Making Independent Decisions Together: Rational Emotions in Legal Adjudication

This article analyzes rational decision-making in court as an emotive-cognitive process formed in and through social interaction. Current theoretical perspectives have shown how emotion and thought are intertwined in the workings of the human brain but have seldom elaborated on the contextual and structural features of rational-emotional decision-making. I propose a model that maps emotional processes and emotional management demands to the temporally extended, stepwise process of rational-legal decision-making. I show that (a) the bounded structure of the decision-making process actualizes different emotive-cognitive complexes at different stages and (b) the demand for objectivity in rational decision-making calls for parallel emotional processes and subject positions to remain independent while sustaining social cohesion.


INTRODUCTION
If my judgments were guided by emotions, I would be revealed straightaway … I stand by my judgements, if I am the least bit uncertain, I acquit. (Judge Hjalmar Forsberg, author translation) These reflections were made by an appeals court judge in an interview for a Swedish daily newspaper (Lerner 2019), and they illustrate the tension between rationality and emotion that still prevails in today's Western courts. Judge Forsberg establishes that emotions have no role in judging and then argues that if he feels uncertain, he acquits. His uncertainty is thus an important guide for him in his judging-he needs to feel certain to convict. This apparent contradiction can be explained both by the persistent demand for dispassion in legal settings (Maroney 2011) and by a narrow understanding in which emotion is limited to overwhelming experience and clear expression. However, many emotional processes are subtle in how they are both experienced and expressed. Looking through the lens of emotions, deliberation (in court and elsewhere) can be seen as a process of going from uncertainty to certainty, feelings that guide behavior without necessarily drawing much attention and that are perhaps not even understood as emotions. But the present paper is not about what emotions are, it concerns what emotions do in rational processes. Emotions drive and orient our actions and interactions in the social world; they are the "motor" of social action, conceptualized as "readiness to act" (Frijda 1986). Another invisible tension in the quote above is the fact that in appeals courts (in Sweden and in many other countries), decisions are made by a panel. Independent decision-making is done together but almost always described in the singular. What emotional dynamics does independent joint decision-making entail?
In research from several disciplines, the traditional dichotomy between emotion and rationality has been replaced by an understanding in which emotion and rationality are seen as complementary (Damasio 1994;de Sousa 1987;Etzioni 1988;Illouz and Finkelman 2009) or even continuous (Barbalet 1998;Barrett 2017;Grossi 2019). Emotions are necessary to determine salience (de Sousa 1987;Morton 2010); to evaluate the importance of different alternatives (Damasio 1994); to motivate or evaluate knowledge acquisition (Arango-Muñoz 2014); and to motivate and facilitate a rational focus (Barbalet 2002). Current theoretical perspectives have shown how emotion and thought are intertwined in the workings of the human brain but have seldom elaborated on the contextual and structural features of rational-emotional decision-making (but see Barbalet 2002;Illouz and Finkelman 2009). Rationality is always performed within a social context that provides norms for how emotions ought to be experienced and expressed, depending on the situation (Hochschild 1983). For example, philosophers have argued that emotions play a vital role when people are seeking and evaluating knowledge (Arango-Muñoz 2014), but they have not placed the not (yet) knowing person in a social situation, which would allow them to understand how context can breed such different emotions as boredom, epistemic anxiety, or curiosity. A judge embarking on her eighth petty-theft trial of the day may have to work hard to fight off feelings of boredom after evaluating a series of similar stories, while a newly appointed prosecutor presenting her first case may juggle feelings of nervous anxiety and curious expectation: Will the evidence be enough for a conviction? Has she prepared for all possible objections from the defense? Here, I elaborate on how a contextual understanding can incorporate the way rational-emotional decisions are formed in and through social interaction (see also Rossner 2019).
Rationality is a broad concept that alludes to diverse forms of optimal problem-solving (Brunsson and Brunsson 2017). Thus, to define what I mean by rationality and delineate the social setting in which it takes place, I use the example of legal decision-making in court. Legal decision-making embraces bureaucratic rationality that builds on information gathering and processing, comparison and weighing of alternatives, and objective reflection "without regard for persons" (Weber 1948:215). 1 In legal decision-making in court, objective and impartial reasoning is stressed and linked to rules; the criteria for how to gather and process information and weigh alternatives should be defined in advance and followed every time to ensure equal treatment. The idea that rational decision-making is a structured process is in line with Pierre Bourdieu's (1986:816) argument that the legal procedure itself legitimizes the judgment and that "the internal logic of juridical functioning which constantly constrains the range of possible actions … limits the realm of specifically juridical solutions." In effect, judicial decision-making is not constituted by one decision but by several smaller decisions made over an extended period of time and by several decision-makers.
In the present paper, I show that mapping emotional processes and emotion management demands to the temporally extended, stepwise process of rational-legal decision-making enables us to unfold the rational emotions involved because different emotive-cognitive complexes are actualized at different stages. Furthermore, rational decision-making is often linked to a solitary, emotionally detached arbiter. If we position rational decision-making in its social setting where decisions are made over time and by several people, then the arbiter is no longer alone but immersed in social interaction. I show that the demand for objective and impartial justice calls for parallel emotional processes and subject positions to remain independent while sustaining social cohesion.
In my effort to incorporate how organizational and ritual arrangements produce rational decision-making, I use Randall Collins' (2004) model of interactional ritual chains (IRCs), which allow me to investigate the emotively-cognitively structured process of rational-legal decision-making. However, IRCs were designed to account for emotional energy (rather than cognition) as a common denominator for rational action (in a broad sense) (Collins 1993). Here, my focus is on rational decision-making (in a restricted sense), and I adjust the IRC model to this bounded process to start developing a model that can integrate discrete emotional processes and multiple subject positions as well as link them to specific phases in the IRC.
A process model of rational decision-making can contribute theoretical knowledge on the social interactional dimensions of decision-making and lay the foundation for more fine-tuned empirical studies meant to identify the discrete emotional processes and emotion management demands of different decision-making phases. Furthermore, the institutions of modern bureaucratic democracies depend on public trust for their ability to perform rational, objective decision-making. This core ideal can be one reason for the persistence of the emotion-rationality dichotomy within the law: How can we retain a measure of sound rational-legal decision-making if or when we abandon the rationality versus emotion dichotomy? A mapping of emotion to specific tasks and phases can be used to identify when discrete emotive-cognitive processes advance sound/rational decision-making and when they do not. In this article, I begin exploring this puzzle by unfolding the emotive-cognitive process of making rational decisions in legal judgments.
Next, I turn to the field of legal decision-making and emotion, where the strains between rational and nonrational processes are further problematized. Second, I present ritual theory as an efficient tool for investigating the emotive-cognitive process of bounded rational decision-making. Third, I investigate the tension between rational independence and social cohesion, showing how their integration demands parallel emotional processes, conceptualized as moral emotions, performative emotions, and epistemic emotions. Fourth, I present a real case to illustrate the consecutive and parallel emotional processes depicted in the previous sections. This fourth section shows how independent decision-making in a group setting requires the display of solitary, collective, and empathic subjects to balance independence and cohesion. Last, I propose a conceptual way forward by developing a process model of legal decision-making as a bounded IRC that is embedded in cultural and relational structures.

EMOTION AND LEGAL DECISION-MAKING
The epitome of rational agency is the judicial system, as can be seen in its main guiding principle: with impartiality and objectivity. This view connects to the ideal of positivistic objectivity, which implies that legal professionals should and could view and review the world at a distance, putting aside their emotions and values. The ideal of "judicial dispassion" (Maroney 2011) has been contested by scholars from a range of disciplines, creating a separate research field (Bailey and Knight 2017;Bandes and Blumenthal 2012;Grossi 2015;Maroney 2016;Wiener, Bornstein, and Voss 2006). Law and emotion scholarship covers a multitude of perspectives and aspects of law and legal work, such as the role of emotion in law-making and legal doctrine (Bianchi and Saab 2019;Nussbaum 1999); expectations concerning defendants' and victims' emotions (Konradi 1999;Rose, Nadler, and Clark 2006;Weisman 2016); and the emotion work of legal professionals (Bergman Blix and Wettergren 2018;Kolb 2011;Roach Anleu and Mack 2005).
When we narrow our focus to legal decision-making in court, two central perspectives emerge: information processing and empathy (for reviews see Bandes 2009;Feigenson and Park 2006;Henderson 1987). Both perspectives engage with the problem of agency for rational decision-making: (How) can a feeling subject make objective decisions (Grossi 2019)?
Regarding the first perspective, social psychological theory predominates, showing how emotion influences judgments by affecting how people process information-how they attribute responsibility and blame and, thereby, the severity of sentencing (Feigenson 2016). The impact of emotion in a legal setting is complex and includes the arbiters' respective moods when coming to court (Semmler and Brewer 2002), their understandings of the evidence (Salerno and Bottoms 2009), and their interpretations of the emotional state of the victims/plaintiffs or defendants, for example, in relation to evaluating credibility (Weisman 2016;Wessel et al. 2006). Although the biasing role of emotion in rational information processes is central to social psychological theory building in this field, there is also emerging research arguing that awareness of and reflection on emotional processes could promote objective decision-making (Gendron and Barrett 2019;Grossi 2019), the implication being that silencing emotions per se disrupts rational reasoning (Maroney and Gross 2014).
Regarding the second perspective, the importance of understanding and evaluating other people's experiences and motives to legal decision-making has inspired theory development on the role of empathy in adjudication. Empathy is a debated concept in law because it implies that the interpretations involved in legal decision-making go beyond pure syllogism (Bandes 2009). Interpretation is done, by necessity, from a subject position (Abrams 2010) and thus involves the risk of biasing empathic perspective-taking. To avoid biased empathy, Martha Nussbaum (1996) develops the idea of narration in interpreting a case. The adjudicator should approach a case as if it were a piece of fiction, that is, engage in all of the characters while being aware that it is not their own lives that are at stake. This implies that objective interpretation demands full subjective immersion, including thick description (Henderson 1987), and systematic use of the imagination to overcome resistance to empathizing with, for example, the motivations underlying gruesome deeds (Del Mar 2017).
Previous research has offered us important insights into various aspects of the role of emotions in legal decision-making, particularly its vulnerabilities, but the field would benefit from engaging in "theoretical debates within emotion theory itself" (Little 2000:974). Studies on decision-making have largely focused on individual jurors or judges, and there is a need to broaden the scope to include other legal actors and the dynamic of collectives (Bandes and Blumenthal 2012). What is more, the predominance of psychological and philosophical theories also tends to result in the neglect of structural and interactional dimensions (Bergman . Nevertheless, the legal field, particularly the courts, represents the archetype of bureaucratic rationality, with a strictly regulated agenda of decision-making that can serve as an "extreme case" or process model for other types of rational decision-making. Next, I turn to interaction theory to integrate a process perspective on rational-legal decision-making into my model (Collins 2004). Interaction rituals, if successful, advance social cohesion, and because the demands for objective and impartial decision-making in law emphasize independence, I focus on how rituals that further cohesion can run parallel to the performance of independence.

INTERACTION RITUALS
It may seem strange to use ritual theory when trying to understand (independent) legal decision-making, as it is commonly used to stress the importance of emotion to social cohesion and consensus. When people come together in rituals, their physical proximity and joint focus can create collective exuberance and shared moral consciousness, thus erecting boundaries to nonparticipants and shared within-group solidarity (Collins 2004;Durkheim 2008). As I establish below, these features are to some extent present in the court interactions required for making legal decisions. 2 The emotion management needed to perform legal decision-making in a dispassionate and objective manner demands "a mood of solidarity" (Fleck 1935 in Barbalet 2002:144), allowing participants to agree on a shared understanding of what is going on and how to move forward.
Rituals generate sacred symbols, which reinvigorate transient collective feelings and thus stabilize social solidarity (Durkheim 2008). A symbol can be an item such as a totem in a religious ritual (Durkheim 2008); it can be a code of statutes, a robe representing impartial justice, or the idea that the individual is sacred in profane rituals (Goffman 1967). Putting the presentation and protection of the (sacred) self at the center of rituals stresses the emotional tension inherent in social interactions, the push of the constant risk of humiliation due to failed presentations (Goffman 1967), and the pull of feelings of pride and belonging when the "social bond" is maintained (Scheff 1990). Rituals thus encompass the many ways in which people align themselves with each other in mundane, often routine activities in everyday life (Goffman 1972). Here, I draw on Collins' (2004:48) definition, according to which rituals include four premises: face-to-face interaction, awareness of the boundaries between participants and nonparticipants, shared focus of attention, and shared emotion or mood. A successful ritual incorporates all four components, while less successful or failing rituals are lacking in the first three components and, thereby, produce less solidarity or even alienation and conflict.
The emotional underpinnings and impacts of interaction rituals also link to cognitive motivation, or Verstehen (Weber 1978). Verstehen (meaning-making) refers to the intersubjective process of understanding our own actions in relation to how we make sense of other people's behavior. Symbolic interactionist theory argues that people act (and make decisions) in accordance with how they understand a situation, and that their understanding arises from the interaction itself. The link between interpretation and interaction accentuates the fact that meaning-making is dynamic and can change depending on how a situation unfolds (Blumer 1969). This process perspective on social action serves two important functions. First, it weds emotional and cognitive functions. People's constant need to adapt and align their interpretations of a situation emphasizes how emotions arise in, are exchanged within, and are shaped by social interaction (Hochschild 1990). Interaction theory investigates how social situations evoke emotion and how emotion orients and informs social action. Second, it highlights the constraining impact of the interaction order (Rawls 1987). All social interaction builds on an inherent, albeit preliminary, agreement about how to interpret the unfolding situation-an agreement that both enables and constrains action: "Social constraints [ … ] are ways in which situations unfold, so that participants feel they have to behave in a certain way, or make amends for not doing so" (Collins 1988:57).
Furthermore, situations are not produced in a vacuum; people enter into them with a set of expectations that frame their interpretations. Collins introduces the notion of IRCs that bridge situations through emotional energy, that is, an extended emotional tone that moves from one situation to the next. High emotional energy entails feelings of pride and confidence that, in turn, influence the expectations and space a person claims in upcoming interactions. In this way, emotional energy can reproduce but also challenge or change how people grant social status (Collins 2004). The prosecutor in the example above who felt insecure in court might eventually expect to be and feel confident with being at the center of attention after being accorded that status in hearing after hearing. This last example stresses the fact that expectations related to a situation also depend on institutional frames. The court procedure builds on elaborate rules, and the participating legal professionals know what is expected of them and how the ritual will evolve. With growing experience, law professionals habituate the social constraints of the legal ritual; their thinking and feeling align with an "emotive-cognitive judicial frame" (Bergman Blix and Wettergren 2018)-a behavioral script that demands subtle emotional expressions and sensitivity to emotional cues in order to sustain an ideal of dispassion. Frames are situational in nature; people walk in and out of frames and manage intersecting or multiple frames during their everyday lives (Goffman 1974). The framing influences how an interaction unfolds. For instance, a prosecutor might joke about a quirky judge during a break with her colleagues and express respect for the same judge at the hearing, adjusting her interpretations and feelings to the different frames.
Collins' model of interaction rituals has been used to study diverse social activities such as the formation of social activist groups (Summers-Effler 2002), children's conflicts in school (Jensen and Vitus 2020), and jury deliberation (Rossner 2019). The present article continues this tradition by moving the analysis beyond the solitary adjudicator studied in most previous research on legal decision-making, to a focus on interaction and how subjects' orientations adjust to the unfolding interactions. It adds to the field by introducing a bounded IRC, where looking at the unfolding of a chain of consecutive, regulated interactions helps us understand the emotional constituents of rational decision-making.

RATIONAL EMOTIONS IN A LEGAL DECISION-MAKING PROCESS
The legal decision-making process in court is fundamentally interactional, both during the information gathering phase of preliminary investigations and hearings, and often in the evaluations made during deliberations. To understand legal decision-making, we need to include interactions and view the subjects as being in continuous dialog with others.

A Bounded Interaction Ritual Chain
The interactions involved in the unfolding of a case in court are part of a strictly bounded process; the rituals follow a predefined order, and the decisions made along the way are restricted by the type of interaction and earlier delineations of the case. A generic sequence of formalized interactions in a legal IRC includes four instances: (a) preliminary investigation and preparatory hearings; (b) main hearings; (c) deliberations; and (d) delivery of a judgment-with (b) to (d) potentially being repeated at the appellate level. 3 In the following, I limit the professional interaction to include the two types of legal actors who are required to make objective decisions: judges and prosecutors. Other types of lawyers represent clients and thereby need to put loyalty above objective reasoning (Flower 2018). The unfolding of cases through consecutive interaction chains demonstrates how the final decision (judgment) is the end product of many smaller decisions along the way, where previous decisions influence and delineate later ones (Bourdieu 1986). It also shows how a decision at one point is measured against future decisions. If, in a criminal case, the court decides to detain a suspect (hearing [a]), the prosecutors gain confidence for the next hearing (another [a] or [b]), increasing the likelihood that they will present the case efficiently. The testing or checking nature of these preliminary hearings thus emphasizes the prosecutors' need to empathically tune into the judge's point of view to check whether their decision and presentation of facts accord with the judge's understanding of the case.
In the trial phase, the prosecutor has made her decision about how to present the case and needs to be self-confident if she is to convince the judge of the accuracy of her decision. At the same time, the judge needs to critically assess the presented interpretation of the evidence, actualizing emotions such as curiosity and skepticism. The deliberation (c) can be an individual activity, but in many systems, different constellations of judges, lay-judges, or jurors interact to make the final judgment. Deliberation can actualize diverse emotional processes, but the overarching process is the route from uncertainty to certainty (elaborated in the section "Collective and Independent Emotional Processes"). The judgment (d) can give rise to pride (Bogira 2005), insecurity, or defensive anger among prosecutors because it constitutes an evaluation of their indictment. In a previous study, judges in lower courts recounted that they seldom get feedback on their judgments from the people involved, but that appealed judgments could give rise to emotions of pride when their judgments were sustained or feelings of insecurity and shame when they were revised. Moreover, even experienced judges described how a few consecutive revised appeals could create long-lasting insecurity in their own ability to make sound judgments (Bergman Blix and Wettergren 2019b).
The rituals described thus far are the formal interactions involved in the decision-making process, but one important feature of the legal IRC is that the formal interactions are interspersed with intraprofessional and to some extent interprofessional informal interactions. Planning, reflecting, and strategizing are fundamental to upholding the legal decision-making process and include efforts to build self-confidence and instrumental focus prior to formal rituals, emotional reflection to evaluate interpretations between hearings, debriefing to manage anger or shame in the event of failure, and pride following success. In the long run, repeated IRC failures on the part of individual legal professionals can threaten the legal decision-making process, as driving this process forward effectively requires some level of emotional energy (Collins 2004) and ease (James 1879).
As we can see, even in objective decision-making-where the goal (at least in theory) is limited to making a correct decision (rather than reaching a specific outcome)-the motivations underlying decision-making are emotionally charged; the hopes or fears related to decision outcomes and how these feed into future decisions need to be taken into account (Zeelenberg et al. 2008). Next, I investigate the potential conflict between rational independence and social cohesion in legal decision-making.

Rational Independence and Social Cohesion
The main argument in ritual theory is that people who participate in successful consecutive rituals develop a shared moral order and social solidarity-an outcome that would seem to contradict objective, that is, independent, decision-making (Weber 1948). On the interactional level, a legal hearing includes three of the four components of a ritual (Collins 2004): face-to-face interaction, boundaries to outsiders, and a joint focus. Do hearings also incorporate the fourth component of shared emotions? By teasing apart the emotive-cognitive complexes of collective emotions, Christian von Scheve and Sven Ismer (2013) link the first component, face-to-face interaction, to emotional convergence through facial mimicry and emotional contagion, speculating that these nonconscious processes are fostered in collectives with a shared expressive repertoire. A foundational part of habituating the "judicial frame" referred to above is mimicking and aligning one's expressive repertoire with the demands of a rational, dispassionate judiciary (Bergman Blix and Wettergren 2019b). This means that face-to-face interaction could foster emotional convergence between legal professionals in court, while the judicial frame's demand for a subtle display of emotion would obstruct convergence with the laypeople present. In generic terms, the more a professional frame deviates from the broader emotional regime (Reddy 2001), the more likely it is for emotional convergence to occur in interactions among professionals, and the less likely it is for emotional convergence to occur with people outside the profession. This would apply particularly to formal interactions with strict display rules.
The habituation of the judicial frame also links cognition, or meaning-making, to emotional convergence through shared beliefs and values. It is more likely that the third component, joint focus of attention, will give rise to shared emotions if the participants' interpretations of the situation are similar. This is accomplished both through the individual process required to join the ranks of legal professionals and through the collective process of participating in particular interaction chains, where the latter eventually produces social solidarity (von Scheve and Ismer 2013). 4 In sum, court rituals can promote the fourth component of an interaction ritual-shared emotions-and thus develop into a shared moral order and social solidarity, but these emotions are presumably restricted to the involved legal professionals (Skolnick 1967). Social solidarity is commonly conceived of as inherently positive. The typical negative outcome of cohesion is that it does not include everyone: A feeling of we depends on a feeling of them. However, one foundation for arriving at a correct legal judgment is that the different parties are unconnected and independent, making social solidarity a complex issue. On the one hand, legal professionals need to share a moral order if they are to agree on how decisions should be made and to collaborate in moving a case forward (Barbalet 2002;Bergman Blix and Wettergren 2019a); on the other hand, independence in decision-making is fundamental both constitutionally (top-down) and in interactional practice (bottom-up). 5 In an effort to tease apart this potential conflict between independence and solidarity among legal professionals, I now look more closely at the emotional processes involved in the legal IRC. In real-life decision-making, the emotional processes of the interaction order and the emotions linked to decision-making are intertwined (Bergman Blix and Wettergren 2018;Goffman 1961), but here I differentiate between them for analytical purposes. The epistemic process of decision-making requires independence, while the interaction order is inherently collective.

Collective and Independent Emotional Processes
As described in the section on ritual theory, the moral emotions of pride and shame make up the push and pull of maintaining the interaction order (Goffman 1956). Moral emotions provide the boundaries for the rational decision-makerboundaries that position the (legally and procedurally) correct process above a potentially attractive outcome. 6 Legal training and professional socialization create an emotional commitment to objective justice. Legal professionals feel proud about representing the law when they act and keep their independence intact (Bergman Blix and Wettergren 2018). In legal decision-making, moral emotions, in turn, encourage a rational/professional demeanor that comprises emotional detachment, seriousness, and determination (Mack and Roach Anleu 2010). These performative emotions are contingent on professional collaboration. From a dramaturgical perspective, an objective performance depends on all legal professionals playing their part in court. Even though they play conflicting roles, argue for or take conflicting decisions, they depend on each other when processing a case through the bounded IRC. Previous research has shown that this requires emotional attunement and situated empathy, interprofessionally as well as with laypeople (Bergman Blix and Wettergren 2019a; Roach Anleu and Mack 2005), but it may also trigger irritation and anger when the other participants fail to collaborate (Maroney 2012). This is by no means a complete list of the emotions linked to the legal interaction order; instead, it provides a taxonomy for understanding the mechanisms of a collective rational ritual. Moral emotions commit participants to the ritual, while performative emotions guide how the ritual unfolds.
The second set of emotional processes in a legal IRC relates to decision-making that requires a high degree of independence. Philosophical research on emotions has identified a set of emotions that are denoted "epistemic" due to their particular intentional character (Arango-Muñoz 2014; de Sousa 2009; Morton 2010). Epistemic emotions are specifically linked to cognitive evaluation by providing information about the quality of one's knowledge and motivating certain kinds of mental action (Arango-Muñoz 2014). Researchers are still discussing which emotions can be considered epistemic, but these include feelings of certainty, skepticism, curiosity, epistemic anxiety, and uncertainty. These feelings all influence deliberation; for example, curiosity opens the door to further examination, while "the feeling of certainty freezes inquiry" (de Sousa 2009:146).
The philosophical underpinnings of epistemic emotions emphasize their function in individual mental processes, but I argue that a social interaction perspective can further our theoretical understanding of how epistemic emotions are influenced by and aligned with social rules and norms. For example, epistemic and moral processes can go hand in hand; the feeling of certainty can be related to both morality and epistemic knowledge. In a legal setting, epistemic certainty linked to moral evaluation is considered suspect and can cause problems in panels consisting of both professional and lay-judges, where professional judges need to manage lay-judges' sometimes blunt expressions of moral certainty ("Of course he did it, he's done it before"). Furthermore, in the bounded IRC in court, a judge would break the rule of impartiality if she were to express the feeling of certainty before all the evidence has been heard in a trial. This procedural rule may, thus, support the epistemic feelings of interest in and curiosity about the presentation of evidence. The bounded IRC model allows us to investigate how different phases and tasks in the legal decision-making process actualize and encourage the different emotional processes linked to deliberation. In the earlier example of empathy, previous research has taken a strong normative perspective, in which empathy, if equally distributed, is perceived as unequivocally positive (Del Mar 2017;Nussbaum 1996) or negative (see further Abrams 2010). However, different levels of empathic perspective-taking can be efficient and expected in different phases of the decision-making process (Wettergren and Bergman Blix 2016). Consider a prosecutor working on an abuse case. At some point during the investigation, she will examine photos of injuries. She needs to take in information about a revolting crime to evaluate the cause of injury. What does the photographic evidence say? Was it the blow the accused struck or the ensuing fall that caused the injury? Here, a nonempathic stance is important to remaining objective. In a later phase, when the same photos are seen while the weeping victim is present in the courtroom, empathic perspective-taking is vital to ensuring an efficient examination. Furthermore, the later phase can sometimes develop into sympathy based on feeling the full suffering of the victim or relatives (Bergman Blix 2019). In the moment, the prosecutor's experience of sympathy might jeopardize an objective decision, but it might improve her abilities in the long run, by preventing her from becoming "case hardened." The example of empathy shows how different phases of the legal process can actualize different emotive-cognitive complexes and different subject positions, such as a solitary subject evaluating evidence in the office and an empathic subject examining a witness in court. Empathy per se is thus neither good nor bad for rational decision-making. Instead, it needs to be understood and evaluated in relation to its function in a particular task or phase of the bounded IRC. When the prosecutor evaluates photos in her office, empathic involvement on her part might jeopardize the defendant's right to an objective evaluation (Bright and Goodman-Delahunty 2006), while a nonempathic stance during examination of the victim could jeopardize the access to robust evidence and the victim's sense of having been given a fair trial (Konradi 1999).
To demonstrate how the two sets of emotional processes can unfold in a social IRC, I now illustrate the argument using an empirical case.

From Speculation to Argument: A Case of Division of Property
In this case, a Swedish panel of three judges sit in trial and thereafter deliberate on a civil case of division of property. In civil legal systems, it is common to have a panel of judges in certain lower court trials; in common law systems such as in the United States, panels of judges preside over appellate and supreme courts on both the state and federal level. This observation extract comes from an ongoing international research project that is investigating, using ethnographic methods, the emotive-cognitive process of judicial decision-making in courts and prosecutors' offices in Sweden, the United States, and Italy. We have shadowed prosecutors and judges while they prepare cases, observed them at hearings and deliberations, and interviewed them before and after hearings, amounting to approximately 200 cases thus far.
I focus on how the process of going from uncertainty to certainty (independent decision-making) and empathic perspective-taking (cohesive interaction order) can unfold in the consecutive interactions of a trial and deliberation and in parallel during deliberation. The parties in this trial are a divorced couple, both represented by lawyers. One of the judges on the panel, Nils (age 40+), is the reporting judge, meaning he has prepared the case and knows the most about it beforehand. He has communicated with the lawyers and received and read all the documentation in the case. The most experienced judge, Kajsa (45+), acts as the chair in the trial. In Sweden, as in other civil law systems, being a judge is a career profession, thus judges are typically younger than in common law systems, where they are selected from the ranks of lawyers. The third member of the panel is Maria (30+), a newly appointed judge. Before the trial, the three judges and a clerk gather in the empty, locked courtroom talking about the case. They all express their preliminary skepticism about the claim of the plaintiff (a woman suing her former husband), and Nils emphasizes in an agitated voice "let's just keep it to property settlement so we don't have to take up all the other shit, the stocks and why he's a scumbag and such!" Kajsa responds that she usually lets people talk … "so you need to show indulgence. But to keep you in a good mood, I won't introduce the panel today," she adds with a laugh. (fieldnotes before trial, property division) In an interview with the researcher a couple days later, Kajsa points out that she did not want to put Nils in a bad mood that day, so she refrained from introducing the judges by name at the start of the trial. She thinks introductions are important but knows that many judges, including Nils, find them irrelevant and inefficient.
The demeanor of the panel changes dramatically when the parties are called in. All judges shift from energetic backstage interaction to calm, neutral stone-faced appearances. The trial goes on for an hour, with the presentation of evidence and closing statements from both sides. We stop shortly here to look at an exchange after the lawyers representing each party have given their closing statements.
The plaintiff suddenly stands up and approaches the defendant decisively with a bunch of documents in her hands. She throws the documents on the desk in front of him exclaiming "Ha!" The defendant brushes the papers off the desk so that they scatter on the floor shouting "I don't want these!" The plaintiff returns to her seat. During the episode, Nils and Maria sit completely still, Nils with a look of controlled fury, while Maria has a neutral expression. Kajsa leans forward toward the parties and say with a calm voice, "You know what, you have to take that outside." She repeats these words several times before they are back on their seats and quiet again. The trial ends and the parties leave. All the judges breathe out, and Kajsa expresses doubt about how she managed the outburst. Maria gives her praise: "you did it just right, you let them talk, but interfered when it became too heated!" (fieldnotes of trial, property division) The posthearing deliberation among the three judges begins. It is clear that Maria and Kajsa have not made up their minds at this point, while Nils is already certain about overruling the plaintiff's claim.
Maria sits in silence with her face screwed into a worried look. Kajsa turns to her and exclaims: "You look very concerned!" Maria develops on her concerns [about interpreting the case in favor of the defendant], while Nils looks her way but slightly downward, frowning with irritation. When Maria stops, Kajsa turns to Nils about a legal issue she is uncertain of. Nils takes over and soon his irritation seems to fade and he looks contented and relaxed. Maria starts to discuss a new aspect of the claim at a slow pace, as if she were testing her line of argument. Nils [with a calm voice and distinct smile]: "I think that you're mixing up different parts. If we disallow [the will] in its totality, then we will have to rewind the process … but read up on it, we all should read up on it. I'll start writing, and I intend to keep it short." Both Kajsa and Maria say they "spontaneously agree" about the proposed conclusion, but that "certain aspects complicate [the decision]." Kajsa: "It's only that you have to find support for your feeling." (fieldnotes of deliberation, property division) On a general note, we can observe the swift and fluent change in demeanor between the more relaxed, energetic backstage before and after the trial, and the formal, strict performative emotions during trial, including stone faces and low expressivity of any emotion so as to display impartiality (Roach Anleu and Mack 2019). Even though all three judges are formally equal in status, Kajsa signals her status as chair before the trial and gives Nils a heads up that she will let the parties speak. Moral emotions provide the boundaries for the rational decision-maker, and Kajsa thinks that, to ensure a fair trial, it is important to let the parties speak. It is her right as the chair to decide how to run her trial, but her status position depends on the other judges' collaboration, so she compromises by promising Nils she will refrain from introductions, thus expressing a collective subject. Nils accepts this and maintains his stone face during the trial, but it is clear that demeanor and emotional experience do not necessarily go hand in hand. Nils is agitated before the trial and subtly signals his strong irritation with the parties during the trial, as well as with Maria at the beginning of the deliberation.
Kajsa, acting as an empathic subject, picks up on Nils' irritation and defuses a potential conflict by asking him a legal question, where he, as the referent in the case, can shine and return to his good mood. Kajsa in effect negotiates some bargaining room for Maria. When Nils is in a good mood, he allows Maria to express and develop her uncertainty about how to interpret the evidence of the case.
Maria continues her noncertain line of thought, expressing calm curiosity that emphasizes her presentation as a solitary subject. Nils finds her thoughts irrelevant, but he has now calmed down and displays two parallel processes. He dismisses Maria's argument ("you're mixing up different parts"). If we follow Nils from before the trial, he already expressed certainty about how to evaluate the evidence. During the trial, he was the judge expressing the most (although very subtle) irritation with the parties. We can see how his feeling of certainty becomes locked in by feelings of disinterest and resistance to questioning that certainty ("we don't have to take up all the other shit," "we have to rewind the process") (de Sousa 2009), thus displaying a solitary subject. But his solitary display is mixed with his acting as an empathic subject -he smiles to defuse his rejection-and as a collective subject -he indicates that he too needs to acquire further knowledge ("we should all read up on it"), thereby managing Maria's potential feelings of rejection and trying to ensure a unanimous judgment. Nils' parallel display of a collective and solitary subject is reflected by Kajsa and Maria, who also mix their display of collective subjects, accepting Nils' suggestion ("spontaneously agree"), while persevering their presentation as solitary subjects in an independent process ("certain aspects complicate"). In the last exchange, Kajsa again displays an empathic subject by managing Maria's potential feelings of anger or sadness about not having gained support for her interpretation ("have to find support for your feeling").
The extract from the hearing also shows the complexity of ritual interactions. They can be successful or fail to different degrees and for different sets of participants. The judges in this case walked away with a moderate reproduction of group solidarity, while the divorced couple, at one point, were approaching violence. The chief judge Kajsa managed the situation and restored the mood, or at least the parties' expressive behavior. The episode left Kajsa worried, but she was reassured by Maria, who empathically assured her that she had handled the situation well. Kajsa's display of an empathic subject in relation to Maria at the end of the deliberation can be understood in light of this exchange.
As we can see, the bounded structure of the legal IRC, which demands strict adherence to a procedurally correct and premeditated process, prompts interactions and gives tight boundaries for their implementation. In the example, the consecutive interactions before the trial, during the trial, and during deliberation all hinged on one another, while actualizing different emotive-cognitive processes and different emotion management strategies. Furthermore, the parallel processes of arriving at independent decisions and preserving social cohesion were intertwined. As chair Kajsa in particular made a clear effort to ensure that the less experienced judge, Maria, could arrive at an opinion of her own. Kajsa, and eventually all three judges, maintained the ideal of three independent decision-makers, while also supporting social cohesion and a unanimous judgment.

TOWARD A BOUNDED PROCESS MODEL OF LEGAL DECISION-MAKING
In the present article, I have investigated the IRCs of legal decision-making (Collins 2004) and how these chains are bounded by institutional constraints and performed as a behavioral script within an emotive-cognitive judicial frame (Bergman Blix and Wettergren 2018).
To develop a full-fledged model of the rational decision-making process, we need to embed the bounded IRC in the relational and cultural practices of the larger society in which they take place (Wettergren and Bergman Blix 2021). These cultural and relational dimensions need further development that goes beyond the scope of the present article, but I want to bring up some examples that stand out in the case illustration. First, the dramatic shift between the judges' demeanor during the trial and that before and after emphasizes how the Western emotional norm system or "emotional regime" (Reddy 2001) of rationality requires a silencing of emotions, including demands for stringent individual emotion management-that is, the varyingly conscious adjustment of feelings and expressions. These demands do not only influence demeanor but also the decision-makers' space for emotional reflection. Although the rational-emotional regime is part of an overarching Western norm system, emotions seep into the courts in different ways, depending on how they are perceived in society as a whole. Preliminary analyses in our ongoing project indicate that the emotionally expressive Italian legal regime allows more deliberate and critical use of emotional clues in the decision-making process than does the restricted and subtle emotional regime in Sweden.
Second, the relational structures of, for example, gender, class, and ethnicity, as well as formal organizational hierarchies, feed into the cultural-emotional regime of a specific society. Even though Sweden is considered a gender-equal country in an international comparison, it is not surprising that the female chair put considerable empathetic effort into calming the male judge and keeping him in a good mood, while ensuring that the more inexperienced female judge had room to voice her concerns. These are both high-status, intellectually adept women, but they nevertheless spend more time and energy as collaborative and empathic subjects than their male counterpart does (Hochschild [1989(Hochschild [ ] 2003. Third, historical research has illustrated how relational arrangements in society as a whole also influence legal ideals (Bailey and Knight 2017). Impartial detachment is valued today, but even recent historical examples, such as the "revolutionary justice" of the early Soviet state, instead portray moral emotion and empathy as being essential to legal decision-making (Vasilyev 2017). Perhaps more subtly, an international comparison of interaction rituals in different contemporary legal settings can uncover variations in the fundaments of objective decision-making. Although present in all Western legal systems, features such as formality, morality, and independence can have different weight and enter at different points in the legal process across different countries or systems. As mentioned in the empirical example, the reference to and importance of formality stand out as especially prominent in Sweden compared with the other countries under study.
These examples all need to be backed up by further empirical studies, but the intention here is to show how these societal structures feed into the micro-level emotive-cognitive frame, including emotional motivations and behavioral expectations/rules, as depicted in the above example of a civil dispute.
The decision-making process unfolds in a bounded interaction chain that provides the emotional energy necessary to move on to the next step (Collins 2004), but it also incorporates discrete emotive-cognitive processes to narrow the decision-making and move it forward. If we use the example of the epistemic process of going from noncertainty to certainty among judges and prosecutors, we can note that legal decision-making expects the process toward certainty to evolve at specific points in the IRC. To ensure objective decisions, prosecutors should remain uncertain until the end of the preliminary investigation, while judges are to remain uncertain until all evidence has been presented during a trial. These prescribed emotional processes are likely to instigate strategic emotion management (see further Bergman Blix and Wettergren 2019b). It is interesting to note that, in many social situations, not knowing/feeling uncertain gives rise to shame due to loss of status (Kemper 2011), but the demand for independence in legal decision-making influences the interaction and emotion management strategies chosen. As we saw in the quote at the beginning of the paper, a judge can take pride in his uncertainty. One can expect these relational dynamics to be of importance in analyses of the collective process of deliberation in courts, which can include actors of different status (lay-judges, jurors, or a panel of judges). Thus, the model opens the door to a new niche of empirical studies that incorporate how interactional and contextual factors, such as power relations and professional norms, influence people's inclination to develop and hold on to certain epistemic feelings and associated subject positions while ignoring others (Terpe 2016).
Furthermore, the incorporation of actors (subjects) into the interaction chain (a deviation from Collins' original chain, which starts at the interaction level) enables a detailed examination of how discrete emotional processes can differentiate between actors with different stakes in the process and how actors combine and weigh subject positions so as to remain independent while maintaining the collective interaction order. For example, the requirement of unanimous decisions in jury deliberations positions the collective emotions of the interaction order over autonomy (Rossner 2019). In jury deliberation, the emphasis is on being judged by one's peers, rather than being judged by an objective (independent) arbiter.
We may expect that when there are demands for objective decision-making that includes rituals in some form, we will find measures intended to keep social solidarity in check. These measures can take on different forms: doctrinal measures, such as the formality of procedural rules that reduce the room for dialog in favor of consecutive monologues, which effectively decrease the potential of "rhythmic entrainment" to intensify a shared mood (Collins 2004:48); organizational measures, for example in Sweden, where the former practice of always letting the same prosecutor litigate with a specific judge has been abandoned in favor of randomly assigning prosecutors-a structure that reduces recurrent interactions with the same professionals; spatial measures, such as the architecture of the court room and separate backstage space for the legal professionals, which minimize informal interactions; and temporal measures, such as splitting up the hearing of a case into several shorter sessions, instead of one longer consistent interaction (Bogira 2005). In sum, even in the case of rational/independent decision-making, a bounded IRC model can analyze how actors organize the decision-making process so as to integrate independent decision-making within a collective social order.

CONCLUSION
In the present article, I have begun to develop a process model of rational emotions in legal decision-making. I have argued that the construction of rational objective decision-making can be analyzed as a bounded interaction chain that unfolds in a prescribed order, actualizing different emotional processes at different stages. A rational IRC incorporates parallel emotional processes and subject positions to ensure social cohesion and independent decisions in individual interactions, as well as demands for particular discrete emotions at different phases in the chain, differently timed for the respective professional actors. The various emotional processes and acting subjects are not separate and discrete but combined in interaction ritual chains. The model stresses the notion that rational decision-making is not a dichotomy between emotional bias and detached fairness, but a delicate balance between having sufficient solidarity to ensure that the ritual is upheld and having enough space for participants to demonstrate their independence. Future empirical studies mapping emotional processes and subject positions in bounded interaction ritual chains could identify vulnerable instances-instances when the construction of an objective process goes off course due to the impact of emotions that do not match the expectations of a particular stage or interaction and that, thus, run the risk of biasing the decision-making.

ACKNOWLEDGMENTS
I thank Alison Gerber, Terry Maroney, Susie Scott, Richard Swedberg, two anonymous reviewers, and the editor Scott Harris for constructive comments on earlier versions of the manuscript. This research has received funding from the Swedish Research Council (2016-01218) and from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (Grant Agreement No. 757625,JUSTEMOTIONS). A Fellowship at Stockholm Centre for Organizational Research (Score) at Stockholm University, Sweden, during 2020 gave me time to write. NOTES 1. Although rules and logic are vital for bureaucratic decisions, decision-making by necessity includes some level of chance: "Something that has already been determined in all respects cannot be decided" (Luhmann 2005:85). Solving an equation does not include decision-making: The correct answer results from following the correct procedure. In bureaucratic decision-making, the outcome may vary when the rules are followed incorrectly (procedural wrongs) or because processing information and weighing alternatives involve different possibilities. A decision always includes more than one possible outcome. 2. Two central principles guiding court proceedings are that the evidence should be presented orally (principle of orality), and the judgment should be based on the facts presented during the trial (principle of immediacy). These principles in effect demand three of the four features of an interaction ritual: face-to-face interaction; boundaries to outsiders; and shared focus. 3. Most cases do not go all the way to appellate level. Different legal systems allow for various alternatives to end cases at earlier stages, for example in plea bargains, but the process applied here exists in this generic form in all Western systems. 4. Here, individual process refers to a process that has evolved separately from the other participants in a particular interaction. Having gone through law school likely implies having developed similar interpretations of a situation in court, but this understanding has not come about as a result of the particular group interaction. In contrast, if someone at a hearing makes a joke and everyone laughs, the shared understanding creates a "we mode" that furthers social solidarity. 5. It is also interesting to note that when applying ritual theory to the court procedure, social solidarity seems to be an outcome of rituals (at least in theory) even when it is not the intended goal of the rituals.