Taylor maintains that the principal issue of contention in whites' orientation to policies of racial integration since the 1950s has been over mandatory versus voluntary compliance (Taylor 1986, 24–38, 191–204). As though to underscore that point, Herrnstein (1990, 6) carefully distinguishes between "desegregation" (by which racial segregation is made illegal) and "integration" (by which equal proportionate representation of groups is mandated). It is well known that Northern whites who opposed slavery in the 1860s did not generally have racial equality in mind as the obvious alternative; likewise in the latter half of the twentieth century, it seems that many Northern whites who opposed the de jure segregation of the Jim-Crow South did not object to the de facto segregation that characterized the neighborhoods and schools of Northern cities.

The attraction of whites to "something in between" segregation and integration suggests that their support for racial integration might most generously be described as soft. Indeed, it may not reflect support for the rights of blacks so much as the rights of individuals to live their lives as unencumbered as possible by the stipulations of others. Individuals should be able to buy and sell in a free housing market and they should

also be able to decide whether they want to have contact with another racial group. No one should be prevented from selling his house to whoever can afford it, and by the same token no one should be forced to have contact with another racial group. Such a view falls considerably short of an abandonment of the racial status quo and implies a larger gap between the racial policy principles of whites and blacks than a cursory inspection of the data might suggest.

Gender

The views of men and women on gender policy principles reveal two important contrasts with the racial policy principles expressed by whites and blacks. First, rejection of discriminatory gender-policy principles lags considerably behind the repudiation of racial segregation found among both blacks and whites. Second, men and women do not differ markedly from each other in the gender policy principles they espouse. Men and women exhibit a roughly parallel division of opinion on the subject. The percentage distributions of men's and women's responses to the three items on gender policy principles are presented in table 6.2.

When asked about responsibility for housework and childcare, there is an almost even division of opinion within each gender group. Close to half of each sex endorses the principle of an egalitarian division of domestic labor; almost as many men and just as many women maintain that domestic labor is primarily the woman's responsibility. When asked about whether women should share some of the financial responsibility for the family (recall the conservative phrasing of this question), the distribution of responses from women is very similar to that elicited from them by the item on housework and childcare. Men lag a little behind, with only about one-third supporting even some diffusion of financial responsibility to women and just over one-half seeing the financial support of the family as the man's responsibility alone.

On both of these items, a small minority of each sex (11–16 percent) broke away from the absolutes that were posed and volunteered conditional answers (such as, for the first item, saying it depended on whether the woman worked, for the second item, on whether the husband's income was sufficient, and for both items, on whether there were children). That some respondents were moved to volunteer such statements without any encouragement from the question itself suggests that such considerations feature prominently in people's thinking. Their effect is largely conservative, and their insertion into the issue implies that others' support for egalitarian principles of labor-division may also be fragile, crumbling easily in the face of specific objections that might be raised in individual cases. Few men or women may believe in an egalitarian division of labor between the sexes as a first-order right of women, but

rather as something that is permissible, if individual circumstances are favorable or if it can be done without violating the primary obligations of either gender.

The third item on gender policy principles directly broaches an issue that is hallowed in the Western capitalist democracies—equality of opportunity—and in this respect its focus parallels the race item on equal housing opportunity. This gender item elicits more affirmative responses than either of the other two items on gender principles, but equal job opportunity for women still does not receive as much support as the principle of equal housing opportunity for blacks receives from both blacks and whites. And as with the other two gender items, this issue does not elicit very different responses from women than from men: 57

percent of men and 64 percent of women endorse the principle that a woman should have an equal opportunity to be considered for any job that interests her.

Class

Egalitarian principles seem even further removed from most people's thinking when confronted with the topic of how incomes should be distributed. And although there is some disagreement by social class, the vast majority of people in all social classes subscribe to the principle of income inequality to some degree. The percentage distributions of people's responses about income inequality, by social class, are given in table 6.3.

When asked in the most general terms about whether there should be a great deal of difference, some difference, or (as the timidly egalitarian option) almost no difference in how much money people in different occupations earn, the variation in responses by social class is limited. The main distinction is that, with descending social class, there is some retraction of support for a great deal of income inequality; the position to which people in lower classes remove themselves, however, is most likely to be support for some difference in income. Among those identifying with the upper-middle class, there is a roughly even division of opinion between those favoring a great deal of difference and some difference in incomes. Among the three lower classes, opinion shifts slightly

to favoring some income difference over a great deal of difference by a ratio of about 2 to 1. The idea of almost no income difference remains largely out of bounds in all social classes: that idea is supported by only about 8–11 percent of the upper-middle, middle, and working classes, and 17 percent of the poor. Data from an open-ended probe asking respondents why they took the positions they did indicate that the overwhelming majority of people in all social classes subscribe to some version of an individualist-achievement rationale (those data are discussed in detail in Jackman and Jackman 1983, 206–213).

The pattern of responses to the income-ratio measure replicates responses to the more global item above. Again, we observe a general aversion to the idea of a very small income gap. Note that the most egalitarian ratio-score category in the table is set very inclusively at merely .58 and above, and even so, it attracts very little support. Again, there is a mild relationship between social class and conceptions of the optimal size for the income gap. With descending social class, there is some retreat from the idea of a very large income gap and a timid lead in favoring a small gap. About 60 percent of the upper-middle class and 40 percent of the middle class advocate an income gap in which the highest occupations earn at least three times as much as the lowest occupations, whereas about one-third of the poor and working class support such a large income gap. Conversely, about one-quarter of the poor and working class favor an income gap in which the highest occupations earn less than twice as much as the lowest occupations, compared with only 4 percent of the upper-middle class.

General Patterns

The general policy principles that are espoused in these three intergroup relationships betray only limited support for egalitarian goals and only restricted disagreement between unequal groups on this score. Egalitarianism in the shape of income equality elicits the least support overall, and there is modest disagreement among social classes only in the extent of their rejection of the principle of income equality. In gender relations, both men and women are divided almost evenly between egalitarian and discriminatory principles on the division of labor between the sexes. It is in the context of racial segregation/integration that one finds the strongest endorsement of affirmative principles, but the race items invoke equality of opportunity rather than racial equality per se. Blacks assert their right to equal treatment more clearly than do women or the poor or working class. Whites also reject the idea of the categorical separation of the races more vehemently than men (or women) reject the categorical separation of tasks between the sexes or than the upper-middle class rejects the notion of wide income differentials. But, nonetheless, whites lag behind blacks in their enthusiasm for

the principle of racial integration , resulting in a larger rift in general policy principles between the races than exists between the sexes or classes.

These patterns suggest that subordinate challenge of categorically unequal treatment may push the dominant group away from a stance that morally endorses such treatment, but the position to which the dominant group retreats is likely to be one of equal opportunity rather than egalitarianism pure and simple. Thus, although whites have largely repudiated racial segregation, they have not embraced racial integration either. Many whites appear to be drawn to an alternative that leaves the rights of individuals confined as little as possible. In a similar vein, the arena of gender relations that shows the highest rejection of unequal treatment is that of employment opportunity, the arena that most directly invokes the principle of individualism. Note, however, that the rights of individuals are not enunciated as vehemently in the context of gender as they are in the context of race. Finally, the almost total absence of support for the concept of income equality is explained by the respondents themselves as stemming from a belief in individualistic-achievement principles.

Data:
Specific Policy Goals

To explore the substance of people's specific policy beliefs, I begin with their ideas about the appropriate course for governmental action, along with the existential beliefs that subjectively anchor those prescriptions—their perceptions of the current level of governmental activity. In the second part of the analysis I build that anchor into my assessment of specific policy goals. I measure the extent to which people advocate affirmative change over the perceived current state of affairs, no change (that is, conservatism), or reactionary change back to a lowered level of governmental intervention on behalf of subordinates' rights.

Measures

Given the importance of the federal government as the most powerful institution in American politics, my measures of specific policy goals all focus on federal government intervention. And in keeping with my concern that people's prescriptions be anchored in their subjective perceptions of the current status of policy activity, respondents were asked about both their perceptions of and prescriptions for governmental action.

Specific policy goals for race, gender, and class were measured with a series of item-pairs: for each policy issue, a question about respondents' existential perceptions of governmental activity was followed by a question about what level of governmental activity they would prescribe. The

pairs of items for policies relevant to race, gender, and class were interspersed among one another, and the series was introduced as follows:

People have different opinions about how much the federal government is doing about various things. People also differ about how much they think the federal government should be doing about these things.

First of all, how much do you think the federal government is doing to make sure blacks and whites go to the same schools? Just look at the card and tell me what you think [A lot, Quite a bit, A little, Nothing].

Now, how much do you think the federal government should be doing about this?

The pairs of questions covered the following policy issues for race, gender, and social class:

(RACE)

. . . to make sure blacks and whites go to the same schools?

. . . to make sure that blacks can buy any house on the market that they can afford?

. . . to make sure blacks have the same job opportunities as whites?

(GENDER)

. . . to make sure women have the same job opportunities as men?

. . . to make sure laws are applied in the same way to women and men?

. . . to provide daycare centers for the children of working mothers?

(CLASS)

. . . to make sure that everyone who wants a job can get one?

. . . to give welfare benefits to people who don't have very much money?

. . . to make sure that everyone has at least a minimum income?

To interpret these items, two factors need to be considered: (1) are they straightforward applications of the abstract principles addressed in the earlier questions?; and (2) to what extent do they address the enforcement of individual rights or equality of opportunity versus group rights or the redistribution of resources from one group to another? I consider these issues for race, gender, and class, respectively.

The specific policy questions for race all involve straightforward applications of the general principle of racial integration. These items address three alternative, highly visible policies that seek to implement racial integration. As with the abstract principle of racial integration itself, these policies do not broach the issue of racial equality, but instead are aimed at the protection of individual rights. But, unlike the questions on abstract principles, these questions do ask about enforcement, and as such, they automatically remove some of the laissez-faire connotations that notions of individual rights and personal freedom have for many

people. In addition, two of the items (on jobs and schools) do deviate somewhat from the abstract principles in that they implicitly entangle redistributive issues. Of the three items, that on equal housing opportunity has the closest correspondence with at least one of the abstract-principle items and it is the most oriented toward individual rights. The item on equal job opportunity also broaches the enforcement of individual rights, but for many whites the issue of enforcing equal job opportunity may conjure up visions of racial quotas or preferential treatment of blacks (cutely termed "reverse discrimination" because they violate whites' individual rights), and to the extent that this is the case, this item embroils a significant redistributive element. Finally, the item on school integration undoubtedly elicits thoughts about school busing: school busing was designed to create equality of educational opportunity for blacks and whites, but the implementation clearly involves some redistribution of valued educational resources from whites to blacks.

The items on specific gender policies are tangential to the basic principles of gender-role definition, but one item corresponds quite closely to the abstract principle of equal employment opportunity for women. As I noted earlier, the principal sphere for the enactment of gender-roles is in intimate personal relationships, conveniently beyond the reach of normal governmental activity. Thus, the question of how to enforce any change in gender-role definitions becomes moot. Instead, these items focus on policies that address some of the fallout from gender differentiation: women's inferior employment opportunities, women's inferior legal rights, and women's need for daycare for their children so that they can participate in the paid labor force. The first two of these issues involve the enforcement of individual rights (with the same proviso as noted for the comparable racial-policy items), whereas the third item involves the redirection of resources to assist women's participation in society on the same terms as men.

The three specific class-policy items are essentially soft applications of the general principle of income equality—to the policies of full employment, welfare, and minimum income. The item on guaranteed employment is the most moderate of the three: although full employment has traditionally been a higher priority of the left-wing, it is regarded as desirable by both the left and the right, especially insofar as it is associated with a healthy economy rather than with governmental intervention through public work-programs. The item on welfare programs probably has the most cogently redistributive meaning, both because of the frequent public discussion of welfare programs and because such programs are generally assumed to connote the direct redistribution of resources at the government's disposal. Note, however, that in keeping with the boundaries of public discussion of economically redistributive measures,

none of the specific policies addressed in these items does more than raise the possibility of minimal safety-net measures. Such measures merely aim to soften the harshest punishments of economic inequality: they fall far short of any attempt to eliminate or even to reduce significantly the gap between rich and poor.

Results: Prescriptive and Existential Beliefs

Race

The disagreement between whites and blacks that showed itself only suggestively in the articulation of abstract policy principles becomes a marked and systematic rift when we examine people's prescriptive beliefs about how active the federal government should be in pursuing specific racial policies and also in their existential perceptions of the current level of governmental activity. The relevant percentage-distributions are presented in table 6.4: the first two columns give the percentages of whites and blacks who advocate that the federal government should be doing "a lot," "quite a bit," "a little," or "nothing" to ensure

racially integrated schools, equal housing opportunity, and equal job opportunities for blacks; the second two columns give comparable distributions for whites' and blacks' existential beliefs about how much the federal government is currently doing on those same racial policy issues.

Whites' support for governmental intervention to promote blacks' civil rights tends to be lackluster—blacks are much more likely than whites to think that the federal government should be intervening vigorously in this area. The gap between the races widens still further when one considers that whites also tend to see the current level of governmental intervention on behalf of blacks' civil rights as much more vigorous than do blacks. I begin with the prescriptive beliefs of each racial group.

The policy that elicits the least support from both racial groups is that of racially integrated schools; equal job opportunity elicits the highest levels of support from both groups. But there the similarity between the two groups ends. Among blacks, about two-thirds to three-quarters think the federal government should be doing a lot to promote these policies, and virtually all the remaining blacks think the government should be doing quite a bit. Among whites, only 14 to 24 percent think the government should be doing a lot on any of these policies. On the issue of equal job opportunity, only slightly over one-quarter of whites go to the opposite extreme of advocating little or no governmental activity, but 40 percent of whites would like to see little or nothing done to promote equal housing opportunity, and on the issue of racially integrated schools, over one-half of whites think the government should be doing only a little or nothing.

These data give us an indication of the different directions that the two racial groups take when they apply their abstract policy principles to specific issues. What looks like a reasonably high, albeit guarded, endorsement of the principle of racial integration among whites is unlikely to manifest itself in a desire for strong governmental measures to bring integration about. But among blacks, endorsement of the abstract principle of racial integration is highly likely to be manifested in a strong commitment to specific policies to achieve that goal. Thus, for example, although 80 percent of whites believe blacks have a right to live wherever they can afford, only 18 percent think the government should be doing a lot to ensure that right. Among blacks, however, 97 percent endorse the same abstract principle and 72 percent think the government should be doing a lot to make it happen. Thus, what appear at face value to be relatively modest differences between blacks and whites at the level of abstract support for racial integration translate into a serious rupture in real policy terms.

When we consider people's existential beliefs about the same three racial policies, the rift between the racial groups deepens. Governmental

activity is most likely to be seen on the issue of integrated schools, with equal job opportunity close behind and the least activity being perceived on the issue of equal housing opportunity. This rank ordering is found in the perceptions of both whites and blacks, and it corresponds reasonably well with the relative public attention that each issue has received in the media (note also the relatively large number of whites—18 percent—who say they don't know how much the government is doing to promote blacks' housing rights). All this might suggest that people's existential beliefs about racial policies are cleanly grounded in a common reality. However, on each issue, whites and blacks diverge sharply in their perceptions of what the current level of governmental activity is.

Among whites, the most frequent perception is that the federal government is currently doing "quite a bit" to promote blacks' civil rights, with "a lot" of governmental activity being the second-most frequent perception on two of the three issues. In all, as many as three-quarters of whites believe the federal government is already doing either quite a bit or a lot to ensure integrated schools and equal job opportunity; about one-half of whites see that degree of governmental activity on the issue of equal housing opportunity. Against this existential backdrop, whites' policy prescriptions are revealed in a yet meaner light.

Blacks' estimates of the current level of governmental activity tend to fall far short of whites': only about one-quarter of blacks think the government is already doing a lot or quite a bit to ensure equal housing opportunity or equal job opportunity for blacks, and even on the highly visible issue of school integration, only about 40 percent think that a lot or quite a bit is being done. The most frequent perception among blacks is that the government is doing only "a little," with about one-half of blacks having that perception; over the three policy issues, the perception that the government is doing little or nothing to ensure blacks' civil rights is held by over one-half to over two-thirds of blacks. With these existential beliefs as the backdrop, blacks' advocacy of governmental intervention is revealed even more urgently. The rift that separates their policy views from those of whites gapes open to a chasm.

These existential beliefs also provide, less directly, contextual information by which we may reinterpret the abstract policy principles of whites and blacks that were reported in table 6.1. If many whites believe that much is already being done to ensure racial integration, this suggests that their conception of what constitutes racial integration is much less far-reaching than what many blacks have in mind. This is consistent with my earlier argument that the large proportion of whites who favor "something in between" segregation and integration (when they are given such an option) suggests that many whites endorse a vaguely conceived, laissez-faire situation of "voluntary integration" that would pro-

tect whites' personal freedom of choice without "forcing" them to conform to the requirements of either racial segregation or integration. Such an arrangement would of course fall seriously short of any guarantee of racial integration, but this shortcoming is doubtless less apparent to whites than to blacks.^[5]

In this regard, bear in mind that the urban residential segregation that characterizes race relations in the contemporary United States has a de facto quality that leaves most individual whites feeling personally blameless. The policies and practices that have brought about residential segregation have had low visibility and they are buried in history, lost behind such positive terms as "suburban growth" and "urban renewal." Most whites are blissfully unaware of the policies pursued by federal and local governments and by the banking and real estate industries that established racially segregated neighborhoods. And whereas the racial segregation of the Jim-Crow South required the active involvement of the white population in order to maintain it, contemporary residential segregation achieves a pervasive separation of the two groups without individual whites having to take any active steps. Thus, much as most whites express a clear personal preference for neighborhoods that are all-white or mostly white (Jackman and Jackman 1983, 197), they are rarely called upon to act on their preferences. Segregation is so well entrenched that most whites can enjoy its benefits without exerting any personal effort. Blacks reside in different locations, and segregation in schools and jobs follow "naturally" from that. In such a situation, a laissez-faire approach by whites is sweetly convenient.

Thus, the so-called gap or inconsistency between whites' racial policy principles and their specific policy views may be illusory. At both the abstract and the applied level, many whites gravitate to a policy solution that puts a priority on individual freedom of movement with as little governmental intervention as possible. With an initially weak conception of what constitutes racial integration, whites are consistently following through (rather than inconsistently backing off) when they oppose specific policy proposals that would accelerate governmentally enforced racial integration.

Gender

Men and women do not differ markedly from one another in their views about governmental gender policies. Table 6.5 presents the percentage distributions of men's and women's prescriptive and exis-

[5] The wishful thinking that is manifested in whites' support for "something in between" racial integration and segregation might be likened to Walter Laqueur's analysis of German citizens' cognizance of the holocaust: "while many Germans thought that the Jews were no longer alive, they did not necessarily believe that they were dead" (Laqueur 1980, 201, quoted in Elster 1983, 152).

tential beliefs about governmental intervention to ensure equal job opportunity for women, equal legal treatment for women, and daycare centers for the children of working mothers.

Of the three issues, it is equal legal treatment for women that elicits the strongest support from both sexes. Almost 40 percent of both sexes think the government should be doing "a lot" and about another 40 percent think the government should be doing "quite a bit" to make sure that laws are applied in the same way to women and men. Both of the other two policy issues elicit somewhat less support from both gender groups, with women's level of support only a shade ahead of men's.

Both men and women are less likely to think the government is currently doing a lot or quite a bit than that the government should be doing that much. On the issues of equal job opportunity and equal legal treatment for women, this tendency is slightly more marked among women, with 46 to 50 percent of women and 56 to 60 percent of men believing

that quite a bit or a lot is currently being done by the government. The issue of daycare centers is less likely to be seen by either men or women as a target of governmental activity, with only about 20 percent of men and 30 percent of women believing that a lot or quite a bit is already being done.^[6]

These figures contrast sharply with those for race. Most notably, women do not deviate seriously from men in their views about governmental intervention in women's civil rights. And although men appear to be slightly less withholding than whites in their prescriptive policy beliefs, this difference is not nearly as large as the difference between women and blacks in the degree to which they advocate affirmative change. Indeed, on the issues of equal job opportunity for blacks and for women, respectively, whites and men exhibit virtually the same levels of support—they differ only in that men are somewhat less likely than whites to believe that the government is already engaging in "a lot" of affirmative activity. Women, however, lag considerably behind blacks in the strength of their support for governmental activity on behalf of their group, and they also display considerably less jaundiced perceptions of the current level of governmental activity on their behalf than do blacks. Whereas almost all blacks advocate either a lot or quite a bit of governmental action to ensure equal job opportunity for blacks and only one-quarter believe that level of activity is already happening, about two-thirds of women advocate a lot or quite a bit of governmental activity to ensure equal job opportunity for women and one-half believe that level of activity is taking place already. The net effect is that only a hint of the rift found between racial groups is to be found between women and men.

Another contrast with the data for racial policy views is that, for gender, the move from abstract principles to specific government policies is not associated with any clearly interpretable changes in the data. In both realms, men's and women's views parallel one another fairly closely, and neither realm elicits systematically more or fewer affirmative prescriptions from either group. Of course, comparison of the two realms is less straightforward because the issues that come into focus are not directly comparable: the most central abstract principles for gender deal with egalitarianism in the conduct of people's "private" lives, whereas governmental intervention is usually raised in relation to the enforcement of individual rights in the public sphere. The former broaches redistributive goals at the sensitive core of the relationship; the latter

[6] But this issue also elicits a large number of "don't know" responses (about 30 percent). Daycare centers cater to a relatively narrow constituency, and the issue may not be very salient or visible to those who do not have preschool-aged children.

raises the threatening specter of enforcement but on issues involving individual, rather than group, rights, and away from the nerve center of family relations. In my study, there are just two items that allow for a fairly clear-cut comparison: those on abstract support for equal job opportunity for women and on support for governmental intervention to ensure that opportunity. On this issue, approximately 60 percent of both sexes support the abstract principle whereas about 30 percent think the government should be doing "a lot" to guarantee the principle and about 40 percent think the government should do "quite a bit."

Class

Social classes present a similar pattern of disagreement to that found between whites and blacks—but in muted form. As with whites and blacks, lower classes are more likely to advocate governmental intervention whereas higher classes are more likely to perceive that such intervention is already taking place. However, existential beliefs are less disparate by class than are prescriptive beliefs, and the pattern of class differences in general is not as dramatic as it is for race. The data for class are presented in table 6.6.

Descending social class is associated with a gradual increase in advocacy for these economic policies, and there is a difference of about 30 percentage points between the upper-middle class and the poor in their propensity to advocate "a lot" of governmental intervention. The magnitude of this class difference is constant across all three issues, even as

the absolute levels of support shift. It seems that the more directly a policy carries redistributive goals, the less support it receives from any class. Thus, the provision of welfare benefits receives the lowest endorsement from any class, with fewer than one-quarter of the upper-middle class and about one-half of the poor advocating "a lot" of governmental action on that issue, whereas the provision of guaranteed jobs receives the highest levels of endorsement, with just over one-third of the upper-middle class and about two-thirds of the poor supporting "a lot" of governmental intervention on that issue. Virtually no one among the poor believes the government should do only "a little" or "nothing" to guarantee jobs or a minimum income and about 13 percent of the poor take such a position on the more sensitive issue of welfare benefits. Among the upper-middle class, between 26 and 40 percent advocate little or no government action on the three issues, with welfare benefits exciting the strongest exhortations to inaction.

These figures (both in the pattern of disagreement among classes and in the absolute levels of support for egalitarian goals) are broadly consistent with those for abstract economic policy principles. If anything, there seems to be slightly less resistance in all classes to specific redistributive governmental policies than to economic egalitarianism in the abstract. As I noted earlier, however, these policies encompass only modest safety-net provisions, and thus they do not broach the issue of economic equality as boldly as did the items on the abstract principle of income equality.

In existential beliefs, the largest class differences occur in perceptions of the government provision of welfare benefits. The upper-middle class is almost twice as likely as the poor to believe that the government is already doing "a lot" or "quite a bit" to provide welfare benefits: among the upper-middle class, about 40 percent think a lot is being done and another 40 percent think quite a bit is happening, whereas only about 20 percent of the poor believe a lot is happening and fewer than 30 percent of the poor think even quite a bit is being done. On the other two economic policy issues, class differences in perceptions are in the same direction but smaller. And as with economic policy prescriptions, the gap between the poor and the upper-middle class is bridged in stepwise increments by the classes that lie in between. In this way, the political rift among classes is fractured rather than presenting alignments on either side of a simple, dichotomous cleavage.

These class divisions in economic policy views fall short of the gaping chasm that exists in both prescriptive and existential beliefs between whites and blacks. At the same time, the differences among classes are persistent and systematic, and far exceed the mere hint of disagreement found between men and women. This muted rift is reflected in the abso-

lute levels of support for affirmative policies that are displayed by the various classes. The upper-middle class demonstrates about the same degree of support for affirmative economic policies as do men for affirmative gender policies. The degree of support gradually increases from one class to the next until, at the other end of the class structure, the poor's advocacy of redistributive economic policies is almost as strong as blacks' advocacy of governmental action on behalf of blacks' rights.

General Patterns

Overall, these data yield very similar patterns to those that obtain for abstract policy principles. It is only in the relationship between whites and blacks that the rift between groups is revealed much more sharply in the context of specific governmental policies. And this shift is solely attributable to one group—whites—whose policy goals appear considerably less affirmative in the context of specific governmental policies. In gender and class relations, the policy goals of the respective groups appear fairly consistent in the abstract and specific contexts. In either arena, there is very little rift between men and women, with both groups indicating moderate levels of support for egalitarian and affirmative policies. Among classes, there is a muted rift that follows a step function from modest support for redistributive policies among the upper-middle class to fairly strong support among the poor. There is, if anything, a slight tendency for social classes to express slightly higher levels of support for specific governmental policies than for abstract egalitarian principles.

Thus, there does not seem to be any necessary relationship between the level of abstraction of policy questions and the degree of support expressed for them. Subordinate groups do not systematically reveal more radical goals in the "real" world of governmental policies. And dominant groups do not display any systematic tendency to espouse lofty principles and then abandon them when they move to specifics. Finally, there is no evidence of a general aversion in any group to the federal government intervening in social or economic issues. The oft-cited "gap" between abstract and applied principles in whites' policy views may simply stem from ambiguities that are inherent to items that measure general principles. Abstractions are by definition nonspecific, and it is only in the context of the specific that their meaning is revealed. Investigators may have optimistically read more into people's responses to general items on racial integration than was really there.

Support for a policy instead seems to hinge on three interrelated factors, none of which is related to the distinction between abstract principles and applied policies. First, policies that strike at the core of an intergroup relationship generate more cautious responses than those that aim at the periphery of the relationship. Thus, enforced racial integration

in schools generates the least endorsement from both whites and blacks because its sensitivity is recognized by both sides: whites guard their schools with particular ferocity because schools constitute the primary institutions by which cultural values and achievement-related skills are transferred to the next generation, and blacks are less likely to be assertive in an area that is most likely to elicit a backlash (and perhaps especially since it is their children who would be on the frontline). Second, by the same token, policies that have a broader scope elicit less support than policies that seek only small changes. And third, people are more likely to shy away from policies that threaten individualistic values, especially freedom of choice. All of these factors are interconnected, and, indeed, the prevalence of individualism lies in its ideological convenience as dominant groups strive to divert energies away from core redistributive issues to those that lie closer to the periphery of their relations with subordinates.

There seems to be a common understanding between contending groups of the moral framework that binds their disputes. As subordinate groups attempt to enhance their control over society's assets, they cope with the uncertainty of the terrain by avoiding dramatic demands that violate the terms of that framework. Dominant groups continue to proselytize the broad outlines of the moral code as they resist the specific inroads that subordinates attempt to forge. Subordinates have no choice but to be sensitive to the hard spots and soft spots of the dominant group and to calibrate their demands accordingly. In this way, the contending groups get locked in a delicate struggle that is several steps removed from the core redistributive issues that define their relationship. The rights of groups appear to lie beyond the confines of legitimate political contention, and the domain of debate becomes restricted to questions about the extensiveness and enforcement of individual rights.

As the contending groups address those questions, a comparison of their prescriptive and existential policy beliefs reveals that they are divided by both. People's sensitivities—whether to the indignities of deprivation or to the threat of lost benefits—affect not only their advocacy but their perceptions of reality as well. Subordinates are generally more likely to advocate government intervention, whereas dominant-group members are more likely to believe that government intervention on behalf of subordinates is already considerable. In this way, the separate existential worlds of dominant and subordinate groups reinforce the rift created by their differing prescriptive views. The dual-edged divide is most pronounced between whites and blacks, somewhat weaker in class relations, and only whispered in gender relations. Men and women do not differ substantially from one another in their prescriptive views, but women do tend to see somewhat less government activity already in place

than do men. These patterns suggest that the import of the policy goals that people articulate cannot be fully assessed unless we know the existential context in which they are framed.

Results:
Affirmative Change, Conservatism, and Reaction

In order to define the lines of contention between groups more sharply, I now synthesize the data on specific policy goals by building people's existential beliefs into the measure of their policy prescriptions. In this way, each individual's policy prescriptions are measured with a built-in reality check—against the existential beliefs that frame those prescriptions. This yields a more sensitive indicator of people's policy goals.

Measures

I take the difference between each respondent's belief about what should be done and her belief about what is already being done by the government for each policy issue.^[7] For each policy measure, this yields scores ranging from -3 (the government should be doing a lot but is doing nothing) through +3 (the government should be doing nothing but is doing a lot). These scores break down into three distinct categories: those who advocate affirmative change (that is, they think the government should be doing more than it currently is), those who take a conservative position (they think the government is doing about the right amount now), and those who take a reactionary position (that is, they would like to see the government do less than it currently is).

By using perceptions of the status quo as the anchor for prescriptions, we can directly assess the degree to which various groups advocate affirmative change. For example, the white who says the government should be doing "a lot" to ensure equal employment opportunity for blacks takes on a less glowing pallor if she also thinks the government is already doing a lot. We can also assess whether resistance to affirmative change takes the form of conservatism or is outright reactionary. In this way, the character of the debate between contending groups may be specified more acutely.

Definition of Policy Domains

I begin by examining the zero-order correlations among the government-policy difference-scores, to check the extent to which the measures for race, gender, and class, respectively, fall into distinct domains. Table 6.7 presents correlation matrices (Pearson's r 's) for the government-policy difference-scores (below the diagonal), as well as for the original prescriptive government-policy items (above the diagonal). The correlations for policies within each intergroup

[7] Respondents who said "don't know" or who had missing data for either item in a pair were excluded from the calculation of the difference score for that pair.

relationship are in boldface. The difference-scores are in their continuous scoring from -3 to +3, while the original prescriptions are scored from 1 ("A Lot") through 4 ("Nothing"). "Don't Know" responses and missing data are excluded on a pairwise basis.

For gender and class, the two sets of measures display similar correlations. For race, however, the policy difference-scores are more highly intercorrelated than are the uncorrected policy prescriptions. This doubtless reflects the double polarization between whites and blacks that is captured when both prescriptive and existential beliefs are incorporated.

Of the three intergroup relations, the policy goals for race show the sharpest internal convergence. The three race-policy difference-scores are all intercorrelated between .56 and .68. These correlations are not only very high in themselves but they all exceed the correlations between any of the race measures and the measures of gender or class policy goals. Thus, the three race-policy measures demonstrate both high convergence as well as clear separation from the policy domains of gender and class.

The three gender-policy measures do not cluster together as sharply. The difference-scores for equal job opportunity and equal legal treatment are strongly intercorrelated (.46) and this intercorrelation is higher than either measure correlates with measures from race or class (although the correlations of the equal job opportunity measure with the race measures are only a shade lower). The daycare measure does not

correlate distinctively with the other two gender issues, and its correlations with the class measures especially suggest that it is seen as a class issue at least as much as a gender issue.

The class-policy issues, not surprisingly, tend to be somewhat entangled with the race issues: interestingly, the welfare issue (the most directly redistributive policy) is also the one that is the most highly correlated with the race issues. Given the disproportionate share of blacks who live in poverty, it seems unrealistic to develop any measures of views about redistributive economic policies that would not be intertwined with views about racial policies. Yet the welfare difference-score still correlates more highly (r = .45) with one of the other class policy measures (guaranteed minimum income) than it does with the race issues. The guaranteed minimum income and the guaranteed jobs difference-scores also correlate more highly with each other (r = .46) than either does with the race issues. It is only the correlation between welfare and guaranteed jobs (.35) that fails to exceed the correlations of welfare with the race issues.

These data indicate that the issues surrounding policies that are aimed at a specific intergroup relationship tend to fan out to other intergroup relationships. Insofar as race, gender, and class all involve alternative manifestations of inequality in the same society, it is perhaps inevitable that there will be spillover effects from policy views in one area to those in another. The intergroup relationship that has had the most sustained policy visibility—race—should be expected to generate the tightest set of policy views, as people get more experience in using contextual information from that intergroup relationship to cue their reactions to any new policy issue that arises. Indeed, habits of thinking coming out of that intergroup relationship then take on psychological primacy and are especially likely to permeate the way people think about policy issues arising in other intergroup relationships as well.

The Lines of Debate

I now evaluate the lines of policy debate between groups. What are the blends of affirmative advocacy, conservative resistance, and reactionary opposition that color each intergroup relationship? Table 6.8 presents the percentage distributions by group for the pertinent policy measures, for race, gender, and class, respectively.

The chasm between whites and blacks is revealed starkly with these measures. Whereas the overwhelming majority of blacks (72–89 percent) want affirmative change on each of the policy issues, whites are almost as overwhelmingly conservative—or reactionary. On the issue that generates the staunchest opposition among whites and the least support among blacks (integrated schools), over 70 percent of blacks want to see affirmative change whereas almost 60 percent of whites desire reactionary change. On the other two race issues, almost 90 percent of blacks

advocate affirmative change, whereas about one-third of whites take a reactionary position and about 40–45 percent are conservative.

For gender, I restrict attention to the two issues that reflect gender concerns relatively cleanly, equal job opportunity and equal legal treatment for women. The opinions of men and women almost parallel one another on these issues: both groups are predominantly divided between conservatism and affirmative advocacy, with women leaning slightly more in the affirmative direction and men's views weighted more toward conservatism. Reactionary views are less frequent, although the issue of equal job opportunity generates stronger resistance than does the issue of equal legal treatment: fewer than 10 percent of either sex take a reactionary position on the question of equal legal treatment, but about one-quarter of men and one-fifth of women take a reactionary position on the issue of equal job opportunity. The general tenor of opinion on these gender policy issues is thus more moderated than for racial issues and involves much less disagreement across group lines. But, nonetheless (as with whites' racial policy goals), support for affirmative change in gender policies is considerably lower than was inferred when people's prescriptive views were considered separately from their existential beliefs. Only slightly over one-third of men favor affirmative change, and women have just nudged ahead of men in advocating such change.

With the class policy measures, one issue—guaranteed jobs—generates fairly high levels of affirmative advocacy from every class, but the other two issues—welfare benefits and guaranteed minimum income—are more divisive. The clear majority of every class (ranging from 57 percent of the upper-middle class to 77 percent of the poor) wants to see the government do more to guarantee jobs, and most of the balance of opinion is conservative rather than reactionary. More governmental effort to guarantee a minimum income is advocated just as strongly by the poor (79 percent), but with ascending social class support gradually slips away to only 43 percent of the upper-middle class. And although most of the balance of opinion in each class is again conservative rather than reactionary, ascending social class brings a steady increase in reactionary opinion, from only 3 percent of the poor to 18 percent of the upper-middle class. The most sharply divisive issue is that of welfare benefits. On this issue, the stance of the upper-middle class is almost as withholding as that of whites on school integration (only 19 percent of the upper-middle class supports affirmative change and 45 percent is reactionary); opinion shifts slightly in the direction of affirmative change from the middle class to the working class, but among the poor there is a pronounced shift in opinion, with 60 percent of the poor advocating that more be done and only 13 percent taking a reactionary stance. In

all, the class rift in policy goals is pronounced, although it still does not equal the rift between whites and blacks.^[8]

These data thus portray rather disparate dialogues across the three intergroup relationships as the contending groups articulate their policy goals. Although subordinates consistently take the lead in the direction of affirmative change, the depth of their rift from their dominant-group counterparts varies dramatically. Between blacks and whites there is a searing disparity, among social classes there is a large but graduated disagreement, and among women there is only a hinted departure from the policy goals of men.

Conclusions

This chapter has been directed by one central question: to what extent do contending groups in an unequal relationship seek to increase their control over resources? Although the question sounds straightforward, the answer is necessarily layered beneath a series of considerations.

Intergroup policy goals have been the object of repeated empirical inquiry, but scholars have varied widely in their assumptions about whether policy goals reflect underlying, irrational, intergroup antipathies (prejudice), the apolitical quest for altered lifestyles ("sex-role attitudes"), the overriding of conflicting interests and intergroup antipathies (political tolerance), or the rational pursuit of group interests (class consciousness). Investigators have also disagreed about how to interpret various measures of policy goals. Attempts to measure intergroup policy goals have been waylaid repeatedly by an observed gap between the views expressed about abstract principles and applied policy goals. In research on prejudice and class consciousness, scholars have debated fretfully about the meaning of that gap. This has led to arguments about what agents and methods of change are most appropriate to achieve particular goals. And throughout, the measurement of policy alternatives has been restricted to those that have been given life in the public debate. Scholars have generally lost sight of the broader spectrum of issues that lies outside the realm of public discourse.

My approach calls for the examination of intergroup policy goals as the rational pursuit of group interests, within constraints. As people attempt to further their interests, they do not operate freestyle. They ma-

[8] Within each class, blacks are more likely to advocate affirmative change in these economic policies than are whites. Further, among blacks there is no systematic relationship between class and class policy goals. Although whites' support for these economic policies gradually diminishes with ascending social class, blacks' support remains at a high level, making the racial difference larger as one progresses from the poor to the middle class. This underscores the intersection of race and class issues in American society.

neuver within pragmatic and moral conditions that render some routes plausible and others futile or even dangerous. The society that houses an intergroup relationship is infused by a general moral climate that has grown out of the various intergroup experiences that have dominated its history: that climate limits the kinds of demands and appeals that have moral credibility. Within that societal context, specific constraints are posed by the internal structure and politics of the relationship itself. The parameters of public policy discussion are set by the conditions that are fused from those two sources.

Out of the growth pains of capitalist democracy, individualism has become established as a bedrock moral tenet. The norm of individualism has infused popular conceptions of intergroup policy goals and twisted the meaning of equality into something more accommodative—equality of opportunity. Attempts to assess popular sentiment for an egalitarian redistribution of resources have repeatedly run aground on the shoal of individualism. The patterns displayed in my data underscore the pervasiveness of individualism as a guiding principle in the articulation of intergroup policy goals. Indeed, the so-called gap between abstract and applied policy views that has persistently intrigued empirical analysts may be nothing more than segmented reflections of the layering of individualism beneath egalitarianism. The silent transformation of egalitarianism to incorporate individualism is masked in many survey questions, leading to apparent inconsistencies in responses to various questions.

On the issue of racial integration, questionnaire items measuring abstract principles have elicited more support from whites because they leave the rights of individuals untrammeled, whereas questions about government intervention to enforce racial integration uncover the conflict between individual rights and group rights. Many whites then back away from enforced racial integration as they opt for the protection of (their own) individual rights. In gender policy goals, the degree of support elicited for different abstract principles varies according to their link with individualistic principles. A general question on equal employment opportunity for women invokes individualistic principles and it elicits much stronger support than do general egalitarian principles for the division of labor between the sexes. And once the unsullied individualism of the general principle of equal job opportunity for women is harnessed to governmental enforcement, support for stepping up that enforcement drops to about half the level garnered for the general principle. In class policy goals, it is the abstract principles of egalitarianism that elicit less support than do questions about specific governmental policies, because in this case it is the abstract principles that state the goal of equality more boldly whereas specific governmental measures merely cast safety nets to cushion the deepest ravines of inequality. And among questions that

ask about various specific governmental policies, the degree of support for those policies is elevated or depressed as a direct function of the extent to which such policies are respectful of individual rights or tilted more toward an egalitarian redistribution of resources between groups.

Hedged in by the pervasive morality of individualism, the contending groups in race, gender, and class relations formulate their policy goals. The issues that are thus breathed into life in the public arena do not call for radical changes in the established division of resources but involve readjustments at the margins. Even those issues, however, do not gain easy acceptance. As dominant groups resist intrusions by subordinates, contending groups become locked in disputes—at times heated, at times mild—about potential alterations to arrangements at the periphery of their relationship.

Within these boundaries, do contending groups seek to increase their control over resources, simply maintain their current holdings, or make concessions to their protagonists? On what kinds of issues do contending groups yield concessions or press harder to advance their control over resources? Are these proclivities affected by the position of the group in the relationship (dominant or subordinate) or by the structure of the relationship between groups? To address these questions, I examined people's prescriptions for policy against their existential beliefs about the current state of governmental intervention. In this way, we can distinguish those who favor affirmative change, those who wish to maintain things as they are (the conservatives), and those who would like to reduce the amount of government intervention (the reactionaries).

To begin, we can observe that in all three intergroup relationships, subordinate groups show little tendency to give up their current holdings: their energies are primarily divided between preserving what little they have and pressing for more control over resources. That is to say, very few members of subordinate groups voice the reactionary view that the government should reduce its intervention on their behalf. Women are the only subordinate group to evidence such a propensity among more than a handful of their members, and even here, it is distinctly a minority view within the group. The policy goals of subordinates generally range between conservative and affirmative, with blacks showing an overwhelming preference for affirmative change, the poor and working class leaning almost as strongly toward affirmative change, and women showing only a slight preference for affirmative change over conservatism.

Among dominant groups, affirmative change usually finds less support than do conservative or reactionary policy goals, but the aversion to affirmative change among dominant groups is less extreme than is subordinates' aversion to reactionary policy views. It might thus seem

that dominant groups are more susceptible to relinquishing resources than are subordinates, but recall that the policies under discussion have already been molded by the time they reach the public arena so that none of them threatens flagrant redistribution. By steering discussion away from the most threatening issues, members of the dominant group can afford to have occasional concessions wrung out of them on the more peripheral issues that become the focus of public debate. Among those issues that do surface, dominant-group intransigence increases as specific policies come closer to touching core redistributive issues. On the issues that are more threatening (governmental intervention to ensure equal job opportunities for women and blacks, racially integrated schools, and provision of welfare benefits), the responses of dominant groups are largely torn between holding on to present arrangements and increasing their control over resources in a reactionary bid. The reactionary spirit is most in evidence in the racial policy goals of whites and least evident in the gender policy views of men. Among men, preserving current policies is the prevailing disposition. In the upper-middle class, the prevailing weight of opinion ripples from affirmative to reactionary, depending on the specific policy at hand. And among whites, the prevailing policy stance is lodged between conservative and reactionary, with the specific tilt dependent on the policy at hand.

There is clearly an interplay between the policy goals of dominant and subordinate groups. Within each intergroup relationship, the issues on which dominant groups are more intransigent are also the issues on which subordinates are less likely to push forcefully for affirmative change. Subordinate groups learn to throw more energy into issues that keep a safer distance from core redistributive concerns. And so the game goes on, with contending groups apparently torn between preserving current holdings and gingerly seeking to increase their control over resources. Subordinates generally yield to their more powerful protagonists by shying away from issues that are more threatening to the relationship. When more assertive policy proposals slip into the public arena, they meet with stiff resistance, or even reactionary counterproposals. But on more peripheral issues subordinate investment is more likely to be rewarded with occasional minor concessions—after exhaustive debate.

This pattern holds across all three intergroup relationships, but it is played out with varying ferocity, according to whether communication channels between groups are personal or aggregated. Between men and women, where the structure of intergroup contact is the most intimate, each group arrives at its mutually negotiated position with only a hinted rift between them. Between whites and blacks, where the physical separation of the groups is the most extreme, the negotiations between groups take the severest form. As the groups communicate across a great divide,

messages must be shouted to be heard. Each group must react to the demands and vulnerabilities of the other in full public view. Finally, in class relations, the physical separation of disparate classes is broken by a step-wise pattern of restricted personal contact between adjacent classes. Both of these channels of communication affect the negotiated policy position of each class. The result is a graduated but resilient rift between contending classes in their articulated policy goals.

In these variants, it is the policy position of subordinates that is more vulnerable. Dominant groups retain the upper hand and they do not stray far from a general proclivity to resist change. At the same time, it should be remembered that it is generally the dominant group's hold over resources that is under more immediate threat in these policy discussions. A defensive position, backed by the full weight of established institutions and their supporting morality, is easier to maintain than is the assertive position of visualizing, demanding, and justifying affirmative change.

The opportunities that are offered to subordinates to advocate affirmative change are never great. They gravitate to causes that do not pose deep risks and that offer some promise of success. That steers the mainstream of subordinate effort inexorably toward less radical issues. But the ability of subordinates to chip away at even the periphery of their relationship with the dominant group is constrained by the structure of that relationship. As the structure imposes increasing physical distance between the groups, subordinates are afforded more opportunity to coalesce around demands for moderate, affirmative change.

Chapter Seven—
The Ideological Molds of Paternalism and Conflict

The prevailing approaches to intergroup attitudes and group consciousness anticipate "consistency" in people's attitudes as the emblem of coherence, constraint, and crystallization in the way people think about their relations with another group. In particular, the policy dispositions of dominant group members toward subordinates should be a direct reflection of how positive or negative their feelings are toward subordinates. The latter, in turn, are thought more likely to assert their political rights when they feel hostile toward those who assert dominance over them. The only "inconsistent" attitude structure that has been the object of sustained scholarly attention is tolerance, that is, the advocacy of equal rights for another group despite negative feelings toward the group.

In this chapter, I explore the relationship between people's intergroup feelings and their policy dispositions. My approach to intergroup ideology does not anticipate a pivotal role for feelings of intergroup hostility. Nor does it anticipate that the magnet that binds attitudes and lends them coherence is "consistency" in the conventional sense. The extent to which groups pursue their interests must be disentangled from the issue of whether feelings of hostility pervade the relationship. The intersection of intergroup feelings and policy views holds considerably more subtlety and implicit finesse on the part of both dominant and subordinate groups than conventional approaches have anticipated.

My thesis is that the beacon to which dominant groups are drawn is the "inconsistent" attitudinal mold of paternalism—the combination of conservative or reactionary policy dispositions with positive feelings toward subordinates. How much paternalism is manifested in the three different relationships under investigation in this book? To what extent is paternalism displaced by "consistent" attitudes comprising negative

feelings and conservative policy views or positive feelings and affirmative policy views? And how empirically viable is the alternative form of attitudinal inconsistency, the configuration of tolerance?

The ideology of paternalism is geared toward eliciting deference from subordinates—warm intergroup feelings and conservative policy dispositions that comply with the dominant design. How successfully do the different dominant groups elicit such deference from their subordinates? What blends of feelings and policy goals are interlaced by subordinates to deal with the rigors of inequality, and how does this vary from one type of intergroup relationship to another?

I begin the chapter by discussing the featured role of consistency in the literature on attitudes and public opinion. This includes a brief discussion of the single empirical concept that has posited an inconsistent attitude structure—political tolerance—to assess its bearing on the prevailing preoccupation with attitudinal consistency. I then consider the factors that have contributed to the binding significance of consistency in attitude theory, and I challenge the underlying assumptions on which that significance rests. Following this, I address two different theoretical concepts that imply attitudinal inconsistency, but which have developed without reference to that issue. The first of these is "hidden" or "everyday" resistance, which depicts the attitudes of subordinates. The second concept is paternalism, which addresses the attitudes of dominant-group members. With those issues delineated, I turn my attention to the intersection of intergroup feelings and policy goals in my data for race, gender, and social class.

I begin that analysis by examining the ways in which positive and negative intergroup feelings are converted into policy goals by people in different kinds of dominant and subordinate groups. I then present a simple empirical scheme that allows us to identify broad configurations of intergroup feelings and policy views. As I address the meaning of each of the categories in the scheme, it becomes apparent that there is not a straightforward equivalence between dominant and subordinate attitudes and that the meaning of particular attitudinal compounds depends on the position of the person's group in an unequal relationship.

This scheme is used to make a baseline empirical assessment of the prevalence of alternative attitudinal configurations in the exchange of ideologies between dominant and subordinate groups in race, gender, and class relations. These data also suggest broader inferences about the dynamics that pervade the articulation of intergroup ideologies by unequal groups. In this way, we can delineate the intersection of the two basic parameters of intergroup ideologies—the expression of hostility or friendship and the pursuit of group interests.

The Issues

Attitudinal Consistency and Ideological Constraint

An assumption has run like a deep dye through public-opinion research that the hallmark of constraint, coherence, and crystallization in people's belief systems is consistency. Since Philip E. Converse's seminal article, "The Nature of Belief Systems in Mass Publics" (1964), students of public opinion have taken it as a given that consistency is the key indicator of whether attitudes have substance and meaning. A number of factors have conspired to foster the preoccupation with consistency in analyses of attitudes and public opinion.

The work that riveted the attention of political-opinion analysts on attitudinal consistency was Converse's 1964 article. His argument was located squarely within the context of mass political opinion in liberal-conservative partisan politics. He contended that most members of the mass public in the United States have only a hazy understanding of the terms liberal and conservative and that they have poorly formulated political opinions of their own. The issue stance of the average citizen fluctuates on the liberal-conservative axis, both over time and from one policy issue to another. Converse demonstrated empirically that correlations were weak between people's issue-stances on the same issues over time and between different specific issues at the same point in time. From this evidence, he argued that belief systems in the mass public generally lack constraint. He reasoned, further, that this lack of constraint derived from the low salience of political issues to most members of the mass public, as well as from poor information about both those issues and the broader political principles that could lend them coherence. This public-opinion muddle was contrasted with the political opinions of political elites: the latter were portrayed as having considerably more constrained belief systems and as being better schooled in the principles that bind specific issues together.

Converse's argument was very compatible with earlier work on voting and public opinion which had already decried the ideological confusion and lack of coherent political thought that reigned in the mass public. For example, the main conclusions of a major empirical study of political opinion by Berelson, Lazarsfeld, and McPhee (1954) were that the average democratic citizen lacks political interest, motivation, or knowledge and casts his vote in a way that is unconstrained by principles and without reference to considered thought about political issues. A similar argument was made in 1960, in a much-cited article in the American Political Science Review by McClosky, Hoffmann, and O'Hara. In an empirical comparison of the policy views of a sample of the mass public and a sample of the participants in the national conventions of the Republican

and Democratic parties in 1958, McClosky, Hoffman, and O'Hara argued that the mass public contrasted sharply with political leaders in having poorly formulated political ideologies and in being poorly informed and having low interest in politics.

Converse's research completed the indictment of the American mass public. This dismal depiction of nonattitudes withering the grassroots of democracy inevitably drew fire. Converse's paper was the lightning rod for a crop of articles that sprang up over the years from scholars who found the picture offensive or unpalatable, or who sought to defend or elaborate Converse's original argument. The ensuing debate has been catalogued by others (see especially Smith 1989). One line of research took off on the question of whether the level of (in)consistency has changed over time (e.g., Bennett 1973; Nie and Anderson 1974; Hagner and Pierce 1983; Kirkpatrick 1976; LeBlanc and Merrin 1977; Nie, Verba, and Petrocik 1976; Piereson 1978; Smith 1989), or whether changes in survey question-wording in the National Election Studies were responsible for the observed changes in consistency (e.g., Bishop, Oldendick, and Tuchfarber 1978a , 1978b ; Bishop et al. 1979; Brunk 1978; Sullivan, Piereson, and Marcus 1978; Sullivan, Piereson, and Marcus 1979). Another line of research raised statistical questions about how best to measure consistency, pursuing such concerns as measurement unreliability, the idiosyncratic properties of various measures of association, and the statistical confounding of true change in attitudes with mindless fluctuations in political views (e.g., Butler and Stokes 1974, 276–295, 316–337; Achen 1975; Weissberg 1976; Krosnik 1991). Meanwhile, Converse continued to develop and buttress his original argument with further empirical work, most notably with two articles titled (meaningfully) "Attitudes and Nonattitudes: The Continuation of a Dialogue" (1970) and, with Gregory B. Markus, "Plus ça change  . . . The New CPS Election Panel Study" (1979).

In short, the publication of Converse's article in 1964 provoked an enduring research agenda and molded the way public-opinion analysts continue to think about attitudes. For all the energy that has been absorbed by the questions of how much consistency there is in the mass public and how to measure it, no one has paused to ask whether consistency is such a vital concept in the first place. It has been accepted intrinsically that consistency is the benchmark by which we judge whether attitudes have meaningful substance and whether they may reasonably be considered to belong to a coherent belief system. Indeed, consistency and constraint have been treated as equivalent terms.

There is only one form of attitudinal inconsistency that has engaged empirical analysts of mass political opinion, and that is political tolerance. This construct has been the subject of sustained empirical inquiry, but,

perversely, that research has done nothing to undermine the prevailing conception of attitudes as governed by the pressure toward consistency. The concept of political tolerance, as it emerged in empirical research on democratic norms, is concerned with the ability of democratic citizens to grant full civil liberties to groups with whom they disagree or whom they dislike. In other words, the concept posits a logically inconsistent attitude structure: can people override their negative feelings toward out-groups and support affirmative policies toward them? The concept is central to empirical democratic theory, and it has generated a substantial body of public-opinion research, starting with Stouffer's classic study, Communism, Conformity, and Civil Liberties (1955), and continuing actively to the present (e.g., Kuklinski et al. 1991). I have discussed that research in chapters 1, 5, and 6. What is of relevance here is that, despite the central position of that research in the literature on public opinion, it has not undermined the prevailing conception of attitude structure as bound by the pressure to be consistent.

There are probably three main reasons for this apparent disconnection. First, although tolerance was conceived as an inconsistent attitude structure, empirical measures initially focused on policy dispositions entirely, without heed to respondents' feelings toward the target groups. Respondents' feelings were simply assumed to be negative. But because feelings were not incorporated in measures of tolerance, the specific composition of the tolerant attitude was not acknowledged explicitly. This element was not highlighted in empirical work until Jackman (1977, 1978) and Sullivan, Piereson, and Marcus (1979, 1982) developed measures of tolerance that incorporated both feelings and policy dispositions. Second, tolerance was initially conceived as a difficult state to attain, something uniquely required and fostered by democratic political systems. The problem was posed in terms of whether people could overcome their natural tendency to express their negative feelings in negative policy dispositions and instead develop the restraint necessary to sustain political tolerance. Tolerance was thus thought of as the exception that proved the general rule of attitudinal consistency. Third, empirical research on tolerance showed that it was, indeed, a difficult state to attain. Although scholars have disagreed about the extent of its dearth (see, for example, Sullivan, Piereson, and Marcus 1979, and Abramson 1980), studies have generally found political tolerance to be present in no more than a minority of the population. By default, then, the consistency rule was reaffirmed by empirical analyses of tolerance.

The intuitive reliance on consistency as the cement of attitudes is not restricted to the literature on political opinion. Students of prejudice have conceived of that phenomenon as a lock-step of negative beliefs, negative feelings, and negative policy dispositions. The unprejudiced

person converts those negative values into positive ones: just as negative feelings fuel discrimination, positive feelings undermine it. Similarly, students of class consciousness have expected a proletarian "class-for-itself" to show its ripened political consciousness by adopting negative beliefs about and hostile feelings toward the dominant class, along with assertive policy goals and combative strategies. In the cases of both prejudice and working class consciousness, feelings feed policy dispositions.

In another variant of the consistency approach, Abercrombie and Turner (1982) reflexively use the presence or absence of a consistently applied moral code as a yardstick to assess the fiber of dominant ideology. Their reasoning is that if the dominant class propagates an ideology as a broad moral code to induce the compliance of subordinates, the same moral precepts will be used consistently to dictate the social behaviors of all members of society. They argue that dominant classes in history have not held their subordinates to the same moral code as they apply to themselves and that this means that dominant classes do not in fact have coherent ideologies that would qualify as viable instruments of social control. As I argued in chapter 2, however, this implicit reliance on consistency as the yardstick for the salience of an ideology is misplaced. Dominant groups are not in a contest for consistency: in the day-to-day business of eliciting compliance from subordinates, rigid consistency may be more of a political liability than an asset. Belief systems that are unerringly consistent are vulnerable to attack from all angles: the demolition of a single segment would cause the entire edifice to crumble. Additionally, there seems little practical compunction to apply the identical moral precepts to the group from whom one desires compliant behaviors as to the group whose design it is to govern: divergent roles in life would seem to call instead for flexibility in the application of an abstract moral code.

The common reliance on consistency in these different variants of research on attitudes, belief systems, and ideology is not a coincidence. It can be traced back to the way attitudes were initially conceived in classic attitude theory. An attitude is a purely theoretical construct that is not susceptible to direct empirical observation or measurement. What can be observed in human behavior is sporadic expressions of beliefs, feelings, and behavioral predispositions: in order to tighten their conceptual grip on such an unwieldy class of phenomena, attitude theorists relied on consistency as a key building block. Most definitions of attitudes have given a central place to an evaluative or affective disposition: the feelings that lie at the core of an attitude are thought to be supported by consistent perceptions and, in turn, to drive a stable predisposition to respond to the object of the attitude (see, for example, Allport 1935; Doob 1947). In this way, some order and predictability were carved out of a complex web of empirical phenomena.

The theme of consistency was to be replayed in a range of theories about the structure and characteristics of attitudes. Heider's balance theory (1946) posited that people press their cognitions into consistent configurations. A state of cognitive balance exists when an entity is perceived as having the same dynamic character in all respects (as when a person both admires and likes an object), when all parts of a unit are perceived as having the same dynamic character (as when they are seen as all positive or all negative), and when entities with different dynamic character are segregated from one another. A state of imbalance produces tension that creates pressure to reshape incongruent cognitions so that they come into line with the perceptual gestalt .

These postulates laid down a framework that was explored, developed, and elaborated in theoretical and empirical work by a large and influential school of researchers (see, for example, Newcomb 1953; Osgood and Tannenbaum 1955; Cartwright and Harary 1956; Rosenberg 1956; Festinger 1957; Abelson 1959; McGuire 1960; Cohen 1960; Abelson et al. 1968). Some of the most influential ideas to come out of this school of thought have been that incongruent perceptions cause dissonance in individuals that they are driven to resolve (Festinger 1957); that the affect an individual feels toward an object is tied logically to specific positive or negative perceptions of the object (Rosenberg 1956); that changes in evaluation are always in the direction of increased congruity with the existing frame of reference (Osgood and Tannenbaum 1955); that the greater the salience of a topic, the less tolerance the individual has for attitudinal inconsistency about that topic (Cohen 1960); and that the more directly two issues are connected with one another, the less the individual is able to tolerate inconsistent cognitions across those issues (McGuire 1960).

It thus became a theoretical maxim that attitudes can be identified by their internal consistency: the more salient a topic, the more the individual thinks about it and the more internally consistent the attitude becomes. A lack of consistency signifies the lack of an attitude. Research on political ideology, prejudice, and class consciousness simply built on that theory. Consistency became the standard currency of that research. People's attitudes are thought to manifest increasing internal consistency as they become more crystallized. Indeed, the standard procedure for testing the validity of specific questionnaire items is to check their consistency with other items on the same topic.

Consistency clearly has intuitive appeal to scholars as a magnet to organize the tangled web of beliefs, feelings, and behavioral dispositions that are observed in human expression. Two assumptions underlie this appeal. First, there is an implicit assumption that attitudes are self-contained and individually driven. This creates the theoretical imperative

to find the source of someone's policy dispositions toward an object within the structure of the attitude itself. Thus, discriminatory policy dispositions are driven by negative feelings, whereas positive feelings remove the individual's motivation for discriminatory policy goals and lay the foundation for affirmative policy dispositions. The feelings, in turn, are logically fed by the person's beliefs about the object. Second, there is an assumption that people have a need to be consistent when they think about something. As a topic becomes more salient, the individual thinks about it more often and more intensely: the pressures increase to mold the various idea-elements into a consistent mass. Inconsistency can only be tolerated on topics that receive little attention and are not salient to the individual.

Both of these assumptions are problematic. To begin, the assumption that attitudes are primarily the property of individuals is questionable. Although individuals are the bearers of beliefs, feelings, and behavioral dispositions, they do not hatch these thoughts and feelings independently. Instead, they borrow from the cultural repertoire that is available to them. This book is premised on the idea that the intergroup attitudes that individuals espouse are not formulated discretely out of their personal experience, but are the communal property of the group to which they belong. Second, as people grapple to understand the world in which they live, they do not work according to the principles of either originality or logic. Instead, they reach for ideas from those that float past them on a daily basis, ideas that are espoused convincingly by others, and that have the ring of truth that only familiar things can have. Various thoughts and feelings are borrowed as needed.

The principles that guide the adoption of specific beliefs, feelings, and behavioral dispositions are convenience and efficiency. People tend to think and feel whatever is convenient, and they tend to deal with an issue as efficiently as possible, thinking about it only as much as is necessary to satisfy the political and practical exigencies of the moment. As the political reality changes, the members of a group will adjust their views and feelings, gradually, in a piecemeal fashion. How those exigencies are seen and interpreted can vary within the group: individual variation could stem from variation in personal experiences or pressures, and from differential exposure to information flows. Thus, for example, the well-educated members of a dominant group may espouse a slightly different variant of their group's intergroup ideology: their prolonged experience in educational institutions molds them into an especially individualistic way of thinking and it also exposes them more directly to information flows that contain the latest ideas (Jackman and Muha 1984). But the prevailing force that shapes an individual's intergroup attitudes is his or her membership in a group that has interests to defend or

advance. The specific interests of the group and the history of its collective experience always limit the range of ideological options available to the individual actor.

Intergroup attitudes that are shaped by political exigencies are not perversely static. If political conditions remain unchanged, the prevailing intergroup attitudes will likewise remain in place. But as people respond to the political dynamics of the intergroup relationship in which they participate, old ideas are abandoned, gradually, if they are found to be less serviceable. New ideas drift into their place. At any single point in time, examination of a cross-section of attitudes in a group most likely uncovers a variety of elements, old and new, surrounding the prevailing theme. Similarly, individuals in the group have amalgams of thoughts and feelings that may contain residues from an earlier era along with the currently adopted line. As new information and ideas disseminate through the community in irregular flows, new patterns of ideological response emerge. An interesting example of this process is provided by Gamson and Modigliani (1987) in their analysis of the various packages of ideas that emerged in the unfolding public debate on affirmative action from the 1960s to the 1980s. In response to the policy initiatives of the federal government in the 1960s, a welter of competing ideas in the media were gradually reduced to an identifiable set of packages, out of which one emerged as dominant—the "reverse discrimination" argument. As I discussed in chapter 6, the "reverse discrimination" argument caught on because it resonated with the morality of individualism, and thus it provided whites with a basis for opposing affirmative action that felt comfortably principled and unbiased.

In such a dynamic and political process, the expectation of a clean consistency at either the individual or the group level defies political or common sense. The structure of intergroup attitudes conforms, not to what is consistent, but to what is politically expedient. Indeed, political expediency may dictate an inconsistent attitude structure.

Inconsistent Attitudes in Ideology

In the theoretical literature on intergroup attitudes and group consciousness, two different ideological modes have been posited that imply technically inconsistent attitudes. One of these, known as "hidden" or "everyday" resistance, is described as politically expedient for subordinate groups. The other is paternalism, and it applies to the politics of domination. The concepts of hidden resistance and paternalism have developed without reference to the empirical literature on attitudes, but each of these concepts has important implications for the structure of attitudes in intergroup ideology. The first of these, hidden resistance, has attracted a good deal of attention from scholars of group conscious-

ness, but it is problematic, with serious measurement ambiguities. The concept of paternalism has attracted less interest, and scholars have been wary of its subtleties. I discuss each concept in turn.

Hidden Resistance

An idea that has gained increasing currency among scholars of group consciousness is that a common, stable response to subordination is "hidden resistance," masked by visible acts of compliance (Genovese 1974; Cloward and Piven 1979; Anyon 1984; Scott 1985; 1990). The essence of the argument is that there is more resistance and alienation among subordinates on an everyday basis than a superficial observation of their behavior would suggest. This approach emerged out of the consternation of scholars as to why there has been little sign of an assertive conflict-orientation among subordinates. The argument is that we just were not looking hard enough. The major acts of compliance that are readily observable are depicted as politically expedient acknowledgments by subordinates of the superior might of their oppressors. Underlying such apparent conservatism are definite feelings of alienation from the dominant group which are expressed openly "in house" and which bubble up in acts of petty sabotage, dissembling, and minor infractions of the dominant will. This everyday resistance lies behind the grand theater of compliance (see, especially, Scott 1990, 1–16, 45–69). The expression of resistance is restricted to individual behavior, and it is channeled into hidden outlets that are safe from the scrutiny of the dominant group, or else it is disguised in such behaviors as fawning, feigned stupidity, or incomprehension.

This argument is interesting in that it highlights the subtleties that are nested in the interactions that take place between unequal groups. It also breaks free of the all-or-nothing approach to intergroup conflict. It is still, however, bound by the general parameters of the conflict model, seeking to bring the search for intergroup conflict to a fruitful conclusion by redefining the nature of conflict. In so doing, it introduces some serious definitional ambiguities that sometimes cloud the delineation of resistance beyond recognition. Major acts of compliance are discounted whereas petty acts of noncooperation or sabotage are given predominant weight in the assessment of subordinates' state of mind. Resistance becomes something individual (rather than a group action) and something private or camouflaged (rather than openly combative). Further, scholars acknowledge that acts of hidden resistance often also function as accommodative acts, buffering their practitioners from the worst indignities of subordination as they comply with its requirements (Genovese 1974, 597–598; Anyon 1984). Many acts of day-to-day "resistance" are safe precisely because they conform to the dominant group's expectations (such as beliefs that subordinates are childlike, irresponsi-

ble, emotional, or stupid), and such acts thus also function to reaffirm dominant ideology and legitimize the regime. Indeed, the line between accommodation and resistance becomes faint, and sometimes evaporates altogether (see, especially, Anyon 1984). Malleable rules about which behaviors are valid reflections of the underlying consciousness of subordinates and about how to read specific behaviors turn the observation of human behavior heavily into a matter of interpretation. This makes it difficult to verify or falsify the concept of hidden resistance.

Despite the ambiguities of the concept, we can extrapolate expectations for the structure of subordinate-group attitudes. It implies that subordinates refrain from threatening the dominant group on issues that embroil the conduct of their mutual relationship, but beneath this conservatism lie feelings of estrangement and alienation from the dominant group. These feelings are a part of everyday discourse among subordinates and are thus a part of their consciousness. What this implies, then, is that subordinates have conservative policy views that comply with the dominant will, but that they retain negative feelings toward the dominant group. Any negativism in their behavioral dispositions is restricted to safe areas that are petty or hidden from view.

Paternalism

Dominant attitudes that are paternalistic combine positive feelings toward subordinates with discriminatory policy dispositions. By taking over the definition of subordinates' interests, dominant groups bring themselves to believe that the inequalities that they seek to perpetuate are actually to everyone's benefit. This frees them to practice the inequality without any sense of unpleasantness or hostility. Instead, they can discriminate against subordinates while professing warm feelings toward them and indulging in a gratifying sense of duty and obligation. It is a central premise of the book that this ideological system is a favorite with dominant groups.

Yet paternalism has received remarkably little attention from scholars. Genovese's rich and sensitive (and controversial) portrayal of slavery in the American South features paternalism as the central organizing principle of that system of oppression (Genovese 1974), and Newby has used the concept with effect to analyze both rural labor relations in Britain (Newby 1977a ) and historical changes in the control of industrial labor (Newby 1977b ). There have also been a few scholars who have made use of the concept of paternalism to throw light on race relations or labor relations (e.g., Blumer 1951; van den Berghe 1967; Burawoy 1984).

There is a strain that runs through these discussions of paternalism, as scholars try to reconcile the dominant group's espousal of positive feelings toward subordinates with the presence of exploitation in the intergroup relationship. The subtle but shattering difference between

paternalism and benevolence lends itself both to a misconstruction of the dynamics of paternalism and to an edginess or defensiveness in using the concept. Analysts are torn between assimilating the many expressions of positive affect that emanate from the dominant group and contemplating the brutalities that are also manifested in such a system of inequality. It would seem that analysts experience some disbelief or puzzlement as they are confronted with these apparently contradictory phenomena. Warm feelings seem incongruous with a discriminatory intent, even in the minds of scholars who work with the concept of paternalism.

Indeed, both Genovese (1974) and Newby (1977b ) have argued that a paternalistic system of control gives subordinates a wedge to extract more concessions from the dominant group than they could otherwise obtain. In essence, they believe that the dominant group's espousal of warm feelings toward subordinates poses an internal contradiction in paternalistic systems of control. This contradiction causes a strain that clouds the relationship and leaves the dominant group vulnerable. They argue that subordinates can take advantage of the dominant group's sanctimonious casting of their relationship in terms of a sense of duty and mutual obligation to make a moral claim on some small transfer of resources. Both scholars see dominant groups who espouse paternalism as being vulnerable to giving away more than they bargained for, in order to preserve the legitimacy of their system of inequality. At the same time, the intrinsically demeaning character of paternalism is noted by these scholars, and Genovese is at pains to point out that paternalistic ideology did not prevent white salve-owners from practicing routine violence against their subordinates. These different themes are not integrated, however, and an ambiguity hangs over these analyses.

But it is precisely this ambiguity and subtle deception that makes paternalism such an insidious form of social control, and that generates the lure that it holds for dominant groups. Paternalism offers a way for them to have their cake and eat it too, a way to enjoy the fruits of expropriation without feeling they are taking anything, without having to wrest it away. Far from causing dissonance, expressing affection toward those whom one exploits comes as naturally as retracting your finger from a flame—as long as subordinates are duly compliant and loyal. If analysts sometimes trip up in the morality of paternalism, even as they are observing it from a distance, imagine how well that morality captures that awareness of the hapless participants who are living by its precepts.

In any system of power relations, it is an implicit rule that the dominant party must give some resources to subordinates. In order to facilitate social control, it is important to make sure that subordinates have something to lose. Groups who have nothing to lose pose a dangerous

threat, since they have no stake in the system, no resources to protect. The trick is to keep the exchange of resources as unequal as possible, that is, to extract as much as possible from subordinates while giving as little as possible in exchange. Paternalism, far from lessening the dominant group's control, actually makes it easier for the dominant group to obtain a more favorable exchange. First, paternalism presents the inequality less harshly and thus masks the inequality of the exchange. This leaves subordinates less offended and therefore less guarded. Second, in a paternalistic system, the dominant group has seized more control over the way subordinates' needs are defined. This makes it less likely that subordinates will demand something that the dominant group does not want them to have. And finally, as its trump card, paternalism trades a distinctly invaluable commodity in return for compliance: friendship. If subordinates are dependent on members of the dominant group for the fulfillment of their emotional needs and for their full acceptance in organized social life, the dominant group has an extremely potent weapon with which to extract their compliance.

The coercive potency of paternalism thus draws vitally on the "inconsistent" attitude structure that lies at its core. In order to grasp fully the special dynamics of this ideological system, it is critical to acknowledge this rudimentary point. Feelings toward a group are important, but not in the conventionally understood way of dictating logically consistent policy dispositions. Instead, feelings that are logically inconsistent can enhance the dominant group's ability to practice discrimination. For this reason, I argue that paternalism can become a highly crystallized ideological form that is hard to dislodge. Thus, attitudinal coherence and constraint should not be measured in units of consistency. Dominant groups have every reason to crystallize their thinking in the form of positive feelings toward subordinates and discriminatory policy dispositions.

Data

I begin my analysis by examining the extent to which people's intergroup feelings seem to influence their policy goals in the conventionally expected, logically consistent direction. Next, I investigate the various ways that feelings and policy goals can intersect, first by considering a simplified empirical scheme, and then by examining how the data for race, gender, and class fall into that scheme.

The Relationship between Intergroup Feelings and Policy Goals

How consistent are the data with the expectation that feelings drive policy dispositions, as dominant and subordinate groups face each other in their ongoing relations? Tables 7.1, 7.2, and 7.3 present data that address

that question, for race, gender, and social class, respectively. These tables show the percentages of each group with affirmative, conservative, and reactionary policy goals, according to the distinctions they draw between their own and the other group in their feelings of warmth.

Intergroup feelings are measured in terms of the distinctions people draw between groups, since they are a more sensitive and cogent indicator of affective boundaries than are people's absolute feelings toward the other group. In addition, recall from chapter 5 that people are generally much more likely to express an affective rift between their own and another group than to express outright negative feelings toward the other group. Thus, use of measures of affective differentiation between groups draws out feelings of rift where no absolute feelings of hostility may be present. They make a less exacting indicator of intergroup negativism than would absolute feelings.

Intergroup feelings are categorized slightly differently across the three tables, since the distribution of feelings varies somewhat from one intergroup relationship to another (see chapter 5). And feelings of warmth (rather than closeness) are presented, since the former seem more pertinent to the rudimentary issue of the presence or absence of intergroup hostility.^[1] Similarly, the analyses in chapter 6 indicated that policy goals are reflected more sensitively when they are measured against people's existential perceptions. Policy goals are defined as in chapter 6: affirmative goals are those that advocate the government should be doing more than it currently is doing, conservative goals are those that urge no change , and reactionary goals are those that advocate the government should do less than it currently is doing.

Conventional conceptions of intergroup attitudes anticipate a substantial difference between the policy goals of those who draw intergroup affective distinctions and those who do not. If feelings drive policy goals, we should expect the percentage supporting affirmative policy goals to shift substantially and systematically as one moves from those who draw affective distinctions to those who do not. However, across the three tables, the data fall considerably short of that. Overall, the relationship between intergroup feelings and policy goals might most generously be described as loose. I discuss the data for race, gender, and social class, in turn, before drawing out the general patterns.

Race

Table 7.1 displays the figures for whites and blacks. Among whites, intergroup feelings range primarily between degrees of prefer-

[1] Intergroup feelings of closeness and warmth are in fact very highly correlated (see chapter 5, footnote 4). And when tables 7.1, 7.2, and 7.3 were reestimated, substituting feelings of closeness for warmth, the results were highly comparable.

ence for their own group and affective neutrality, with just a tiny minority of whites (1.4 percent) saying they feel warmer to blacks than to their own group. Among blacks, intergroup feelings vary between degrees of preference for their own group and affective neutrality. In neither group does this variation in intergroup feelings bear distinctly on racial policy dispositions.

Among whites, the relationship between feelings of intergroup warmth and racial policy goals is nonmonotonic and of only moderate strength. There are only modest shifts in the percentage supporting affirmative policy goals (shifts of 10 to 20 percentage points) and somewhat larger shifts in the percentage advocating reactionary goals (15 to 30 points) between those who feel much warmer toward whites and those who draw no affective boundary. Policy goals shift more sharply among those few whites who go so far as to prefer blacks to whites: among those whites, the percentage supporting affirmative goals shifts about another 30 points beyond those who draw no affective boundary. Such an abundance of good feeling, however, is not only rare but it is more than should be required to remove any affective basis for racial discrimination. The three lefthand columns in table 7.1 represent a wide range of inter-

group feelings, and yet the shifts in policy goals that accompany that range of feelings are fairly limited.

Whites who express affective neutrality are, indeed, more likely than those who prefer their own group to take an affirmative stance on racial policies, but these differences are modest. And unless they go so far as to express an affective preference for blacks, it remains only a minority of whites who can bring themselves to support affirmative racial policy goals. The effect of feelings is consistently more pronounced in drawing whites away from reactionary policy opinions. But dislodgement from a reactionary position does not translate into a comparable increase in support for affirmative change.

Among blacks, the overwhelming preference for affirmative racial policies is budged only a little by their interracial feelings. Between those who draw larger and milder affective distinctions, there is no difference in their policy dispositions. And the move from making slight affective distinctions to making no distinctions in feelings of warmth between the two races brings but a moderate shift (of 8 to 20 percentage points) away from affirmative policy goals.^[2]

Gender

Table 7.2 presents the data for gender. Recall from chapter 5 that feelings between men and women span a smaller range than for race or class. In general, neither gender draws an affective boundary between the groups, and feelings do not stray far in either direction. Although women show a stronger tendency than men to prefer their own gender group, women's gender feelings have no bearing on their policy goals. Among men, gender feelings have a modest effect on their policy goals: between men who feel warmer to their own group and warmer to women there is a shift of 11 and 21 points, respectively, in the percentage supporting affirmative policies to provide equal job opportunity and equal legal treatment for women. Most of that difference occurs between those who prefer their own group and those who draw no affective distinctions.^[3]

Class

The data for class are presented in table 7.3. For ease of presentation, interclass feelings are dichotomized into those who feel

[2] Measures of association for the cross-tabulations in table 7.1 tell the same story of only moderate associations between interracial feelings and racial policy goals. For whites, Goodman-Kruskal gammas range from .27 to .40, and Kendall's tau-b's range from .16 to .26 (recall that these measures of association include those whites who express an affective preference for blacks over whites). For blacks, gammas range from .29 to .34, and tau-b's range from .12 to .20. Gammas are of course sensitive to the presence of cell N 's that are zero or small; the more stable tau-b's range from small to modest.

[3] Measures of association for the cross-tabulations in table 7.2 are uniformly trivial: for men, gammas range from .08 to .13, and tau-b's range from .05 to .08; for women, gammas range from .01 to –.04, and tau-b's range from .00 to –.02.

warmer toward their own class and those who make no affective distinction or who prefer other classes; recall from chapter 5 that most people falling in the latter category have neutral interclass feelings (rather than preferring the other class). For respondents in the upper-middle and middle classes, the referent classes are the poor and the working class; for those in the latter two classes, the referent classes are the middle, upper-middle and upper classes. In both cases, people's feelings between their own and the referent classes are averaged.

Interclass feelings have only a modest impact on policy goals. Among the upper-middle class, there is a difference of 24 percentage points between those who prefer their own class and those who do not in their likelihood of taking an affirmative position on the issue of government-guaranteed jobs; but on the other two issues, the comparable percentage difference is only about 9 points. Among the middle class, interclass feelings show even less connection to people's policy goals. The impact of feelings is virtually restricted to drawing people away slightly from reactionary goals on two of the issues (providing welfare benefits and guaranteeing a minimum income)—and the percentage difference there is only about 11 points and 5 points, respectively.^[4]

Among the working class, the effect of feelings is again quite modest: between those who express warmer feelings toward their own class and those who do not, there are percentage differences of just 6 to 10 points in the likelihood of taking an affirmative position on any of the three policy issues. Among the poor, the effect of feelings is just a notch stronger than in the working class.^[5] Regardless of their interclass feelings, very few poor people endorse reactionary policy positions, but they are slightly more likely to move away from conservative policy positions to affirmative ones (by 9 to 21 percentage points) when they have warmer feelings toward their own class.^[6]

[4] Data from tables that trichotomize interclass feelings indicate that, in the upper-middle class, there is little difference in policy views between those who prefer lower classes and those who have neutral feelings. In the middle class, however, those who prefer lower classes are more likely to take an affirmative policy stance whereas those with neutral feelings are little different from those who prefer their own class.

[5] Because so few members of the poor and working class express warmer feelings for higher classes than for their own class, an attempt to cross-tabulate policy goals with a trichotomized version of interclass feelings resulted in cell N 's that were too low to yield reliable data for those who preferred higher classes.

[6] Measures of association for the class cross-tabulations in table 7.3 are as follows: for the upper-middle class, gammas range from .13 to .44 and tau-b's range from .07 to .23; for the middle class, gammas range from -.02 to .16 and tau-b's range from -.01 to .08; for the working class, gammas range from .08 to .22 and tau-b's range from .05 to .10; and for the poor, gammas range from .19 to .50 and tau-b's range from .10 to .23. The occasional higher gammas are attributable to the presence of empty or near-empty cells in those cross-tabulations. The more stable tau-b's remain in a range from zero to modest.

General Patterns

Overall, the relationship between intergroup feelings and policy goals ranges from zero to moderate. The policy dispositions of whites show more of a connection to their intergroup feelings than is the case for other groups, but even among whites, the relationship is only moderate. Nor does the relationship between intergroup feelings and policy goals fall into a straightforward monotonic pattern that affects the main body of responses in a single thrust. In addition, there is no systematic tendency for the relationship between feelings and policy dispositions to be affected by the position of the group to which the person belongs (that is to say, dominant or subordinate), or by the type of policy issue at hand (that is, moderate versus more redistributive policies).

Instead, the impression these tables convey is that the prevailing preference of subordinate groups for affirmative policies and aversion of dominant groups to those same policies is buffeted only a little by the intergroup feelings that individuals hold. Clearly, people's intergroup feelings are too loosely coupled with their policy goals to be construed as the force that is driving the formulation of those goals.

Alternative Attitudinal Compounds

The analysis thus far has conceived of the relationship between intergroup feelings and policy goals in the conventional way—that is, in terms of the "consistency" between the two. That approach looks for positive feelings to go with affirmative policy goals and negative feelings to go with conservative or reactionary policy goals among the members of dominant groups and for the opposite relationship to hold among subordinate group members. I ask the reader now to abandon that mold of thinking. It restricts our thinking about intergroup attitudes to a narrow track that bypasses the very types of attitudes that are, I argue, the primary draw for groups. I suggest a new approach that delineates the various ways that intergroup feelings and policy goals may intersect.

An Empirical Scheme

In order to clarify the issues, figure 7.1 presents a simplified scheme that depicts the main ways that intergroup feelings and policy goals can intersect, for people in dominant and subordinate groups, respectively. Each axis is dichotomized. On the horizontal axis, intergroup feelings are divided into (1) those making no distinction between groups or even showing a preference for the other group (labeled inclusive feelings), and (2) those expressing a preference for their own group over the other (labeled estranged feelings). On the vertical axis, policy goals are divided into (1) those advocating affirmative change (that is, that the government should do more than it currently is doing), and (2) those taking conservative or reactionary positions (that is, that the government should do no more, or less than it is currently doing). Four

Figure 7.1
Paternalism and conflict: Alternative configurations of intergroup feelings and policy goals for dominant and subordinate groups.

distinct compounds of feelings and policy goals may thus be identified in dominant and subordinate groups.

The first thing that is highlighted by this scheme is that the meaning of particular configurations is altered entirely according to whether one is examining the attitudes of dominant-group members or subordinates. Among the members of dominant groups, those configurations that fall on the diagonal going from top left to bottom right (revisionist and conflictive attitudes) are the "consistent" ones that have been the target of scholarly scrutiny. They combine inclusive feelings with support for affirmative policy change or estranged feelings with conservative or reactionary policy goals. These are the attitudinal configurations that are commonly termed "unprejudiced" and "prejudiced," respectively.

Of the two "irregular" cells, one—the compounding of estranged feelings with affirmative policy dispositions (tolerant )—has received considerable attention from students of political tolerance. The other formally inconsistent cell—the compounding of inclusive feelings toward subordinates with conservative or reactionary policy goals (paternalistic )—has been bypassed in conventional approaches to public opinion and has fallen by the wayside as an awkward anomaly. It is this configuration, however, that holds the most promise for scholars in unraveling the dynamics of intergroup ideologies.

Among the members of subordinate groups, it is the configurations that fall on the opposite diagonal, going from top right to bottom left (conflictive and deferent attitudes), that have conventionally been treated as consistent. Subordinates are logically expected either to take a conflictive approach by expressing estranged feelings from the dominant group and demanding affirmative change, or to be deferent toward the dominant group by espousing inclusive feelings toward them (possibly even showing a higher regard for the dominant group than for their own) and adopting conservative or reactionary policy views.

Neither of the two "irregular" cells has been the object of investigation in public-opinion research on subordinate group consciousness. However, the accommodative cell (that combines estranged intergroup feelings with conservative or reactionary policy dispositions) is certainly implied by theories that postulate that subordinates express "hidden resistance" (Genovese 1974; Cloward and Piven 1979; Anyon 1984; Scott 1985, 1990). Such theories imply that although subordinates generally comply with dominant group demands, their compliance is nothing more than a tacit acknowledgment of the superior power of the stronger group that masks an underlying alienation in spirit that is manifested in small, hidden acts of resistance. If the compliance of subordinates is merely a veneer, we would expect them to populate the accommodative cell, espousing conservative or reactionary policy views but retaining alien-

ated feelings from the dominant group. Accommodation among subordinates might be regarded as the logical counterpart to tolerance in the dominant group: in each case, estranged feelings must be overridden to behave in a supportive way toward the other group.

The opposite "irregular" cell—the combination of inclusive feelings toward the dominant group with calls for affirmative change—I have labeled integrationist , because it reflects the spirit of the movement for racial integration in the United States in the 1950s and 1960s, as advocated by Dr. Martin Luther King, Jr. Despite the prominent position of the civil rights movement in recent U.S. history, this attitudinal compound has had no place in conventional analyses of group consciousness, since it is logically inconsistent. This configuration of feelings and policy goals might be regarded as the subordinate counterpart to paternalism in dominant groups: subordinates might find it strategic to moderate their demands for change with assurances that they feel no hostility toward those from whom they wish to extract concessions.

The simplified scheme that I have outlined identifies four main ways that dominant-group members and subordinates may blend their intergroup feelings and policy goals. The meaning of any specific compound of feelings and policy dispositions cannot be interpreted out of context but depends on whether the exponent is a member of a dominant or a subordinate group. Only one term is applicable to both kinds of groups—conflictive attitudes—but note that it is constructed quite differently, depending on whether one occupies a dominant or a subordinate position in an intergroup relationship. All the other terms are unique to dominant or subordinate groups. For example, it is not germane to ask whether subordinates are willing to override their estranged feelings and grant equality to the people who exceed them in status and perquisites. Similarly, the attitudinal state of deference is a prospect that members of dominant groups are spared.

Fitting the Data

The next step is to apply this empirical scheme to the data for race, gender, and class. To that end, figures 7.2, 7.3, and 7.4 present the percentages of the pertinent dominant and subordinate groups expressing each of the four possible attitudinal compounds, for race, gender, and social class, respectively. Intergroup feelings are measured in terms of warmth or coldness. And recall that since feelings are measured in terms of affective differentiation (rather than absolute feelings toward other groups), the effect is to create more respondents who are categorized as having "negative" feelings, since more people draw affective distinctions between groups than express outright negative feelings toward the other group. Policy goals are reported separately for each specific policy, in order to capture the variant reactions that

different policies elicit. Thus the patterns of attitudes can be mapped in the form in which they find expression in the evolving policy debate—around specific policy issues.

The construction of these figures focuses our attention on the structure of intergroup attitudes in a different way than the conventional approach. Instead of asking what percentage of a group with particular intergroup feelings support logically "consistent" policy goals, we now catalog the varying ways that intergroup feelings and policy goals intersect. The concern becomes, what percentage of each group compound their intergroup feelings and their policy goals in various alternative ways?

Two tiers of questions await us. First, what proportion of responses are logically consistent or inconsistent, and does that vary between dominant and subordinate groups or across the three intergroup relationships? Second, we move to the more important set of questions, which are unconstrained by concerns with logical consistency: What kinds of attitudinal compounds prevail among dominant and subordinate groups? What are the attitudinal blends by which dominant groups rule or concede and by which subordinate groups comply or resist? And what is the fit between the attitudinal compounds that are favored by dominant groups and the subordinates they seek to control?

Consistency

First, what percentage of responses fall into the logically consistent cells?^[7] About 51 to 56 percent of whites, about 40 percent of men, between 41 and 63 percent of the upper-middle class, and 42 to 56 percent of the middle class espouse feelings and policy dispositions that are either consistently positive or consistently negative. Among subordinate groups, the comparable figures for those taking either a consistently conflictive or consistently deferent position are about 60 percent of blacks, about 55 percent of women, 50 to 59 percent of the working class, and 58 to 68 percent of the poor.

In general, then, the proportion of responses that could be considered conventionally consistent falls in the range from somewhat below one-half to just over one-half. The proportion giving consistent responses is somewhat less in the context of gender relations than for either race or

[7] Scholars have conventionally relied on measures of association to assess attitudinal consistency, but, as Weissberg (1976) has suggested, a more direct measure of attitudinal consistency is the percentage of respondents giving consistent responses—that is, the percentage giving responses that fall on the main diagonal when two attitude items are cross-tabulated. As Weissberg points out, measures of association can be depressed when there is restricted variance on the items (even if responses are very consistent), and two items can be strongly associated in the absence of consistency if responses on two items fall in the same rank-ordering.

class relations. And in all three intergroup relations, dominant groups show a slightly greater aversion to logically consistent responses. However, the proportion of logically consistent responses does not vary greatly from one intergroup relationship to another, or between dominant and subordinate groups.

Alternative Compounds

Overall consistency may not vary dramatically across groups or intergroup relationships, but there is substantial variance in the types of consistent responses that prevail. We turn now to the second tier of questions: What blends of feelings and policy goals are espoused by dominant and subordinate groups in the three intergroup relations? I start by comparing the prevalence of alternative types of "consistent" responses, and I then consider the prevalence of alternative "inconsistent" compounds. Finally, I evaluate the way dominant and subordinate attitudes dovetail with one another to form a patterned exchange in each intergroup relationship.

Among whites, consistently conflictive responses far outweigh consistently revisionist responses, by a ratio of at least two to one. And on the issue of integrated schools, conflictive responses outnumber revisionist responses among whites by almost four to one. Among men, however, consistently revisionist responses outnumber conflictive responses by more than ten to one: almost no men take a conflictive stance in gender relations. In the upper-middle and middle classes, the ratio of revisionist to conflictive responses fluctuates, depending on the extent to which the specific policy is redistributive. On two of the policy issues, revisionist responses outnumber conflictive ones in both the upper-middle and middle classes, by ratios between two to one and four to one. But on the issue of welfare benefits, conflictive responses either equal or slightly outnumber revisionist responses.

On the subordinate group's side, consistently conflictive responses far outnumber consistently deferent ones among blacks and the poor and working class. Among blacks, conflictive responses prevail by more than two to one on the issue of school integration (the most sensitive of the three race policy issues) and by more than five to one on the other two race policy issues. Only between 8 and 17 percent of blacks assume a deferent orientation in their racial feelings and policy goals. Among the poor and working class, the margin favoring conflictive responses over deferent ones is slightly smaller than for race. And on the most sensitive class policy issue (welfare benefits), the margin in favor of conflictive responses drops to about two to one among the poor, and it disappears completely among working class respondents. The poor and working class thus manifest just slightly more deference than do blacks, with the figures ranging from 14 to 27 percent among the working class and 12

to 19 percent among the poor. Among women, the ratio of deferent responses to conflictive ones flips around completely from that observed for blacks and the poor and working class. For women, deferent responses outnumber conflictive ones by more than four to one, with almost one-half of women taking a deferent position.

Thus, over the three intergroup relations, conflictive orientations are most heavily in evidence among whites and blacks: approximately 40 percent of whites and 44 to 54 percent of blacks take a conflictive stance. Gender relations manifest the smallest evidence of conflictive orientations: fewer than 4 percent of men and about 10 percent of women take this stance. In class relations, the poor manifest about the same proclivity as blacks for a conflictive orientation, and the working class lag just slightly behind them. The upper-middle and middle classes, however, do not match the propensity of whites for a conflictive orientation: among the upper-middle class, about 17 to 25 percent take a conflictive stance, and among the middle class, conflictive responses drop to between 10 and 23 percent. Note that in all three intergroup relations, even in gender relations, subordinates are more drawn to conflictive responses than are the members of dominant groups. I return to this point below.

The response pattern that draws dominant groups more than any other is a technically inconsistent one: paternalism. Paternalistic attitude blends are most in evidence among men, of whom about 60 percent evince such a response pattern. Among the upper-middle and middle classes, paternalism accounts for about 25 to 56 percent of responses, with the sensitive welfare-benefits issue drawing the most paternalistic responses. Finally, even among whites, whose propensity for conflictive responses exceeds that of any other dominant group, still between 34 and 44 percent take a paternalistic position—roughly the same percentage as take a conflictive position.

The alternative inconsistent attitude blend for dominant groups—tolerance—is the one that has received much more attention from students of intergroup attitudes, but it is the least common response pattern among dominant groups in all three intergroup relationships. Only about 5 to 10 percent of whites and a minuscule 1 to 2 percent of men indulge in affirmative policy dispositions with feelings that are estranged from the other group. Among the upper-middle and middle classes, as many as 10 to 20 percent take a tolerant position on the two relatively tame policies of guaranteed jobs and guaranteed minimum income, but only 4 to 7 percent ascend to a tolerant position on the more threatening issue of welfare benefits. It seems that it is far easier to discriminate against groups one likes than to give equal treatment to groups from whom one feels estranged.

A similar but less extreme phenomenon occurs among subordinates. In general, it seems to be easier to demand affirmative change from a group one likes (the integrationist response) than to accept the status quo from a group from whom one feels estranged (the accommodative response). Among blacks, integrationist responses sharply outnumber accommodative ones, with about one-third of blacks taking an integrationist position and only between 2 and 12 percent taking an accommodative stance. Interestingly, it is on the explosive issue of integrated schools that blacks yield the most. The figures are similar for women, with about one-third or more of women taking an integrationist position and about one-tenth taking an accommodative posture. Among the poor, integrationist responses have a smaller edge over accommodative ones, with about one-quarter of the poor expressing integrationism and about one-tenth being accommodative on the two tamer policy issues and about 20 percent taking each position on the tougher issue of welfare benefits. Among the working class, integrationism loses its edge over accommodative responses. On the two less demanding issues, about one-quarter of the working class take an integrationist stance, with either a smaller percentage or the same percentage taking an accommodative stance. And on the tougher issue of welfare benefits, integrationist responses are far outweighed by accommodative ones in the working class, with only 13 percent being integrationist and 37 percent being accommodative. Thus, overall, subordinates find it more difficult to be accommodative than to be assertively upbeat—but on issues that find stronger resistance among dominant groups, subordinates are somewhat more likely to yield, despite estranged feelings.

Finally, I consider how dominant and subordinate response-patterns dovetail with one another. Surely the response pattern that dominant groups crave from subordinates is deference—the willing and friendly accession to the modus vivendi. To that end, dominant groups drift easily into paternalism, extending the hand of friendship as they withhold resources. How successful are they in their endeavor? Is paternalism rewarded with deference, or is it met with rebuff?

It is in gender relations that paternalism has its fullest flowering, with about 60 percent of men expressing a paternalistic compound of warmth and discrimination. Consistently positive responses are the only other compound to be expressed by any significant number of men, with about one-third being revisionist and only minuscule numbers taking either a tolerant or a conflictive approach. In short, men's gender attitudes commonly include discriminatory policy goals, but estranged feelings from women are virtually absent. And it is in gender relations that deference is most in evidence: close to half of women reward men's paternalism with deference. Some women seek affirmative change, but when they

do so they are much more likely to take the route that is the subordinate counterpart to dominant paternalism—integrationism—than to be flagrantly conflictive. About one-third of women are integrationist. Only about 10 percent of women venture into a conflictive mode, and another 10 percent take the accommodative route. Note, however, that much as deference is the prevailing mode among women and conflictive attitudes uncommon, still deference is less in evidence among women than paternalism is present among men. And, by the same token, a few more women than men launch into a conflictive approach. It seems that the paternalistic-deferent mode of interaction holds somewhat more appeal for men than for women.

In class relations, the upper-middle and middle classes appear primarily as either revisionist or paternalistic, depending on the policy issue at hand. On the two more moderate issues, revisionists tend to equal or outnumber paternalists, but on the tougher issue of welfare benefits paternalists lead the way, with over half of these classes taking a paternalistic line on the issue of welfare benefits and only 15 to 19 percent venturing into revisionism. The poor and working class are much less likely to show deference than are women, although deference is slightly elevated on the sensitive issue of welfare benefits. And as in gender relations, fewer subordinates exude deference than seems called for by the rate of paternalism among dominant group members—indeed, this imbalance is stronger than in gender relations. Over the three policy issues, between 12 and 27 percent of the poor and working class express deference. Matching the class imbalance between paternalism on the one hand and deference on the other, the poor and working class are also considerably more likely than the upper-middle and middle classes to engage in a conflictive approach. Note, however, that on the tougher issue of welfare benefits, the conflictive tendencies of the upper-middle and middle classes are somewhat heightened and the temerity of the poor and working class slips somewhat. Over the three policy issues, between one-tenth and one-quarter of the upper-middle and middle classes take a conflictive stance, whereas the comparable figures for the poor are about 40 to 55 percent and for the working class about 23 to 46 percent.

The exchange of attitudes between blacks and whites is especially revealing. As whites face blacks, they find a group that is more combative than any other subordinate group. Among blacks, conflictive responses are at about the same high level as among the poor (44 percent on the sensitive issue of integrated schools and 51 to 54 percent on the other two issues). In addition, the level of deference among blacks is lower than in any other subordinate group (17 percent on the issue of school integration and less than 10 percent on the other two issues). The response among whites is that a higher percentage is drawn to the conflic-

tive mode than is the case for dominant class or gender groups—approximately 40 percent of whites assume a conflictive orientation. At the same time, however, it is interesting that whites still do not match the level of conflictive orientation found among blacks. In addition, despite the low levels of deference and high levels of conflictive attitudes among blacks, still about 40 percent of whites cling to a paternalistic mode—just as many whites as take a conflictive approach. Despite the failure of paternalism to elicit any significant level of deference from blacks, many whites still cling to this comforting ideological mode.

In each case, then, it is subordinates who seem to lead the way on the path to conflict—albeit with caution. In gender relations, this is manifested in women's slight lag in returning deference for paternalism and in their timid lead in venturing into conflictive dispositions. In class and race relations, where paternalism has less of a grip, these asymmetries between dominant and subordinate groups are more pronounced. The middle and upper-middle classes remain more likely to express paternalism than outright conflictive dispositions, and whites are as likely to express paternalism as conflictive dispositions, even as these dominant groups are confronted with subordinates who generally fail to be deferent and who seem more drawn to the conflictive path.

Conclusions

Consistency has been the centerpiece of theory and empirical research about attitude structure, mass political opinion, prejudice, and class consciousness. Attitudinal expressions that reflect something tangible have been routinely expected to follow the precepts of consistency. Only one attitudinal construct that involves a logically inconsistent configuration has been subjected to sustained inquiry, and that is tolerance—the extension of equal rights to subordinates despite negative feelings toward them. This construct has generally been regarded as the exception that proves the rule, a testament to the power of democratic norms to reshape humans' basic proclivity to form policy dispositions toward a group on the basis of their feelings toward the group.

Outside the bounds of attitude research, two concepts have been introduced that have important implications for the way intergroup attitudes should be conceived, but these implications have not been fully realized. The two concepts are hidden resistance and paternalism. Hidden resistance implies that subordinates retain negative feelings toward the dominant group as they yield to its will. Paternalism implies that dominant groups fuel their discriminatory dispositions toward subordinates, not with negative feelings, but with positive ones.

My approach, abandons the underlying premise on which rests the commonly held tenet that consistency is the glue of well-formed attitudes. Instead of portraying attitudes primarily as a property of individuals, I treat them primarily as the property of groups. The attitudes that prevail in a group are shaped by its collective experience: the moral framework imposed by societal institutions, and the political pressures and opportunities emanating from the structure of the intergroup relationship and the behavior of the other group or groups in that relationship. Individual group members are the agents of ideological forms, not the autonomous creators of those forms. Attitudes are selected from the available repertoire, not for their tidy logic, but for their political utility. As exigencies shift, so too does the array of attitudes in a group: new patterns of response work their way through information channels and gradually supplant older forms that no longer prove serviceable.

The data in this chapter indicate that, in all three intergroup relationships, and among both dominant and subordinate groups, the direction of intergroup feelings has only a moderate influence on policy dispositions. Like many other students of public opinion, I have found that consistency is not the hallmark of popular attitudes. However, I do not infer from this that popular attitudes are unwieldy or incoherent. Instead, I have argued that the question of how feelings and policy goals intersect should be laid open empirically. I outlined a simple empirical scheme that identified alternative configurations of intergroup feelings and policy goals. Each configuration holds a distinct meaning for the ideological messages that flow back and forth through an intergroup relationship.

An examination of the data through this lens demonstrates that formal consistency does not mechanically stamp the intersection of people's intergroup feelings and their policy goals. Instead, the importance of consistency appears stronger or weaker, as a function of the direction of people's feelings. Among both dominant and subordinate groups, negative intergroup feelings are more likely to find expression in logically consistent policy dispositions than are positive feelings.

Thus, the members of dominant groups find it very difficult to maintain affirmative policy dispositions toward subordinates unless they hold inclusive feelings toward them. The inconsistent attitudinal compound of tolerance does seem to stretch most people's capabilities too far: estranged feelings from subordinates are hard to repress when dominant group members formulate their policy goals. Conservative and reactionary policy dispositions, however, do not require estranged feelings to fuel them, and, indeed, they are more often fueled by inclusive feelings. The inconsistent attitudinal compound of paternalism draws as many whites as does a flagrantly conflictive approach, and it occupies a slight

lead over conflictive attitudes in the upper-middle and middle classes and an overwhelming lead over conflictive attitudes among men. Overall, inclusive intergroup feelings are more likely to accompany discriminatory dispositions than are estranged feelings, and holding inclusive feelings toward subordinates is more likely to result in discriminatory policy dispositions than in affirmative ones.

Among subordinate groups, alienated feelings from the dominant group are more likely to produce a consistent pattern of advocating affirmative policy change than they are to be found simmering quietly behind conservative acceptance of the status quo. It seems to be almost as difficult for subordinates to keep their negative feelings tamed in the formulation of their policy goals as it is for dominant group members, and unhappy accommodation to the dominant will is almost as unusual among subordinates as is tolerance in dominant groups. But, as with dominant groups, inclusive intergroup feelings provide less of a guide than do estranged feelings as to the way policy goals will be formulated. Inclusive feelings toward the dominant group sometimes lead to a deferent acceptance of the status quo and sometimes accompany a call for change. Indeed, in race relations, integrationist responses resoundingly outnumber deferent ones, and in class relations, integrationism usually predominates over deference. In gender relations, where deference is more common than in either of the other two intergroup relationships, integrationist responses are clearly outnumbered by deferent ones, but still a substantial minority of women (one-third or more) match inclusive feelings toward men with the advocacy of affirmative change.

We may think about integrationism as the subordinate counterpart to paternalism in dominant groups, whereby each group packages the pursuit of its interests with the expression of positive feelings toward the other group. Just as paternalism is an intuitively shrewd way to soften the presentation of dominant group interests, integrationism relies on the expression of affective inclusiveness toward the dominant group to sweeten the advocacy of affirmative change. It is interesting that dominant groups seem more drawn to such an ideological strategy than are subordinate groups.

We know from the data in chapter 5 that subordinates are more likely to withdraw emotionally behind group lines than are the members of dominant groups. The experience of underprivilege is more of a stimulant to the development of emotional boundaries between groups than is the experience of privilege. When subordinates develop estranged feelings from the dominant group, they find it more difficult to contain those feelings and continue endorsing the status quo than to translate them into an assertion of affirmative policy goals. Subordinates seem more inclined to engage in open resistance than hidden resistance. But

even though the dominant group is getting less compliance from subordinates than it seeks, dominant-group members generally resist the pull toward either affirmative change or intergroup conflict. Confined by their unwillingness either to grant concessions or to engage in flagrant conflict, the members of dominant groups find refuge in the haven of paternalism.

In an intergroup relationship that is structured with frequent, intimate, one-to-one contacts across group lines, paternalism holds the day. Almost two-thirds of men are paternalistic, and they are rewarded with deference from almost half of women, considerably more deference than is found in any other subordinate group. In race and class relations, the structure of interaction across group lines is less favorable to paternalism. The extreme spatial segregation of blacks from whites has seriously restricted whites' opportunities to invade the cognitive and emotional sensibilities of blacks, and blacks show less inclination to be deferent than does any other subordinate group. Yet even as about half of blacks take a conflictive orientation and only a tiny minority are deferent, still whites lag behind blacks in their readiness to be openly conflictive, and they are as inclined to cling to paternalism as they are to be conflictive. The structure of class relations is marked primarily by spatial segregation and aggregated communication channels between classes, especially between the highest and lowest social classes, but this pattern is broken by limited opportunities for personal contacts across the boundaries of adjacent classes. The former factor undercuts the suitability of class relations for the practice of paternalism, but the latter introduces some opportunities to penetrate the awareness of at least those subordinates who are most adjacent to one's own class in the step-wise chain of class contacts. These dual factors work to mitigate the tendencies toward conflict: even as the poor, and to a lesser extent the working class, manifest a pronounced proclivity for conflictive attitudes and only slightly more deference than do blacks, the upper-middle and middle classes retain a greater propensity for paternalism than for conflictive attitudes.

In the ongoing political game between dominant and subordinate groups, it is subordinates who lead the way to conflict, but even their efforts in that direction are wary. As each side attempts to engage the other while trying to avoid a strong counterreaction that could jeopardize current holdings, bold moves are a rarity. Dominant and subordinate groups demonstrate their mutual sensitivity in the way they shift their positions in step with each other as they move from one policy issue to another that is more, or less, threatening to the core of their relationship. On those issues that have less redistributive implications, such as guaranteed jobs in the class arena, dominant-group members are more inclined to be revisionist and subordinates push a little harder. On issues that are

more threatening, such as the class issue of welfare benefits, dominant group members move from tolerance to conflict and from revisionism to paternalism: subordinates retreat accordingly from conflict to accommodation and from integrationism to deference. By responding differentially to the various policy issues that are raised in an intergroup relationship, dominant groups can thus steer their attentive subordinates away from those issues that pose the greatest threat to dominant interests.

As the game continues, flexibility and differentiation are more common traits on both sides than is an unerring consistency. Far from indicating a lack of coherence or salience, these traits reflect the acute political sensitivities of both sides as they pursue their interests with intuitive care. For the same reasons, the ideological patterns that prevail are not flagrantly conflictive. Both dominant and subordinate groups show an inclination to package the pursuit of their group's interests in positive, rather than negative, feelings. This task is more difficult for subordinates, who must bear the daily offenses of underprivilege, and who are thereby often provoked into feelings of estrangement from the dominant group. But for the members of dominant groups, who can usually contemplate their daily experiences with greater equanimity, paternalism holds a magnetic lure that lingers on even as they are jostled by the wary and fitful ventures of subordinates into conflict. It proves difficult to abandon the trappings of an ideological system that so delicately obscures the raw pursuit of your own group's interests.

Chapter Eight—
The Cognitive Embroidery of Intergroup Relations

The way people portray the vital characteristics of their own and other groups has drawn the attention of scholars of prejudice, gender ideology, and class and race consciousness. Two kinds of cognitive attributions have received particular attention. First, there is a large and venerable literature on personality trait attributions to groups, commonly termed stereotypes. Second, there has been a growing literature on the way people explain the relative standing of contending groups, with particular interest in whether people locate the causes in the internal characteristics of the affected groups (such as inherent biological attributes or subcultural mores) or in biases of the social structure (such as the structure of rewards and opportunities).

Research on stereotypes has been tied most closely to the analysis of racial and ethnic prejudice. Recall that research on prejudice has been guided by Allport's classic definition of that phenomenon as "an antipathy based on a faulty and inflexible generalization" (Allport 1954, 9), thus giving stereotypes a pivotal role in the generation of prejudice. This role for stereotypes has been absorbed into their intrinsic conception, even when they are studied in other contexts, such as in attitudes toward women or the poor.

The way people explain differences between groups has been regarded as critically important by a growing number of scholars in predicting policy dispositions toward the respective groups. Research on popular causal attributions to groups has been directed primarily to the explication of racial ideology and opinions about poverty. The classic archracist or class elitist is expected to believe in immutable, biological differences between the pertinent groups, whereas the person who seeks change is thought to draw on a cognitive scheme that posits a critique of the way structural arrangements disburse resources to contending groups.

The recurrent concern has been the extent to which subordinates are set apart, categorized, and derogated in popular beliefs. The underlying purpose has been to distinguish "good" beliefs from "bad" ones, to assess the extent to which people indulge in beliefs that are irrational, untrue, exaggerated, negative, simplistic, and inflexible. The identification of these elements in popular beliefs is deemed critical because they are thought to play a pivotal causal role in the endorsement of discriminatory practices. It is held that among dominant-group members, cognitions that are derogatory and categorical fuel hostile feelings and discriminatory dispositions; among subordinates, the acceptance of such beliefs undermines self-esteem and thus renders them passively uncritical of discriminatory arrangements. In short, specific kinds of beliefs are identified as damaging to the subordinate group—if those beliefs could be erased, the cognitive basis for the inequality would be dissolved.

I approach intergroup attributions from a different perspective. Although individuals' beliefs about groups surely serve to justify their behavioral predispositions and to frame their expectations about group members, I hold that it is a mistake to attach any genuine causal significance to those beliefs. People's intergroup policy goals are not driven by the kinds of intergroup beliefs they hold—instead, policy goals derive directly from the pursuit of group interests within a specific political context. Intergroup beliefs serve as cognitive props to explain, justify, and perpetuate those goals. In this framework, the significance of intergroup cognitive attributions is as cultural artifacts that embellish the way one group orients itself toward another. Cultural beliefs about the salient characteristics of unequal groups symbolize each group's attempts to define and explain social reality—they embroider the ideological representations that are exchanged between groups. The beliefs that unequal groups have about each other are subject to the same constraints as other aspects of their intergroup attitudes, and, thus, self-serving beliefs do not inevitably translate into hostile beliefs. Indeed, the pernicious quality of intergroup beliefs does not lie in any fixed, constantly identifiable quality—but rather in their malleability.

The focus of past literature on categorical, derogatory beliefs has misdirected our attention from more subtle and insidious forms of cognitive attribution that can pervade cultural interpretations of inequality. The problem is not whether the members of dominant groups categorize and derogate subordinates and whether subordinates accept or reject their assigned image as inferior and less worthy. History should have taught us that the reasoning that supports systems of inequality rarely follows such a straightforward design. Instead, the cognitive struts of inequality follow various and often convoluted paths, as people grope for stories that address the sundry tensions pervading an intergroup relationship.

Stories are framed in terms that conform to the current moral code and they shift to keep up with altered exigencies. In this ongoing process, those stories are more notable for their slippery variegation than for their rigidity, constancy, or logical tightness of reasoning. After all, there is more than one way to skin a cat. As we sift through the tangle of popular narratives about the pertinent characteristics of unequal groups, we must treat those narratives for what they are—interpretive stories about the modus vivendi that reveal the underlying pressures and strains demanding popular explanation.

What are the alternative modes of cognitive attribution that dominant groups develop, under different constraints, to interpret their unequal relations with another group, and how effectively can they limit the cognitive reactions of subordinates? As we move from paternalistic to more conflictive relations, what kinds of definitions and explanations of group differences gain popular credence among members of dominant groups, and how persuasive are those messages with subordinates?

In this chapter, I assess the cognitive attributions that shade different kinds of intergroup relationships. This begins with a discussion of the concerns, assumptions, and measurement strategies that have governed past work on cognitive attributions to groups, and I then delineate my approach. With those issues in hand, I turn to my data. The ensuring analysis of the cognitive props of race, gender, and class relations focuses on two sets of questions. First, I examine the ways in which people describe the salient personality characteristics of their own and other groups. Specifically, I examine the extent to which such descriptions entail derogation, sharp intergroup distinctions, and categorical imagery. Second, I investigate the perception of intergroup differences, more broadly conceived, and the popular causal attributions that are adduced to explain such differences. These two kinds of data inform us about the types of cognitive representations that prevail among dominant and subordinate groups as they face one another under varying constraints. What does the cognitive embroidery of an intergroup relationship tell us about the specific sensitivities and moral proclivities of the affected groups? More generally, what can we learn about the quicksand of folk wisdom as it absorbs the problematics thrown up by different forms of social inequality?

The Issues

Stereotypes of subordinates and folk explanations of perceived group differences have both been assumed to play a portentous role in motivating negative feelings and discriminatory policy dispositions toward subordinates. The intimate conceptual link between stereotypes and preju-

dice continues to impel much research on stereotyping among students of prejudice as well as in some applications to gender and class attitudes. More recently, interest in intergroup beliefs has broadened to include folk explanations of group differences, and there has been a growing body of research on how people account for the differences they perceive between unequal groups.

The exact role played by intergroup beliefs in people's attitudes toward a target group has not usually been specified, and some unacknowledged and unresolved ambiguity hovers over this issue. The prevailing theme has been that such beliefs arise out of ignorance and parochialism and that, once installed in the person's mind-set, they establish expectations about the members of the target group and they thus function to motivate the person's feelings and dispositions toward the group. This theme is consistent with Allport's definition of prejudice as "an antipathy based on a faulty and inflexible generalization" (Allport 1954, 9). The term prejudice itself is derived from the notion of prejudgment and carries a strong connotation of a preexisting bias based on a prejudgment. Some scholars have attempted to test specific models based on this conception (for example, Rosenberg 1956; Rokeach, Smith, and Evans 1960; Stephan and Stephan 1984; Kluegel and Smith 1986), but more commonly this conception of the causal dynamics has simply underwritten empirical research on prejudice.

For example, the two most widely suggested policy interventions for eradicating or diminishing prejudice are built on that causal conception. The contact theory of prejudice postulates that with high levels of personal, equal-status contact between groups, dominant-group members learn the invalidity of their stereotypes and thus shed their prejudice (for example, Deutsch and Collins 1951, 1965; Wilner, Walkley, and Cook 1955; Miller and Brewer 1984; Stephan and Stephan 1984; Hewstone and Brown 1986; Hewstone 1989). And a policy that has been rapidly gaining attention in college campuses throughout the United States is the introduction of required ethnic studies courses in undergraduate liberal arts curricula: here the purpose is again to eradicate stereotypical thinking about minority ethnic groups in order to undermine prejudice. In short, stereotypes are considered the direct product of ignorance and parochialism: by reeducating people, stereotypes can be dispelled and thus the cognitive basis for intergroup hostility is removed.

At the same time, there has been a secondary theme that intergroup beliefs are rationalizations, adduced by the individual to justify preexisting discriminatory feelings and policy dispositions. This secondary theme is sometimes injected into discussions of stereotypes. If taken literally, it contradicts the primary theme noted above, since it suggests that inter-

group feelings and policy dispositions temporally precede, rather than follow, beliefs about the group. This, in turn, has distinctly different implications for theories of prejudice. However, this conception of the causal direction is not usually introduced as an alternative to the primary conception, but rather as a supplement. For example, in The Nature of Prejudice , Allport credits stereotypes with a pivotal role in causing intergroup antipathy in his basic definition of prejudice in chapter 1 (quoted above), but in his subsequent discussion of stereotypes in chapter 12 he proceeds with the following elaborations:

. . .a stereotype is an exaggerated belief associated with a category. Its function is to justify (rationalize) our conduct in relation to that category.

In chapter 2 we examined the nature of categories; in chapter 10 we explored the cognitive organization that builds itself around categories. In the preceding chapter we stressed the importance of the linguistic tag that designates our categories. At the present time we are completing the story by talking about the ideational content (the image) that is bound in with the category. Thus category, cognitive organization, linguistic label, and stereotype are all aspects of a complex mental process.

. . . . A stereotype is . . . a fixed idea that accompanies the category.

The stereotype acts as both a justificatory device for categorical acceptance or rejection of a group, and as a screening or selective device to maintain simplicity in perception and in thinking. (Allport 1954, 191–192)

Perhaps the best way to reconcile these contradictory causal conceptions is to extrapolate a model in which negative beliefs are given a primary causal role but in which there is some feedback from feelings and policy dispositions that reinforces the negative beliefs. In such a model, negative beliefs foster negative feelings and policy dispositions: once the latter are enacted, the person then reaches for the beliefs to provide the justification for his feelings and policy dispositions. In this way, a loop is set up in the person's mind that can perpetuate itself indefinitely, resulting in beliefs becoming rigidified as they become emotionally invested. This model captures the various strands that have been emphasized by analysts (see, for example, Myrdal 1944, 101–112; Rose 1964, 36–42; Williams, Jr., 1964, 36–38; Tajfel 1969, 1982; Pettigrew 1979).

In any event, both stereotypes and folk explanations of group differences are seen as critically important in the formulation of prejudice. Although some murkiness surrounds the exact role in these cognitions in prejudice, most research has proceeded on the assumption that negative images and explanations are the product of ignorance, and that, once installed, such cognitions rigidly feed intergroup hostility. Both forms of cognition involve the delineation of images of groups and causal attributions about them, although stereotypes involve the latter more implicitly and folk explanations invoke imagery more implicitly. In either case,

the primary concern has been with the degree to which the members of dominant groups indict subordinates by portraying them as deficient in the personal attributes that are associated with success in life. I turn now to a brief discussion of the ways in which that theme has been pursued in research on first, stereotypes, and then, folk explanations of group differences. I then present my conception of the nature of intergroup beliefs in ongoing relations of inequality.

Stereotypes

Stereotypes occupy a venerable position in research on prejudice. Their invidious character has been so well established that the term immediately conjures up trait attributions to groups that are categorical, rigid, and invective. Stereotypes have been inextricably bound to the concept of prejudice—they are "the language of prejudice" (Ehrlich 1973, 21).

In identifying those aspects of stereotypes that are especially damaging to intergroup relations, analysts have focused primarily on the following: the categorical attribution of personality traits to a group, the sharp setting-apart of a subordinate group from the dominant group, the derogation of subordinates, and the reliance on erroneous information (folk knowledge) rather than rational, scientific information about a group. These properties have been regarded as so endemic to stereotyping that researchers have felt no compunction to assess their presence empirically in standard measures of stereotypes. Instead, these properties were simply loaded into the two prevailing measures of stereotypes (stereotype checklists and agree-disagree categorical statements) rather than providing subjects with the opportunity to express more modified intergroup beliefs. I briefly discuss analysts' concern with each of the key properties of stereotypes, in turn, and then outline how the standard measures of stereotypes have served to reinforce, rather than to test, common assumptions about the content and form of damaging intergroup beliefs.

Categorical Attributions

For many scholars, the description of groups in categorical terms is perhaps the most fundamentally damaging feature of stereotypes. Most definitions of stereotypes emphasize categorical attributions as a critical element (for example, Allport 1954; Fishman 1956; Richter 1956; Vinacke 1957; Secord 1959; Koenig and King 1964; Williams, Jr. 1964; Sherif 1966, 38–41; Campbell 1967; Harding et al. 1969; Tajfel 1969, 1982; Cauthen, Robinson, and Krauss 1971; Brigham 1971; Mackie 1973; Hamilton 1981; Hamilton and Trolier 1986). The tendency to simplify and classify the world is often regarded as a basic perceptual operandum as well as a functional coping mechanism for bringing some order to the confusion of reality (Lippmann 1922; Allport 1954; Tajfel 1969; Ehrlich 1973, 38; Hamilton and Trolier 1986). In

the context of intergroup beliefs, however, this process is believed to become emotionally invested, exaggerated, and rigidified.

Use of the term stereotype to denote trait attributions to groups is borrowed from the printing industry where it means "a one-piece printing plate cast in type metal from a mold taken of a printing surface, as a page of set type" (Webster's New World Dictionary 1968). Stereotypes thus preempt raw perception with a rigidly fixed stamp. The categorical perception of groups "provides the mold which gives shape to intergroup attitudes" (Tajfel 1969, 91).

A frequent line of reasoning has been that any perception that unequivocally lumps all the members of a group into one unvaried category is intrinsically damaging because it credits group membership with an overwhelming role in determining individual personality characteristics, and because it provides the perfect grounding for discriminatory policies and practices. Most important, the categorical description of a group represents a closed and insensitive perception that routinely pigeonholes group members and leaves the perceiver unreceptive to contrary evidence, which is either overlooked or dismissed as a mere "exception." Categorical descriptions of groups are thus seen as inherently inaccurate and irrational (Lippmann 1922; Allport 1954; Richter 1956; Williams, Jr. 1964, 36; Ehrlich 1973; Mackie 1973; Tajfel 1969, 1978; Prager 1982). Thus, although there has been more interest in categorical attributions that are explicitly derogatory, any kind of categorical attribution is regarded as damaging, whether the specific traits are positive, negative, or even vague. An extreme example of a belief item that reflects this perspective is found in Wilner et al.'s study (1955, 62): "There may be a few exceptions, but in general, Negroes are pretty much alike." A similar item is found in the ethnocentrism scale of Adorno et al. (1950).

It is frequently assumed that such categorical attributions are more probable when people are describing groups other than their own, especially groups with which the individual has little or no personal familiarity (for example, Allport 1954; Pettigrew 1982; Hamilton and Trolier 1986; Linville, Salovy, and Fischer 1986; Miller and Brewer 1986; Hewstone 1989). This assumption ties in with the recurrent theme that prejudice is a product of ignorance. It is assumed that familiarity would yield the inescapable observation of variation within the target group, and categorical attributions would thus crumble.

Setting Groups Apart and Derogating Subordinates

A second feature of stereotypes that has been regarded as damaging is the attribution of positive traits to one's own group and negative qualities to other groups. Such attributions set groups apart and assert the superiority of one group over another in a single maneuver (Vinacke 1957; Allport 1954; Williams

1964, 40; Sherif 1966, 37–38; Campbell 1967; Harding 1968; Tajfel 1969, 1978; Eldridge 1979, 23–27). In Tajfel and Turner's theory of social identity and social comparison, it is postulated that individuals seek to enhance their self-esteem by comparing themselves favorably to other individuals and their group favorably to pertinent out-groups: this poses a problem for subordinated groups, whose self-esteem is damaged by their position as the object of consensual dominant derogation (Tajfel and Turner 1979).

The extent to which people distinguish between the basic personality attributes of their own and other groups provides the meaning for group labels. In addition, such distinctions carry an implicit evaluation of each group's worth and serve as part of an attributional process that accounts for the varying fates of different groups. A positive self-image is believed to come easily to dominant groups as a product of their privileged position in society; besides, assertions of their superior endowment of personality traits constitute a convenient rationale for their privileged position. Thus, the common assumption has been that dominant groups will assert large and invidious distinctions between their own and subordinate groups as part of a logical attempt to vindicate the status quo unequal relations between them.

The extent to which subordinate groups accede to or deny dominant group assertions of superiority has been regarded as more problematic. This is a critical question, since the former implies a consensus about fundamental perceptions that bolster the status quo, whereas the latter implies a rift between groups that would provide a force for social change. It is generally assumed that the dominant view, because it emanates from the group with more power and prestige, has the advantage, and that it is difficult for subordinate groups to counter with claims of their own superior or even equal endowment of personal attributes. The extent to which a subordinate group does challenge the dominant group's definition of the distinctions between them gives a fundamental indication of their development of positive group consciousness as well as of the dominant group's ideological control over the relations between groups.

Erroneous Beliefs

A common but often unspoken assumption is that stereotypes are erroneous, and that a good part of the damage they wreak comes from their unscientific, irrational genesis. Students of prejudice generally place implicit faith in rational, scientific knowledge as the enlightened replacement for parochial folk knowledge. In Allport's initial delineation of prejudice, he illustrates the problem by drawing a pivotal distinction between two different types of avoidance behaviors: his first case is the anthropologist who does not allow his children to have

contact with the American Indian tribe he is studying, and his second case is the hotel proprietors who refuse accommodation to "Mr. Greenberg" because of his Jewish name. Allport reasons that the anthropologist's behavior is untainted by prejudice because it is based on a realistic assessment of the risk of his children contracting tuberculosis from the Indian children in the village, whereas the hotel proprietors are acting out of prejudice because their behavior is based on an irrational, categorical rejection.

If the innkeepers were basing their rejection on facts (more accurately, on a high probability that a given Jew will have undesirable traits), their action would be as rational and defensible as the anthropologist's. . . . Certainly they had not consulted scientific studies concerning the relative frequency of desirable and undesirable traits in Jews and non-Jews." (Allport 1954, 5–6)

Allport's exposition is consistent with the reigning assumption that prejudice is born out of ignorance. Although there is some ambiguity as to the source of such ignorance (that is, whether it is imposed on the individual by circumstances beyond her individual control, or whether the individual fails to seek out available information about the target group), ignorance itself is regarded as a key to the problem of prejudice. In 1944, Myrdal wrote:

The race prejudice of the typical Northerner . . . is based mainly on ignorance, both simple and opportune. . . . The Northerner seldom gets a chance to see the Negro's good points, and he does not understand the social background of the Negro's bad points. The Southerner's prejudice also has much ignorance in it, but the Southerner's ignorance is more opportune because it is tied to fundamental motives. (Myrdal 1944, 1142)

The theme has persisted: stereotypes contradict knowledge and reason (see, for example, Harding et al. 1969; Selznick and Steinberg 1969; Simpson and Yinger 1972; Pettigrew 1982; Stephan and Stephan 1984).

Despite this emphasis, the verity of stereotypes has been a troublesome issue that has thrust an uncomfortable wedge into scholarly inquiry from time to time. Some scholars have wrestled with the "kernel of truth" hypothesis that stereotypes are exaggerated attributions based on a germ of truth (Allport 1954, 189–196; Mackie 1973). Others have pondered the distinction between incorrect beliefs and negative beliefs (Schuman and Harding 1964). Discussions of Jensen's argument (1969) that ethnic groups vary in IQ also have hinged on whether his argument is true or false. And moral judgments about Jensen's line of research have hinged on whether it is scientifically motivated or racially motivated, as though science automatically cleanses human observation. This issue has been

troublesome because stereotypes have been regarded as the antithesis of science and "rational" knowledge.

Measurement of Stereotypes

Two measures of beliefs about group traits became the standard, the stereotype checklist and agree-disagree categorical statements about a group's attributes. Both measures reflected analysts' assumptions about the hallmark characteristics of stereotypes. Rather than testing the properties of people's intergroup beliefs, analysts instead relied on measures that had the expected properties loaded into them. Reliance on these measures thus served to perpetuate analysts' preconceptions about the specific properties that go into the manufacture of discriminatory intergroup beliefs.

The most frequently replicated research on trait attribution to groups has used the stereotype checklist, usually administered to small samples of select subgroups (generally college undergraduates). This measure presents respondents with a long list of traits and instructs them to assign the traits to specific groups; the positive or negative values attached to the traits are often obtained from another group of respondents (see, for example, Katz and Braly 1947; Meenes 1943; Gilbert 1951; Sherriffs and McKee 1957; Karlins, Coffman, and Walters 1969; Broverman et al. 1972; Smedley and Bayton 1978).

In cross-sectional, public-opinion surveys, the measure that has been used most frequently presents respondents with categorical statements about the trait attributes of a group (such as "poor people are lazy," "blacks are musical"), and respondents can express agreement or disagreement with those statements (see, for example, Adorno et al. 1950; Deutsch and Collins 1951; Wilner, Walkley, and Cook 1955; Marx 1967; Selznick and Steinberg 1969).

These two standard measures share one important feature—they both force respondents to express themselves in categorical, all-or-none terms, thereby precluding the expression of more qualified beliefs.^[1] Because of this constraint, these measures cannot reflect the degree to which people do in fact think about groups in a categorical or qualified way. The concern of analysts with this specific issue has exacerbated the problem by encouraging interpretations that far outstrip the information yielded by these measures. Thus, it is commonly assumed that respondents who agree with the statement "blacks are lazy" think all blacks are lazy. Similarly, the very name of the "stereotype checklist" carries the clear assumption that, in requiring respondents to label groups with

[1] Procedures used by Broverman et al. (1972) did allow respondents to make finer distinctions between men and women, but these distinctions were ignored in the variables that were constructed from the data.

traits, the analyst has captured stereotypical thinking. Indeed, one analyst has made this assumption explicit: Ehrlich argues that "the statement 'Atheists are cynical' may be correctly transliterated 'All Atheists are cynical'" (Ehrlich 1973, 21).

The categorical phrasing of the measures not only restricts the information they can yield, but it may also make them more susceptible to response-bias effects. The blatant character of the task may cue some respondents (especially the well-educated) to social-desirability pressures, and make them unwilling to present their beliefs in an unfavorable light.^[2] Meanwhile, other respondents with simpler cognitive styles (especially the poorly educated) may be trapped by the loaded and unbalanced construction of the measure into saying "yes" or into labeling groups with traits because these response options are "near enough." Thus, these measures may be seriously confounded with response sets caused by varying cognitive styles or susceptibility to social-desirability pressures (Campbell et al. 1960; Ehrlich 1964; Jackman 1973; Jackman and Muha 1984).

Several analysts have registered dissatisfaction with these measures of stereotypes. As early as 1951, LaViolette and Silvert noted that "the attributes of stereotypes have not been examined critically by social psychologists" (1951, 260). In 1954, Harding et al. pointed out that standard stereotype measures do not differentiate between respondents who are indeed ignoring individual differences and those who are merely making a statistical generalization. These complaints have been reiterated over the years (Hyman 1969, 10; Brigham 1971; Jones 1972, 70; Jackman and Senter 1980, 1983; Ashmore and Del Boca 1981). Ehrlich (1964) experimented with the introduction of response options to the standard public-opinion measure that allowed for the expression of strong or mild agreement or disagreement, indecision, and "don't know": he reported that the standard format overstates the degree of acceptance of prejudiced statements. Ehrlich and Rinehart (1965) experimented with a variant of the stereotype checklist that used an open-ended format: they reported that the standard format elicited the assignment of more traits to groups and more consensus among respondents, and that the two formats produced different listings of traits. Brigham (1971) suggested that respondents should be asked to assess what percentage of a group has a particular trait, in order to get a more valid measure of the form of people's attributions to groups. McCauley and Stitt (1978) and McCauley, Stitt, and Segal (1980) have suggested that the standard stereotype check-

[2] Some evidence of this is provided by the replications of the original Katz and Braly design among Princeton undergraduates. Gilbert (1951) and Karlins, Coffman, and Walters (1969) report that many subjects protested the unreasonableness of making such simplistic generalizations.

list oversimplifies human perception and that stereotypes can be understood better as probabilistic predictions that distinguish one group from another. More recently, the General Social Survey (Davis and Smith 1990; Bobo and Kluegel 1991) measured stereotypes of whites, blacks, Hispanics, and Asians with 7-point bipolar scales in which respondents indicated whether they thought each group was characterized more by one trait (for example, hardworking ) or by its opposite (for example, lazy ). But despite the criticisms and suggested revisions over the years, the standard measures of stereotypes have not been seriously discredited. Studies using the traditional measures continue to be cited as classic evidence of the pervasiveness of stereotyping, and, as Ashmore and Del Boca (1981) report, the Katz and Braly checklist has continued to be the reigning measure in the prolific literature on stereotypes.

As a consequence, the complaint made by LaViolette and Silvert in 1951 that "we may call their [stereotypes'] characteristics 'claims' rather than established attributes" (1951, 260) remains as true today. First, as already noted, the standard measures do not yield direct evidence on the extent to which people's images of social groups are indeed categorical or qualified. Second, we lack critical information on the pattern of categorical thinking. Is categorical thinking more evident when people are describing groups other than their own, or do people tend to describe their own group in positive categorical terms and other groups in negative categorical terms? In other words, are categorical descriptions used to differentiate the homogeneity of another group from the individual variance within one's own group ("they are all alike"), or to maximize the difference between the bad qualities of the other group and the good qualities of one's own group ("they are all bad, we are all good")? Third, by the same token, the standard measures cannot give information on the degree to which people draw distinctions between groups. Do people tend to draw large or small distinctions between their own and other groups? Do people tend to minimize within-group variance among outgroups and exaggerate between-group differences, as Tajfel and others have claimed? Fourth, the stereotype checklist is often accompanied by an evaluative scale for the traits on the list, but the standard measure used in public-opinion surveys makes no direct assessment of the derogatory or flattering intent of a person's beliefs.

Folk Wisdom about Group Distinctions

As students of prejudice began to focus increasingly on whites' policy dispositions toward blacks (and especially support for policies of racial integration and affirmative action), their interest in intergroup beliefs broadened to include people's causal attributions about the differences they perceive between groups (Tajfel 1969; Schuman 1975; Pettigrew

1979; Apostle et al. 1983; Sniderman and Hagen 1985; Kluegel and Smith 1986; Kluegel 1990). At the same time, there has been a parallel and sometimes overlapping interest in how people account for the socioeconomic condition of the poor (Feagin 1972, 1975; Williamson 1974a , 1974b ; Nilson 1981; Kluegel and Smith 1981, 1986; Kluegel 1987). In both cases, the essential quest has been to find something in the individual's package of beliefs that could explain why he took a position for or against policy intervention on behalf of a specific group.

This research developed as an application and extension of attribution theory from social psychology. Pettigrew (1979) drew on the idea of "the fundamental attribution error" (Heider 1958; Ross 1977) that observers consistently underestimate the effects of situational factors and overestimate the influence of personal, dispositional factors on actors' behavior. A logical parallel in intergroup relations would be the propensity of dominant group members to attribute the unhappy fate of subordinates to personal, dispositional factors rather than to situational factors beyond the control of subordinates. Pettigrew dubbed this proposed phenomenon the "ultimate attribution error."

Various studies have employed somewhat different terminologies and have highlighted slightly different distinctions among competing popular attributions. However, the central issue has been whether the individual explains group differences in a way that lays the blame for those differences on biases in the social structure (thereby facilitating support for affirmative policy interventions) or on the subordinate group itself (thereby alleviating the dominant group of responsibility for affirmative policy intervention).

Two other themes are reiterated in much of this research. First, several analysts have discussed the pervasive effects of individualism on the way dominant-group members explain group differences. Whether the American commitment to individualism and self-reliance is depicted as a merely technical obstacle to equality (Sniderman and Hagen 1985) or as a deeply entrenched value-system that runs directly counter to the concept of equality (Kluegel and Smith 1986), analysts have argued that individualism further encourages Americans to hold subordinates responsible for their own fate. Second, a key distinction is often drawn between two alternate ways that subordinates may be held responsible for their own fate: are group differences attributed to immutable, biological factors or to distinctive cultural mores and values? The former view is the one traditionally treated as the hallmark of racism and prejudice (Myrdal 1944, 97–99; Rose 1964, 37; Sherif 1966, 39; Tajfel 1969, 94; Schuman 1975; Apostle et al. 1983; Pettigrew 1985, 343): it unambiguously explains the current position of the subordinate group as fixed by the group's inborn characteristics. The invocation of cultural differences

is more ambiguous and complex: subordinates are still burdened with the primary responsibility for their fate, but their culture and values may be amenable to change with appropriate resocialization. For this reason, analysts have usually regarded this view as less venomous or hopeless than the invocation of immutable differences between groups.

Attributions As Embroidery

I view intergroup attributions in a different light. Dominant groups' beliefs about the vital attributes of subordinates have their genesis in political exigencies rather than in ignorance per se. The importance of these beliefs is not as a causal force impelling negative feelings and discriminatory policy dispositions. Instead, intergroup beliefs are the embroidery of intergroup attitudes, and they are important for what they tell us about the sensitive spots in the relationship and the values that infuse and protect the inequality.

Ideas about the characteristics of subordinates emerge out of the political demands imposed by the need to justify and clothe the rude facts of expropriation. Thus, the perceptual distortions and lacunae that are manifested in the intergroup beliefs of dominant groups should not be expected to follow any one prescribed pattern or to be locked into any particular configuration. They are more likely to have a chameleon quality, with various ideas and modes of attribution flourishing or fading, dependent upon the demands of the relationship at any given time and the broad moral themes that have contemporary currency.

In this process, logical inconsistencies abound. As with other aspects of people's attitudes, the essentially political nature of intergroup beliefs means that they do not need to fulfill any logical consistency requirements to be satisfactory to those who hold them.^[3] Indeed, rigidity and consistency turn out to be poor political weapons. As Myrdal (1944) and Rose (1964) long since observed, the overwhelming requirement of intergroup beliefs is that they address the various stress points that exist in the intergroup relationship. Inconsistencies may lie quietly unresolved and untroublesome to their adherents as balm is offered to the complicated sources of stress in an unequal relationship.

Frederickson's essay (1988) on white slave-owners' beliefs about their slaves in the Old South provides a pointed illustration of the way inconsistent beliefs can cohabit comfortably within a single culture, and, indeed, within a single mind. On the one hand, slave-owners depicted slaves as

[3] Interestingly, Adorno et al. (1950, 75–76) note an inconsistency between the anti-Semitic images of Jews as both intrusive and seclusive. But they interpret this inconsistency as evidence for the irrational basis for stereotypes: that stereotypes fulfil the warped personality needs of the individual.

Sambo (happy-go-lucky, lazy, content, friendly, warmhearted, childlike), and on the other hand they drew a more menacing image of the black savage (threatening, violence-prone, emotionally volatile, sexually aggressive). The inconsistency in these images apparently mattered little to those who held them, since both images addressed the sensitive spots in their relations with slaves. The former image assuaged slave-owners' half-dormant fears about black resistance to slavery by declaring blacks to be contented, and it also justified slavery by declaring blacks to be in need of white tutelage to take care of them. This all contributed to the general goal of portraying the master-slave relationship in benevolent terms. Nevertheless, residual fears about the dangers posed by having an exploited labor force in residence on the premises of the master's own home surfaced in simultaneous accounts of the black savage. The latter image also worked to justify the institution of slavery by claiming that it protected whites (as well as blacks themselves) from blacks' underlying savagery. Slavery was thus deftly recast from an institution of exploitation into a benign institution that elevated blacks from savagery and gave them a chance to travel the long road to civilization.

If consistency is not the glue that makes certain beliefs stick together, what is? To elucidate the inherently political nature of intergroup belief systems, I now reconsider each of the qualities that have traditionally been identified as the hallmarks of discriminatory intergroup beliefs: that they attribute traits to subordinate groups in ways that are erroneous, derogatory, categorical, and sharply differentiated from the dominant group, and that such beliefs draw an immutable, genetic boundary between the groups. I argue that dominant groups are rarely at liberty to manifest all of these qualities in their intergroup beliefs. As they fashion their stories about the attributes of their own and other groups, they are buffeted by the daily pressures that they encounter in their attempts to govern and explain the relationship from which they benefit.

Science, Folk Knowledge, and Human Error

The traditional distinction between beliefs that are rational or "scientific" and those that are irrational or false is irrelevant. It requires only cursory reflection to recall the venerable history of science in the propagation of harmful beliefs about various social groups. Consider, for example, the beliefs about women that were vigorously promoted by American physicians and medical researchers in the last quartile of the nineteenth century—that women's reproductive function placed such stringent demands on their energy that attempts on their part to engage in sustained intellectual work would impair their health and weaken or destroy their reproductive capacity (Rosenberg 1982). These ideas, emanating as they did from legitimate scientific sources, were taken very seriously in educational cir-

cles. Consider also the beliefs that were espoused at the turn of the century by prominent scientists about head size, brain size, and genetic intelligence among various ethnic groups. Such respected figures in the British scientific community as Francis Galton (Darwin's half-cousin) and Karl Pearson did not hesitate to issue scientific judgments about the relative physical and intellectual capabilities of various ethnicities (Hirsch 1973). More recently, in World War II, Nazi racist ideology in Germany employed scientific delineations of Aryan facial and body features and their relationship to personality attributes; and many scientists, academics, and physicians participated voluntarily and actively in the persecution of Jews, Gypsies, the mentally ill, and gay men, both in delineating "theoretical" materials and enacting policies (see Burleigh and Wippermann 1991, especially pages 51–56, for a chilling account of scientific and medical complicity in Nazi policies of racial purity). Meanwhile, on the other side of the war, psychiatrists in the United States' armed forces actively promoted the definition of homosexuality as a mental illness, and they devoted vigorous effort to the scientific detection of homosexuality, including such procedures as examining recruits' physiques, analyzing their urine, administering psychological tests and "tongue depressor tests," and developing a profile of the homosexual "personality type" [stereotype?] (Berube 1990, 128–174). And in the present era, respected scientific figures such as Arthur Jensen and William Shockley have asserted the scientific basis of their claim that IQ test scores provide a valid measure of intellectual capacity and that various ethnic groups differ in their average, genetically determined, intellectual capacity (for example, see Jensen 1969; Shockley 1971, 1972a, 1972b).

Would Allport's infamous hotel proprietors have been cleansed of impure thoughts if they had consulted any of those scientific studies to get their "facts"? Reflection on the history of scientific thought hardly makes one sanguine about the ability of science to withstand either the foibles of human perception or the political pressures to which all other participants in an unequal intergroup relationship are subject. Folk knowledge may be articulated with less polish but it seems no more vulnerable to the political exigencies of the moment.

Malevolent intergroup beliefs are popularly validated by drawing on sources that carry contemporary credibility as independent fonts of knowledge and morality. The possessors of such beliefs reach for validation from whichever source offers the surest authority. When religion held sway, dominant groups drew on religious precepts and beliefs to validate their claims about ethnic group attributes. For example, the American proslavery apologists of the mid-1800s relied heavily on Christian theology to validate their claims about the attributes and needs of enslaved Africans (see Fitzhugh 1854; Elliott 1860), and Jordan's de-

tailed account of the historical genesis of English attitudes toward Africans in the 1500s and 1600s indicates the central role of religious beliefs in the formulation and justification of those beliefs (Jordan 1968). With the demise of religious authority and the advent of science, the preferred source of validation for intergroup beliefs likewise shifted. Both religion and science have proved to be rich and conveniently malleable sources of validation. Both offer a set of precepts and a method of viewing the world that are authoritative at the same time as they are sufficiently broad and encompassing as to lend themselves conveniently to a variety of interpretations.

More generally, it is misleading to try to evaluate intergroup beliefs by their apparent verity. It is not so much whether beliefs are true or false or contain a "kernel of truth"—it is more a matter of which factors are selected for emphasis and what values are attached to those factors. For example, in contemporary Western achievement-oriented society, it is hardly surprising that intelligence is a highly salient personal attribute and that groups who have experienced more socioeconomic success should find themselves in possession of more of the golden attribute than less "successful" groups. For similar reasons, dominant classes and races have historically had a penchant for describing the lower orders as lazy. Is this attribution made because subordinates have a bad habit of contributing their labor to the dominant group with less alacrity than the dominant group deems appropriate, or because it is convenient to regard one's comparative well-being as a hard-earned success rather than the spoils of a biased reward structure?

Another interesting intergroup comparison that has been reiterated through history is the depiction of subordinates (peasants, lower classes, blacks in slavery, and, in contemporary America, women) as more emotional or volatile than the members of dominant groups. Presumably, those in a position of command find it necessary to curb their emotional expressiveness, since such expressiveness would be more self-revealing and hence make them more vulnerable (Henley and Freeman 1979). For these reasons, it might be advantageous to dominant groups to have subordinates reveal their emotions (thus making them more vulnerable) while controlling their own emotional expressiveness. This is somewhat akin to the power differential between one person being clothed and the other naked. Are there "real," systematic behavioral differences between dominant and subordinate groups in emotionality? Or do groups find different avenues for emotional expression—for example, open expression of anger may be more acceptable for dominant-group members whereas open expression of warmth or laughter or emotional distress may be more acceptable for subordinates? Or do both groups exhibit the same range of emotional expression, but dominant group members

are unduly sensitive to "inappropriate" emotional displays from subordinates (such as anger) that are threatening to authority relations between the groups? Or, finally, is the group difference portrayed in order to establish an ideal to guide each group's behavior?

Ultimately, it matters little whether the imputed differences between groups are "real" or false. It is not self-evident how one would "scientifically" measure alleged differences between groups in any case. One would need to incorporate and observe all manifestations of a specific trait (for example, the many forms of "emotionality"). Further, observations of that trait would need to be weighted within the full range of personality attributes on which individuals and groups might potentially be compared. Such a task is not only daunting—it is beside the point. What is more revealing is which traits are popularly selected as salient comparison-points between groups and the way in which those traits are used to delimit the place of each group in society. Attention to these issues can direct us to both the stresses and the moral basis of an unequal relationship.

The Derogation Game:
Flattery and Social Control

The traditional emphasis on the derogation of subordinates is also misleading. Derogation of subordinates has indeed been a frequent recourse of dominant belief systems, but this recourse is not always politically convenient. And even when derogatory distinctions are drawn, they are unlikely to comprise the complete picture that the dominant group draws of subordinates. More likely, invidious distinctions are softened with other distinctions that credit subordinates with the advantage in some positive attributes.

The derogation game is further complicated by two other factors. To begin, certain traits can be portrayed in alternative ways to turn them into assets or detriments, or as impediments or advantages in the fulfillment of particular roles in life. Gloria Steinem's clever article, "If Men Could Menstruate" (1978), illustrates this point well. As Steinem adroitly observes:

Male human beings have built whole cultures around the idea that penisenvy is "natural" to women—though having such an unprotected organ might be said to make men vulnerable, and the power to give birth makes womb-envy just as logical.

In short, the characteristics of the powerful, whatever they may be, are thought to be better than the characteristics of the powerless—and logic has nothing to do with it.

In this spirit, do we depict a group as lazy or as laid-back, as hardworking or uptight, as consistent and rational or rigid and cold, as emotional and temperamental or warm and expressive, as absentminded or flakey?

A second and more elusive point is that traits may sometimes be regarded as an asset when they are possessed by a person from one group but as a detriment when exhibited by someone from the other group. Such conditional value-shifts are to be particularly expected in intergroup relationships that are governed by role segregation. Thus, for example, individual women are praised for being warm and expressive (designated role-appropriate female traits) but criticized for being detached and authoritative (designated male traits). This leaves women who seek positions of authority with an awkward choice: either fulfill the classic female attributes (and thus retain one's feminine identity but fail to command authority), or mimic the classic male traits (and command more authority but incur disapproval as a "cold bitch"). And the unhappy possibility remains that women may not be able to command authority even when they forfeit their "femininity" and mimic male traits: women behaving in this way run the risk of being regarded as such deviant personalities that their claim to authority is undermined.

These considerations suggest that derogation often takes a more roundabout route that camouflages the invidious character of intergroup distinctions and nestles them softly into the pattern of everyday life. When groups are at war, blatant derogation of one's opponents serves a useful political purpose, for the overriding task then is to incite one's own ranks to unbridled animosity. But the maintenance of peaceful expropriation requires attention to the sensibilities of both groups, and the purpose of intergroup attributions then is to weave a case for the modus vivendi that will be persuasive to all the participants. Blatant derogation is hardly up to the task.

The overwhelming pressure exerted on dominant groups is to mold the attributed characteristics of each group to be consistent with the organization of everyday life. In this endeavor, it is preferable to weave a story that melts naturally into the setting it seeks to preserve rather than to draw starkly invidious contrasts that might jar the senses. Of course, the superiority of the dominant group must be worked into the story, since the dominant group has an obvious interest in promoting its higher status—but there is no need to overdo it. Indeed, there is every reason to credit subordinates with a superior endowment of certain attributes that are consistent with the tasks or the position assigned to subordinates and inconsistent with the tasks or position claimed by the dominant group for itself.

As with other manifestations of hostility, derogation of subordinates is fraught with risk. The members of groups that benefit from relationships of inequality are aversive to strategies that carry a high risk of alienating the very people whose continued cooperation is essential to the perpetuation of that relationship. In relationships based on role seg-

regation, flagrant derogation is likely to be particularly counterproductive. In such relations, the high level of interpersonal contact across group lines increases the political pressure to sweeten the exchange with a little flattery. As any job supervisor ought to know, the alienating effects of insults are revealed especially swiftly in interpersonal interactions, whereas the impact of the derogation is less immediately and less vividly felt when the object of the derogation is spatially separated from the subject. In any case, the primary need in relations organized around role segregation is to distinguish clearly the attributes of the respective groups to match the segregated tasks to which they have been assigned.

How sweet it is to praise subordinates warmly for their enviable possession of those personality traits to which one's own group makes no claim. Thus, women may be praised copiously for their abilities and achievements as mothers or as family cook or housekeeper. Status distinctions between the two groups can be maintained unobtrusively but securely by assigning greater prestige, power, and material rewards to the tasks for which the dominant group appears more aptly fitted. And besides, what more gratifying way is there to achieve a feeling of comfortable superiority than by making it your business to define the needs and attributes of your subordinates? And then to provide for their defined needs furnishes an incomparable opportunity for amiable condescension. The result: a cozy feeling of magnanimous superiority without a single derogatory stone being cast. How much more subtle and effectual is this strategy than would be the clumsy excesses of direct derogation.

Unequal relations marked by spatial segregation do not pose the compelling interpersonal constraints against direct derogation that exist in relations that have high one-to-one contact across group lines. In addition, spatial segregation affords subordinates more opportunity to distance themselves from the warm clutches of the dominant group and hence to begin to mount more opposition. In such circumstances, not only are there fewer constraints against derogation, but as subordinates introduce some conflict into the relationship the good old paternalistic way of sweet praise for the successful execution of narrowly defined tasks no longer quite works. Besides, what more natural response is there to groups who get uppity than to remind them of their deficiencies? When subordinates challenge arrangements, it becomes necessary to put them in their place—however regretfully. But even here, the dominant group is unlikely to rush headlong into derogation—ways will be found to soften the blow even as it is administered. After all, the dominant group has no interest in heightening hostilities, but rather in putting a stop to them.

The Politics of Setting Groups Apart

The common assumption has been that harmful intergroup beliefs set groups apart sharply and attribute

traits categorically to the subordinate group. As with derogation, this assumption is misleading, and for similar reasons. Although it is important for dominant-group members to maintain group boundaries, there are a variety of ways of accomplishing this end without invoking exaggerated descriptions of the groups. And it is not always opportune to categorize the members of a group flatly or to draw abrupt distinctions between groups in their attributed characteristics. Indeed, as McCauley, Stitt, and Segal (1980) have pointed out, probabilistic attributions should be less subject to everyday disconfirmation than are flatly categorical attributions. This should make probabilistic attributions more resilient than categorical ones, and thus more serviceable politically.

Malevolent intergroup thinking does not express itself irrepressibly in sharp distinctions and categorical attributions. Such attributions will surface only when they meet the demands of the situation—otherwise, popular thinking will travel along other more convenient routes. In order to be politically serviceable, attributions to groups need only to conform to the demands of the relationship and be formulated in a language that is culturally compelling at the moment. This permits intergroup attributions to assume a variety of specific forms.

In relations based on role segregation, there is a pressing need to differentiate the traits of the participating groups sharply: a benign explanation must be established for the inflexible way tasks are allocated to each group. In addition, the high degree of interpersonal contact across group lines that often accompanies role-segregated relationships might pose some risk of muddying group boundaries if the participants were not reminded of the boundary that exists between them on the basis of their group membership. As I outlined above, there is also pressure in such relationships to avoid outright derogatory attributions, and so the combined effect is to encourage evaluatively neutral but sharp distinctions to be invoked between the groups—"we are good at this, you are good at that." Finally, an essential element of the paternalistic ideology that emerges under these conditions is the presumptive definition of the subordinate group's needs—and in order to define the group's needs, one must first define the group's attributes. All these constraints push the dominant group toward categorical descriptions of subordinates and sharp differentiation of subordinates from the dominant group in their personality traits.

In relations based on spatial segregation, a different set of constraints is operative. In such relations, the inequality does not rest on the groups performing categorically different tasks in life that must then be justified. Nor do such relations present much threat of the members of the unequal groups crossing over group lines: the spatial separation of the groups effectively rids the relationship of opportunities for confusion

about which individuals are in which group. Instead, strong systems of spatial segregation pose a different problem for the dominant group—how to maintain links between the groups so that the relationship does not rupture completely. Without the existence of a complementary, group-based task structure that instructs the participants in how their group is related to the other, the prevailing risk is that subordinates might drift away from dominant control entirely. There are not daily opportunities for the dominant group to invade the consciousness and emotional vulnerabilities of subordinates, and the latter are thus able to begin formulating an independent interpretation of their experiences. To compound the problem, the spatial separation of the groups reveals the accompanying disparities in status, power, and economic well-being more rudely than when they were nestled in the intimate daily lives of the participants.

Under these constraints, the dominant group would be ill advised to heighten group distinctions by describing subordinates as categorically different. As subordinates decry the group basis of social life, ways must be found to deny and obscure it. And since spatial segregation effectively ensures that the groups will remain separate, there is little risk that the ideological softening of the group basis of social life will result in a wholesale breakdown of group distinctions—especially if the strategy is pursued with care. Under these constraints, it behooves the members of the dominant group to assert the principle of individualism.

As I argued in chapter 2 and chapter 6, individualism provides an effective, principled way of undermining the moral credibility of claims and grievances made on behalf of groups while at the same time reaffirming a competitive norm that works quietly to preserve inequalities. The individualistic goal of equality of opportunity preempts egalitarianism, making outright redistribution of social assets a dead issue and morally endorsing the idea of a competition—in which there will be winners and losers (Schaar 1967). What implications does this have for the shape of intergroup beliefs?

Individualism spawns discomfort both with blanket categorizations of subordinate group members and with attributions that draw impermeable distinctions between groups. Categorical thinking becomes a liability. Instead, the pressures are toward emphasizing individual variation: distinctions between groups must still be made, but they must be couched in the language of individualism. Dominant group members drift toward muted differentiation between groups—individual variation within groups is readily acknowledged at the same time as average or modal distinctions between groups are reaffirmed. This more probabilistic mode of differentiation is also consistent with scientific methods and language, and this adds to the credibility of such attributions. Arthur

Jensen's discussion of differences in IQ between ethnic groups (1969) epitomizes this approach. The essential process of group attribution is retained, allowing people to make the critical distinctions they are impelled to make, while seeming like the product of careful, objective observation—what could seem more reasonable or undeniable? Thus, the political dynamics of spatial segregation push dominant-group members more toward the derogation of subordinates, but the same dynamics also force them to retreat from categorical attributions and to edge toward the more guarded delineation of group differences within a framework of individual variability.

Fingering the Causes of Group Differences:
The Political Uses of Equivocation

As with other aspects of intergroup beliefs, folk wisdom about the causes of group differences is likely to veer away from the blatant, the definite, or the unremitting disparagement of subordinates. Several scholars have noted the ambiguity of some popular explanations of group differences and the tendency of many people to espouse seemingly contradictory explanations of group differences (Apostle et al. 1983; Sniderman and Hagen 1985; Kluegel and Smith 1986). Although the classic, dyed-in-the-wool racist has been expected to show himself by a belief in immutable, genetic differences between groups, unequal relations may rarely present circumstances where such a flagrant position is politically opportune. The invocation of cultural differences between groups, although usually treated by scholars as less invidious than the allegation of immutable differences, may actually fulfil the political requirements of group domination much more satisfactorily.

Cultural explanations still lay the blame for group differences squarely in the subordinate camp—such explanations make it clear that subordinates lack the requisite values and attributes for success because their culture fails to instill them. And cultural explanations still assert dominant superiority over subordinates, as well as extolling dominant culture and values as the ideal. Thus, cultural attributions achieve the most important goals of a discriminatory belief, but in a way that has less edge to it than the assertion of genetic differences. Indeed, the reaffirmation of dominant culture and values that is built into such attributions is more likely to entrap subordinates' thinking. The assertion of cultural deficiencies in the subordinate group takes it as a given that the dominant culture is indeed the superior and desirable one. The terms of debate are then set in terms of whether and to what extent subordinate-group members manage to manifest the cultural attributes of the dominant group—and kept safely at bay is the prospect of any direct challenge to the reigning cultural values of the dominant group.

At the same time, cultural attributions sound much more reasoned,

reasonable, and open to falsification than does the rigid assertion of genetic differences between groups. Cultural explanations imply that if change occurs in the subordinate group's culture or if some individuals from the subordinate group manifest the appropriate dominant values, it will be noticed and there will then be no further obstacles to success and integration into the dominant stream. And yet the lack of specificity of cultural attributions and their reference to such intangibles as personal motivation and willingness to defer gratification actually renders them immune to any threat of falsification. In addition, such attributions astutely concede the possibility of change in the future while simultaneously making such change seem arduous, remote, and improbable.

Cultural explanations of group differences also offer more versatility of application to different kinds of intergroup relationships, precisely because they are more vague and less flatly disparaging than is the assertion of immutable, biological distinctions. Within a cozy paternalistic framework, the impact of biological attributions is softened: dominant-group members can feel confident about defining the needs of their immutably inferior subordinates and they may magnanimously assume control of a group that has been defined as weaker by nature than themselves. But in a more conflictive political context, the assertion of immutable, genetic differences between groups is too rigid and harsh to carry much political value—such assertions would only serve to heighten hostilities when the primary game-plan for the dominant group is to curb conflict.

Cultural attributions, on the other hand, can be tailored for either paternalistic or conflictive relations. Allusions to cultural deficiencies among subordinates fit comfortably into a paternalistic framework. Such explanations depict subordinates as requiring the benefit of a proper education in the dominant values in order to "civilize" them or resocialize them. With supercilious generosity, the dominant group may then take that burden upon itself. When stripped of a paternalistic support system and thrown into a more conflictive setting, cultural attributions retain their political value, as the dominant group reaches for stories that seem universalistic, open-minded, and reasonable. How can you expect to do as well as us when you don't speak proper English, you don't defer gratification for long-term benefits, you are unlawful, and you fail to maintain a decent family life? Be more like us!

Interestingly, the proslavery tracts of the mid-1800s in the South were as likely to allude to potentially correctable cultural deficiencies in their African-American slaves as to refer to biologically immutable attributes (see, for example, Fitzhugh 1854; Elliott 1860). The attribution of cultural deficiencies to African Americans allowed the defenders of slavery to define a singularly satisfying role for that institution: slavery could be

represented as a vehicle for teaching people who were less culturally developed in the ways of Western civilization while protecting them from full participation in a society for which they were deemed as yet culturally ill-equipped. In a similar vein, white South Africans have often been heard in the past few decades arguing that the blacks are "not ready" for democracy and that further time (left gloriously unspecified just how long) would be needed to socialize them into the norms and values of Western democracy.

In short, the assertion of cultural differences between groups offers the dominant group all the advantages of genetic assertions—and more. As with other aspects of their intergroup beliefs, the members of dominant groups may display more political acumen than analysts have credited to them.

The Agility of Intergroup Beliefs

The most enduring feature of malevolent intergroup beliefs is not their rigidity, negativism, persistence, or consistency—it is their elasticity. Dominant groups gravitate toward this or that explanation of the vital attributes of their own and other groups according to the political exigencies of the moment. Their beliefs must serve a political purpose, and they are continually molded, reformulated, and reshaped to meet that purpose. The particular ideas that have prominence in a given era and the way they are expressed and validated have no intrinsic meaning or significance. They are a patchwork of thoughts, accumulated from an aggregate of people, past and present, who are continually responding to the pressures of the unequal relationship in which they participate.

Various personality traits and values gain or lose salience. Different modes of attribution flourish and decline. Different factors gain or lose credibility in popular explanations of group differences. Alternative sources of validation are called into service. As the members of dominant groups reach for persuasive explanations of their privilege, they glide easily into patterns of thinking that fit the needs of the moment. Dominant groups focus automatically on those personal attributes that seem to bear meaningfully on their relationship with subordinates. They allude to values that strike a cord in themselves and their contemporaries. They express themselves in ways that are persuasive and compelling in the light of contemporaneous pressures. And they draw on sources of validation for their stated beliefs that carry authority in their era.

In this ongoing process, the behavior and reactions of subordinates feed into the ideas mill. Through trial and error, the members of the dominant group learn what kind of stories not only are persuasive within their own ranks but also produce the desired response from subordinates. Thus, the part played by subordinates is not as mere passive vic-

tims. Subordinates react in one way or another to the stories they are told, depending upon the opportunities posed by the structure of their daily interaction with the dominant group and the degree to which the dominant group has fashioned a story that fits the contingencies. Of course, subordinates are no more at liberty than are members of the dominant group to be freely expressive in their reactions, but the dominant group has every reason to be finely attuned to their mood. Accordingly, the nature of the subordinate response shapes the unfolding storytelling efforts of dominant-group members.

And because the line of defense is constantly shifting, the amalgamated ideas at any one time contain internal inconsistencies and a deliberate lack of clarity. Precision and consistency are only important if one wishes one's ideas to be potentially falsifiable, but popular intergroup beliefs are of more use to their proponents if they are not falsifiable. Thus, the language of intergroup beliefs will tend more toward the slippery and imprecise than toward the extreme or definite. And the package of beliefs to which people adhere will conveniently contain inconsistencies. Loose-jointed belief systems are less vulnerable to attack, since one component can be given up or reformulated, if politically convenient, without jeopardizing other components that still prove serviceable.

The resultant mix of intergroup beliefs that exists at a particular historical moment has accumulated artlessly from the reactions of the individual participants. And yet, because individuals' perceptions are molded by the pressures that bear on them, the emergent composite proficiently addresses the stress points in the relationship.

Data:
Trait Attribution to Groups

We may now assess the content and form of the intergroup beliefs that embroider race, gender, and class relations in the United States. The analysis begins by examining the way salient personality traits are ascribed to the unequal groups, for race, gender, and class. In the next section, I take up the issue of how distinctions between these three sets of unequal groups are popularly explained.

To what extent do dominant groups invoke categorical imagery of subordinates and draw large and invidious distinctions between their own group and subordinates? And do subordinates accede to the images, distinctions, and values that are set by the dominant group? If subordinates succeed in breaking away from the dominant story, what kinds of moves do they make and what aspects of the dominant story are most likely to be rejected? As we compare data from the three different intergroup relationships, what do they suggest about how patterns of domi-

nant assertion and subordinate reaction are affected by the structure of contact between groups?

For race, gender, and class, respectively, I assess how key personality traits are attributed to groups. I examine the extent to which cognitive distinctions are drawn between groups in salient personality attributes and the prevalence of categorical imagery in trait attributions to groups. I also evaluate the extent to which cognitive attributions involve the derogation of subordinate groups.

This analysis addresses questions at two levels. First, what is the general character of intergroup trait attributions? Are stereotypes, as traditionally conceived, an important feature of intergroup imagery, or are intergroup perceptions more qualified and malleable? Second, with these broad issues resolved, I turn to the shifting character of intergroup trait attributions as we move from one intergroup relationship to another. How are popular beliefs about the salient attributes of unequal groups molded by the specific exigencies of an intergroup relationship?

Measures

Trait Attributions

My measure of trait attributions was designed to maximize information on the degree to which people describe their own and other social groups in categorical or qualified terms, while minimizing social desirability and other response-set pressures. During the interview, respondents were handed a booklet with the following introduction by the interviewer:

There are many groups in America and people have different opinions about what these groups are like. This booklet contains a list of some of the ways groups are described. In each case, you are asked how many people in various groups you would say are like this. The more people you think are like this, the higher the number you should choose from the scale at the top of the page—the fewer people you think are like this, the lower the number you should choose.

This doesn't need to take much time. We're just interested in your first reaction to these questions.

There were thirty-two questions of the following form in the booklet: "How many (group ) would you say are (trait )?" For example, "How many women would you say are emotional?" Questions were grouped by trait, with the question about the subordinate group preceding the parallel question about the respective dominant group (a given question about blacks preceded the one about whites, and so on). Each trait was introduced by a subheading and a definition of the trait in parentheses. Respondents selected their responses from a 9-point scale extending from left to right and printed at the top of each page of the booklet, and they

recorded responses in the box next to each question.^[4] The lefthand pole of the scale was labeled 0-NONE, and the righthand pole was labeled 8-ALL. The intervals between were marked 1 to 7; no labels were attached to them. Attention was drawn to the midpoint of the scale by short vertical lines above and below the 4. Respondents were also provided explicitly with a "don't know" response-option: to the right of the scale was written "X = DON'T KNOW."

As with the measurement of intergroup feelings, questions about trait attributions to groups were presented in a self-administered booklet in order to maximize the respondent's sense of privacy and reduce social pressures.^[5] The construction of the items also presents no obvious source of bias, since a variety of response options (including "don't know") are presented in an equally balanced way.^[6]

The specific traits that were used in the questions were selected for their pertinence to each relationship of inequality and their prominence in research on race, gender, and class stereotypes. At the same time, traits could only be used if they could be comprehended readily and interpreted commonly by a cross-section of the population which includes people with widely varying levels of education and life experiences. In this way, we selected the following traits. First, the trait INTELLIGENT (or bright ) was included for all three intergroup relationships: in a society obsessed with IQ tests and individual achievement, the trait of intelligence seems centrally significant to popular discussion of any form of inequality. Other traits varied across the three intergroup rela-

[4] Brigham (1971) has suggested that respondents be asked to indicate what percentage of a group they think has some trait. However, we found in a pretest that this format elicited a large number of response-set answers where respondents repeated the same percentage throughout the booklet. Apparently, the implication of precision in the use of percentages scared some respondents away from the task.

[5] To increase the respondents' sense of privacy and to reduce social pressures further, interviewers were instructed to keep busy with other editing tasks while the respondent was completing the booklet and to put the booklet into an envelope as soon as the respondent was finished with it.

[6] The administration of the trait-attribution booklet went smoothly. On the 1,914 people in the sample, only 42 (2.2 percent) did not answer the booklet at all. The median time spent on the booklet was five minutes. No response-set problems were apparent. In a check for blatant response-set, we counted the number of times each numerical response was repeated throughout the booklet. The middle category on the scale (the 4's) was used most often, but even here only 5 respondents used this response exclusively and only 68 respondents used this response twenty-one times or more over the thirty-two questions. The "don't know" response was chosen for all thirty-two items in the booklet by only 20 respondents; 69 respondents gave this response twenty-one or more times. No other points on the scale were used more than twenty-three times by any respondent. We may assume these counts constitute an upper-bound estimate of this kind of response-set since we cannot separate out content-free repetition from "true" responses.

tionships. For race, the two additional traits of LAZY (or not hardworking ), and DEPENDABLE (or reliable ) were included. Both of these traits have figured centrally in whites' beliefs about blacks from slavery to the present. For gender, the two additional traits were EMOTIONAL (or quick to show feelings ), and TALKATIVE (or talk a lot ). Both of these classically "feminine" traits relate centrally to the expressive role that has been assigned to women. For class, three additional traits were included, all of them relating in some way to the business of individual socioeconomic achievement: LAZY, THRIFTY (or careful with money ), and SELFISH (or concerned only with one's own problems ). The trait lazy has featured prominently in popular discussions of the poor much as blacks have been labeled with that trait, and a lack of thrift is another deficiency that has been commonly attributed to the poor. The trait selfish was added to the set on class since that attribute has variously been attributed to the working class (selfishly pursuing their own interests rather than the common interest of society) or the upper class (selfishly ignoring the economic plight of the remainder of society).^[7]

These measures permit two different kinds of assessments of the way in which people attribute key personality traits to social groups. First, each respondent's attribution of a specific trait to a subordinate group was subtracted from his or her attribution of that trait to the pertinent dominant group. This allows us to assess the degree to which the respondent draws a difference between the two groups being compared, and the resulting difference-scores can range from -8 (all of the subordinate group and none of the dominant group have the trait) through 0 (no difference drawn between the groups) to +8 (all of the dominant group and none of the subordinate group have the trait). Second, people's responses to each trait item are examined singly to ascertain the nature of their imagery about each group: of particular interest is the extent to which people describe a group in categorical terms or see internal variation within the group.

Evaluative Overlay of Trait Attributions

In addition to the trait attributions themselves, I gathered data on the evaluative connotations of the traits involved, with an independent measure of the respondent's evaluation of each of the traits used in the booklet. This information is critical to the assessment and interpretation of trait attributions. The items on the evaluative overlay of the traits were administered thirty-four pages after the beliefs booklet in the interview schedule.

[7] Other traits were pretested, but they proved unsatisfactory, either because of poor respondent comprehension (e.g., logical for gender) or because they seemed to produce discomfort/evasiveness (e.g., musical for race.)

I am going to read a list of things people use to describe each other. For each one, I am going to ask how much you generally like or dislike people like this. Here is a scale that ranges from 1 for "dislike very much" to 9 for "like very much." For each characteristic I read, please tell me the number that comes closest to how you feel.

How much do you like or dislike lazy people?

The question was repeated for the remaining six traits used in the trait booklet. The poles of the scale that was shown to respondents were labeled as described above, and the midpoint of the scale (5) was labeled "neither dislike nor like."

The mean evaluations of the traits are presented in table 8.1. The traits vary considerably in how they are valued, with dependability being rated considerably more positively than any other trait, and selfishness and laziness being valued considerably more negatively than any other trait. Although the data are not presented here, these traits are valued in much the same way by different groups and by people who attribute the traits differently to the pertinent groups. These data highlight the values that are at stake, on both sides of the relationship, when people are comparing different types of dominant and subordinate groups. Race comparisons carry strong value-connotations, since they involve the two most highly valued traits (dependable and intelligent) and one very negatively valued trait (lazy). Similarly, class comparisons involve two positively valued traits (intelligent and thrifty) and the two most negatively valued traits (lazy and selfish). Gender comparisons involve one positively valued trait (intelligent) and two fairly neutral traits (talkative and emotional).

Results

Tables 8.2, 8.4, and 8.6 report the cognitive distinctions that are drawn between groups, for race, gender, and class, respectively. Each table gives

the percentage distributions, by group, of the trait difference-scores in collapsed form, as well as the mean trait difference-scores (based on the full scales). For race and gender, the scales are collapsed into 7 points, whereas for class they are collapsed into 3 points (to spare the reader's eyesight). Tables 8.3, 8.5, and 8.7 report the manner in which the traits are assigned to each group for race, gender, and class, respectively. These tables give the percentage distribution and mean score for each trait as it is attributed to each group, by group. For class, the data are again presented in reduced form because of space limitations: for class, only the tails of the distributions are presented, along with the mean scores for the full scale.

A quick perusal of the data reveals two general points. First, it is rare for large trait distinctions to be drawn between groups, in any of the three intergroup relationships. Instead, people who differentiate between groups in salient personality traits tend to portray those distinctions as small or moderate. Second, in the same vein, categorical images of groups are also far from commonplace. Although categorical descriptions of groups are more in evidence than the attribution of large differences between groups, there is considerably more popular recognition of internal variation within groups than the literature on stereotypes might lead us to expect. Thus, the beliefs that accompany these three relationships of inequality do not seem to be scored very deeply by two of the most central conceptual elements of stereotypical thinking—the attribution of extreme differences between groups and the categorical description of groups.

What, then, are the characteristics of the intergroup trait attributions that accompany race, gender, and class relations? I turn to each of those now, in turn, before assessing the general patterns that are suggested by these data.

Race

Two main factors distinguish the trait attributions that accompany racial inequality. First, it is in the race context that the classic elements of stereotypical attributions are least in evidence. Trait distinctions are less likely to be made between racial groups than between gender groups or classes, imputed trait distinctions are the smallest, and categorical imagery is the least likely to be invoked to describe either racial group. Second, the race context is also marked by more disagreement between groups than one finds for either gender or class in the way traits are attributed to the pertinent groups. Whites tend to draw small to substantial, but invidious, distinctions between the races, whereas blacks tend either to deny that there are differences between them or to make counterclaims that blacks have a superior trait endowment to whites. The combination of clear disagreement between the races and restrained

cognitive expressions on the part of whites suggests that as contending groups diverge in their cognitive attributions, dominant-group members become self-conscious and cautious about the way the subordinate group is portrayed.

Whites are more likely than blacks to think there are trait differences between the races, and whites who see race differences generally say there are fewer intelligent or dependable blacks and more lazy blacks. Between 37 and 57 percent of whites make these invidious distinctions

between the races. Interestingly, whites indicate greater reluctance to distinguish blacks on the negative trait lazy (37 percent) than to credit whites with the positive traits intelligent and dependable (57 and 51 percent, respectively). It may seem less blatantly inflammatory to give one's own group the advantage in positive attributes than to depict the other group as prevailing in negative traits. In keeping with that cognitive caution, fewer than 3 percent of whites draw extreme differences between the races (that is, differences of 5 to 8 points), although the majority of whites who draw racial distinctions do describe those differences as substantial—that is, as 2 to 4 points apart. And as further evidence of caution in whites' racial trait attributions, virtually all whites refrain from describing blacks categorically: fewer than 4 percent of whites say that all blacks are lazy or that no blacks are intelligent or dependable. Trait attributions tend to be anchored around or above the middle of the scale (with the negative trait lazy being attributed to fewer members of both groups than are the two positive traits), indicating that whites are careful to acknowledge internal variation within each group, even if they draw distinctions between them.

The racial trait attributions of blacks differ from those made by whites in three main respects. First, blacks are less likely to believe there are any differences between the races in their personality traits: between 56 and 66 percent of blacks assert that there are no differences between the races in intelligence, dependability, or laziness (as compared with 42 to 50 percent of whites who say there are no racial differences in those traits). Second, when blacks do see racial differences in personality traits, they are somewhat more likely to assert opposite differences from those claimed by whites than to concur with the views of whites. This tendency is most pronounced with the negative trait lazy, where blacks are about twice as likely to assert that laziness is more characteristic of whites as they are to concur with the more common white claim that laziness is more characteristic of blacks. With the two positive traits, blacks are about equally likely to assert differences that favor their own group as they are to concur with the prevailing white portrayal of racial differences, but blacks whose perceptions favor their own group tend to assert larger differences between the groups. The net result is that only 11 to 20 percent of blacks accept the invidious trait distinctions that are drawn by 37 to 57 percent of whites. Third, blacks do not avoid categorical or near-categorical descriptions of either racial group quite as assiduously as do whites. Almost 10 percent of blacks categorically attribute the two positive traits, intelligent and dependable, to all whites and blacks; another 21 percent claim that almost all whites and blacks are intelligent and about another 8 percent say that almost all whites and blacks are dependable. Thus, blacks seem less aversive to making sweeping trait

attributions to racial groups, but these sweeping attributions are as likely to be made to either racial group.^[8]

Overall, blacks diverge less sharply from whites in their racial trait attributions than they do in their intergroup feelings or in their racial policy goals, but the differences between the trait attributions made by the two groups are still pronounced. Only 11 to 20 percent of blacks accept the invidious racial trait distinctions that are drawn by 37 to 57 percent of whites. Between 56 and 66 percent of blacks deny that there are any trait distinctions between the races, and another 16 to 25 percent make the more assertive claim that the trait distinctions favor blacks rather than whites. Faced with this opposition, whites appear to proceed cautiously rather than to launch into more extreme claims. Whites avoid categorical attributions or the assertion of extreme differences, and yet a sizable proportion of whites claim that there are small-to-substantial differences between the races in their endowment with key personal traits. Thus, whites respond to blacks' challenge by phrasing their comparisons of the two groups with care, moderating their claims of invidious intergroup distinctions with acknowledgments of intra group variation. In this way, whites are able to interject qualifications without yielding their claims of critical, invidious distinctions between the groups.

Gender

Trait attributions to gender groups contrast with racial trait attributions in four main ways. First, women and men do not differ from each other in their trait attributions to gender groups. Second, trait distinctions between the sexes are made more frequently, and these

[8] The tendency is somewhat more pronounced among black women than among black men. Indeed, black women are more likely than black men to make categorical attributions to all race, gender, and class groups.

distinctions also tend to be larger than those made between the races. Third, women are considerably more likely to be described categorically than are blacks. And fourth, although both men and women differentiate between gender groups fairly sharply, these attributions do not carry invidious overtones.

Table 8.4 shows that few people think there is any difference between the sexes in intelligence: about 70 percent of both women and men see no difference here (roughly comparable to the percentage of blacks who see no race difference in that trait). However, 50 to 60 percent of both sexes think there are more talkative women than men, and over 70 percent believe there are more emotional women. Although extremely large distinctions (of 5 to 8 points) are as uncommon between the sexes as between the races, moderately large differences (of 2 to 4 points) are seen by about one-third of both sexes in the trait talkative and by about one-half of both sexes in the trait emotional. These trait distinctions are, respectively, as large as and larger than the race differences seen by whites alone in the most differentiating race traits, dependable and intelligent.

The data in table 8.5 indicate that the images of each gender group that underlie the imputed distinctions between them are that men vary a good deal in their personal attributes whereas women tend to be fairly homogeneous. The trait intelligent elicits few categorical attributions, and between half and three-quarters of both sexes are commonly described as intelligent. But the traits emotional and talkative are attributed to all women by about one in six respondents. And about one-third of both sexes believe that all or almost all women are talkative (scale values

of 7 and 8), and almost one-half of both sexes make such sweeping statements about the proportion of women who are emotional. Consistent with these results is the low amount of uncertainty ("don't know's") expressed by either men or women when they are asked to describe the sexes—lower than one finds for either race or class attributions.^[9] All indications suggest that both men and women feel a high degree of confidence in describing the trait attributes of gender groups.

The trait distinctions that are so sharply drawn between women and men do not carry invidious value implications. The mean evaluations of the traits emotional and talkative hover around the neutral point (see table 8.1), and most respondents who draw a distinction between the sexes on these traits are implying an evaluation of women that is mildly negative, neutral, or mildly positive (data not shown here). Differentiation between the sexes in the trait intelligent does carry more invidious implications, but only a small minority draw such a distinction. In the paternalistic setting of gender relations, the definition of large, neutrally valued distinctions between the groups in personal attributes serves both to validate the group basis of social life and to neutralize it: the features that are alleged to distinguish the two groups are presented in a positive, rather than a negative, light. The success of this presentation can be gauged by its near-consensual endorsement by both groups.

Class

Trait attributions to social classes involve a more complex set of comparisons, since there are five, rather than just two, groups. To sharpen the focus, table 8.6 presents only those trait comparisons made between one's own class and other classes, rather than all possible class comparisons. Table 8.7 presents the images that are held of each class by the members of each class. These data reveal several features in the way people portray the salient personality attributes of different social classes.

First, trait distinctions are drawn frequently between classes, especially between classes from opposite ends of the continuum, where distinctions are sometimes larger and more frequent than between race or gender groups. As with interclass feelings, the frequency and size of distinctions that are drawn between classes increase incrementally as one moves from adjacent classes to classes that are further apart. Both the direction and the evaluative connotations of these attributed trait distinctions vary. Second, categorical attributions are also made relatively frequently, although it is the two highest social classes that are most likely to be described

[9] Uncertainty about how to describe a group might be seen as diametrically opposed to the categorical description of a group. Indeed, certainty is sometimes explicitly included in definitions of a stereotype (Lippmann 1922; Richter 1956; Harding et al. 1969).

categorically. The upper class is described categorically about as frequently as are women, with the upper-middle class not too far behind. Further, the class context is unique in that categorical exclusions of traits from a class (for example, "no poor people are selfish") are also made by more than a handful of respondents. Third, the amount of disagreement among the social classes is not pronounced, although it is more in evidence than it is between gender groups. Across all classes, the same general patterns of trait attributions are made, but there is some tendency for distinctions to be asserted more frequently in the class that is reflected favorably in the comparison and less frequently in the class that is reflected unfavorably.

Americans commonly believe that both intelligence and selfishness are found increasingly with ascending social class. Differences in intelligence are attributed more frequently than are differences in selfishness, but note that although the former belief puts higher classes in a more favorable light the latter belief puts lower classes in a more favorable light. Class distinctions in the traits thrifty and lazy are just as likely to be seen as in the traits intelligent and selfish, but there is less convergence within each class as to the direction of these differences. The prevailing attributions in each class are that the poor have the most lazy and the fewest thrifty people, with the upper class ranked next, followed by the upper-middle class and then the other two classes. These two trait attributions both cast the poor in a more negative light than other classes. At the same time, they ironically characterize the upper class almost as negatively as the poor, whereas the working class and middle class are portrayed in a relatively positive light.

Classes are portrayed as differing sharply in intelligence. Indeed, differences in this trait are drawn more frequently between classes from opposite ends of the continuum than race differences in this trait are asserted by whites. With decreasing social class, there is a gradual drop in the percentage of people subscribing to this belief, but even in the poor and working class, the ascription of greater intelligence to higher classes is widespread. Thus, in the upper-middle class, 84 percent assert that their class has more intelligent people than does the poor and 80 percent compare their class as favorably vis-à-vis the working class. Among the poor, 70 percent concur with the upper-middle class's assertion of the difference between them in intelligence, and 66 percent of the working class concur with the upper-middle class's asserted superiority over them in intelligence. The negative trait selfish is also attributed more frequently to higher classes (although not as overwhelmingly as is intelligence)—this attribution reflects less well on higher classes, and in this case it is the higher classes that tend to minimize the distinction.

There is considerably less agreement, both within and across classes,

about which classes are more characterized by the traits thrifty and lazy. Again, the clearest distinction is made between the poor and other classes: in comparisons between the other classes, people from each class are almost as likely to see differences in one direction as in the other, although there is a slight tendency for self-flattering attributions to prevail and for attributions that reflect negatively on one's own class to be minimized. In the upper-middle, middle, and working classes, the prevailing assertions are that the poor have the most lazy and the fewest thrifty people, with the upper class ranked next. In addition, there is a slight tendency for people in these three classes to see their own class as having the fewest lazy and the most thrifty people. Note that, in comparing their own class with the poor, about one-half of the people in each of these three classes assert that the poor are more likely to be lazy—considerably more than the percentage of whites (37 percent) who distinguish blacks from whites on this negative trait. Among the poor, there tends to be concurrence with the view that higher social classes have more thrifty people than does their own class (although with less vigor than this distinction is asserted by higher classes), but the poor tend to reject the portrayal of their class as the most characterized by laziness. Although most poor people do see a difference between their own class and higher classes in laziness, they are divided almost evenly about whether it is the higher class or their own that is more characterized by this negative trait.

Table 8.6 does not display data on the size of asserted class differences, but large distinctions are drawn more frequently between the classes than between racial or gender groups (and this is reflected in the size of the mean difference-scores). Indeed, although extremely large distinctions (±5 to ±8) are made infrequently, they are made more often than with race or gender, and moderate-to-substantial distinctions (±2 to ±4) are more likely to be seen between classes from opposite ends of the continuum than between the races or sexes. For example, when comparing the poor and the upper class on the trait intelligent, more than 12 percent of each class see extreme differences, and as many as 48 to 59 percent of each class see moderate-to-substantial differences (data not presented here).

Table 8.7 indicates that the images of each class that underlie these trait distinctions include a relatively high incidence of categorization. It is the upper and upper-middle classes that are most likely to be described categorically. The upper class is described categorically about as often as are women, with the upper-middle class not far behind. Poor people are slightly more likely than people from other classes to attribute traits categorically to all classes (including their own), especially positive traits. As with blacks' images of blacks and whites, poor people's categorical

descriptions of classes do not seem to be part of an attempt to differentiate between them. And the stronger proclivity of the poor to make categorical trait attributions is offset by their considerably higher expression of uncertainty. When asked to describe any class, about one-tenth to one-fifth of the poor say "don't know," whereas people from other classes give this response with approximately the same infrequency as in the race context.

Intelligence is thought to characterize all or almost all of the upper class by fully one-half of the poor and working class and by almost as many people from the middle and upper-middle classes. The other positive trait (thrifty) is attributed categorically to the upper class by over one-third of the poor, whereas the negative traits selfish and lazy are attributed categorically to the upper class by about one-quarter and one-sixth of the poor, respectively.^[10] The proportion of the poor making such sweeping attributions to the upper-middle class is only somewhat lower. People from other social classes are as likely or almost as likely as the poor to attribute these traits to all or almost all of the upper class,^[11] but considerably less likely to make such sweeping attributions to the upper-middle class. A markedly smaller proportion of people from any class perceives the other three classes in categorical terms. In this regard, note that the negative trait lazy is attributed categorically or near-categorically to the poor with about the same infrequency as it is attributed in this way to blacks (between 9 and 14 percent of each class). The positive trait intelligent, however, is attributed categorically to the upper class much more often than it is attributed categorically to whites (by about one-half of each class, compared with about 16 percent of whites and 29 percent of blacks who attribute that trait categorically to whites).

General Patterns

Portrayals of the trait attributes of unequal groups seem to be characterized more by moderation and caution than by unrestrained assertions. Extreme trait distinctions between groups are made by only a small minority of respondents, and categorical attributions to specific groups are also the exception rather than the rule. In addition, these assertions do not always carry value connotations that are either strong or negative toward the subordinate group. Indeed, the size of the distinctions that are made and their positive or negative value connotations vary independently of one another. Although the gender context is marked by the relatively frequent perception of substantial trait distinc-

[10] Discussion here is based on belief-scale values of 7 and 8. As with race and gender, "pure" categorical thinking (scale values of 8 or 0) is generally less prevalent than "almost all" or "almost none" statements (scale values of 7 or 1).

[11] In fact, people from all classes are about equally likely to attribute the two traits intelligent and selfish to the upper class.

tions between the sexes, these distinctions rarely imply a negative evaluation of women. In the race context, the perception of trait distinctions is considerably less frequent and imputed distinctions tend to be moderate rather than extreme—but the evaluative connotations of the distinctions that are made by whites are sharply negative toward blacks. Trait differences among the social classes are seen frequently and they are often substantial distinctions, but the evaluative connotations are mixed rather than uniformly positive or negative. Thus, intergroup attributions do not seem to conform to the specifications of the stereotype concept in any of these three intergroup relationships.

In addition, none of the intergroup relationships is marked by dramatic disagreement between groups about their distinguishing characteristics. In no case is the rift between groups such as to produce radically different portrayals. Subordinate groups do appear to find it relatively difficult to challenge dominant group definitions of the distinctions between them. It seems that dominant groups are generally quite successful (although to varying degrees) in defining the way in which the contending groups are distinguished in salient personal attributes.

The infusion of subordinate thinking with dominant beliefs is most successful in gender relations. An amicable consensus appears to envelop gender groups about their defining characteristics. Men assert that there are substantial but evaluatively neutral differences between the sexes, and women do not disagree. A challenge to dominant assertions is most likely to be made in the race context. Most blacks deny white assertions about black inferiority and assert their equality with (and sometimes their own superiority to) whites. Faced with this challenge, whites become skittish about making dramatic assertions of racial distinctions: some whites disclaim racial differences altogether, and the majority who do draw distinctions do so with care. In class relations, substantial personality-trait distinctions are widely perceived and they more often favor higher classes over lower ones. These differences tend to be minimized somewhat by lower social classes or even refuted with minor competing claims of their own superior endowment, but the dissent is modest.

These differences across the three intergroup relationships reveal the variability and the malleability of stories that are woven to characterize unequal groups. In the sticky web of paternalism that envelops gender relations, men praise women for their distinctive personality attributes and women graciously accept the confining compliments. In the more conflictive political environment of race relations, whites have met the challenge from blacks by retreating from sharply categorical distinctions as they attempt to preserve a blurred definition of group attributes that does not violate their heightened commitment to the norm of individualism. In class relations, shaped by a dual history of muted conflict and

paternalistic domination, subordinate classes are both derogated and flattered. Sharp class distinctions in key achievement-related attributes have been embossed on the public consciousness with only petty dissent from the less-favored classes. At the same time, subordinates are granted superiority on an attribute that is irrelevant to the business of socioeconomic achievement. Class images are further muddied by the fact that class relations involve multiple groups in which people in all classes except the poor and the upper class schizophrenically experience deprivation and privilege as their vision swivels upwards or beneath them. Even those who enjoy the safehouse of upper-middle-class privilege are not immune to occasional moments of resentment against a class higher than their own. Images of the low and the mighty classes are thus torn by these competing pressures.

Data:
Popular Explanations of Group Differences

I turn now to the folk wisdom that accompanies group attributions. To what extent do people think there are important differences between their own and other groups, and how do they account for those differences, for race, gender, and social class?

Measures

Immediately following the administration of the beliefs booklet, respondents were asked the following set of questions, beginning with gender, followed by race, and then class:

Apart from differences in appearance between men and women, do you think there are many important differences between men and women, some important differences, a few important differences, or no important differences?

If respondents gave one of the first three response-options, they were asked the follow-up:

People disagree about why there are differences between men and women . Which of the statements on this card comes closest to what you think? Just tell me the letter of your answer. (Interviewer instruction: READ CARD. MARK ALL LETTERS MENTIONED.)

X. Most differences are there because they're born different.

Y. Most differences come from the way they're brought up at home.

Z. Most differences come from the different opportunities they have in America.

The wording for the comparable questions for race and class was identical, except that the lead-in for the class questions began:

Remember I asked you about the poor, the working class, the middle class, the upper-middle class, and the upper class. Do you think there are many important differences between these classes, some important differences, . . . [and so on].

The first item in each set takes a broader gauge of the potential group distinctions that people might see than do the personality-trait attributions analyzed in the previous section. The respondent was free to draw on any characteristics that she felt were variant across groups: this might include personality attributes, socioeconomic differences, crime rates, or anything else that the respondent considered salient.

The second question in each set provides for three primary explanations of perceived group differences: groups are inherently or biologically different (born different ), groups are raised with different cultural values (home upbringing ), and groups have unequal opportunities in society (opportunities ). These three response-options cover the essential distinctions about which analysts have been concerned. The first explanation (born different ) is the one associated with a classic racist position: it makes the group's condition immutable and therefore beyond the reach of public policy. The second explanation (home upbringing ) alludes to differences in culture and values between groups, but it takes on a slightly different meaning for gender than for race and class. Since both gender groups are raised in the same homes by the same people, reference to the way people are brought up at home has more discriminatory implications. (Note, however, that the critique is at best ambiguous, since it is the subordinates—women—who have primary responsibility for bringing up children of both sexes, and private homes have been defined as safely outside the domain of public policy.) For race and class, the meaning of the second proffered explanation is more clear-cut, since the pertinent groups are brought up within their own ranks: thus, any differences in their upbringing at home are the responsibility of the group itself. This explanation casts the blame for any differences between race or class groups on the subordinate group, and it implies that any policy intervention would have to include such paternalistic measures as resocialization of subordinates or developing special programs that would counteract subordinates' unfortunate values. The third explanation (opportunities ) casts the blame for any group differences on the social system, and it implies that to eliminate group differences, biases in the structure of society must be eradicated.

Results

Table 8.8 presents the percentages of each group who believe there are no important differences, a few, some, or many important differences between contending groups, and table 8.9 presents the percentages of

the respective groups who endorse various explanations of those differences.

In all three intergroup relationships, the overwhelming view is that there are at least a few important differences between the contending groups. And in every group except one (blacks), at least two-thirds claim that there are more than just a few important differences between the pertinent groups. Blacks are the only group to diverge from this generally held view, and they diverge only slightly. About 30 percent of blacks think there are no important differences between the races, as compared

with 14 percent of whites who hold this view, about 12 percent of men and women who think there are no important gender differences, and about 7 to 10 percent of the four social classes who think there are no important class differences.

The folk explanations that are adduced to account for these alleged differences do show more variance across the three intergroup relationships. Table 8.9 presents the percentage of each group who espouse each of three alternative explanations for group differences: that the groups are "born different," that there are differences in "the way they're brought up at home," or that they have "different opportunities." A few respondents preferred to cite more than one explanation, and those responses are recorded in the bottom rows of the table, along with the "don't know's."

Group differences are not generally seen as inherent. The explanation that the groups are "born different" is particularly out of favor in class relations, where fewer than 8 percent of any class endorses this view. In race relations, it is also the least favored of the three possible explanations, but it gains more credence than in the explanation of class differences, especially among whites: 18 percent of whites and 11 percent of blacks see race differences as inherent. It is in gender relations that this explanation has most credence, especially among men: 28 percent of men and 21 percent of women say that most important gender differences are inherent.

The explanation of important group differences in terms of "different opportunities" is found most frequently in class relations, although endorsement of this view varies by class, decreasing gradually from 57 percent of the poor to 35 percent of the upper-middle class. Blacks are less likely to attribute race differences to this cause than are poor, working-class, and middle class people to explain class differences this way, but blacks are still somewhat more likely to blame differences in opportunities than are whites: 42 percent of blacks and 29 percent of whites take this view (making whites just slightly less likely than the upper-middle class to blame opportunities). Differences in opportunities are least likely to be cited in the gender context, where this explanation has less credence than either of the other two possibilities. Only about 15 percent of either men or women adduce this explanation for the important gender differences that they see.

Attributions to "the way they're brought up at home" as the main cause of group differences are found most frequently in gender relations, where this explanation is also cited more often than either of the other two possibilities by both sexes: 43 percent of men and 52 percent of women take this view. As noted earlier, this explanation has a somewhat more critical ring in the context of gender relations, since men and

women are not brought up in separate homes, but, even in this context, it is considerably less critical than the allusion to opportunities, since it is the subordinate group—women—who are assigned the main responsibility for how children are brought up. In race and class relations, this explanation is cited almost as often by whites and by the upper-middle class as it is by men, but blacks and the poor are less likely to accede to this explanation. About 37 percent of whites and 29 percent of blacks say that home upbringing is the cause of most important race differences, and it is cited as the cause of most important class differences by 37 percent of the upper-middle class, 34 percent of the middle class and working class, and 23 percent of the poor.

Thus, the repertoire of folk wisdom about the causes of group differences varies somewhat across the three intergroup relationships. In the explanation of class differences, the dialogue is almost exclusively about whether it is the cultural socialization that takes place within each group or the societal structure of opportunities that bears primary responsibility. Opinion in the upper-middle class is split pretty evenly between those two views, but with descending social class, opinion gradually shifts more in the direction of blaming opportunities: among the poor, the opportunities explanation is favored over home upbringing by almost three to one.

In the explanation of race differences, the same two alternative explanations are again the most commonly cited, but these two explanations do not hold such a near-exclusive reign as they do in class relations. A significant minority of whites (almost one-fifth) subscribe to the view that racial differences are immutable, and, although blacks are less likely than whites to share this view (just over one-tenth), they are more likely than the poor to attribute pertinent group differences to inherent causes. Of the two main explanations that are adduced in the race context, whites lean somewhat more toward cultural attributions whereas blacks lean more toward explaining race differences in terms of the structure of opportunities. Whites and blacks differ less sharply in their racial explanations than do the social classes in their explanations of class differences. It seems that whites have been more successful than the upper-middle class in steering the thrust of folk wisdom away from explanations that challenge structural arrangements and toward interpretations that lay the blame outside the purview of public policy. The economic basis of class distinctions appears to have a tangibility that is more difficult to obscure (as we found with the perception of class interests).

In the folk explanation of gender differences, structural factors are further obscured, and the stories told by the two groups are more closely allied than in race relations. Gender differences are attributed to the structure of opportunities with about the same infrequency as race dif-

ferences are attributed to inherent causes, and this holds for women's explanations as much as men's. Popular discussion of gender differences is centered primarily on whether they are inherent or caused by differences in socialization, with the latter interpretation carrying overtones that are ambiguously critical of societal arrangements while still being outside the scope of public policy. Both men and women favor the latter view over the attribution of inherent differences, although the margin of preference is slightly greater among women than among men. In the context of gender relations, where paternalism holds the strongest sway, causal allusions to the structure of opportunities have been relegated to a fringe status, whereas the attribution to inherent causes remains a respectable option. As long as hostility is kept at bay, the dominant group may claim inherent and indelible differences between the groups with relative impunity, and the framework of popular discussion is shifted away from the ogre of outright challenge. Women's mild digression from the prevailing path is to venture slightly ahead of men into an interpretation that alludes to social rather than biological causes but which also lays the responsibility for group differences as much with women themselves as with men.

Conclusions

Cognitive attributions about social groups have been a focal point of research on intergroup attitudes. A particular kind of intergroup belief, the stereotype, has occupied a venerable place in that research. Intimately linked to the concept of prejudice, the stereotype has been the template used to assess the quality of intergroup attitudes. Researchers have targeted cognitive attributions that are categorical, sharply differentiating, derogatory, inflexible, and untrue as the kind of beliefs that damage subordinate groups. Ignorance and parochialism are thought to provide the breeding ground for such beliefs, which then set the mold for hostile feelings and discriminatory dispositions toward subordinates while eroding the self-esteem of subordinates themselves. The stereotype has so dominated scholarly thinking about intergroup cognitive attributions that researchers have felt little compunction to validate the concept through measurement. Instead, the prevailing measures of intergroup trait attributions have been phrased so as to constrain responses to simplified, categorical terms.

There has also been growing interest in the popular explanations that are adduced to explain the differences people perceive between groups. Folk wisdom about group differences has been measured in a variety of ways, but researchers have been especially concerned with distinguishing among popular explanations that treat subordinate groups as biologically

distinct, as culturally different, or as the victims of biased social arrangements. The primary purpose has been to ascertain the extent to which people have a cognitive basis for supporting policies geared toward affirmative change. In this endeavor, the cultural type of explanation has occupied a relatively ambiguous middle-ground, one that still casts blame within the subordinate group but with less rigidity than the biological view, which is seen as the clear hallmark of the bigot.

My approach to intergroup attributions takes a different course. Instead of being the fuel for discrimination, individuals' intergroup beliefs are molded by the system of inequality that links their own group with another. Rather than being the naive driver of inequality, intergroup cognitive attributions are themselves driven by the political exigencies arising from the system of inequality. These beliefs are the cognitive embroidery of intergroup relations, with the patterns tacked in to meet the demands imposed by the pursuance of group interests within a specific set of constraints.

The perennial search of scholars for a specific type of intergroup belief as the bane of intergroup relations has diverted us from the variegated forms that such beliefs may assume, as the beneficiaries of inequality tell stories that advance their self-serving interpretation of their privilege. Various traits may be credited to or debited from groups, and various modes of attribution may be employed, depending on the political exigencies that are operant and the moral code that holds sway. Intergroup beliefs are of interest for what they reveal about the points of tension in an intergroup relationship and the moral umbrella that is unfurled over the inequality.

These constraints are unlikely to produce unmitigated hostility. As the members of the dominant group attempt to explain expropriative arrangements as fair and reasonable, they are channeled into perceptions that seem compelling from their own perspective and that also seem to appease subordinates. Insofar as dominant-group members continue to seek the compliance of subordinates in expropriative arrangements, they can ill-afford to be impervious to subordinates' sensibilities. Various stories and variants of stories circulate: through trial and error, some are relegated to a marginal position or cast off altogether, and others become broadly accepted. The diversity or singularity of opinion within and across the ranks of each group gives an indication of the extent to which stories have evolved that "work" for all the participants, or whether alternative packages of ideas are vying for primacy.

As the beneficiaries of inequality seek to promote their cognitive attributions, they draw on recognized fonts of knowledge and established moral tenets to validate their claims. In the contemporary era, science has largely displaced religion as the legitimate basis of knowledge, and

individualism has become the governing moral precept. Both of these factors influence the content and language of intergroup beliefs. As the specific exigencies of an intergroup relationship shape the contours of dominant stories, scientific and individualistic principles infiltrate their language and expression.

As various stories circulate, the beliefs that gain the broadest acceptance are those that seem to flow most naturally from the constraints of the relationship and the prevailing morality. Such beliefs have credibility because they seem to be substantiated by the daily "facts" that the relationship creates. Thus, the beliefs that develop with the most ease are also the beliefs that offer the most compelling explanation of current arrangements. For this reason, it is misleading to ask whether dominant portrayals of subordinates are "true" or "false." It is more significant to ask why certain personal attributes are emphasized in intergroup beliefs and how much consensus there is within and between groups in those ascriptions. In relations marked by high personal contact across group lines, the two groups are daily presented with the same set of stimuli and they are exposed to the same communication channels. Thus, both groups are routinely presented with the same "confirmation" of dominant beliefs. In relations marked by spatial segregation, the lives of the two groups diverge into separate daily spheres, and the way is opened for subordinates to develop communication networks that are less tied to those of the dominant group. In this context, the unequal groups are more likely to differ in both the stimuli they encounter and the way those stimuli are interpreted.

Thus, the way unequal groups depict their vital attributes is both political and interactive, as the members of each group react to the circumstances in which they find themselves and the claims of the other participants. Dominant beliefs cannot be understood without an assessment of the political climate that pervades the intergroup relationship, and subordinates' views are an important element of that climate. At the same time, the structure of the relationship provides subordinates with varying opportunities to break loose from the symbolic framing of the relationship that is imposed by the privileged participants.

When role segregation constitutes an important basis of group differentiation, it is convenient for dominant-group members to depict the affected groups as sharply distinctive in their vital attributes in order to make the role divisions seem natura. The more central role segregation is to the conduct of the relationship, the more the dominant group feels moved to depict subordinates as sharply and categorically distinct and to explain group differences as self-perpetuating, either through biological or cultural processes. And when the groups are in frequent personal contact, as is the case in relations that are governed by role segregation,

it is convenient to describe subordinates in distinctive terms simply in order to brace the boundaries between groups. In the paternalistic atmosphere that is encouraged by the practice of role segregation with high one-to-one contact across group lines, the clear delineation of the subordinate group's attributes also serves the dominant group's interest in defining subordinates' needs . By describing subordinates as distinctively different from the dominant group in their personal attributes, the way is cleared to define subordinates' needs as distinct from those of the dominant group as well.

In this endeavor, however, there is every reason to avoid the hostile derogation of subordinates. By avoiding derogation while sharply distinguishing the subordinate group, the role segregation is sustained in the cozy spirit of "Vive la difference!" As subordinates comply with the demands of the relationship and the groups thus routinely perform distinctive tasks and behaviors in life, it does not take any contrivance for the members of the dominant group to believe that the groups are indeed distinctive in their personal attributes. Subordinates, too, need only to look at the behaviors of those around them for daily confirmation of the dominant beliefs.

When spatial segregation is the primary differentiating mechanism, there is less need for the categorical differentiation of the subordinate group's personal attributes. The spatial boundary alone accomplishes the clear separation of the groups, and the dominant group thus feels less of an imperative to underscore the boundary line between groups. In addition, the spatial separation of the groups permits subordinates to break away from the dominant group's cognitive and emotional grip and to incubate a challenge to the modus vivendi. Dominant-group members are thus pushed into a defensive position, in which they feel impelled to assert their superiority more overtly, but in which continued emphasis on the group basis of social life has become incendiary. In this more unwieldy situation, dominant-group members prefer to shift the moral tenor of debate away from groups to individuals, even as they are pressed to portray subordinates in a more negative light.

In the contemporary era, the morality of individualism has already evolved and ripened from the competitive praxis of the capitalist economy: individualistic principles govern economic life, and they fall conveniently to hand when they are needed for the exegesis of other intergroup relations. As the members of the dominant group reach for the moral umbrella of individualism to undermine the credibility of group-based demands and complaints, categorical attributions become a liability and they atrophy. Categorical attributions are reshaped into probabilistic ones, even as dominant-group members are pushed by subordinate challenge to assert their superiority more overtly. The language of science,

with its emphasis on such methodological principles as variability, precise measurement, and central tendencies, provides a credible and prestigious basis for probabilistic intergroup comparisons. In the context of spatial segregation, such comparisons are perfectly serviceable for explaining the differential experiences of the unequal group. At the same time, probabilistic comparisons combat an incendiary atmosphere with a language that seems reasoned, moderate, and unbiased.

The patterns of intergroup beliefs observed for race, class, and gender relations illustrate the politically adaptive quality of those beliefs. When we rely on more refined measures than have generally been used in the past and we compare across three intergroup relationships marked by different structural exigencies, we find a distinctive pattern of intergroup beliefs in each case. The prevalence of categorical descriptions, large cognitive distinctions, derogatory characterizations, and biological, cultural, and structural explanatory schemes varies substantially across the three intergroup relationships. Further, these various elements of intergroup beliefs are not locked together, as the stereotype concept would lead us to assume. Instead, they vary flexibly to create configurations of beliefs that are uniquely adapted to the political context in which they are formulated.

In gender relations, the lives of men and women are intricately bound together within the rigid confines of role segregation. Here, the contending groups manifest the strongest convergence in their intergroup beliefs, and the categorical differentiation of subordinates occupies a central place in those beliefs. At the same time, the evaluative overtones are kept on a positive note. Women are warmly congratulated for their distinctiveness in personal traits that are appropriate to the tasks and behaviors assigned to them and to which men have no aspirations. And although biological attributions are generally out of favor in all three intergroup relationships, gender differences are attributed to immutable, biological causes by over one-quarter of men and one-fifth of women—more than for race or class, and more often than differential opportunities are invoked to account for gender differences. The most common explanation of gender differences is to allude to the way people are brought up at home. Thus, the overwhelming majority of both men and women believe women are clearly different from men—in positive ways—and that those differences are caused by factors outside the domain of governmental intervention: responsibility is generally attributed to either the physical makeup or behavioral proclivities of women themselves.

Class relations are marked by an intermixture of role and spatial segregation, with a step-wise pattern of contact that brings people into considerable personal contact with those in adjacent classes but little or no

contact with people in more distant classes. Class is also distinct in that there is an absence of rigid, social rules for group ascription: instead, there has been an emphasis on the importance of individual, achievement-related factors as the basis for socioeconomic standing. In the resulting hierarchy of self-defined classes, people who are at neither the pinnacle nor the base alternately occupy a dominant or a subordinate position, depending on whether they cast their vision beneath or above them.

In this context, large and invidious differences are drawn among the classes in the vital achievement-related trait of intelligence. This attribution prevails in all social classes, although its popularity decreases slightly with descending social class. Categorical attributions are invoked as frequently as in gender relations, but, interestingly, the brunt of those attributions is softened by describing the two highest classes in positive categorical terms rather than portraying the lowest classes as categorically deficient in this vital trait. This use of categorical attributions thus pays homage to individualism both by crediting higher classes with a clearly superior endowment of this achievement-related trait and by acknowledging individual variability in the ranks of the lower classes. To soften the sting of this invidious class comparison further, the lower classes are simultaneously credited with superiority in another trait—selfish—that is not pertinent to the justification of socioeconomic inequality. On two other achievement-related traits—thrifty and lazy—the poor are once again compared unfavorably to higher classes, but in these traits people in the working, middle, and upper-middle classes also tend to portray higher social classes as less well endowed than their own. Thus, a variegated pattern of trait attributions emerges among the social classes, in which categorical attributions, positive and negative evaluative connotations, and various class distinctions are intermixed.

Although lower classes have absorbed much of the unflattering portrayal of them in achievement-related personal traits, by and large they have not been asked to see themselves as biologically inferior. The reluctance of dominant classes to declare rigid class groupings makes it awkward for them to draw immutable, biological differences between classes. In the folk explanations that are invoked to account for class differences, biological factors are barely mentioned, and the debate instead centers on cultural factors versus the opportunity structure. Among the poor, there is a strong preference for the argument that the opportunity structure is to blame for class differences. With ascending class, the popularity of that argument diminishes as more people are drawn to the position that responsibility lies with the cultural attributes of classes—a factor that is potentially subject to change but which remains beyond the scope of any obvious governmental intervention.

In race relations, the widespread residential segregation of blacks from whites in urban America has accomplished a pervasive spatial separation of the two groups. Past policies and practices by the federal government, local governments, and the banking and real estate industries have made residential racial segregation such an embedded feature of contemporary urban life that it now requires little active engagement on the part of individual whites to maintain it. Thus, most whites do not feel pressed to explain the residential segregation that insulates them from blacks—it is conveniently a fait accompli. But the lower socioeconomic achievement of the subordinate racial group presents itself as a glaring issue that requires explanation. Indeed, the spatial segregation of the two groups highlights the socioeconomic differences between them, as well as offering blacks more opportunity to break loose from dominant communication channels and articulate a grievance.

In these circumstances, the primary symbolic task for whites is to draw distinctions between the two groups in personal attributes that can account for the differences between them in achievement. As whites accord superiority to themselves in such achievement-related attributes as intelligence, dependability, and laziness, they are generally careful to differentiate the average characteristics of the two groups without making sharp or categorical distinctions. Rather than pouring venom into an already endangered relationship by describing blacks as categorically deficient in achievement-related traits, whites simply credit themselves with a somewhat greater endowment of those traits. And although a significant minority of whites claim the differences between the races are immutable, whites are more likely to allude to cultural factors in accounting for racial differences.

The moral umbrella of individualism is unfurled squarely over the relationship and encases the challenge that is articulated by blacks. Blacks are unlikely to challenge the values attached to achievement-related personal attributes. A minority asserts superiority to whites in those attributes, but blacks' challenge to whites' symbolic portrayal of racial differences is largely confined to denying racial differences in key personal traits and to explaining racial differences in terms of the structure of opportunities more often than do whites. Overall, blacks' cognitive attributions diverge on their own path more notably than do those of other subordinate groups in this study, but they are necessarily contained within the same symbolic framework as that put forward by whites.

These variant patterns of intergroup beliefs in gender, class, and race relations demonstrate the malleability of those beliefs in the face of political pressures. As people try to pursue their interests and as they grope to interpret the circumstances in which they find themselves, they succumb to the relentless pressure of the stimuli that press on them. The

structural arrangements that shape their lives lend credibility to alternative cognitive attributions. Various stories circulate to address the exigencies that people face. In their ongoing interest in communicating with and persuading others, the bearers of those stories are obliged to heed the stories and the morality that others espouse and to frame their own assertions in a common language. The resulting commonality or diversity of stories within and across groups reveals the extent to which structural arrangements create common, interrelated, or divergent conditions of life and lines of communication. Intimate relations based on role segregation nourish paternalistic consensus, and one finds the simple categorization of subordinates, rendered in benign terms. In relations based on spatial segregation, the members of the dominant group retreat from simple group differentiation as they feel the sting of subordinate dissent and hostility creeps into the relationship.

Thus, the cognitive attributions that are formulated by the members of the dominant group are not fixed but are continually reshaped in an interactive process with other members of their own group, with subordinates, and in response to any alteration in circumstances. In this way, dominant-group members retain the upper hand in framing the symbolic portrayal of their relationship with subordinates. In relations with high spatial separation of the groups, the dominant group's communication lines do not envelop subordinates as effectively, but even here, subordinate dissent from the main symbolic themes is contained. Stories are advanced that seem credible, and those stories flourish, fade, or are remodeled, according to their fitness in addressing the stresses that beset the various participants in the relationship.