A Theory of Justice

In Theory, Rawls’s theory of justice and his account of moral psychology run on overlapping tracks. According to Rawls, the theory of justice can “generate its own supports” because it is aligned with “the principles of moral psychology.”⁶² He aimed to indicate the major steps whereby a person could acquire both an understanding of and attachment to the principles of justice. His approach to moral psychology is stylized but, importantly, it is not strictly an exercise in ideal theory; in fact, it is characterized by a strange mixture of ideal and “realistic” psychology. Theory’s aim is to indicate an attachment to the principles of justice for a person growing up within an idealized, well-ordered society, yet Rawls states that the psychological account of moral learning had to be “true and in accordance with existing knowledge.”⁶³ The laws of moral psychology cannot be purely the product of an idealized social order because they are also the unavoidable ground for social relations at their most basic level.⁶⁴ In other words, without a realistic moral psychology, Rawls implies, the theory of justice cannot get off the ground. Rawls, however, is inconsistent on this point, writing that the principles of moral psychology refer to “an institutional setting as being just” and that these principles reflect deep psychological realities without which “our nature would be very different and fruitful social cooperation fragile if not impossible.”⁶⁵ This inconsistency, as we will see shortly, allows Rawls an ideal theory escape hatch, when the resources of his realistic moral psychology run out.

Rawls’s theory of moral psychology includes three successive stages, which he refers to as the moralities of authority, association, and principles. In developing this account, Rawls leans on two broad paradigms of moral learning. The first, which he associates with Freud, is a deficiency model where learning results from the supplying of missing moral motives. The second tradition, which Rawls associates with Rousseau, Kant, and Piaget, sees moral learning less as a form of imposition than as the free development of innate moral capacities, including the “innate susceptibility to the pleasures of fellow feeling and self-mastery.”⁶⁶ Although Rawls claims to combine these approaches, he clearly has deeper debts to the latter model. The moralities of authority and association, for instance, are reflective of an innate-capacities approach. During the stage of authority-morality, the child—while driven by excessive and unruly desires—is primarily motivated by a reparative impulse to atone for any transgressions and to “seek reconciliation” with his beloved authority figures.⁶⁷ For Rawls, these reparative gestures are at the root of moral sense. They are innate capacities cultivated through repeated experiences of love, trust, and affection; they reflect the “deep psychological fact” of reciprocity, or the desire to give back for what we have received.

Authority-morality flows naturally into the stage of associational morality. As the child begins to associate and identify with a wider range of others, he or she develops a sense for fellowship and the virtues and ideals of being “a good student and classmate … a good sport and companion.”⁶⁸ Social cooperation and competition at the associational level instills—through reliance on the deep psychological fact of reciprocity—a feeling of commitment to a larger group. Identification with the group also creates the possibility of associational guilt, when the individual “fails to do his part.”⁶⁹ The expansion of commitment to an ever-widening circle of associates sets the stage for the next and highest step in moral development—the morality of principles. Rawls argues that the taking on of a succession of “more demanding roles” leads to a gradual expansion of perspective-taking ability. However, unlike Jürgen Habermas, who bases his dialogic theory of ethics on a similar expansion of perspective-taking ability, Rawls emphasizes the subjective recognition of being the beneficiary “of an established and enduring just institution” as the basis of a passionate attachment to being “a just person.”⁷⁰ Here the abstract principle of reciprocity returns as the unshakable ground of Rawls’s argument. Just as the child received and gave back love and fellowship in the stages of authority- and association-morality, the adult feels tugged at by a love of mankind that is “continuous … with the sense of justice.”⁷¹ A sense of and passionate commitment to justice must be the result of experiencing justice within a well-ordered society.⁷²

Before showing how this emphasis creates problems for Rawls’s account, I note at this point how Rawls explicitly sets his ontogenetic account of justice against Freud’s. For Freud, the sense of justice is a reaction formation rooted in envy. Because envy cannot be fully expressed, lest personal and social relations deteriorate to the point of nothingness, it is transformed into a social feeling that insists on equality.⁷³ On Rawls’s account, the insistence on equality within justice as fairness is the progeny of a marriage between reciprocity and anthrophilia, but for Freud it springs from the sublimation of aggression against others into a feeling of group spirit, guided by a vengeful superego that twists our envy and resentment of others back into a critical voice of self-control. The superego is a harsh and unimpeachable authority because it must hold back the malevolent desire to spoil others’ access to the good. If this desire were liberated, all hell would break loose.

Rawls acknowledges Freud’s hypothesis about the ontogenesis of justice but sets out to show that envy is not a problem for the theory of justice as fairness. According to Rawls, Freud’s account suffers from a serious “defect.” Namely, it fails to distinguish between what Rawls refers to as “emulative envy” and “rancor.”⁷⁴ Emulative envy, which is “benign,” arises when we aspire to achieve the good virtues or qualities that others display. It is a “socially beneficial” form of striving and in no way leads us to a twisted desire to deny these goods to others. Rancor, on the other hand, is “what emulative envy may become under certain conditions of defeat and a sense of failure.”⁷⁵ Rancor is a “collectively disadvantageous” form of misanthropy, and it leads to a willingness to deprive others of their goods even if we have to sacrifice some of our own.⁷⁶ Yet because rancor is a product of bad circumstances, in well-ordered societies it would not find hospitable soil. Hence Rawls, sliding into the mode of ideal theory, can set socially destructive envy to the side. The reasoners within the original position can bargain without rancor, and the principles of justice remain untouched by it.

This argument against rancor, however, shows clearly that Rawls’s realistic moral psychology and his theory of justice are no longer running together. Our moral sense is a “natural” and “fundamental” attitude and consists of incontrovertible psychological “facts.”⁷⁷ Yet to dispel the strongest anxieties surrounding the viability of this account, Rawls retreats to the image of idealized, well-ordered societies. Envy is a problem within damaged life, but it is not a problem for ideal theory. And this slippage into ideal theory with regards to the problem of envy betrays further inadequacies in Rawls’s account of the ontogenesis of the morality of principles. Namely, unlike the moralities of authority and association, Rawls cannot explain the development of the principled moral stance without recourse to an idealized social order. Reciprocity cannot explain the morality of principles without a vicious form of circularity: If justice is returned for justice received, then where did the “first” justice originate? And while Rawls also claims an innate capacity for anthrophilia, it is hard to see how the principle of love can be extended to the universal human community without some bridging principle. Something essential seems to be missing, something that for Freud was provided by the superego: an agency that turns back our aggressive impulses in ways that make social life possible.⁷⁸

What, then, does Rawls put into the place of this superordinate agency? On my reading, the original position—Rawls’s thought experiment by which the principles of justice are laid down—fulfills this function. The original position is a construct of ideal theory, but unlike the sense of justice resulting from an upbringing in a well-ordered society, the original position is available to reasoners within damaged life. As Rawls puts it, the original position is a “perspective” or gaze that is accessible “at any time.”⁷⁹ This perspective, moreover, although it is designed as a device for reasoning, cannot be reasoned or bargained with itself. The strict rules that structure the original situation make its edicts unequivocal and final. Regardless of how many times we may enter the position, “the same principles are always chosen.”⁸⁰ The univocal, final court of appeal provides conclusive reasons that can then become a means of self- and social-critique.⁸¹ By turning back our penchants for partiality, the original position is the means by which those of us living in unjust circumstances could nevertheless obtain “purity of heart.”⁸²

Rawls is obviously hostile toward Freud’s pessimistic ontogenetic account of justice, but has Rawls really done away with the harsh superego, or has he merely refurbished it? The fact that Rawls ends Theory on the note (however hopeful) of “purity” demonstrates clearly that his project is pressurized by paranoid-schizoid anxieties.⁸³ Fully admitting that the resources within a realistic theory of moral psychology were insufficient to secure the development of social justice, Rawls invokes a kind of Isles of the Blessed from which abstract reasoners could deduce the correct principles of action. Rawls seemed to intuit, in Kleinian fashion, the need for a situation whereby persecutory or disintegrative anxieties could be worked through. This is why, as C. Fred Alford sees it, there are “sound psychological reasons” for constructing the original position in the way it is, because it could allow the “self to begin to think about justice in more abstract … [and] other regarding terms.”⁸⁴ Yet Rawls’s broader social theory does not carry through on these intuitions. He does not look for actual social practices or spaces that might assuage or work through persecutory dread, but instead turns to an always-available superego to overwhelm fantasies of omnipotence rooted in narcissistic partiality. The location of an unimpeachable procedure, however, actually intensifies the very pressures they are meant to alleviate. Because Rawls could only conceptualize the problem of justice “as a problem whose solution is … universal, and categorical,” his theory remains stuck within the paranoid-schizoid position. As Rawls puts it, the original position must be narrowly construed and geared toward unanimity, lest it “lead to endless wrangling … setting the stage for the nastiest individuals getting more than their share.”⁸⁵ The search for purity drives out the dangerous elements within the self but forestalls the recognition of these dangerous elements and the broader ambivalence of social life. Yet it is this work of recognition that makes possible political judgment and “depressive” forms of agency in which we can share power with others.

This is the secret yet strong source of connection between Rawls and the Periclean mode of mourning. Like the funeral oration, Rawls acknowledges envy and “nastiness” but dissolves them through the production of a superego-like mechanism of surveillance that keeps these fractious forces at bay. In this respect, Rawls’s theory of justice, like the Periclean mode of mourning, is motivated by what Loraux calls the fiction of the “unanimous assembly.”⁸⁶ The dream of purity of heart is the dream of moral blamelessness, of acting consciously in light of universal moral principles to the extent that one would know and feel, with certainty, that one has acted correctly and justly. Yet, as Bernard Williams and others have argued, the very idea of blamelessness denies the fact that human life is lived in a tragic world, in which even morally coherent actions may—and even should—inspire regret, mourning, and reparation.⁸⁷ The original position qua unanimous assembly, then, is a reflection of paranoid-schizoid anxieties, in which we require an ideal object to rid us of the persecutory dread that arises when the good and bad elements of life begin to run together. Paranoid-schizoid defenses provide some measure of relief from this anxiety by reinforcing the split within the self and between self and others, but this relief is both shallow and short-lived, and it leaves us ill prepared to negotiate with the complexity or messiness (not to mention the “nastiness”) within self, other, and world.

Political Liberalism

Rawls’s late work extended and refined his early reflections on liberal conceptions of justice, while overturning key precepts of those original arguments. The basic content of justice as fairness—society as a system of fair cooperation between free and equal citizens, each engaging others in concordance with the principle of reciprocity and the duty of civility—emerges in his later work relatively unscathed. Yet the context of articulation for the theory of justice has been altered by Rawls’s recognition of irreducible pluralism in late-modern democracies—specifically, a pluralism of potentially incompatible comprehensive doctrines. These doctrines are comprehensive in that they offer their adherents both a fundamental orientation to the world and a conception of the good. Such worldviews are comparable in their scope to the salvific religious doctrines that clashed violently across Europe in the centuries following the Reformation—competing doctrines whose claims could not be settled or adjudicated on the political bases of compromise or deliberation. These conflicts generally were settled only by “exhaustion and circumstance,” or through the development of a modus vivendi social order based on a precarious balance of power between competing doctrines.⁸⁸ Rawls begins Liberalism by declaring that a plurality of these irreconcilable comprehensive doctrines is not an accident of history but the “normal result of the exercise of human reason.”⁸⁹ The enduring presence of conflicting doctrines represents the haunting possibility of a slide toward violent conflict, and, therefore, it cannot be lightly passed over. The problem posed by the Reformation is a sobering one: “How is society even possible between those of different faiths?”⁹⁰

Famously, at this stage Rawls no longer believed that the full theory of justice as fairness, as articulated in his early work, could provide the terms for a comprehensive reconciliation of reasonable but incompatible doctrines. The moral “baggage” of Theory would have to be bracketed so that the theory of justice can be reconstructed as a “free-standing” and purely political conception.⁹¹ In other words, justice as fairness was to be stripped of its comprehensive philosophical presumptions in order to serve its reconfigured purpose—the securing of a stable and just pluralistic society.

As a central part of this revised project, Rawls crafted a “political” conception of the person, which, he argues, avoids any controversial assumptions about human nature or moral psychology. As he puts it, “Justice as fairness starts from the idea that society is to be conceived as a fair system of cooperation and so it adopts a conception of the person to go with this idea.”⁹² Rawls cannot make this move without making some claims about human capacities, of course; as he puts it, human nature and psychology “are permissive: they may limit the viable conceptions of persons and ideals of citizenship, and the moral psychologies that may support them, but do not dictate the ones we must adopt.”⁹³ Psychological permissiveness means that there are no structural impediments hindering diverse subjects from “impersonating” the free and equal citizen; abstract citizens can each passionately identify with the principle of reciprocity as fair treatment of (free and equal) others, regardless of their background cultures or “nonpolitical” psychological capacities, needs, or anxieties. Instead of his earlier insistence that our moral psychology gives license to this form of civic attachment, now Rawls only claims that there are no insurmountable obstacles in its path.

The political conception of the person is tied to Rawls’s newly adopted “method of avoidance,” the result of which is a self split between a “political” or public identity guided by public reason and a “nonpublic” or “background” self in which the passions and partial attachments are given free play.⁹⁴ In making these arguments, Rawls fully retreats to an idealized account of the person, crafted to fit “the practical needs of political life and reasoned thought about it.”⁹⁵ In other words, the theory of moral psychology is developed to fit the model of a stable and just liberal society—the two tracks have been reduced to a single, idealized model. However this model suffers from the same Freudian slip as did Rawls’s original moral psychology. As I argue in this section, Liberalism in effect replicates, at the social level, Freud’s structural model of the psyche (in which the ego is torn between the disruptive forces of the id and the controlling edicts of the superego).⁹⁶

Rawls sets the stage for the work of reasonable justification within Liberalism with a reminder about the “extreme violence and increasing destructiveness” of the twentieth century, and his method of avoidance is inspired by the specters of religious warfare and constitutional collapse. In his published lectures on political philosophy, Rawls elaborated on the possibility of disintegration within a society’s basic structure.⁹⁷ Using the example of Wilhelmine Germany, Rawls argues that the German constitution was ripe for collapse because the country’s political parties were a “fragmented” array of “pressure groups … [that] held exclusive ideologies which made compromise with other groups difficult.”⁹⁸ Because they “never aspired to govern,” these pressure groups always acted “as outsiders petitioning the chancellor.”⁹⁹ Like the unreasonable comprehensive doctrines described in Liberalism, political pressure groups “cannot support a reasonable balance of political values.”¹⁰⁰ They press inexhaustible demands that cannot be met by the liberal order.

The basic structure (call it the Rawlsian ego) cannot bear up under this pressure unaided. Democratic citizens must therefore adopt a normative schema of reasonableness and use it “to express our moral and political thought and action.”¹⁰¹ Passionate identification with the “reasonable,” however, seems to represent the return of the (repressed) Freudian superego. Just as the original position—the always-available gaze through which reasoners can develop attachment to the principles of justice—appears when Rawls’s moral psychology can no longer support his theory of justice, so does the agency of “public reason” function as a fallback for Rawls’s theory of liberalism. Public reason does not claim sovereignty over the entire basic structure, but only “public political forums,” such as the spaces of law courts, government affairs, and campaigns for public office. It also applies only to certain questions within these contexts, such as questions of “constitutional essentials and matters of basic justice.”¹⁰² Public reason, then, does not eliminate the fractious desires and attachments across the array of comprehensive doctrines, any more than the Freudian superego eliminates the pressing demands of the id. It merely restricts or redirects those demands to the “background culture” of civil society, where they are given free reign.

We can call this, with only slight strain, Rawls’s structural model of the polity: the constitution is the container of the basic rights and liberties of the citizen body, the “nonpublic” demands of civil society persist in a free-wheeling “background culture,” and deliberation and decisions in constitutional bodies are constrained by the demands of public reason. Public political forums are buffeted on all sides by the demands of parties and pressure groups, but these groups are disciplined and their demands sublimated through the operations of a superordinate guiding voice. Recall that Freud saw the ego in a similar middle position, charged with managing the excessive demands of the id and the superego. For Freud, the ego is a “constitutional monarch”; it is not the actual seat of power but rather the space of an aspirational coherence.¹⁰³

Within Liberalism, the structural model of the polity is mirrored by Rawls’s political model of the person. The constitutional container is attended to and guarded over by free and equal citizens who are each committed to self-containment as dictated by the “ideal” of public reason. This ideal elevates citizens until they become hypothetical legislators and judges themselves, thereby capable of articulating their political beliefs and desires in the language of public reason guided by the duty of civility and the principle of reciprocity.¹⁰⁴ Because human nature is permissive, Rawlsian discipline encounters no significant obstacles in this work of political subjectification.

On my reading, Rawls’s Liberalism replicates the paranoid-schizoid defenses that were manifest in his original work; in fact, it seems to intensify them. In Liberalism both society and self are split between public and nonpublic sides, the former guided by public reason and thus oriented toward stable agreement and just reconciliation, the latter a supposedly more raucous and contestatory space of competing doctrines and visions full of encumbered and motivated selves posturing, performing, and protesting. This background noise is tolerated until it threatens the constitutional order (and hence the sanctity of the split itself). When it begins to influence the settled constitutional bodies, either as external pressure or internal interference, then all good, reasonable selves must step forward to reestablish the discipline demanded by the ideal of public reason and the conception of justice as fairness. Public reason becomes a civic superego that disciplines the passionate and fractious voices that, without such a superordinate principle, would drag the polity toward disintegration.

The difficulties with this account do not stem from Rawls’s supposition concerning the destructiveness of the twentieth century, which is undeniable, or from the specters of religious warfare and constitutional collapse, which only a cursory reading of history would convince us are serious and significant threats. The difficulty stems from the mechanism by which Rawls attempts to disperse these intense anxieties. Namely, Rawls attempts to dissolve the anxieties of loss and trauma through the invocation of a consensualist object of attachment. Public reason serves as a mechanism of self-control that merely (and in passing) mentions, but does nothing to effectively work through, dramatic political conflicts.¹⁰⁵ Rather, it acknowledges these conflicts only to insist that they can be resolved through what Rawls refers to as an “overlapping consensus.”¹⁰⁶ The unnerving pluralism within the polity—the ever-present threat of a slide toward violent doctrinal conflict, or toward oppression by unreasonable or mad conceptions of the good—turns out to be an illusion resting on a partial perception of social reality. While each regnant comprehensive doctrine has aspects of belief that fall outside the contours of public reason, there is a happy space of convergence that “when worked up into a political conception of justice” turns out to be sufficient to underwrite a just constitutional regime.¹⁰⁷ Instead of unnerving or dangerous pluralism, we have in front of us (“embedded” in public political culture) the basis of a “reasoned, informed, and willing agreement.”¹⁰⁸ Political reconciliation is by this conjuration not only possible but is already implied by the practices and principles built into the political culture. Hence, like Pericles, Rawls acknowledges the presence of discord and fractious desire only to resolve them through the construction of a superordinate, univocal mechanism of self-control.¹⁰⁹ Again, Rawls’s (Kleinian) intuition regarding the intensity of anxiety surrounding pluralistic selves and societies is reflected but not mitigated by the elevation of public reason as an ideal of social discipline. As Jonathan Lear has argued with regards to Freud’s theory of the superego, its forbidding and unyielding critical voice not only prevents cruelty but also “has its own cruelty in it.”¹¹⁰ While Rawls’s political principles aim to be reasonable above all else, the reason on display is the anxious, persecutory reasoning of the paranoid-schizoid position. The challenge remains, then, to develop Rawls’s powerful psychological intuitions in ways that effectively modulate rather than exacerbate the anxieties and conflicts inherent to social life in disordered societies.

Engaging the Transference within Contemporary Racial Politics

How does this interpretation add to or modify the critical race and agonist critiques of Rawls, as represented here by Mills and Stow? Undeniably, Charles Mills is correct that Rawls’s work does little to theorize historical and enduring legacies of racial domination in the United States. Yet perhaps this limitation of Rawls’s work does not represent an ideological obfuscation so much as an insight into the difficulty of negotiating fair terms of social cooperation in the face of deep and persistent inequalities. Mills and others describe Rawlsian theory as suffering from a pernicious form of amnesia and denial, but perhaps there is a hidden form of denial within this very critique of denial. In short, Mills might neglect the issues of the “transference” circulating within discourses and interactions in our racialized polity. Transference was for Freud and his followers the clearest demonstration that the analyst and the analysand are never the only people in the room during a session. Our relationships with others are mediated by the deep deposits of previous relationships and, pace Klein, by the cognitive-affective schemas that reflect different levels of anxieties and our various capacities for working through them. It is true that Freud defined the work of analysis as the removal of amnesias, but the theory of transference complicates such work. Successful interpretive work is uneven and rarely guaranteed of success because of the inexhaustible unconscious and the oft-times unpredictable patterns of interaction within and between selves. The telos of analysis is not a life of purity or blamelessness, but rather, in Freudian terms, liberation from unmitigated misery to “common unhappiness” and, in Kleinian terms, development of an enriched ego in better touch with the “poignant” and ambivalent nature of internal and external realities.¹¹¹

The theory of transference has direct relevance to contemporary racial politics. Scholarship of racial attitudes in the United States has recently undergone a paradigm shift in how it approaches questions of prejudice and bias. Despite persistent levels of racial inequality, overt levels of support for racial equality in the United States have increased dramatically over the past half century since the passage of the Civil Rights Act in 1964. Yet measures of implicit racial bias reveal a stark divergence between the explicit acceptance of racial equality and unconscious levels of racial prejudice across (and within) different racial groups.¹¹² Racial prejudice in the United States appears to have been sublimated and is now pushed through an ideological filter of values (such as the work ethic) that, in turn, have become racially coded.¹¹³ Because it is increasingly socially unacceptable to express explicit racial bias, “old-fashioned” racism has partly shifted toward symbolic and even unconscious registers.

This shift from explicit to implicit bias has significant implications for Mills’s theory and for his critique of Rawlsian abstractions. Namely, the sublimation of racial prejudice both increases the saliency of racial priming and increases the anxieties surrounding racially coded politics and policies. Perversely, the stricter social superego surrounding racial equality, by shaming the expression of racial bias, appears to repress these thoughts in ways that intensify rather than diminish them. Racial priming can then tap into this unconscious affective layer, and, in doing so, actually increase resistance to policies that might address racial disparities.¹¹⁴ As the literature on affective intelligence has demonstrated, when individuals feel threatened and anxious, their resistance to redistributional politics increases significantly.¹¹⁵ The symbolic racism literature, moreover, shows that this anxiety is more likely as the gap between implicit bias and explicit norms grows ever wider. Paradoxically, then, the direct thematization of racial disparities and inequalities can interact with this hostile unconscious brew in ways that actually decrease the likelihood of making progress in the areas of racial disparities. A Millsian truth-telling, antiamnesic theory of racial domination and enduring injustice, despite its honorable intentions, risks the provocation of a transference storm in the polity at large. Racial politics then oscillates between manic triumph and the most hostile resentment.

Progressive racial politics, in this light, has to confront and work through the racial transference rather than presuming that clear and obvious truths surrounding racial inequality and injustice will, by themselves, cultivate a cross-racial political coalition capable of addressing these social pathologies. In fact, the recent scholarship on unconscious racism and affective intelligence leads us to assume the opposite—that a righteous insistence on racial justice will actually increase resistance and decrease the likelihood of making progress. Klein’s concept of projective identification becomes a valuable explanatory tool in light of contemporary racism’s increasingly implicit basis. Projective identification initially results from the paranoid-schizoid defense of splitting, in which the good and bad aspects of experience are isolated from each other and threats are projected out onto an external object. The other becomes the repository for the aspects of the self that could not be tolerated, such as our aggression or our vulnerability. Racial differences create a ripe playing field for these fantasies of splitting and the corollary phenomena of idealization and demonization.¹¹⁶ Social institutions and cultural norms are deployed to reinforce these defense mechanisms, operating as a screen to obscure intense sub rosa anxieties.¹¹⁷ As the conflicts and prejudices surrounding racial difference become increasingly subject to social shaming, these anxieties are exacerbated and the need for social institutions to provide this fantastical screening is intensified. Our interactions with others are mediated by these intense, largely unconscious anxieties, promoting, once again, either a politics of manic triumph or of racial resentment (cue liberals and tea partiers after the election of Barack Obama). These outcomes offer temporary satisfaction through release of the pent-up tension, just as Pericles’s discourse promotes a temporarily satisfying identification with the idealized polity that simultaneously disavows and expels the dangerous or evil aspects of the in-group onto a safely hated other.

In the late twentieth century, James Baldwin provided eloquent testimony to the idea that race is an illusion born of projective identification, even if Baldwin did not deploy a psychoanalytic framework to do so. Race, for Baldwin, was an “illusion” rooted in basic human needs for security and the fight against anxieties of disintegration. For Klein, similarly, the paranoid-schizoid position is characterized above all by defenses against fears of disintegration—feelings of absolute insecurity and instability that, in turn, call forth defenses such as splitting, idealization, and demonization. Baldwin echoes this idea when he writes that what “drives people” to racist behavior and attitudes is “pure terror.”¹¹⁸ This terror has both psychological and historicopolitical roots. In the American context, the specific terror that drives people is rooted in the (disavowed) knowledge of the historical crimes of white supremacy. As Baldwin puts it, “whites … know the crimes they have committed against black people. And they are terrified that these crimes will be committed against them.”¹¹⁹ In other words, Americans’ violent racial history locks white citizens into a kind of paranoid-schizoid functioning, in which the terror of their collective inheritance has to be split off or denied. Baldwin sees this nexus of defenses as the strongest source of support for group identity in the racialized polity. As he puts it, “it is … comparatively easy to invest a population with false morale by giving them a false sense of superiority,” and this sense of superiority is predicated on and held in place by racial inferiors, who serve as a receptacle for the terror that cannot be acknowledged.¹²⁰ Identity cannot be imagined outside of these polarized terms, as evidenced by the fact that, as Baldwin puts it, “progress” is imagined only in terms of “how quickly a Black kid can become white.”¹²¹ Yet this perverse form of progress implicitly relies on a racial other remainder, who will reinforce the basic boundaries even as those boundaries are surmounted by a fortunate few.

In short, American identity is built on projective identification and paranoid-schizoid defenses, and these defenses are rooted in unarticulated terror: “Americans are terrified … they know that they are capable of genocide. History is built on genocide. But they can’t face it. And it doesn’t make any difference what Americans think they can think—they are terrified.”¹²²

“Whiteness,” then, is a paranoid-schizoid “state of mind.”¹²³ It is rooted in a terror that cannot be expressed because to do so would expose the work of paranoid construction that holds it together. All of this is encapsulated by Baldwin’s notion of “innocence.” Innocence is both a practiced amnesia over historical violence and an immobility of identity. Innocence is “based on a lie, the lie of Manifest Destiny. [America is] a country immobilized, with a past it cannot explain away.”¹²⁴ Yet innocence is not simply the refusal to accept historical crimes or to admit fantasies of expansion that accompanied and often justified such crimes; innocence is also an omnipotent refusal of interconnection, an insistence on what Baldwin calls the “dream” or illusion of separation.¹²⁵ As Baldwin puts it, “people in America have not been able to accept the fact that their reality is entangled and almost defined by the black reality.”¹²⁶ The very categories of white and black in fact ensure that the entanglements and blood-ties of American history will be denied. Social practices of separation—white flight, gentrification, and racialized drug and prison policies—reinforce these categories and prolong the recognition of entanglement. All of this, however, comes with a significant cost; as Baldwin puts it, “you can’t deny your brothers without paying a terrible price for it.”¹²⁷

For Baldwin, the innocence of American life is in many ways a product of what Klein called projective identification. For instance, Baldwin repeatedly interpreted race not only as a certain way of seeing but also as the production of certain sights—namely, “the nigger” and the associated stereotypes of black life such as “Uncle Tom, Aunt Jemima, and Little Black Sambo” (now we might say the black “supercriminal,” or the “welfare queen”).¹²⁸ Whiteness as a state of mind is predicated on the projection of intolerable aspects of self and history into the receptacle of “the nigger.” Baldwin argues, moreover, that the production of such characters is intertwined with an inability to countenance deep anxieties and terrors of life. As Baldwin puts it, “I am not a nigger … [and] if you think I’m a nigger, it means you need it. Why? That’s the question you have got to ask yourself.”¹²⁹ Baldwin focuses as much on the need as the defensive production of the image and the social consequences of its production. White people “have a black citizen locked up in their mind … and that is why they treat them as they do.”¹³⁰ The “nigger” in the mind of the white citizen means that “what they do see when they do look at you is what they have invested you with … all the agony, and pain, and the danger, and the passion, and the torment—you know, sin, death, and hell—of which everyone in this country is terrified.”¹³¹ For Baldwin, then, the historicopolitical practices of white supremacy are inextricably connected to the “heat, horror and pain of human life itself” and to the relatively undeveloped human and civic capacity to face down this horror.¹³²

Can paranoid-schizoid projections that reinforce punitive social practices and serve as omnipotent defenses against interconnection be unwound? Stow’s depiction of a “tragic” mode of public mourning can be seen as one possible response to this challenge. In distinction from the romantic mode of mourning, which intensifies normalizing pressures within political life and marginalizes disagreement and conflict, tragic styles of mourning, according to Stow, generate an agonistic understanding of democracy “in which conflict and disagreement are recognized as central to democratic politics.”¹³³ Tragic public mourning is “pluralistic, critical, and self-consciously political.”¹³⁴ It eschews a maudlin sentimentalization of loss in favor of a tragic mode of response that emphasizes paradox, frustration, and conflict. Above all, a tragic style of public mourning “seeks to generate ambivalence in its audience as a productive response” to loss.¹³⁵ Ambivalence—or the “prevalence of duality over unity”—casts into doubt both romantic readings of loss and the unity or coherence of a collective subject on which such readings are typically founded.¹³⁶

It is on the basis of the distinction between romantic and tragic mourning that Stow describes an African American mourning tradition that is both “tragic and self-consciously political.”¹³⁷ This tradition grew out of the experiences of oppression under slavery and coalesced into a political force that galvanized struggles against de facto and de jure segregation and discrimination. By offering its audience the doubled consciousness of violent oppression alongside a hopeful expectancy of justice, African American mourning practices embodied a democratic pedagogy, which generated solidarity among the oppressed and galvanized countless acts of courageous resistance to legal and social discrimination. They also engendered a mode of “critical remembrance” that challenged dominant interpretations of American history and identity.¹³⁸

For Stow, the historical and contemporary exemplars of this African American funeral tradition—from Frederick Douglass to Joseph Lowery—have been insistent critical voices, continuously calling citizens to respond to the tragic side of American life. Frederick Douglass’s multiple funeral orations for Abraham Lincoln, for instance, were agonistic reminders that fought against both “historical amnesia” and the tropes of romantic eulogy. In the years following the Civil War, Douglass repeatedly refused a hagiographic frame for Lincoln’s life and death. For instance, in a speech before Congress in 1876, Douglass excoriated Lincoln as “the white man’s president.”¹³⁹ In so doing, Douglass “performed … [a] productive disorientation” that forced his audience “to reconsider their own position.”¹⁴⁰ Douglass’s funeral orations provoked dissonance and revealed ambivalence—the combination of stirring promises and pernicious realities.

Stow’s disaggregation of public mourning into competing traditions, his development of a twofold typology that links these traditions to distinct conceptions of democratic politics and identity, and his subtle treatment of African American mourning practices as an instantiation of the tragic mode of mourning are an invaluable contribution to debates about the link between mourning and politics. Moreover, Stow’s description of a democratic pedagogy drawn from an acceptance of ambivalence and conflict has great potential for engaging the transferential battlefield surrounding racial politics in the contemporary United States. Anti-Periclean, dissonant mourning practices can act as a counterweight to idealized accounts of national histories and civic norms that feed cycles of manic denial or omnipotent triumph. Because these countermemorial discourses are rooted in a tragic awareness of fundamental ambivalence, they are also less likely to enflame persecutory anxieties that actively work against cross-racial sympathies and coalitions.

Yet there remain some difficulties within Stow’s account. For instance, Stow acknowledges that there is “something of a paradox” with regard to his claims about Frederick Douglass, whose “unequivocal commitment” to a definite view of the Civil War seems to conflict with the emphasis on tragic ambivalence and the essential partiality of any particular viewpoint.¹⁴¹ It is clear that Douglass’s discursive register—as much as it emphasizes a critical uprooting of convention—also had moments of closure and a drive toward a (reconfigured) social consensus. This paradox creates an internal conflict that causes Stow’s essay to rove across the typology of public mourning he has described—alternately tragic and romantic, agonistic and consensualist. For instance, he describes Douglass’s eulogies as demanding “that those who had hitherto considered their position natural, given and unassailable, defend it against his more compelling arguments. By inviting his white audience to engage him in reasoned argument, moreover, Douglass enmeshed them in a performative contradiction: undermining white supremacist claims by forcing them to recognize him as a person worthy of reasoned engagement.”¹⁴²

Stow’s emphasis on “reasoned” and “compelling” arguments, the work of “performative contradiction,” and the telos of “reasoned engagement” reveals a tacit consensualism within his agonistic approach. Agonists cannot fully quit the dangerous game of consensus without their own performative contradiction. Somewhere there lurks a good object, which licenses the critical uprooting of rooted biases. The good object, for Klein, is a dynamic psychic entity that has a characteristic influence on the individual’s way of experiencing life and relating to others.¹⁴³ In the depressive position, the heretofore-idealized “good breast” is restored to wholeness and morphs into the good object, which still carries a measure of idealization but whose ideal features are now filtered through an awareness of the object’s overall imperfection and complexity. The good object is what allows us to appreciate the “trauma of a flawed world” without that experience tilting us toward endless cycles of despair, fear, or defeat.¹⁴⁴

Klein’s idea of the good object has relevance beyond the intimate world of analyst/analysand in which she was enmeshed. The basic insight contained within the concept of the good object is that the experience of tragic dissolution cannot be borne without a corresponding source of reassurance. In order for us to mourn and accept the imperfections of self, other, and world, there has to be a reliable, whole object that gives us hope for the continuity of life and the possibility of understanding or mutual recognition. We can only be disappointed in an object that we also need; we can only experience disappointment (instead of living it) if at some very deep level we also experience security, comfort, understanding, and hope. All of these experiences are very fragile, and that itself is a tragic recognition that requires a certainty that they are possible, perhaps even unavoidable.

Therefore, instead of thinking about the praxis of mourning from the perspective of separate ideal types, I suggest that we see consensualist and agonist modes of public mourning as ultimately inseparable and complexly intertwined. The work of agonist or tragic disruption cannot function purely on its own power. Tragic and moral narratives must necessarily shadow one another because each makes the other possible. For Klein, the acknowledgment of the tragic (our conflicted and fractious self) stands on and educates the moral or normative core within her theory, represented by the depressive position and the hope for reparation undertaken through the work of acknowledging and accepting the imperfections of self, other, and world.¹⁴⁵

The danger, then, resides not within the game of consensus or the agonistic clash of wills in themselves, but where either of these narratives manically denies the other. As Meira Likierman argues, morality for Klein “must assume the possibility of irrevocable loss all the time” if it is to retain its sense and its connection with lived experience.¹⁴⁶ Morality must be disciplined by tragedy if it is to avoid the rigid moralism of the paranoid-schizoid position, which serves as a protection of the fantasized sanctity of the self and precludes open interaction and engagement between selves. If we build from Klein’s insights, the challenge for democratic theory and practice is to locate good objects and potential spaces that license the acknowledgment of tragedy—to search, in essence, for means of mourning that do not eliminate either agonistic struggles for recognition or the consensualist search for background norms but holds these moments together in tense coexistence. In short, the challenge is to combine the tragic and the moral in ways that mitigate one-sided agonism or consensualism.

Let us go back to Rawls, then, and to what I have described as the Kleinian intuition behind his ideas of the original position and public reason. Instead of asking, pace Mills and Stow, whether or not a theory such as Rawls’s promotes denial, perhaps we ought to ask if the denial at work is pernicious or democratically adaptive. In other words, do the abstractions of Rawls’s approach help us to negotiate the transferential field surrounding racial discrimination, prejudice, and inequality? I want to argue that this is (potentially) the case. With the increasingly symbolic and implicit nature of racial bias, a direct rhetoric of aggressive and unvarnished truth telling may, perversely, have a backfire effect on support for policies that could ameliorate racial injustices. Because racial prejudices are both intense and partly unconscious, Rawls’s intuition to base public reasoning on deracialized stick figures has some merit. The original position and public reason, in this light, might provide some measure of relief from the intense anxieties surrounding race in the contemporary polity, leading to the development of terms of negotiation and engagement that could form the basis for political coalitions forged in the teeth of the racialized transference.¹⁴⁷

However, as previously discussed, Rawls’s theory cannot fulfill this function because of the narrow way in which the original position and public reason—as civic superegos—are theorized. I argued that the reason was Rawls’s Freudian slip toward a severe, univocal superego within his explicit and implicit theories of moral psychology. In Rawls’s lectures on Joseph Butler, for instance, he describes the work of conscience as a “supreme principle” that specifies “conclusive or decisive reasons for what we are to do.”¹⁴⁸ Conscience, like the original position, is always available and accessible; we come to it within what Butler called a “cool hour” of reflection.¹⁴⁹ From this space of solitary consideration, we can come to understand how we should act, and from these judgments “there is no further appeal.”¹⁵⁰ Here again, we detect the specter of Loraux’s unanimous assembly and the search for moral blamelessness or purity of heart. Yet, according to Klein, the univocal and severe superego is precisely that which intensifies paranoid-schizoid defenses such as splitting, idealization, and demonization. The search for blamelessness is a symptom of desperate omnipotence, and purity of heart is a pernicious fantasy that cuts the ambivalent self off from its aggression and hatred. The alternative to the unanimous assembly fantasy is an understanding of conscience as a polyphonous assembly of diverse voices and ambivalent objects.¹⁵¹ This involves the activation of the depressive position as a cognitive-affective schema by which we can acknowledge the “poignant psychic reality” of whole, ambivalent others and by which we can face up to the conflicts that persist within the self and between self and other—preparing all the while for inevitable regrets and the work of reparation.¹⁵² What makes this possible is the interminable work of mourning where we work through, again and again, the differences and distances between and within selves across the battlefield of transference and amid the uneven, unequal, and contested terrain of the social. Abstract others and univocal principles cannot accomplish this work. Only by engaging concrete, whole others can we effectively modulate the anxieties and pathologies plaguing both intersubjective and social life.

Yet all of these possibilities are denied to the representatives in the original position, which is intended only as a “guide to intuition” or a Butlerian “cool hour” of reflection—a process of introspection that all rational agents can take up and practice. Hence it has limited value. In the absence of concrete others and actual social practices whereby paranoid-schizoid defenses can be countered, anxious fantasies of persecution will continue to plague us. The original position cannot, by itself, effect a transition from paranoid anxiety to depressive anxiety; this happens only when our internal objects and fantasies come to more closely match our external experiences. Klein did not extend her theory to social-political practices, but by helping us to see and accept the ambivalence and conflicts within the self while mitigating persecutory fantasies of dissolution or idealized dreams of perfect union, she, in the words of Isaac Balbus, has “taught us something new and important about the emotional demands of democracy.”¹⁵³ As Hanna Segal puts it, Klein’s description of the transition from the paranoid-schizoid position to the depressive position is the “evolution from an insane world determined by misperceptions into a saner world … in which conflict and ambivalence can be faced.”¹⁵⁴ This transition requires the repeated presence of concrete others who can hold and effectively dispel characteristic defenses against anxiety such as splitting and demonization. Yet Rawls neglected to theorize the political practices and experiences that could replicate—and make real—the psychological and cognitive development that he seemed to hope would take place in the original position. Rawls’s sound psychological presumptions (that we need distance from persecutory anxiety in order to sympathetically engage with others in the difficult work of sharing and re-creating a common world) seem to be compromised by the most “liberal” of fears: the fear of politics.¹⁵⁵ But it is perhaps only through such concerted political engagement with whole, ambivalent others that we can achieve integration.¹⁵⁶ It is to this possibility that I now turn.

Conclusion: Viable Others and the Democratic Superego

In her study of the Athenian funeral oration tradition, Nicole Loraux asks, “Is there a democratic way of speaking about democracy?”¹⁵⁷ This project asks a similar question: Is there a democratic way to mourn in a democracy? The Periclean mode of mourning, whose tropes persist, seems problematic on this score. By transforming the democratic polity into an idealized, harmonious whole, Periclean-styled civic discourses and rituals surrounding public traumas seem to obscure ongoing conflicts and injustices, which impoverishes public discourse and impairs political judgment. Guided by Melanie Klein’s theories of the paranoid-schizoid and depressive positions, I have argued that these mourning rituals are politically felicitous and effective because they trigger psychological defense mechanisms such as splitting, idealization, and denial. Yet these defenses have pernicious consequences. Consensualist political theories feed this same cycle of denial and persecutory anxiety. What Periclean mourning and consensualist approaches to public life have in common is the reliance on a superordinate, univocal principle of self-control and social order. Far from providing the means of moral-psychological maturity, however, the univocal superego reflects and feeds the anxieties of the paranoid-schizoid position, which isolates subjects from each other and from the less-tolerable aspects of their selves—their hatred, envy, vulnerability, and fear. When social life is mediated by a severe, unequivocal superego, the intense anxieties that circulate around public trauma and enduring injustices are exacerbated rather than being effectively worked through. The result is an autistic politics characterized by episodes of manic triumph and hostile resentment.

The approach to mourning proposed here, however, provides room for an alternative. With her idea of the depressive position, Klein describes a schema by which the conflicts within and between selves could be acknowledged and the capacities for tolerance, guilt, and reparation could be cultivated and practiced. Klein’s simultaneous tragic and moral narratives license both an agonistic struggle for recognition and a consensualist search for defensible social norms—but more importantly, they warn against the excesses or pathologies attendant on each of these struggles in isolation from the other. The challenge, then, is to develop Klein’s insights by anticipating and describing the political institutions and social practices that could provide space for what I am calling the democratic work of mourning.

In part we can do this by developing Klein’s concept of the superego as a fractious assembly. Unlike the Periclean civic superego, which emphasizes and enforces social consensus and coherence, Klein’s concept of “depressive” conscience takes into account the differences and distances between and within selves.¹⁵⁸ For Klein, the acceptance of ambivalence—the expression of hatred and disappointment toward loved objects, for instance—is the key component of moral-psychological maturity. By accepting ambivalence, we mitigate the temptations toward the temporarily satisfying defenses of idealization or demonization. By facing up to the otherness within the self, we can increase our tolerance toward actual, whole others. This is the integrative work of mourning in the crucible of the depressive position. It does not eliminate conflicts within and between selves, nor does it make injustice, violence, or terror impossible, it merely and temporarily (and this is still quite a bit) helps us to clarify the sources of conflict and to mediate this conflict through a depressive or mournful awareness of complexity and ambivalence.

What makes this work of mourning possible? Thomas Ogden argues, following Winnicott, that the “containing” function of external others is crucial.¹⁵⁹ What we might call “viable others” unsettle the perceptual-affective habits within the paranoid-schizoid position. They simultaneously hold and disrupt the fantastical projections emitted by the defenses of splitting, idealization, and demonization. Viable others, as Klein described them, are whole others; they reflect ambivalence and fractiousness in ways that objects within the paranoid-schizoid position cannot. In reflecting ambivalence they help to inaugurate the depressive position, which disalienates the anxiety, frustration, and fear that had been heretofore split off and disavowed. From within the terrain opened up by viable others, reparative impulses can be liberated and tolerance of self and others can be strengthened.

Moving beyond an anxious emphasis on consensual theories (or rituals) of democracy, then, we need to identify viable others and spaces in which interaction with such others can cultivate a more polyphonous, democratic superego. Once again, on this front we have much to learn from analytic approaches. Freud, for instance, defined the work of analysis by invoking Oedipus’s gradual recognition of his tragic origins. The work of analysis proceeds in like fashion: engaging the transference in ways that allow the analysand to recognize herself, to re-cognize her thoughts, actions, history, and future by a new light. Klein, however, moves us beyond recognition toward productive or generative reversal. Reversal occurs when the projected or disavowed aspects of experience are returned to the self. Reversal happens when the internalized objects talk back, when they object to the flattening and obscuring discourses that embellish them (either as angels or demons). Reversals can themselves be disavowed, of course, which starts the paranoid-schizoid cycle afresh. Yet reversals can also inspire what James Baldwin referred to as a “good-natured reconciliation” to the complexity and ambivalence of self and other.

How are these viable others made to appear? In part, viable others appear from the rubble of a destroyed or overturned projection. Take, for instance, the work of perceptual-affective reversals in Michelle Alexander’s The New Jim Crow, which powerfully details the affects imbuing racialized policing and prison policies in the United States, ranging from the easy scapegoating of black and brown criminals to the cognitive dissonance-inspiring efforts to exclude white, middle-class, or suburban drug users from the stigma of criminality.¹⁶⁰ Racial discrimination finds a ready outlet in the officially color-blind but intensely racialized war on drugs. Alexander provides a thorough genealogy of these policies, revealing them to be related to and in some respects an outgrowth of historical patterns of racial stigmatization. Although Alexander does not mention Klein or projective identification, what she describes as the collective American “permission to hate” criminals can clearly be read as an accessible outlet for paranoid-schizoid defenses such as denial, splitting, and demonization.¹⁶¹ Alexander also models viable otherness with several reversals that hold and disrupt penchants for racialized disavowal. For instance, she argues that—if justice were really color-blind—the drug war could have been waged in a very different manner:

From the outset, the drug war could have been waged primarily in overwhelmingly white suburbs or on college campuses. SWAT teams could have rappelled from helicopters in gated suburban communities and raided the homes of high school lacrosse players known for hosting coke and ecstasy parties after their games. The police could have seized televisions, furniture, and cash from fraternity houses based on an anonymous tip that a few joints or a stash of cocaine could be found hidden in someone’s dresser drawer … all of this could have happened as a matter of routine in white communities, but it did not.¹⁶²

Alexander’s racial transposition reverses the easy projections and disavowals that stand behind prison and drug policies in the United States. Because Americans have permission to hate criminals, and because those criminals are so persistently coded as racial others, the cognitive dissonance that occurs when the frame is switched can challenge or dislodge the nexus of anxious projections that maintain a rigid split between innocence and guilt. In this example, the white, suburban, middle-class drug users are viable others insofar as they disrupt easy stereotypes and complicate the perceptual-affective schemas on which those stereotypes rest. Viable others become the basis for a more honest dialogue about the enduring traumas and intense anxieties surrounding racial politics in the United States. Even more importantly, honest dialogue or cross-sectional deliberation can reveal those viable others.¹⁶³

Rawls, it turns out, had it right the first time around: the struggle for justice not only relies on psychological capacities but calls for a kind of soul-craft. James Baldwin never shied away from this claim, or from the idea that there is a connection between a political struggle for racial justice and a human struggle against the terrifying aspects of our interconnected, ambivalent lives and the desperate strategies we deploy to deny our terror. The struggle against racial injustice is more than a battle against unjust laws, exclusive institutions, or even “racism.” As Baldwin puts it, “The battle is elsewhere. It proceeds far from us in the heat and horror and pain of life itself where all men are betrayed by greed and guilt and blood-lust and where no one’s hands are clean.”¹⁶⁴

For Baldwin, viable others that clarify the “heat … horror and pain of life itself” were what he often referred to as “poets.” The poet, according to Baldwin, does not simply reflect his or her own experience but gives witness to the pathologies and disasters of the world in ways that bring a semblance of order and understanding to those disasters. As Baldwin puts it, in describing Billie Holiday: “Billie Holiday was a poet. She gave you back your experience. She refined it, and you recognized it for the first time because she was in and out of it and she made it possible for you to bear it. And if you could bear it, then you could begin to change it. That’s what a poet does.”¹⁶⁵

Poets “hold us together,” that is, they make experience bearable in part by offering new ways in which individuals can understand themselves, new forms for experience that make this experience, and the experiencing self, more coherent, whole, enriched—assembled. Viable others, however, are necessary but insufficient for the ongoing praxis of the democratic work of mourning. The limitation of the poet is that while they may indicate blockages in particular vocabularies, they cannot create a new vocabulary by themselves. Only groups in assembled spaces can create—and contest, and re-create—a vocabulary.¹⁶⁶ Therefore, in chapters 4 and 5, I look beyond both Antigone and Pericles toward spaces of acknowledgment and assembly through which citizens might mourn democratically in the name of democracy.