When John Rawls crowned justice the ‘first virtue’ of social institutions, stability was its necessary presupposition. For what worth is there of just ideals if the social order they structure will flicker out of existence when under pressure? As Rawls (2005) wrote Political Liberalism, stability became a central concern for his theory of justice, and the “overlapping consensus” is his answer to this problem. The same concern about stability can be found in the works of Habermas, another key theorist of liberal democracy. Habermas (1988, 1998) describes how the only viable source of political legitimacy in the modern world is the socially integrating networks of communication. In his more recent works, he even considers liberal democracy the only viable institutional arrangement that can secure stable political coexistence in our conflict-ridden world. Yet, really existing liberal democracies are far from stable. Followed by decades of neoliberal reform in major liberal democracies, public accountability of governments soon gave way to accountability to private shareholders of multinational capital. Inequalities were staggering, leaving many on the verge of destitution and precarity (Milanović, 2019; Streeck, 2016). Decades after neoliberal reforms have taken root and wreaked havoc, democracies are “undone.” The Left is now disoriented, while angry, disenfranchised masses are ‘re-politicizing’ the privatized world with a vengeance, turning to right-wing populisms of hatred, chauvinism, xenophobia, and misogyny (Brown, 2015; Mouffe, 2018). For most ideal theorists, the problem with existing democracies is that liberal democratic ideals are misapplied.1 They believe that as long as we reattune democracies to their ideals, inequalities will be kept in check, toxic populisms will disappear, and democracies will be stable once again. However, this account seems increasingly untenable: First, politically, the rise of populism in the liberal democratic West shows that politics guided by rationalist ideals are becoming unrealistically “utopian.” Second, these populist currents demonstrate how negative affects such as hatred, jealousy, and paranoid anxieties powerfully shape political life, calling into question the negligence of negative (especially antipathic) affects in ideal theories (Mouffe, 2005, 2009). Thus, if one's theory aims for stable democracies, then one must go beyond ideals, and the ‘affective deficit’ of rationalist ideal theories must be addressed. Some currents in political thought try to overcome this affective deficit. For instance, Nussbaum (2013, 2018) supplements liberal theory with her account of political emotions. She discusses negative emotions such as disgust, anger, and fear, and argues for the need to foster love and forgiveness, redirecting our emotional energies to productive channels. Axel Honneth from the Frankfurt School is also aware of the limitations of pure ideals. His works supplement Habermas's discourse–theoretical ideals with an emphasis on affective dimensions of social recognition (Honneth, 1995). In The I in We (2012), Honneth even raises the need for an understanding of unconscious affects in politics. Yet, despite engaging with negative, antipathic affects, their political prescriptions remain optimistic (sometimes moralisingly) and idealistic. They make it seem as though antipathic and negative affects can be tamed by a mere combination of conscious will, improvements in democratic institutions, and public education.2,3, 2,3 Something seems amiss when we turn to existing political phenomena for a reality check, given the seemingly insurmountable difficulty of resolving antipathic affects. As I shall argue in this paper, populist and fanatical political movements speak to much deeper psychological realities than what simple reforms in democratic institutions and public culture can address. Populist movements (whether of the toxic or progressive types) are resilient to conscious reform due to how they function as unconscious psychic defenses against severe anxieties for their participants. This means that unless the psycho-defensive nature of these movements is tackled, conscious reform will be met with great unconscious resistance. On the other side of the spectrum of political theory, Ernesto Laclau's and Chantal Mouffe's studies of populism eschew political optimism by taking inspiration from Freudian and Lacanian readings of antisocial passions (Laclau, 2005; Mouffe, 2018). Taking psychoanalytic assumptions of the persistence of the death drive and Lacan's understanding of impossible jouissance, they consider antagonisms constitutive of all political relations (Laclau, 2005; Mouffe, 2005). Furthering this view, Laclau (1996) argues that democratic reconciliation and human emancipation are impossible, and political conflicts are necessary. The best way to deal with conflicts, then, is not to repress them but to channel them to critical and democratic causes (Mouffe, 2009). Although Laclau's and Mouffe's shared approach avoids naïve optimism, they seem to overemphasize the reality of antagonisms at the expense of thoughtful considerations of how democratic institutions can be stabilized. Laclau's own theorizations offer little room for understanding how more democratic political arrangements can be stabilized against potential deterioration. Mouffe (2005, 2009, 2018) goes further than Laclau and proposes the ideal of an “agonistic” democracy that aims (i) to redirect antagonistic drives toward a Left populist cause (against neoliberalism), and (ii) to construct a shared symbolic space around liberal democratic ideals open to conflicts in their interpretations. Yet, besides being mostly inchoate, these proposals are suspicious, for they concern only symbolic means to contain antagonistic drives, leaving behind complications at the psychodynamic level. Indeed, if antagonisms are so constitutive as her reading of Freud suggests, what prevents them from overspilling the symbolic framework of an agonistic liberal democracy? To address the affective deficit of democratic theories, I shall turn to the aspects of psychoanalysis that these political theories have neglected. Psychoanalysis is particularly suited to understanding political passions because it systematically accounts for the ubiquitous “irrational” and passionate moments of human sociality (Allen, 2015; Honneth, 2012, pp. 195–196). Inspired by observations in the analytic setting, the psychoanalytic approach takes seriously the fact that (i) negative affects cannot be easily overcome by conscious will, and (ii) that psychotherapeutic interventions targeted at resolving long-standing defenses can only be effective when they work on the unconscious levels. These two features of a psychoanalytic approach counter both the facile optimism in rationalist theories of political affects and the Mouffe's and Laclau's lack of thoughtful consideration of the transformation of antagonisms. Regarding the latter point, psychoanalysis offers significant therapeutic insights that may be useful in helping us see how antipathic affects can be contained and transformed. The possibility of the (social) transformation of affects, not adequately considered by most political theorists, may hold the key to understanding how democracies can be stabilized amid the challenge of toxic populism and sharp antagonisms. In the coming sections, I will first briefly review psychoanalytical theories of society since Freud, arguing that a comprehensive account of the psychodynamics of (political) groups necessitates the study of preoedipal, psychotic mechanisms (Section 1). After surveying Melanie Klein's account of preoedipal psychic processes in terms of the paranoid–schizoid and the depressive positions (Section 2), I will propose a reading of populism as a brand of paranoid politics (Section 3) and show how the productive aspects of the depressive position (i.e., mourning and reparation) can be practiced socially (Section 4). Such psychodynamic reconceptualization of political affects brings into light a new challenge to achieving democratic cohabitation in the real world, which I shall outline in Section 5. Finally, Section 6 will extend the Kleinian model to study resistance movements in a nondemocratic setting. Freud was not only the pioneer in a psychoanalytic theory. He also offered an insightful theorization of group psychology. His famous study in Group Psychology and the Analysis of the Ego (Freud, 1955) argues that the primary motivating force that underlies large-group participation is the libidinal ties between group members and the leader serving as a paternal figure. This love toward a shared paternal ideal often leads to narcissism and aggression, as the group is often intolerant and violent against out-groups. This observation about the affective and aggressive dimensions of groups is also featured in his Civilisations and its Discontents (Freud, 1961), where he proposes the life and death drives as the ineradicable instinctual basis for all forms of human coexistence. Bold and pioneering as Freud's social analyses are, his speculations on human aggression and group psychology are underdeveloped. As Lear (2000, 2005) observes, Freud took the death drive as a mysterious force of aggression constantly welling up along the crevices of social life without sufficient psychodynamic explanation.4 (This theoretical blindness is inherited in the works of Mouffe and Laclau, who draws inspiration uncritically from him.) To address the limitations of previous theories, this paper aims to seek a psychoanalytic theory that (i) helps us diagnose antipathic affects in social reality while (ii) providing directions for a meaningful social transformation of affects essential for democratic stability and reform. This requires us to go beyond Freud. In particular, the Freudian approach misses two important aspects of group psychology. First, as subsequent psychoanalysts point out, Freud's model of groups, centered around the paternal ideal, misses the important preoedipal psychotic dynamics in regressed groups. They propose that, instead of seeing the group as merely centered around the totem pole of the father, groups also perform important “maternal” functions, as group phenomena demonstrate features of our anxiety-ridden, preoedipal object relations with our earliest caregivers (Alford, 1989; Balbus, 2005). These earliest relations are ridden with primitive defenses of denial, splitting, projection, and introjection, which Freud gives insufficient attention to in his sociological writings (Bion, 2004; Jaques, 1953; Kernberg, 1998). As such, our psyches’ preoedipal, psychotic functioning plays an important role in a comprehensive diagnosis of group phenomena.5 Second, Freud's sociological works also fall short of the therapeutic aim of psychoanalysis. Freud's cultural solutions to human narcissism and aggressiveness are dim, and his study of groups focuses mainly on groups in regression. Interesting is how there is no direction for a potential cure outlined in Freud's sociological discussions.6 If the theory of drives and groups is to be properly psychoanalytic, then not only should it explain regressive groups, such a theory should also be able to inspire practices that may bring about (therapeutical) progress. This pitfall is also addressed by later analysts, for whom group formations are not necessarily regressive. Groups, when functioning well, perform great “work” functions—that is, when it allows group members to cooperatively advance conscious, productive aims (Bion, 2004; Rice, 1969). Besides, as Winnicott (2005) proposes, cultural practices may even be the only reliable place where adults can engage in great creativity to work through their losses. Considering the potentially therapeutic nature of groups thus seems indispensable for our search for meaningful social solutions to bitter social antipathies. Given how the shortcomings of Freud's works necessitate an understanding of the psychodynamics of psychotic defenses in group phenomena and a corresponding account of (group-based) affective transformation that points a way out, Melanie Klein's psychoanalytical study of early infants offers great inspiration. Although Klein is not a group analyst, her idea of the paranoid–schizoid position offers a psychodynamic account of how antagonisms and aggression can be read as paranoid defenses against anxieties, while her ideas of reparation and mourning in the depressive position offer insights into social practices that can transform these antagonistic affects. These ideas will prove useful in helping us outline in later sections how democratic stability may be achieved. Beginning her description of the newborn infant, Klein believes that the death drive and frustration of the infant's needs (for nutrition, intimacy, warmth, etc.) haunt the infant's psyche from the moment it is born. Transient bodily states of satisfaction and frustration, pleasure, and pain heavily color the infant's internal and external worlds. When it is fed and well-nourished, it experiences the world as all-good. Yet, when it is frustrated and experiences anxieties, the world becomes menacing, and it is confronted with a profound fear of annihilation that it struggles to defend against. Idealization of the good object: Toward the “good” side of the split world, the infant fantasmatically projects the good, life-preserving parts of the self onto good objects, idealizing them as the source of “unlimited, immediate and everlasting gratification” (Klein, 1984a, pp. 63–64). Then, the infant fantasmatically takes the good object back in by introjection. This returned good object then becomes an internal source of safety and confidence for the infant, paving the way for developing a stable ego. Demonization of the bad object: Facing the bad object, the infant projects displeasure and (inner) anxieties outward to the bad object, demonizing it in the process. Through projection, the infant sees the object's presence as the source of all its suffering, thus assigning the object the role of an evil persecutor in fantasy. Although such projective fantasy creates an image of menacing evil objects, it helps the infant deal with anxieties by externalization.7 Through projecting the bad, internal conflicts are externalized into a persecutory setting, effectively transmogrifying the inner fear of annihilation into “persecutory anxiety” that can be warded off by omnipotent fantasies of control or motor discharge. For Klein, the splitting and projective/introjective orchestration of the good and the bad in the paranoid–schizoid position allows the developing infant to master the internal fear of annihilation arising from the death drive.8 The paranoid–schizoid position is very much naïve and reality distorting. The good and bad objects are reality-distorting fantasies, and they function in the infant's mind as mere containers of the drives and anxieties of the infant.9 The mother, as with other objects in the world, is seldom just the fantasized witch or angel. Despite such naivete, constructing such unrealistic fantasies may be the only way for the early infant to tolerate large volumes of anxieties without complete paralysis and disintegration of the ego.10 Loss of ideals: Since the object now contains both the bad and the good, the idealizations associated with the good object the subject has clung on to so strongly in the paranoid–schizoid position will now prove themselves unrealistic and must be given up. Even if no actual object may be lost during this realization, the developing child is now still faced with great grief, for s/he will be required to mourn the loss of the idealized object that has hitherto been their only support against persecutory anxieties of the paranoid–schizoid position.12 Guilt for object: The pain of losing one's ideal is further magnified by the fear of being deserted and rid of the only sources of goodness. While in the paranoid–schizoid position, the subject has (fantasmatically) attacked the bad object, the reunification of the good and the bad in the depressive position confronts the subject with the horrible fact that what s/he has demonized and attacked (in the paranoid-schizoid position) is at the same time the only source of nourishing goodness s/he has relied on and can still depend on. Acknowledging one's aggression toward the object produces painful feelings of guilt. Faced with the lost ideal and the whole breast destroyed (“in bits”) in fantasy due to aggression, a new type of anxiety emerges—this time not paranoid, but arising from a concern for the object (Klein, 1984b, p. 269).13 Klein refers to this as “depressive anxiety”—anxiety surrounding the loss of ideal goodness and the guilt that one had irrecoverably destroyed the good object. In the early stages of the depressive position, depressive anxiety is very distressing, often forcing the subject to fall back on paranoid defenses or attempt manic denial (Klein, 1984b, p. 271).14 Paranoid defenses against depressive anxieties occur when anxieties (now both persecutory and depressive) force the subject to continue splitting, projection, and introjection in the paranoid–schizoid mode. The paranoid cycle continues, and the idealization of the good and demonization of the bad escalate in ever-greater intensities. Manic denial, on the other hand, limits the force of depressive anxieties by denying the feeling of loss and guilt and renouncing one's dependence on the (whole) object. In mania, the subject narcissistically acts to disparage and express contempt for the object to ward off any feelings of dependency and guilt. So long as the infant is held in a relatively loving environment, the need for manic and paranoid defenses is phased out in the child's normal development. This happens when the “tragic” anxieties of the depressive position15 (loss of ideal and feared loss of object) abate under a constructive metabolization of guilt in mourning and reparation. For Klein, depressive guilt need not be paralyzing, for it can be the source of the desire to repair the damaged relationship—to mend the object the subject has attempted to destroy. This, for Klein (1984b, p. 311), is observable in infants when they demonstrate a “profound urge to make sacrifices” and “a strong feeling of responsibility and concern for [damaged objects].” When such reparative tendencies take root in the infant's psyche and are confirmed by the infant's loving environment, the child may grow to trust his own reparative impulses and the loving goodness of the world (Klein, 1984a, p. 75). Love and concern from the infant's environment mitigate paranoid and depressive fears, helping it realize that the “love object inside as well as outside is not injured and is not turned into a vengeful person.” In this way, the developing infant becomes more capable of loving, confident about its capacities for moral concern, and can express “genuine sympathy” (Klein, 1984b, pp. 311, 342–343). No longer paranoid or guilt-ridden, the subject can also gain a more realistic perception of him/herself and the external world without the need for reality-distorting manic or paranoid fantasies. Love in the tragic realization of the depressive position will therefore provide room for subjects to feel secure about the (internal) good object and better align their perception with the world (Klein, 1984b, pp. 346–347). As the object becomes whole in the depressive position, paranoid aggression gives way to reality acceptance, increasing capacity for love, moral concern for others, and tolerance for inner anxieties. To argue for the depressive position as a developmental advance over paranoia may seem to promote a jump out of the frying pan into the fire. The depressive position is not a bed of roses. To call people exhibiting paranoid–schizoid tendencies to plunge into the depressive position seems only advisable when depressive anxieties can be worked through. Still, dangerous as it may be, central to Kleinian psychoanalysis is what can be called the productivity of the depressive position. If paranoia displaces our inner anxieties into ego-disintegration and unresolvable antagonization of the world, the depressive position offers a way out. As Butler (2020, pp. 86–96) argues, paranoia performatively creates social antagonisms ex nihilo when paranoid subjects initiate cycles of aggression by “pre-emptively” striking an object that it considers a threat in fantasy. Cycles of aggression may thus result from such “pre-emptive” strikes when the other strikes due to this provocation, making the paranoid fantasy a self-fulfilling prophecy. Reparation in the depressive position breaks paranoid cycles. In Klein's reading, guilt in the depressive position is not unproductive stasis. It compels self-reflection, halts automatic aggression and alerts us to repair the damages we have done, and allows us to begin constructive dynamics characterized by better, more mature relations with others. If cycles of reparation get off the ground in the company of good-enough objects (who can be an analyst, a lover, or even a community of solidarity), subjects will gradually be able to acknowledge the other's lack of ill-will and be confident of his/her ability to repair and love. Besides, they may also be able to internalize the goodness of others without the need for (over-)idealization. Confident about oneself and the world, reparation allows the subject to demonstrate openness and moral sympathy, tolerate the smears in the world, mourn the loss of his/her idealizing fantasies, and live better with their endogenous anxieties and others. There will be less need to act out aggression in fantasies of persecution. This is how the tragedy of the early depressive position transitions into a productive state of openness and tolerance. Depressive but reparative accomplishments, as we shall see, may become psychological prerequisites for stable democratic cohabitation. Klein (1984a, p. 233) made a general claim about how paranoid–schizoid defenses can reemerge when anxiety level increases in adulthood.16 As later analysts observe, paranoid dynamics are observed in adults not only in individual psychotic and borderline conditions, but also in socially sanctioned paranoid group dynamics. The Kleinian studies of groups of various sizes by Bion (2004), Jaques (1953), Menzies (1960), and Turquet (1975) give concrete descriptions of paranoid dynamics in groups and how homogenizing belonging in groups helps its members deal with psychotic anxieties. Jaques (1953), in particular, proposes that groups can have important psychic functions, serving as “social defense systems” against psychotic anxieties. Groups differ in the volumes of anxieties they can help the individual defend against. In work institutions (such as factories, schools, and workplaces), too much “politics” (like internal and external schisms of the group) may cause the group to fail its “work function.” On the opposite extreme, then, are political and activist groups, as the “work” aim of political participation is much less defined, and politics is the focus of these groups (Segal, 1997, pp. 133–134). When charged with great anxieties, groups can regress and “behave in a way that would be considered mad if any normal individual did the same thing” (Quinodoz, 2008, p. 149)—grandiose, paranoid, and narcissistic. These regressive tendencies are observable in toxic populism and revolutionary fanaticism. Anxieties and traumas across social classes define our age of neoliberal capitalism, and these anxieties feed into the formation of political groups.17 When these neoliberal anxieties and frustrations pile up, subjects may experience anxieties not dissimilar to those experienced by the early infant. The unconscious fear of annihilation may reanimate psychotic anxieties and unresolved trauma at the personal and collective levels.18,19, 18,19 Psycho-defensive (paranoid) functions of political groups may thus be animated. In a paranoid–schizoid mode, contemporary experiences of frustrations and anxieties are instrumentalized to fuel discourses of splitting—discourses of (self-)victimization (“we the oppressed People”)20 and social antagonization (“They the evil oppressors”). To be sure, one must not reduce discourses of self-victimization and antagonization to paranoiac fantasies. It would be a huge mistake to silence victims, explaining away discourses of victimization as mere paranoia. However, political frontiers seldom map the real causes of social misery. Aside from recent histories of political conflicts, antagonization and victimization frontiers may be heavily tainted by our unconscious fantasies and traumas. For instance, McAfee (2019), following Winnicott (1950), argues that sexist tendencies may be related to how our earliest dependency on our mothers creates a fantasy of a “fearsome woman” that men will seek to defensively overcome. In Winnicott's eyes, “actual domination” can be “derived from a fear of domination by fantasy women” (pp. 182–183). Besides, Balbus (2005, ch. 8) and Volkan (2020) propose how cultural upbringing may predispose subjects to select a racial or ethnic other as the container of unintegrated drives and anxieties.21 These cultural fantasies may be at work in the isolation of women, Blacks, refugees, and other minorities as scapegoats for social problems. Once a frontier is stabilized and the enemy selected, the mechanisms of projection, introjection, idealization, demonization, and manic denial begin. These may work to produce fantasies of conspiring, evil social others on the other side of the antagonistic frontier. Fantasies exaggerate the badness of the other and the goodness of the self—making the other appear extremely bad, dangerous, and conspiring, while the in-group (or a leader, projectively identified) is morally perfect, good, and victimized. At times, a leader of a political movement may deepen the paranoid moments by setting himself as an example, demonstrating infantile narcissistic enjoyment and manic contempt toward outsiders. Such is how paranoia functions in populist persecutory narratives that completely miss the real causes of social injustices, demonize anyone who disagrees, and idealize the moral purity/glory of the “us” group and “our” collective past. Paranoid defenses are not present only in xenophobic, racist, or sexist populisms. Indeed, even socially progressive movements can exhibit paranoid and manic dynamics. Even if these movements may have identified “real” injustices of the social world, orientation to political reality is seldom the sole basis of the movement.22 So long as the need for populist narratives partly originates from the people's need to deal with anxieties (i.e., their function as psychosocial defenses), they are also vulnerable to paranoiac defenses. Crimp's (1989) and Balbus's (2005) respective reflections of 1980s ACT UP activism and 1960s student protests underline how paranoia under political fanaticism may paranoically deny self-reflection, despair, and grief. As Balbus (2005) describes, piling, unmetabolized anxieties in the movement caused activists who continued their fight to idealize their movement by seeking to idealizingly purify themselves, redirecting their rage toward demonized others who do not share their idealized cause.23 Furthermore, when expressions of guilt and despair are barred by fanaticism, McIvor (2016, p. 13) describes how activists may become forever uncompromising due to the formation of a “rigid moral–political identity” based on the resistance. When the psycho-defensive function of victimization/antagonism takes over the movement, the psycho-defensive function of paranoia may overshadow its progressive goal. Paranoid defenses need not be bad, for one cannot underestimate the power of fear, exaggeration, rage, and distrust in fomenting resistance. Perhaps some form of paranoia, mania, and idealization (when not geared to narrow ideas of racial, sexual, or national glory) is essential for critical consciousness and inciting resistance to an unjust world.24 However, despite their utility, they may well hinder democratic progress. Indeed, since paranoid logics automatically displaces our inner anxieties onto ready-made bad objects, it has little room for the self-reflections required for the realistic perception of the social world. This bites emancipatory projects because when defenses prevent a reasonable perception of the (real) social causes of injustice, the movement may fail its emancipatory mission. Even in relatively democratic settings, paranoid defenses can derail progressive projects due to their tendencies to incite distrust and fear. Indeed, that paranoid group defenses displace anxieties means that people may continuously seek new objects for projection when old antagonisms do not allow as much venting of negative affects. In this case, political witch hunts may become the norm with the help that the inner anxieties of subjects are temporarily relieved, but, in the long run, creating deep divides in the citizenry, making it impossible for any lasting political alliance to form. Worse still, it is not hard to imagine how displaced anxieties can performatively create and deepen social antipathies that may play out in increasingly antidemocratic directions. Against this possible deterioration of democracies, it is therefore important to think of ways to deal with paranoia and antagonisms along properly psychological lines that weaken these defenses and transform antipathic affects into more productive ones. Klein's solution to paranoid defenses is the depressive position. Yet, seeing how paranoid–schizoid defenses are present in populisms and fanatical political movements, we can understand why this solution may be hard