The Wolf We Feed: Democracy, Caste, and Legitimacy

Benjamin Justice* and Tracey L. Meares** June, 2021

Legal authority rests on enactment; its pure type is best represented by bureaucracy. The basic idea is that laws can be enacted and changed at pleasure by formally correct procedure. The governing body is either elected or appointed and constitutes as a whole and in all its sections rational organizations. . . . Obedience is not owed to anybody personally but to enacted rules and regulations which specify to whom and to what rule people owe obedience. The person in authority, too, obeys a rule when giving an order, namely ‘the law,’ or ‘rules and regulations’ which represent abstract norms. The person in command typically is the ‘superior’ within a functionally defined ‘competency’ or ‘jurisdiction,’ and his right to govern is legitimized by enactment.
—Max Weber¹

It must be remembered that the white group of laborers, while they received a low wage, were compensated in part by a sort of public and psychological wage. They were given public deference and titles of courtesy because they were white. They were admitted freely with all classes of white people to public functions, public parks, and the best schools. The police were drawn from their ranks, and the courts, dependent upon their votes, treated them with such leniency as to encourage lawlessness. Their vote selected public officials, and while this had small effect upon the economic situation, it had great effect upon their personal treatment and the deference shown them. White schoolhouses were the best in the community, and conspicuously placed, and they cost anywhere from twice to ten times as much per capita as the colored schools. The newspapers specialized on news that flattered the poor whites and almost utterly ignored the Negro except in crime and ridicule.
—W.E.B. Du Bois²

Introduction

Procedure is central to American public legal discourse.³ From the soaring rhetoric of the Declaration of Independence to the Due Process Clause of the Fourteenth Amendment, the American legal tradition rests on the principle that law must be both derived and applied according to fair process.⁴ Consider that in the 2020 election the Trump Administration resorted to fervent and false allegations of widespread voter fraud⁵—that the election process was fundamentally unfair—in order to weaponize Republican voters’ ostensible commitments to fairness against what was, objectively, one of the least procedurally unfair elections in history.⁶ Yet the four-year period of the Trump Administration (2017–2021) also saw the rise of overt and deliberate racist politics and mounting evidence that a universal commitment by all to fairness for all across the United States is a mythical framing of the American creed.⁷ One can look from the apparent lack of justice for unarmed Black civilians killed by police officers to a sitting President’s affirmative support for white supremacist groups to observe the doublespeak associated with fascist regimes: claiming to be restoring law and order while backing away from commitments to due process and equal protection.⁸ And yet, simultaneously, we have also seen the mainstreaming of a successful oppositional politics, including Black Lives Matter, which in June 2020 enjoyed a peak 67 percent approval rating among American adults in a national survey.⁹ Even as white nationalism flourished under the Trump Administration, polls indicate that increasing numbers of Americans acknowledge that our society is unfair to racial minorities.¹⁰ As Americans stand at the perpetual racial crossroads of the twenty-first century, how much does the legitimacy that they accord their government depend on the procedural justice it delivers to all?¹¹

The question of state legitimacy is the province of many disciplines. In this Essay we consider the question from a social-psychological perspective and focus on the centrality of process to the public’s perceptions of fairness. What we find is a contradiction. The social psychology of procedural justice focuses empirically on what people say matters to them in coming to conclusions about whether an authority, institution, or social arrangement is right, appropriate, and just.¹² This scholarship tells us that people, no matter their group membership, reach conclusions about the fairness and legitimacy of legal authorities by relying much more on how individuals are treated by legal authorities than on the outcomes of decisions those authorities produce. Moreover, this research supports the notion that members of one group believe that members of groups to which they do not belong should also be treated by authorities in the ways they want authorities to treat their own group, registering less support for authorities that do not.¹³ The social psychology of intergroup relations, however, undermines the ideas of universality that procedural justice research promotes. For example, psychological experiments demonstrate that, when primed, white respondents (because they are members of the dominant group) view Black people with overt and subconscious suspicion and are more likely to support harsher punishments for them—two clear examples that potentially call into question the belief by all in fairness for all racial groups.¹⁴ One implication of these and other studies is that whites may well support different treatment of Blacks than whites along lines that the procedural justice research seems to refute, and, moreover, consider such different treatment in their judgments about whether the government is legitimate.

This apparent paradox reflects a deeper rift in theories of state legitimacy. The study of procedural justice research flows from the Western (white) political-philosophical tradition fully articulated in the early twentieth century by Max Weber, which presupposes that liberal democracies derive their legitimacy from universally fair treatment, such that disparate treatment of any group erodes the legitimacy of state authority and weakens the rule of law.¹⁵ The analysis of racial bias in civic identity and behavior flows from the African American philosophical tradition, fully articulated in the early twentieth century by W.E.B. Du Bois, a contemporary of Weber’s. This tradition presupposes the United States to be a caste society in which disparate treatment of racial minorities is an expected role for state authority, observing that caste was written into law such that the democratic norms of white society operated in uneasy tandem with the antidemocratic demands of white supremacy.¹⁶ Du Bois also noted that the problem of anti-Blackness was not national but global, sitting right alongside the supposed march of progress through European imperialism.¹⁷ In the late twentieth century, Critical Race Theory emerged to understand the contradiction Du Bois condemned—how democratic legitimacy can sit side by side with caste legitimacy, each indicating a set of deeply ingrained expectations of the law even after overt racial caste law has been abolished.¹⁸

Does state legitimacy in the United States depend on democratic norms of procedural justice, or caste norms of white supremacy maintenance? Our answer to the question is, in each instance, “probably yes,” and we do not attempt a definitive answer. Within this apparent disjuncture in theory and scholarship, however, we see opportunity. It is clear from our review of the literature that the study of legitimation mechanisms, especially in psychology, has yet to fully account for the ways in which people’s commitments to procedural justice relate to their commitments to racial hierarchy maintenance. Understanding more clearly how and in what circumstances these legitimizing evaluations are enacted could contribute to better strategies for political discourse and lawmaking generally and better responses to the urgent need for police reform and reconceptualization of public safety.¹⁹

Our exploration proceeds in three Parts. In Part I, we briefly sketch the genealogy and current state of empirical research on democratic legitimacy, with special attention to procedural justice. In Part II, we similarly examine social-psychological research on caste legitimacy, with an emphasis on four topics of special relevance: commitments to hierarchy, racialized suspicion, punishment, and deservingness. In Part III, we propose a research agenda for bridging the divide we identify and highlight promising emerging scholarship and its implications for policy reform.

I. The Theory and Social Psychology of Procedural Justice

The idea that procedural fairness is necessary to the legitimacy of democratic forms of government is an old one in European political theory. As early modern European empires rode the rising tide of global, imperial capitalism, European philosophers in the seventeenth andeighteenth centuries developed political taxonomies describing the relationships between rulers and the ruled, including the possibility that whole societies could, in the right conditions, rule “themselves.”²⁰ While social-contract theory argued that what made any form of government legitimate to the governed was the protection of their natural rights, it was Montesquieu who suggested that people actually learned from the character of government itself (not just the outcomes it produced). In a republican form of government, he argued, the people must learn to love the laws that govern them.²¹ The reasonable and predictable pursuit of justice through law, he argued, was essential for that love.²²

By the early twentieth century, Max Weber fully realized a theory of proceduralism in democracy: “Obedience is not owed to anybody personally but to enacted rules and regulations which specify to whom and to what rule people owe obedience. The person in authority, too, obeys a rule when giving an order, namely ‘the law,’ or ‘rules and regulations’ which represent abstract norms,” he wrote. “The person in command typically is the ‘superior’ within a functionally defined ‘competency’ or ‘jurisdiction,’ and his right to govern is legitimized by enactment.”²³ In 1971, John Rawls’s publication of A Theory of Justice marked another watershed.²⁴ Rawls theorized that fairness is something best generated “behind a veil of ignorance,” whereby people agree on principles of justice before knowing what their later position will be in the application of those principles.²⁵

Importantly, over the centuries these and other (white) theorists also developed mechanisms for rationalizing the unprecedented scale of exploitation, suffering, and illegitimate governance that fueled the material enrichment and democratization of European societies and their white colonies. As Charles Mills has argued, the European “social contract” tradition in Western philosophy that so deeply shaped the legal system of the United States implicitly and explicitly relegated colonized and nonwhite people to subpersonhood outside the civic ken, reflecting a durable racial contract.²⁶ Locke, Hume, Kant, Jefferson, and other Enlightenment thinkers foundational to American political philosophy explored racial taxonomic justifications for white settlerism and enslavement.²⁷ These justifications emphasized the innate or cultural inferiority of nonwhites and contributed to the rise of scientific racism and social Darwinism over the long nineteenth century.²⁸

By the early twentieth century, prominent theories of liberal democracy conceptually erased or simply ignored the antidemocratic political and economic systems that they were built upon. In his celebrated works on political theory, for example, John Dewey wrote nothing accounting for American’s racial caste system in his theory of democracy as a “mode of associated living.”²⁹ His contemporary, Max Weber, moved away from biological racism over the course of his career, but continued to hold ambiguous, neoracist beliefs with regard to culture, and adopted a race-blind stance in his account of the workings of liberal democracy rooted in proceduralism.³⁰ John Rawls’s monumental work on justice not only imagined an ideal model free from race, but also spawned decades of philosophical churn that, itself, also ignored race, racism, and racial justice.³¹

The study of justice within psychology is relatively recent.³² Tom R. Tyler has described this engagement as a series of “waves,” beginning in the 1950s and 1960s with the study of relative deprivation and distributive justice.³³ These studies, often facilitated by large federal grants, reflected postwar efforts to understand and avoid authoritarianism.³⁴ They posed questions such as: What led individuals to become politically dissatisfied or to identify with authoritarian regimes? Why, and under what circumstances, did they accept or reject inequality? Justice in these studies assumed an instrumentalist orientation and focused primarily on objective and subjective outcomes.³⁵ Rawls’s innovation—the development of an idealized mechanism for maximizing fairness by randomizing future position—should be understood against the background of this research.

Procedural justice as a distinct field within psychology was itself a reaction to Rawls, and like Rawls’s work was uninterested, at least initially, in America’s distinctive racial caste system.³⁶ In 1974 Thibaut, Walker, LaTour, and Holden published Procedural Justice as Fairness, a provocative study that presented itself as a significant corrective to A Theory of Justice.³⁷ Rather than defining justice in terms of the anticipation of outcomes from third-party adjudication, Thibaut and colleagues focused instead on the formal procedures that people used to determine outcomes.³⁸ They found in their experiments that certain types of procedures produced higher or lower perceptions of fairness, depending upon whether people knew their position or not.³⁹ This focus on procedure-as-justice initiated a new branch of social psychology, which Thibaut and Walker sketched out the following year in their seminal book, Procedural Justice: A Psychological Analysis (1975).⁴⁰

The social-psychological study of procedural justice has exploded since Thibaut and Walker’s seminal work. E. Allan Lind and Tom R. Tyler advanced the project significantly in their 1988 book, The Social Psychology of Procedural Justice.⁴¹ Their work shifted the theoretical basis of an individual’s interest in the value of process from controlling outcomes in contexts of uncertainty, an instrumental approach, to an identity hypothesis emphasizing social relations within a group, which emphasizes relational concerns.⁴² According to this “Group Value Model,” people look to key aspects of process to gather information about how authorities view them and the group or groups to which they belong.⁴³ The Group Value Model stands in stark contrast to Thibaut and Walker’s interpretation of people’s interest in process. Lind and Tyler conclude that relational considerations generally matter more to people than instrumental ones as they reach conclusions about the fairness of authorities.⁴⁴

Procedural justice researchers have identified four factors that people focus on when assessing the fairness of authorities.⁴⁵ The first two factors concern aspects of treatment that people perceive as fair. People care about voice and participation, and they also care about being treated with dignity and respect.⁴⁶ People report higher levels of satisfaction in encounters with authorities when they have an opportunity to explain their situation and perspective on that situation, and this is true even when people are aware that their participation will not impact the outcome.⁴⁷ In addition to being taken seriously and listened to, people also desire to be treated with dignity, with respect for their rights, and with politeness. The next two factors pertain to the fairness of decisionmaking by authorities. That is, people tend to evaluate the decisionmaking process of authorities, and they also attempt to evaluate the trustworthiness of the decisionmakers’ motives.⁴⁸ With respect to decisionmaking fairness, people look to indicia of decisionmaker neutrality, objectivity and factuality of decisionmaking, consistency in decisionmaking, and transparency.⁴⁹ For example, it is important that, in an interaction with a member of the public, a legal authority takes the time to explain that the motivations of the authorities are sincere, benevolent, and well-intentioned. Basically, members of the public want to believe that the authority they are dealing with believes that they count.

Writing alone, Tyler connected the procedural justice research to public perceptions of legitimacy of authorities. In Why People Obey the Law, Tyler showed empirically that public reliance on procedural justice factors could lead to legal compliance.⁵⁰ Later work expanded the conception of legitimacy beyond compliance to include cooperation and engagement.⁵¹ As a general matter, research on procedural justice has grown to explore its salience as a critical aspect of how people come to conclusions about the fairness of both legal authorities and the law itself. For example, while Tyler’s initial model was developed through a panel survey of Chicagoans, further work has established the importance of procedural-justice-based fairness assessments to individuals residing in twenty-six countries with very different types of government structures,⁵² to minority group members reporting negative experiences with police and courts,⁵³ and even to individuals with serious criminal legal system exposure.⁵⁴

We think it is fair to say there is a broad, solid research basis for the proposition that there is widespread consensus among different groups and across cultures about what constitutes fairness—at least insofar as people consider their views of the fairness of authorities along the lines of procedural justice theory.⁵⁵ But, the research on racial bias referenced at the outset of this piece raises an important question: Despite the fact that different groups appear to agree about what constitutes fairness in how authorities treat them, do we also find that members of groups, in coming to conclusions about the fairness and legitimacy of authorities, want those authorities to treat members of groups to which they do not belong consistent with those same principles? In short, do people want procedural justice for others?⁵⁶

There’s a lot riding on this question. Recently, policy circles have emphasized the possibility that procedurally just policies, strategies, and training might address or ameliorate negative interactions some groups, particularly groups of color, have with legal authorities such as police.⁵⁷ One would hope that the research base concerning the relationship between procedural justice as fairness and intergroup relations would be extensive. Surprisingly, though, the research here is not very robust. A handful of studies by Tyler and various colleagues explore the extent to which a commitment to justice across groups will predict that when people see police injustice, whether directed at members of their own group or at members of other groups, they will view police as less legitimate and will be therefore less likely to cooperate with them. Studies in two different contexts, racial profiling of ordinary crime and policing focused on terrorism, support this prediction.⁵⁸

A study by Huo outside of the legal-authority context is also relevant.⁵⁹ Huo conducted a lab study among university students to understand when, why, and in what ways people deny certain kinds of goods to members of groups to which they themselves did not belong.⁶⁰ Participants had to make decisions about allocating three types of goods to groups described as having values that either aligned or conflicted with their group identity.⁶¹ The first group was an honor society-type group open to students interested in intellectual exchange and especially encouraging minority group members’ and women’s participation. A second group was described as politically oriented, discouraging academic focus, and dedicated to promoting white supremacy. The goods in question were monetary support (economic), an opportunity to distribute flyers in support of one’s position (procedural “voice”), and an assessment of how authorities should treat members of different groups (relational).⁶² Huo found that identity concerns did drive instrumental judgments: participants discriminated against the unlike group and withheld monetary support.⁶³ Yet Huo also found that study participants were least likely to withhold relational goods from the out-group.⁶⁴ With respect to relational goods, there was almost no distinction between the respondents’ assessment of the kind of treatment authorities should afford the two different target groups.⁶⁵ Additionally, respondents were reluctant to deny procedural goods such as “voice” to the out-group, but the effect was not quite as strong as it was concerning relational goods.⁶⁶

This study is the strongest piece of evidence that we could find demonstrating that members of one group support relational goods for members of another group—even a group representing values for which respondents profoundly disagree, which in this case is white supremacy. There are limitations, however, to applying this study to our notion of caste legitimacy. We do not know whether the way participants considered the misalignment between themselves and the fictional white supremacist group in the study operates in the same way that racial identity in the real world does. There are further questions about whether the participants are representative and their behaviors generalizable. The fact that the study is singular also gives us pause. These are all issues of external validity.

To review, the caste thesis of legitimacy implies little to no identification of in-groups with out-groups. And caste legitimacy in U.S. law has proceeded in explicit racial terms, as we noted at the outset of this Essay. Applying these ideas to Huo’s work, we might expect respondents in Huo’s experiment to be less willing to extend relational and procedural goods to a Black Nationalist group as opposed to the fictional target group in the study. On the other hand, the small group of studies conducted by Tyler in the policing context suggests that members of majority groups extend their commitments to justice to include procedural justice for minority groups. To us, the psychological research pertaining to how people think about race is critical to adjudicate an answer to the external validity question given the clear paucity of research. We turn to that in the next section.

II. The Social Psychology of Racism

There has been a large, diverse body of resistance thought in oral tradition, literature, art, music, and other forms of human expression countering the racist epistemologies and practices of imperialist European liberalism. Much of this work has been carried out in the United States, by people enslaved or encasted by American law and custom. Frederick Douglass, the best-known orator of the nineteenth century, exposed the fallacies of scientific racism, the myth of American exceptionalism, and most of all, the paradox of chattel slavery within an allegedly Christian republic.⁶⁷ At the same time, Sojourner Truth, and later, Anna Julia Cooper, articulated theories of intersectionality, reframing gender and race as co-constructing categories of domination that also belied liberal republican claims to benevolence and fairness.⁶⁸ Through music, literature, and activism, Sioux Zitkála-Šá explored the duality of liberalism and cultural genocide in her own Native American identity formation.⁶⁹ And by the mid-twentieth century, Black radical scholars including Claudia Jones, Oliver C. Cox, and others exposed the connections between capitalism and racism, while Franz Fanon and others critiqued racial coloniality.⁷⁰

In terms of the early twentieth century development of the theory and social science of racism, W.E.B. Du Bois built the most robust body of academic scholarship, deploying social science and historical analysis on the contours of the American caste system, simultaneously disproving and explaining the workings of racial caste within an ostensibly democratic republic.⁷¹ His work proposed two important psychological theories: first, the two-ness of civic experience as a person labeled as Black, which he described as life “[b]eneath the veil;”⁷² and second, whiteness as a kind of “wage” or psychological benefit to people labeled as white, for the maintenance of which they will support policies that run counter to their own economic self-interest.⁷³ Over the course of his life and scholarship, Du Bois’s analysis of American caste shifted from an optimistic account of the duality of American caste and American democracy to a pessimistic view that the latter would never overcome the former (a shift paralleled in the work and life of Derrick Bell).⁷⁴ Critical Race Theory, so-named, developed among law school students and faculty in the 1970s and 1980s in response to the failures of the Brown decision and other liberal legal reforms to successfully dismantle racial caste after the Supreme Court repudiated its place in formal law.⁷⁵ In essence, the question we pose in this Essay reflects that enduring tension in these traditions: that American political culture is, at its core, a struggle between democratic ideals and the maintenance of white settlerist privilege.⁷⁶ Though we agree with Black feminist tradition that oppression operates as a matrix of intersectional identities, for reasons of economy, our focus in this Essay is the social psychology of racism.⁷⁷

Throughout the twentieth century the branch of social psychology called “intergroup relations” engaged race and racism directly. In the early decades, American psychologists played an ignominious managerial role in white supremacy, often in relation to white domination of education for Black people. Pseudoscientific claims of race psychology reinforced racial conceptions of intelligence, argued that segregation by race was natural, and buttressed a softer social Darwinism that framed European culture and society as the apex of human social evolution.⁷⁸ During the interwar years, however, overtly racist psychology declined in the face of withering scientific critique, alarm at European fascism, and a growing willingness among psychologists to engage in public activism.⁷⁹ For example, Otto Klineberg’s research on Native American and Black student achievement exploded myths of race-based intelligence,⁸⁰ and John Dollard exposed the benefits of everyday racism to whites in his psycho-ethnography of a southern town.⁸¹ In the decades after the Second World War, psychological research on the intersection of individuals and antidemocratic political systems included significant work on racial prejudice—most conspicuously Gordon Allport’s The Nature of Prejudice (1954).⁸² While much more sophisticated than the early race-based pseudoscience, the second generation researchers’ conception of prejudice was based on overly optimistic assumptions regarding the inevitable triumph of democratic values, the ascendancy of research-informed policy, and the desire of out-groups to assimilate into the dominant social paradigm.⁸³ Even during the civil rights era, the dominant view among intergroup psychology researchers was that legal reform would inevitably create the necessary social context for the steady erosion of racism.⁸⁴

More recently the psychological study of racial prejudice has exploded. This research comprises many hundreds of studies using multiple methods, generating endogenous subfields within psychology and also raising profound challenges to exogenous fields such as law, philosophy, education, criminal justice, and more.⁸⁵ Much of this work is framed by the paradox that, since the 1970s, overt racism has declined in public discourse of legitimate governance (the fascist politics of the Trump-era Republican Party notwithstanding), yet racial inequality endures and, by some measures, has worsened.⁸⁶ A critical observation in the work on the psychology of racism is its recognition of the centrality of “sociostructural forces in creating and maintaining racial bias and biased outcomes.”⁸⁷ We see race, at its core, as a social construction that serves the psychological purpose of justifying inequality. In the twentieth century legal transition from white democracy to color-blind democracy, Americans rewrote laws to eliminate overt racialism in text but did not address the cultures or structures of caste that the law had already aided and abetted for centuries and now maintains in its ordinary operation.⁸⁸

The study of justice in a liberal democracy presupposes universal commitments to its tenets; but the social psychology of racism presents significant challenges to this supposition. We briefly highlight four areas of social-psychological theory and topical research that raise questions about whether the group value thesis of procedural justice will extend across racial groups and, in particular, extend from white people to Black people. The thrust of this work potentially impairs the proposed value of policy innovations to promote procedural justice and even democratic approaches generally, which has important implications for how Americans think about justice writ large.

A. Commitment to Hierarchy

In the last thirty years, researchers have developed two theories in particular that raise questions about broad commitments to democratic legitimacy. The first, Social Dominance Theory (SDT), is a comprehensive theory that seeks to explain how group-based social hierarchies (not exclusively race-based ones) are maintained through individual discrimination, institutional discrimination, and behavioral asymmetry that is largely guided by ideology.⁸⁹ The theory posits that members of socially dominant groups will legitimize and maintain their dominance by supporting policies that reinforce their group advantage, seeking out social roles where they can police the interests of their group status (including, importantly, working as police officers),⁹⁰ and engaging in overt and discrete acts of discrimination against nondominant groups.

Ideologies are central to SDT, and researchers explore how dominance is supported by legitimizing myths, such as race, gender, and meritocracy. Empirically, researchers have developed scales of Social Dominance Orientation (SDO) to measure the extent to which individuals prefer hierarchical versus egalitarian group relations. These scales are highly predictive of a person’s identification with a hierarchical group position, their commitments to legitimizing myths, and their judgments of the undeservingness of other groups they view as threatening to their position.⁹¹

In contrast to SDT, Aversive Racism Theory (ART) seeks to account for the discriminatory behavior of individuals who hold egalitarian values. This body of research demonstrates that white people who identify as socially liberal and do not hold explicitly racist views can nevertheless score high in implicit bias measures and behave in ways that reinforce the racist systems they oppose.⁹² As Dovidio and Gaertner explain, aversive racism is not driven by the desire to dominate Black people, but by subconscious negative feelings toward them (such as discomfort, disgust, and fear) that result in preferential treatment toward whites.⁹³ Aversive racism is most salient in situations where rules and procedures are not clear, ⁹⁴ suggesting an interesting and important relationship between ART and theories of process-based fairness.

Unlike older theories of racism, both Social Dominance Theory and Racial Aversion Theory posit hierarchy-maintaining behaviors as natural, not pathological.⁹⁵ Both social dominance and aversive racism result in the maintenance of white supremacy. But it’s not clear what their relationship is to conscious and subconscious commitments to procedural justice. An interesting study by De Cremer, Cornelis, and Van Hiel demonstrates that people with high SDO scores value voice, one component of procedural fairness, in their interactions with authorities more than those with lower scores.⁹⁶ This study notwithstanding, there is not a research base exploring the intersections of SDT, ART, and procedural justice. We think the field is ripe for this kind of intersecting exploration— especially as it pertains to policy-relevant studies.

We next review studies in policy-relevant domains that we think raise critical questions about the power of advancing procedural justice strategies and policy in criminal legal reform by demonstrating the persistence of conscious and unconscious forms of caste-maintenance. To us, the question is the extent to which these conscious and unconscious commitments to hierarchy can be overcome by democratic legitimacy.

B. Suspicion and Danger

There is nearly eighty years of psychological research documenting stereotypical association in the United States between Black people and criminality.⁹⁷ In a classic 1980 study, Sagar and Schofield found that Black and white male sixth graders interpreted Black and white stimulus performers engaged in ambiguous aggressive behavior differently—both sets of children rated the Black actors’ actions as being more mean and aggressive than those of the white actors.⁹⁸ In a 2017 study, Wilson and colleagues found that white people perceive images of young Black men as being physically larger and more dangerous than those of young white men, and consequently favored greater hypothetical use of force against them (the effect was less pronounced among Black participants than white ones, especially in perceptions of dangerousness).⁹⁹ The advent of subconscious processing research has taken this work a step further: even thinking about people who are Black can lead to thoughts of crime and to split-second misapprehensions of their behavior as more suspicious and more dangerous.¹⁰⁰ Likewise, thinking about crime and violence (primed subconsciously with an image of a gun) can lead to thinking about Black people.¹⁰¹

Insofar as procedural justice relies upon the notion that those who interact are reliably ordered individuals, perceptions of a group as being inherently disordered also raise potential problems. In their study of 500 city block groups in Chicago, for example, Sampson and Raudenbush found that all racial groups associated the presence of racial minorities with a form of disorder, irrespective of observable physical characteristics of the place.¹⁰² Moreover, these same forms of mistrust and suspicion can be activated politically through the use “dog whistles” that tap into white racial fears—of Black Americans, of Spanish-speaking immigrants, or of Muslims, for example. While the use of symbolic political language has occurred across national contexts and across forms of difference, its use in activating caste legitimacy has been a key feature of American politics since the late 1960s.¹⁰³

C. Punishment

In a broad study of intergroup dynamics, Newheiser, Sawaoka, and Dovidio found that research participants who believed out-group members were highly cohesive and like-minded tended to be more suspicious of those out-group members’ moral commitments and, as a result, recommended harsher retributive justice for culpable individual members of that group.¹⁰⁴ In a society where many white people live largely segregated lives and hold disproportionate power, this form of generalized out-group suspicion and proclivity to punish can be especially costly to racial minorities, and this study is hardly singular.¹⁰⁵ Many studies find that white Americans support harsher punishment for Black people relative to white people. Using a nationally representative sample including 735 white Americans, for example, Rattan and colleagues found that participants judged adolescents who are Black to be more culpable and favored sentencing them more harshly than white adolescents.¹⁰⁶ Holding facts of the cases constant, Glaser, Martin and Kahn found that juries were more likely to find Black defendants guilty for the same crime when the death penalty was an option; that effect was the opposite for white defendants.¹⁰⁷ Moreover, these associations with blackness are not binary but colorist, so that even among people identified as “Black,” darker skin tone and certain facial features attract increased punishment.¹⁰⁸ In a study of capital cases in Philadelphia, Eberhardt and colleagues examined photos of defendants and found that in cases in which the victim was white, the extent to which the defendant appeared more stereotypically Black predicted imposition of the death penalty.¹⁰⁹ In two linked studies, Hetey and Eberhardt found that the white voters they interviewed in New York and San Francisco expressed more support for harsh laws when presented with visual or statistical evidence that the population affected by the policies was more “Black.”¹¹⁰

D. Deservingness

At the same time that Thibaut and Walker were developing a distinctive field of procedural justice, psychologists who studied the effects of relative deprivation on fairness judgments theorized “deservingness” as a key modulating mechanism. In her classic study of working women and job discrimination, for example, Faye Crosby found that despite objective evidence of extreme wage discrimination, her participants modulated their judgments through a sense of that they thought they deserved as members of their in-group.¹¹¹ The effect of deservingness has been found in multiple justice contexts in which people evaluate injustice for others, too. For example, people have been found to justify injustice to other individuals by activating out-of-context evidence—making appeals to “ultimate justice” in the long term, even if the particular result is unjust, or “immanent justice,” by which a current injustice is justified based on causally unrelated perceived past misdeeds.¹¹²

Deservingness may also affect fairness judgments about whole groups. In a 2019 Daedalus essay, Bloemraad, Kymlicka, Lamont, and Son Hing developed a multidisciplinary account for why, across Western societies generally, increasing diversity and inclusivity in the social sphere in the last fifty years has been accompanied by a curtailment of the social rights of citizenship for the poor and racialized minorities.¹¹³ In short, they find greater diversity increases the weaponization of deservingness arguments against distributive justice by race and social class.¹¹⁴

Does deservingness affect people’s evaluation of the legitimacy of authorities based on how authorities treat racial others? While we do not know, it seems plausible and would lend credence to the idea that caste legitimacy could continue to play a significant role in how Americans think about procedural justice.

III. Where Do We Go from Here? Research and Reform

Much of the research we have reviewed in this Essay is either directed toward or designed to influence policy in ways that mute or disable caste legitimacy. This is true whether the research is based in procedural justice or focused on problems of intergroup relations pertinent to prejudice and discrimination. Our view is that reading the lines of research together, along with a thoroughgoing commitment to theory, can better inform policy instruments, strategy, and proposals in ways that can make them more successful. One contribution that scholars of procedural justice have made to criminal legal system reform is to question the theoretical premises of its operation, whether stated or unstated.¹¹⁵ Like the early social psychologists referenced in this Essay who believed that it was possible to overcome the social dynamics that lead to race discrimination through law and policy, we still are hopeful that law, policy and institutions properly constructed can advance the project of democratic legitimacy. But those interventions must be attentive to the dynamics that impede them. Our primary goal in this last section is to raise some questions for a research agenda explicitly targeted at the disjuncture we have identified and discuss some implications of such an agenda. We then conclude with an old saw.

A. Asking the Right Questions

There is one obvious first step in carrying out research that is attentive to potential impact of both American traditions of legitimacy. Procedural justice researchers must account in their models for the ways in which the persistent structure of racial injustice specifically limits the extent we can expect legal authorities who are committed to procedural justice tenets to carry them out. While it is true that procedural justice is primarily an intragroup theory, researchers could be more sensitive to intergroup concerns when designing studies, especially if their goal is to be more policy-relevant. The operation of criminal legal systems, for example, necessarily implicates how in- and out-groups view one another, and it is perhaps more important in these contexts to consider how social-dominance concerns affect results.¹¹⁶ As an example, procedural justice researchers could test explicitly the extent to which research subjects from one group, and especially a group considered dominant in accord with Social Dominance Theory, are willing to accord procedural justice “goods” to down group members along the lines of the hypothetical twist on Yuen Huo’s Berkeley study we suggested above.

In turn, researchers focusing primarily on intergroup relations might consider whether and how being attentive to process concerns and fairness ameliorates the negative consequences of bias. In a series of studies, Goff and colleagues have demonstrated that the fear of being perceived as racist actually provokes the use of force among police officers.¹¹⁷ The authors argue that self-legitimacy, where legitimacy is grounded in ideas akin to procedural justice, is an important predictor of resolving interactions with little to no force, because “officers must wield power in a way that mirrors their self-beliefs about their position in society as fair and equitable agents of the rule of law.”¹¹⁸ Consistent with that idea, the question was whether self-legitimacy would blunt the impact of stereotype threat on use of force. The authors found that it did, but only partially.¹¹⁹ More work in this vein would be helpful to test the power of procedural-justice-based interventions, strategy, and policy.

Additionally, there is a need more generally to assess the explanatory power of both racial bias and procedural justice contextually and situationally. Much psychological research is carried out in laboratory settings, and this makes it difficult to assess the extent to which the important dynamics these researchers uncover operate in the day-to-day. With respect to this last point, we take Jill Swencionis and Philip Atiba Goff’s framework for understanding racial bias in policing as a potential model, which they described in a recent article.¹²⁰ There the authors point out that both personal and situational factors present risks of discrimination in policing. They helpfully array such psychological routes to discriminations such as social dominance orientation, aversive racism, and Black-crime associations (to name three categories discussed in this Essay) against common policing situations that present potential for officers to enact discrimination, such as high-discretion decisionmaking, cognitive demand, and whether the officer is a novice.¹²¹ To this two-by-two model researchers could add a third axis, Z, and assess how procedural justice impacts and, by hypothesis, potentially ameliorates the impact of a potential route of discrimination.

B. Thinking About Method

Much of the psychological research on prejudice, whether social dominance or implicit bias, relies on primes. Priming is a form of stimulus that focuses the brain in ways that shape the types of associations it makes (ideally in the same modality, such as visual primes in advance of visual information). In a sense, priming is a form of manipulation—shaping both the information we select for noticing and the ways we interpret that information.¹²² Some of the work on the psychology of race we have reviewed here uses photos of people or other images to prime research subjects.¹²³ Dovidio and colleagues primed subjects through auditory means.¹²⁴ While we have seen that priming can work to activate prejudice, it stands to reason that priming also can be deployed to counter it. We think it is important to explore this possibility. For example, there are studies of the relationship between forgiveness and justice in which researchers attempt to assess whether thinking about justice—both procedural and distributional—advances or impedes forgiveness.¹²⁵ But we have not uncovered research relying upon procedural justice primes to assess how it might mitigate the risk of caste legitimizing behavior. Following this line of thinking, we hypothesize that a targeted procedural justice prime could activate egalitarian commitments that presumably are already quite strong in some groups, as the research on aversive racism indicates. The question would then be whether priming procedural justice among members of this group could overcome strong implicit bias. Moreover, utilizing primes in this way could also facilitate more contextual assessment of both theories of justice and discrimination in conjunction, as Goff and Swencionis urge. Policy instruments, we think, could potentially operate as “primes” themselves.

C. Thinking More About Time

Because laboratory research focuses on understanding mechanisms, one perennial criticism pertains to whether these laboratory findings are externally valid—namely, can lab findings be replicated in the real world. Researchers seek to address these concerns by relying on subjects of concern in lab settings or by taking experiments out of the lab to large scale survey samples involving heterogeneous groups as opposed to just university undergraduates. Even these laudable approaches have weaknesses. The many apparatuses necessary to disable the pernicious effects of naturalized approaches to maintaining hierarchy will require intervention and work at the individual, institutional, and societal levels, and the effects of these various strategies will impact individuals in different ways. This might just be a complicated way of recognizing that time is a relevant dimension for thinking about how the psychology of procedural justice and racial bias interact.

Consider some evidence concerning this last point. At the individual level there is, for example, evidence that training people in authority, especially criminal justice workers, to recognize and manage their trigger points in their professional interactions holds promise for increasing the quality of interactions with civilians of color.¹²⁶ And on the macro, sociological, and political levels, developing more sophisticated understandings of the activating mechanisms for democratic or caste legitimacy potentially assists our ability to counter inflammatory political rhetoric. Important questions with respect to both of these strategies pertain to time. How much training? How long does the training last? How powerful and durable are our methods to engage countervailing pressures in support of caste legitimacy? All of these questions implicate the reality that whatever we do in this arena, we will not see results overnight. We cannot reverse the structures that reinforce the negative consequences of intergroup relations on a dime.

Conclusion

The individual and structural dynamics that have built both the cultural determinants of our preferences for process-based fairness that support democratic legitimacy and also laid the foundations of caste legitimacy are long-standing, durable, invisible, and yet pervasive. Recognition of this dynamic requires persistence and perseverance. Any strategy we adopt to fully entrench an ideal version of justice will require a list of approaches writ large and writ small—all of which must accrete over time. There is no other path forward. A popular parable attributed to the Cherokee and Lenape oral traditions captures the nature of this work. In every person’s heart, the story goes, there is an ongoing fight between two wolves. One wolf is evil, representing all the worst attributes of humanity: rage, arrogance, self-pity, cruelty, deceitfulness, fear, and greed. The other is good, representing kindness, humility, peace, joy, truthfulness, hope, and generosity. Who wins the fight in our hearts? The wolf we feed.

* Professor of Education and (by courtesy) History, Rutgers University.

** Walton Hale Hamilton Professor and Co-Faculty Director of the Justice Collaboratory, Yale Law School. The authors would like to thank Phillip Atiba Goff, Paul Gowder, Elizabeth Hinton, Leah Mirakhor, Jennifer Richeson, Tom Tyler, and participants in a workshop on the Reckoning and Reformation journal symposium. All errors are our own.