1  At first glance, the notions of negotiation and adjudication have little, or nothing, in common. Most textbooks devoted to dispute resolution depict negotiation and adjudication as two clearly distinct ways of dealing with conflict. Negotiations are reached through discussions between the parties or their representatives, without the involvement of a third party. They are voluntary and non-binding. Adjudication, however, implies the presence of a neutral third party, usually an independent judge or a designated arbitrator, who gives a decision which is binding unless or until revised in litigation or arbitration. There are, in the general view, further subtle differences. Negotiation implies bargaining and compromise, while adjudication can be defined as the interpretation and application of a pre-existing internationally relevant set of rules.

2  The aim of this entry is to challenge this clear-cut distinction by exploring the scope and limits of negotiation theories in order to explain some specific stages of international adjudication. The intention is not to reduce the complexity of adjudicative deliberations to pure negotiations, but to investigate signs of a negotiation process in international adjudication situations.

3  The underlying hypothesis of this research is that some adjudication decisions result from a situation that has all the characteristics of a negotiation. From this perspective, negotiation theories can enrich our understanding of the choices made by judges or arbitrators. These theories seem to be particularly enlightening in two specific circumstances. The first of these is in the increasingly frequent cases of amicable dispute settlement, or règlement amiable, which imply both a negotiation process between the parties and the presence of an adjudicator. The second concerns the negotiation process inside judicial chambers. This process is inevitable in legal contexts where adjudicators deliberate on the basis of the unanimity rule. It also happens where the majority rule applies. Even if a judge or arbitrator can write a dissenting opinion, and is in the minority, they can implicitly enter into a type of negotiation that can mitigate the final outcomes, for instance, the amount of damages. In all these cases, the ‘parties’ are not the parties to the dispute, but the adjudicators themselves. This entry does not focus on both these circumstances, ie amicable settlement and deliberative adjudication processes. Knowing that amicable settlement among the parties is a more intuitive and widely studied issue, this entry concentrates on the negotiation process inside judicial chambers.

4  Drawing a parallel between adjudicative deliberations and negotiation processes raises several questions. Can we speak of a balance of power in a court chamber setting? Do such deliberations involve all sides making concessions, although not necessarily to an equal degree? What is the impact of the adjudicators’ personalities on negotiations leading to a jointly adopted decision? Are the existing typologies that categorize various conflict styles, including accommodating, avoiding, competing, cooperating, and compromising as discussed below, relevant to adjudicators and their teams? Does culture affect the outcomes of the negotiations taking place in international adjudications? If so, to what extent?

5  These questions indicate new avenues for studying existing practices in international procedural law. They allow us to bridge two disciplines that are intimately intertwined: international law and political science. Numerous articles are devoted to the links between these two disciplines (see eg Maveety, 2003; Pollack, 2013; and International Balance of Power; Political Theory of International Adjudication). Many of these studies discuss the fundamental epistemological distinction between law and political science. This analysis pursues a much more precise and modest goal. It aims to generate a better understanding of judicial decision-making, by questioning the significance of negotiation theories when these are applied to particular adjudication practices.

6  This aim implies a particular research posture. It attempts to move beyond the competing views on adjudication: on the one hand, the ‘legal model’, according to which adjudicators decide cases based on the facts of the case in light of the relevant legal provisions in a neutral and objective way (Cook, 1977, 551) and, on the other hand, the attitudinal and strategic models in which the adjudicators are merely—if not exclusively—motivated by policy. The research posture that is adopted here is not over-idealistic vis-à-vis adjudicative deliberations; nor does it arrogantly depict such deliberations as a simple bargaining game. The purpose is to consider both the legal reasoning, which remains the fundamental basis of adjudication decisions, and the ways in which representations of the facts and interpretations of the law are debated and even negotiated. In this sense, the goal is not to evaluate judicial behaviour, but to create bridges between disciplines in exploring further the process of deliberation.

7  In terms of methodology, this study is exploratory rather than prescriptive. Since such procedures are governed by principles of confidentiality, we cannot fully ‘unpack’ the black box of deliberations (Secrecy of Deliberations). In the absence of documents that shed light on internal deliberations, it is difficult for scholars to depict the interactions among judges in a court. However, this limitation does not prevent us raising and addressing new questions (see Kornhauser and Sager, 1986). Focusing on collegiality as the ability to generate compromises does not imply a ‘pathologization’ of the process. The intent is not to distinguish between ‘good’ and ‘bad’, ‘fair’ and ‘biased’ practices. It is, rather, to stress the ambivalence of deliberative processes which are neither positive nor negative per se—meaning that their value fundamentally depends on the objective pursued. At this initial stage of the research, the question is not ‘how should it work?’, but ‘how does it work?’. Or, in other words, how do judges and arbitrators actually make decisions?

8  The analysis is divided into six parts. The first clarifies the concept of negotiation. The others focus on the five main theoretical approaches to negotiation (Bercovitch and others, 2008; Rosoux, 2013). The behavioural approach emphasizes the attitudes of the parties. The strategic approach is inspired by game theory. The processual approach highlights the importance of procedures. The structural approach stresses the notion of power. The cultural approach insists on values, beliefs, and even memories. As we will see, these approaches are not incompatible, but rather, complementary. While each of them emphasizes specific aspects of adjudicative deliberations, they all feed into one another. Only by combining them can we approach a comprehensive understanding of the practice.

9  Negotiation is both a timeless process and an increasingly frequent practice. Negotiation is an enterprise as old as humankind. From time immemorial, negotiation has governed people’s lives. In this regard, it is emblematic that a biblical scene already depicts the bargaining process between Abraham and God in regard to Sodom and Gomorrah (Genesis 18). Similarly, the negotiations between Greeks and Trojans as to the return of Helen of Troy failed and tragically provoked the Trojan War. Moving beyond myths and legends, all human societies, including the most traditional ones, seem to be somehow anchored in a form of negotiation (Strauss, 1978).

10  The negotiation process is both ancient and highly topical. The end of East/West antagonism favoured the worldwide development of the market economy. The financial dynamics of mergers and acquisitions helped to undermine national boundaries. From a technological perspective, the development of information and communications technologies has enabled us to spread news almost instantaneously, so that more and more actors, such as companies, non-governmental organizations, religious actors, and even individuals, have become involved in international politics. From a political vantage point, the processes of regional integration, the proliferation of international organizations, and the significant global challenges facing us have radically modified the attitudes of States, which were traditionally much more autonomous. All these factors and more have led to a situation that is often qualified as an era of ‘permanent negotiation’.

11  In the specific field of law, negotiation is sometimes depicted as the ‘lawyer’s principal occupation’ (Teply, 2005, 1). Lawyers negotiate to create legal relationships, effect transactions, resolve legal disputes, or favour settlement agreements. Can we, however, consider that the notion of negotiation is relevant to understanding the practices of international adjudication? To address this question, it is crucial to clarify the notion of negotiation and to determine which aspect of international adjudication might possibly be related to it.

12  Negotiation is often defined as a process by which two or more parties interact in order to accommodate their conflicting interests in a mutually acceptable solution (Dupont, 1994, 19). This definition stresses the four main components of any negotiation. (1) The first concerns its relational aspect: negotiating inevitably involves interaction between two or more parties, and, as such, more or less formal communication. In his novel Saint-Germain ou la négociation (Prix Goncourt in 1958), Francis Walder insists subtly on the dense network of human relationships. This Belgian novelist was a former military officer who was involved in diplomacy throughout his career. In his view, it is not possible to give a true account of the negotiations between the French crown and the Huguenots that led to the sixteenth century Peace of Saint-Germain-en-Laye without exploring this human dimension. As one of his characters explains, ‘in this intermediary phase of the talks, we had to search for a human contact … I was repeating to myself: “Nevertheless, these are men, men, men, therefore sensitive to what is humane and the destiny of [our] negotiations depends on [our] sensibilities”’ (Walder, 2001, 64–5).

13  International adjudication is not radically different; the human dimension is also decisive. Interaction takes place not only between litigating parties as amicable settlement, but also and above all between judges or arbitrators themselves. Thus, the fact that adjudicators are adopting decisions collectively means that those responsible for preparing the bench’s written opinion must convince a majority of their colleagues that the crafted opinion they present is the appropriate one. It is therefore not surprising that as early as 1922, Charles G Haines considered that a judicial decision ‘involves factors, personal and legal, which carry us to the very roots of human nature and human conduct’ (Haines, 1922, 116). From this perspective, adjudicative deliberations can easily be understood as a back-and-forth communication between individuals ‘who know, like, and depend on each other’ (Richman and Reynolds, 1996, 273).

14  (2) The second element lies in the divergences that result from disagreements between the parties about facts, aims, or methodologies. In this respect, the most critical element is not always the existence of objective differences, but the belief that these divergences exist. As the American psychologist Jeffrey Rubin explains, ‘what ultimately matters in negotiation, are the perceptions, beliefs, and assumptions that protagonists bring with them into the fray, rather than any objective measure of difference’ (Rubin, 1991, 220–1). In this regard, international adjudications can involve differences of opinion. They can even become divisive. Such divergences, actual or perceived, do not usually concern objectives or methodologies. Adjudicators share in principle the same objective: delivering quick, fair, and impartial decisions. In terms of methodology, the legal reasoning is generally agreed upon. Adjudicators are applying ‘complicated precedents to even more complicated facts’ and strive ‘to get the right answer’ (Wittes, 2001, A25). However, adjudicators are not strictly bound by the diktat of a predetermined set of abstract rules. As Anand explained in 1965, ‘law is not an exact science or a series of IBM or push-button machines from which answers come tumbling out when right levers are pressed. There are two sides to almost every issue that comes before the Court and a group of five, seven, nine or fifteen judges, with varied backgrounds and beliefs, with different legal, social, economic and political philosophies, cannot be expected to think alike’ (Anand, 1965, 790).

15  Adjudicators may differ regarding the interpretations to be given to the facts presented to them, on the existing rules of law or the choice of the relevant legal precedents and principles that are the most appropriate to the cases. In reasoning and debating means and ends, adjudicators have a certain discretion. This ‘privilege’—which is also a ‘responsibility’ (Pannier, 1981, 623)—does not mean that the outcome of the deliberation depends on a purely subjective sense of appropriateness. This criticism is particularly acute with reference to the United States, where the existence of ‘ideological camps’ among judges has often been highlighted by scholars and practitioners themselves. In this study, the notion of divergence does not refer to this kind of bias, but rather to the initial diversity of perspectives that characterizes the members of a bench. The process through which these perspectives can eventually balance out one another is a form of negotiation, however limited.

16  (3) Despite these divergences, parties in a negotiation have common interests. The third main component of any negotiation situation is a certain level of interdependence. None of the parties sitting at the table can reach a satisfying outcome without the others. In international adjudication, judges and arbitrators are de facto interdependent. Their mutual aim is to apply the law. As Judge Edwards noted, ‘when I speak of a collegial court, I do not mean that all judges are friends. And I do not mean that the members of the court never disagree on substantive issues. That would not be collegiality, but homogeneity or conformity, which would make for a decidedly un-healthy judiciary. Instead, what I mean is that judges have a common interest, as members of the judiciary, in getting the law right, and that, as a result, we are willing to listen, persuade, and be persuaded’ (Edwards, 2003, 1645). In other words, ‘judging is an inherently interdependent enterprise’ (Edwards, 2003, 1645).

17  (4) The last characteristic of any negotiation situation is that parties seek to reach a mutually acceptable solution. This kind of outcome does not exclude the possibility of an unbalanced agreement (Zartman, 1997). In international adjudications, divergent views among judges or arbitrators can often end up as dissenting opinions. The mere possibility of issuing dissenting opinions alleviates the pressure related to the absolute need when deliberations are based on the unanimity or consensus rule, and the relative need when deliberations are based on the majority rule, for the panel of adjudicators to negotiate. However, the opportunity to express dissent, even anonymously, does not prevent judges and arbitrators from addressing the arguments expressed by the members of their bench. In collegial courts, each judicial voice matters. That means that until a final judgment is reached, adjudicators participate as equals in the deliberation and can therefore attempt to negotiate the outcome of the deliberation.

18  For centuries, practitioners have been exploring the variables that explain the success or failure of negotiations. In 1716, for instance, in a different context, French diplomat François de Callières (1645–1717) elaborated on this issue in a long letter, which later became a book: On the Manner of Negotiating with Princes. Along the same lines as Niccolo Machiavelli’s The Prince, de Caillières described the key qualities of a successful negotiator: listening rather than speaking, utter discretion, and ‘a patience which no trial can break down’ (de Caillières, 2000, 25). To this expert, ‘the best way in which the negotiator can establish good relationships is to prove to both counts that their union is of great mutual advantage’. In his view, the building of long-term relationships with one’s partners cannot begin by ‘promises which cannot be redeemed’, and ‘success won by force or by fraud stands upon a weak foundation’ (at 76).

19  Since then, a multitude of how-to-negotiate books have provided straightforward tips and tools for effective negotiation. All of them stress the same advice: preparation is the secret to success. This is also true in international adjudications. Admittedly, procedures can vary greatly, for instance, regarding the possibility of issuing separate opinions. Yet, despite this diversity, the growth in the number of international adjudication bodies, and the expansion of their jurisdictional mandate, the quality of decisions taken in all international courts and tribunals largely reflects the quality of the preparation phase.

20  To give only one example, the deliberation that takes place within a Chamber conference at the Court of Justice of the European Union (‘CJEU’) directly depends on the quality of the preliminary report drawn up by the Judge-Rapporteur (Judge-Rapporteur: Court of Justice of the European Union (CJEU)). The more precise and robust the report in its discussion of the relevant issues of fact and of law raised by the case, the less chance there is that negotiations will become intractable. The specific role of legal secretaries (Référendaire: Court of Justice of the European Union (CJEU)) is particularly crucial. In providing assistance to the members of the court, they can have a critical influence on the drafting of preliminary reports and judgments. In this respect, it is interesting to note that the Court of Justice has introduced a series of measures to reduce the time spent dealing with cases. The stated objective of these measures is to identify and prevent procedural delays, which sounds perfectly reasonable. It might however be useful to question the consequences for preparation of these efficiency gains. Such an analysis would require consideration not only of the average case duration or the number of cases heard by the Court, but also of the length and quality of the preparation phase (Case Management). Having said that, we should not exaggerate the role of this initial stage. As has already been evoked, five main approaches are usually emphasized to explain the outcomes of any negotiation. These can easily be summarized by the following key factors: behaviour, strategy, process, power, and culture. The next sections will examine the extent to which these approaches help to give us a better understanding of international adjudication.

21  One of the most influential approaches to international negotiation emphasizes the relational aspect of any negotiation. Contrary to laboratory experiments, actual negotiations occur between parties who have a substantial history and anticipate a future relationship. In international adjudications, the principle of collegiality and the subsequent interdependence between members of the same bench makes this dimension particularly significant (Behavioural Approaches to International Adjudication).

22  The behavioural approach explains the negotiation outcome by concentrating on the psychological and cognitive dimension of negotiation. Personality psychologists often highlight the importance of individual styles and dispositional traits. In addition to a wide variety in modes of behaviour, key personality factors are often summarized by the Five-Factor Model of personality based on broad categories. There are five categories: extroversion—are the parties sociable, talkative, and assertive?; agreeableness—are they flexible, trusting, and cooperative?; conscientiousness—are they responsible, achievement oriented, and organized?; emotional stability—are they secure and confident?; and openness—are they curious, imaginative, and broad-minded? (Goldberg, 1993; Lewicki and others, 2006, 396–7).

23  The traits of the ideal negotiator may sound inspiring. However, they could suggest the existence of a rigid dichotomy between those with the qualities needed to negotiate ‘well’ and those without such qualities. Research carried out into the psychological and cognitive aspects of negotiating is not limited to the qualities of negotiators who are born gifted. It shows that personality is often less important than the cognitive ability of negotiators (Barry and Friedman, 1998).

24  From a cognitive perspective, practitioners and scholars examine the influence of (mis)perceptions, biases, and emotions. Perceptual errors such as stereotyping, selective perception, or projection often impact negotiation processes in a detrimental way. It seems therefore fundamental to understand how negotiators use information to make decisions. Far from being perfect processors of information, they can commit errors based on irrational escalations of commitment; overconfidence, ie the certainty that one is right and is adopting the best strategies; mythical fixed-pie beliefs, ie approaching opportunities as zero-sum situations where a gain for one party is a loss for the other; the law of small numbers, eg if negotiators extrapolate from their own limited experience; and self-serving biases, eg overestimating the degree of support or consensus that exists for one’s own position (Lewicki and others, 2006, 144–55).

25  Beside perceptual and cognitive biases, few international negotiators can escape dealing with expressions of anger, resentment, fear, or even hatred, especially in a war or post-war context. Negotiation is not, as we could expect, a rational and calculated exchange of information. Rather, dealing with one’s own emotions and with those of the other party is a major task which must be taken on in order to reach balanced and lasting agreements. In the view of some psychologists, the emotional intelligence of negotiators, defined as their ability to perceive and regulate emotion, is often a sine qua non condition to break a deadlock (Mayer and others, 2000).

26  To what extent can this approach be applied to international adjudicators? Handbooks and guides giving tips on how to become a good legal negotiator emphasize an ideal profile. This is almost always the same: the ideal negotiator must be ‘thoroughly prepared’, ‘legally astute’, ‘realistic’, ‘analytical’, ‘creative’, and ‘adaptable’ (Teply, 2005, 55). Other checklists mention the importance of being ‘experienced’, ‘honest’, and ‘self-controlled’ (Williams, 1984, 12 and 77). As well as this way of differentiating between ‘effective’ and ‘ineffective’ legal negotiators, they also draw a distinction between ‘cooperative’ and ‘competitive’ negotiating styles. Most practitioners, however, tend to avoid such binary categories, and describe effective negotiators as, to some extent, hybrid: they are both competitive when it comes to maximizing their own side’s returns, and problem-solvers or cooperative when it comes to maximizing the joint returns achieved by the parties (Craver, 2016). This attitude is sometimes captured by the formula ‘WIN-win’—‘big win for your side, little win for theirs’ (Shapiro, 2015).

27  Despite the success of such formulas, they do not help much in determining the outcomes of adjudicative deliberations. Among the research carried out from a behavioural perspective, it is worth mentioning the insights of social and organizational psychology scholarship that focuses on the behaviour of small groups, and emphasizes the influence of leadership on judges’ attitudes. Since the pioneering work by G Schubert (1965), F Murphy (1966), and S Ulmer (1971), there has been a long tradition of considering adjudication fora, such as courts or even specific chambers, as groups of individuals who ‘interact on a regular basis, have affective ties with one another, share a common frame of reference and are behaviorally interdependent’ (Levine and Moreland, 1994, 306; Martinek, 2010, 73–83).

28  Relationship dynamics are particularly important when the members of the group interact often. They are therefore paramount in the case of chambers that gather several times a week. Repetition of the interaction allows the adjudicators to think about long-term incentives and not only immediate gains. If we consider the specific example of an ongoing relationship between the members of the same chamber, it is also useful to consider the existence of previous and future relations between them. In this regard, individual trajectories and professional or even social networks can be decisive in terms of reputation.

29  The reputation of international adjudicators can be depicted as twofold. Reputation in the eyes of the general public depends on a consistent attitude, which is harder to detect in the absence of dissenting opinions. Reputation among colleagues is not really based on consistency. It depends, rather, on how members of the bench remember their past experience with each other, during or even before serving time at court. This aspect raises the question of social recruitment (Cohen, 2010). It also reminds us that the interdependence of members of the same bench explains the particular importance of maintaining constructive relationships. Over time, a ‘don’t-rock-the-boat’ mentality can even come to characterize certain international adjudications (Richman and Reynolds, 1996, 324).

30  Various studies have demonstrated the long-term impact of reputation. Five main reputation types are usually emphasized: liar/manipulator, tough but honest, nice and reasonable, cream puff, and no reputation (Glick and Croson, 2001). Once developed, reputations are hard to change, and especially difficult to ‘repair’. An adjudicator who has developed a reputation as a distributive hardliner early on will find it very difficult to modify this.

31  As early as 1921, the American Judge Benjamin Cardozo described courts as ‘microcosms’ where ‘deep below consciousness are other forces, the likes and the dislikes, the predilections and the prejudices, the complex of instincts and emotions and habits and convictions, which make the man, whether he be litigant or judge’ (Cardozo, 2010, 68). As Cardozo continued, although judges tend towards an ‘ideal of objective truth’ and often ‘like to figure to [them]selves the processes of justice as coldly objective and impersonal’, they cannot get rid of what he called ‘inherited instincts’: education and expectations in terms of human relationships (2010, 10). Obviously, ‘the great tides and currents which engulf the rest of men do not turn aside in their course and pass the judges by’ (Cardozo, 2010, 68–9). This description confirms the significance of the behavioural approach. However, can it explain the whole range of deliberative dynamics?

32  The strategic approach, rather than emphasizing the relational aspect of negotiations, insists on the fundamental strategies of distributive or integrative bargaining processes. In this view, the outcomes of negotiations do not merely reflect the attitude, motivation, and personality of the actors, but rather the offers and demands made by the parties in order to obtain concessions. Game theory, economic theory, and social psychology are brought in to study comparative costs. This homo economicus model of rational choice is often used to analyse the game of judging (Posner, 1993).

33  This school of thought examines the motivations of the various players and asks fundamental questions about their choices, the conditions affecting those choices, and the level of trust between parties. It presents actors as motivated by preferences and outcome-oriented. Their main objective is to maximize their pay-offs. One illustration is the study of the prisoner’s dilemma. This has led to the development of a theory of cooperation based on a form of conditional trust, otherwise known as tit for tat. In this view, the most compelling long-term strategy is to start by cooperating and in the next phase to respond in the same way as the other party (Axelrod, 1997).

34  The strategic approach cannot predict the actors’ behaviour, however, nor the outcome of their interactions. The negotiations themselves are based on calculations and perceptions that make their progress and conclusion unpredictable. This observation reflects the delicate balance between notions of rationality and predictability. The fact that the parties strive to act rationally, and that in retrospect the results of their negotiations also seem rational, does not mean that the results can be calculated in advance. Historians have repeatedly evoked this point: in considering decisions taken by actors, it is key to keep in mind the context of these decisions, made up of ambiguity and risk rather than certainties and foregone conclusions.

35  In 1959, Glendon Schubert applied game theory to judicial behaviour in his book Quantitative Analysis of Judicial Behavior. Schubert, known as the creator of the attitudinal model, argues that policy preferences and attitudes are the primary determining factor of judicial decisions. In this view, judges and arbitrators act strategically to attain their goals and to maximize their own ideological preferences. That means that adjudicators are not only looking for a sound basis for decision-making, but also, and above all, for a strategy to achieve their preferred result. Sophisticated techniques, mathematical models and regression analysis tend to measure attitudes and ideologies in order to calculate instrumentalist rational self-interest. Since empirical evidence demonstrates that judges act in accordance with their personal ideologies, the attitudinal model tends to predict, in a probabilistic way, judicial behaviour.

36  The strategic model, in contrast, does not see judges as individual actors making decisions independently, but as interdependent actors who are strongly influenced by their panel colleagues. In this regard, international adjudicators make choices not only in line with their own preferences. They also have to take into account the preferences of other actors. Interactions within each panel are therefore critical to understanding adjudicators’ choices (Spiller and Gely, 2008). In a chamber setting, for instance, the repetition of the game explains the development of long-term strategies between members of the same bench. Thus, judges who know that they will meet regularly and that they depend on each other usually favour package settlements, which help them achieve their objective on key issues. To package issues effectively, judges may vote against their own preferences in cases that are not perceived as top priority, hoping that their colleagues will reciprocate in future cases on which they do feel strongly (Epstein and Knight, 1998). This kind of panel-effect does not prevent the existence of distributive processes. Nonetheless, the vast majority of authors emphasize the impact of collegial deliberation and dissent aversion. As in the attitudinal model, the strategic model of judging relies above all on quantitative data to assess judicial decision-making.

37  This methodology has faced one recurrent criticism: that empiricists have attempted to ‘computerize’ judicial decision-making and to reduce it to a matrix. They have not, it is said, seriously considered the qualitative aspects of judicial interaction, such as the parties’ arguments and the contents of the deliberations (Tacha, 1995, 585–6). Besides this limitation, three main issues could be raised with regard to the premises underlying these two models. The first regards rationality. The strategic approach usually assumes rationality. However, how can we deny that individual interests and values are shaped by a complex range of social, institutional, and historical factors (Davis and others, 1993)? The second is related to the measurement of ‘preferences’. Since judges generally do not publicly discuss their ideological preferences, how can we avoid reducing actual preferences to projected preferences? The third concerns the assumption that judges have ‘perfect information’. How can we establish that they really have perfect information about each other’s pay-offs? These questions show why other approaches are useful to fully understand the complexity of the deliberative process.

38  A third approach addresses neither the actors’ behaviour, nor their respective strategies, but rather considers the negotiating process itself. It tends to explain how the interaction between parties changes over time. After examining actual negotiations, distributive or integrative, scholars closely observed the type of communication used by the parties throughout the process, and concluded that negotiation proceeds through distinct stages.

39  The processual approach divides negotiations into successive phases. It differentiates between the initial contact and the phases of information, argumentation, and adjusting positions, concluding with the phase of shaping the agreement. The value of this distinction is to pinpoint the functions, tools, and qualities required in each phase. Yet, in the field, the negotiating process rarely progresses in a linear or ideally ordered manner. There is much back and forth, and it is neither irreversible nor systematically structured in clearly identifiable phases.

40  Such irregularities explain why some authors prefer to highlight three main stages of preparation—formula—details, rather than five (Zartman, 1988). From this standpoint, the first major stage is the pre-negotiation phase of identifying the conditions to be met in order for discussions to progress further. Whether by setting up a communication channel or by gathering the information needed to make a diagnosis, these conditions are identified: demands considered high-priority by each party and those most likely to be so for the other party. Once the diagnosis has been made, the parties generally enter the second major stage of establishing jointly how the agreement should be formulated. Long discussions then take place to determine the terms of the exchange. Will it involve negotiating resources for money, live prisoners for dead bodies, territory for a secure withdrawal? All these formulas are imaginable. It is very important, therefore to agree on the type of exchange that is most capable of satisfying all parties at the table. The third and final phase deals with the details of the transaction. This is the time for fine-tuning positions, calibrating concessions, and specifying the terms of the exchange leading to finalization of the agreement. The care taken in each of these phases determines to a great extent whether it will be possible to implement the agreement reached. As well as proclaiming this stage-by-stage negotiation model, the processual approach also emphasizes the significance of procedures (Albin and Druckman, 2014). It shows that choices concerning language, timetables, mandates, or agendas determine the results of a process.

41  In international adjudications, various stages can be identified, from the drafting of opinions to the final decision. During the deliberative process, each adjudicator presents his or her ‘position’. Concessions can be made. While judges argue the merits of their positions, they can make offers and counteroffers, proposing or demanding changes in drafts, circulating revised drafts. The give and take of collegial deliberation has been depicted by many practitioners. For instance, the American judge Harry Edwards placed particular emphasis on the significance of this process:

As judges engage with their colleagues on a case, from oral argument and conference to opinion drafting and revising, their views evolve out of an interdependent push and pull. They do not misrepresent or suppress their ‘sincere’ views to further a ‘strategic’ purpose. In fact, it is specious to distinguish one judge's ‘sincere’ views from another's when all are working as a group to fashion as correct, accurate, and clear a holding as possible. If the end product looks different from what a judge had in mind at the beginning of the process, that fact reflects the very nature of the group process in which each judge can only contribute to a group product that is ultimately attributable to the court (Edwards, 2003, 1661).

42  This description evokes the stages, such as initial draft, oral arguments, and opinion revisions, that characterize any mutual adjustment and concession-making process. The collegial process determines the evolution of what the negotiation experts call the ‘bargaining range’, ie the spread between parties’ resistance points or, in other words, the difference between the preferred opinions of each party. Nevertheless, one major difference between the deliberative process and most international negotiations lies in the absence of a BATNA—best alternative to a negotiated agreement—for adjudicators. Unlike States or companies, judges and arbitrators cannot accept a deadlock. They must reach an agreement that can be accepted by the majority of the bench members since there is a duty to decide. In the field of international politics, and especially in peace settlements, most negotiation processes fail. The Israeli-Palestinian and Syrian cases are emblematic in this regard. They are reminders that negotiators and mediators can decide not to decide because they are not able to reach an agreement. Judges cannot.

43  From a processual perspective, the structure and the rules of the process are crucial. As in any negotiation, the agenda setting, the language that is chosen, and the number of members of a particular adjudication can affect the ultimate outcome in a decisive way (Ulmer, 1971). In this respect, the distinction between judicial and arbitral bodies is critical. Judicial bodies pre-exist the dispute and remain relatively stable. Members of the same court—or in some cases the same chamber—meet every week, if not every day. The repetition of the game is a decisive aspect of the collegiality principle. In contrast, arbitral panels are nominated after the dispute arises, and disbanded after the award is rendered. While some arbitral bodies take place within institutionalized frameworks, which explains the existence of some repeated players in international arbitration as well, ad hoc arbitration is a one-time exercise. This repeated or less-repeated dimension of the process has a direct impact on the degree of interdependence between adjudicators and therefore on the dynamics as a whole.

44  Beside this major distinction, other procedural elements should be highlighted, such as whether or not dissenting opinions can be issued, the nomination and appointment of international judges and arbitrators, the degree of transparency in the identification of individual judicial positions, whether or not an amicable settlement is possible, and the terms of office served by judges and arbitrators (on some of these criteria, see Dunoff and Pollack, 2017). Among these elements, the key point is whether or not dissenting opinions can be issued. The negotiation process that takes place between the members of an adjudicatory body, to take only one example, is inevitably influenced by whether decisions are made (1) under the rule of unanimity or majority, (2) with or without dissenting opinions, and (3) with public or anonymous dissent. These procedural specificities allow us to differentiate between international adjudications, according to a certain gradation in terms of negotiability. The rule of consensus decision-making implies a high degree of negotiation, while if decisions are taken by a majority, with an option to issue anonymous dissents, there is not the same pressure to reach an agreement acceptable to all. Thus the CJEU, where the judgments are issued per curiam—‘by the court’, without including separate dissenting opinions—seems to be in an intermediate position: the majority rule leaves some leeway for those who do not accept the decision proposed by the majority of the members, but the absence of dissenting opinions maintains a certain level of pressure on those who disagree with the decision.

45  This tension can be summarized by a continuum between two situations. At one end of the scale are all adjudications based on the majority rule, favouring quicker decision-making processes, resulting in clear decisions and in potential shifts in case law. Minority positions are actually not taken into account, even though they can have an influence—if only by inciting to strengthen the reasoning. At the other end of the continuum are adjudications based on the unanimity or consensus rule, which implies longer decision-making processes leading to decisions that might seem more complex particularly when two positions are reflected in the same decision, and to a more stable case law. In some circumstances, the complexity of a decision is the only way found to overcome a deadlock. Rather than conceding, parties can agree to reach a deal based either on silence—with all controversial paragraphs or sentences being progressively erased, which often contributes to the complexity of a decision—or on a ‘constructive ambiguity’, a term first employed in the 1970s by Henry Kissinger. According to the former American Secretary of State, who observed and influenced the world of diplomacy for decades, ambiguously worded decisions can create opportunities for advancing the interests of all parties to a negotiation. If we try to apply this principle to an adjudication setting, it means that the same notion can be interpreted in different ways by the members of the bench. The advantage of this constructive ambiguity is that it enables a decision to be taken, despite the existence of tensions or even contradictions between the parties. However, the disadvantage of such a technique, at least in international relations, is the difficulty it entails in terms of implementation. The question remains open in legal terms.

46  Besides the significance of agency, strategies, and processes, a fourth approach reminds us that the outcomes of any negotiation flow directly from asymmetries of power. In the view of the realist school of international relations, outcomes systematically confirm the initial distribution of power among the parties, as the most powerful is in a position to shape the process.

47  The notion of power is usually defined as the ability to influence others’ behaviour, ie to get them to do what they would not otherwise do (Dahl, 1957). In agreement with Thucydides, Machiavelli, and Hobbes, realist scholars consider that the quest and struggle for power lies at the core of relations among States. They insist on the competitive and conflictual aspects of the international arena. Various sources of power are usually mentioned: material resources such as wealth, property, or natural resources; authority related to a specific role, status, or function; skills and knowledge; human resources in demographic terms or in terms of support for a particular group or leader; or even intangible factors such as ideology or a particular attitude toward obedience.

48  In all its facets, the notion of power is undeniably ubiquitous in political science. Nevertheless, despite its constant use, the concept has certain limitations. Two of these are particularly significant. First, it remains a fundamentally relative concept. Aside from the actual power of each protagonist, often measured in figures based on strategic, economic, and demographic elements, it is also—and perhaps above all—vital to consider power as perceived by the parties. It is also worth taking a serious look at the mechanisms enabling the weakest parties, or those deemed to be so, to modify the initial balance of power. In international negotiations, weak can become strong mostly through three main mechanisms: the formation of a coalition of actors sharing, even temporarily, the same interests; the intervention of third parties, allied States, favourable to the weakest player; and support provided by NGOs and the media, both often able to influence the global arena. Thus, negotiations cannot be reduced to a strict balancing of each party’s initial material resources. Although the least powerful are often at the mercy of the strongest, this is not systematically the case.

49  Second, rather than focusing on the notion of ‘power over’ others—dominance and coercion—it is useful to consider a different perspective, not related to eagerness to control, but to willingness to cooperate. When the negotiation context is not distributive, but rather integrative, the notion of ‘power with’ others is particularly relevant. Since most negotiation processes have distributive and integrative aspects, both perspectives should be borne in mind (Lewicki and others, 2006, 184).

50  In the context of international adjudication, the notion of power is generally less explicitly used. However, some scholars consider that the realist theory is not totally irrelevant to the study of international law-making, interpretation, and application (Steinberg, 2013). More specifically, it is often argued that judges and arbitrators are motivated by a desire to exercise influence (Kapeliuk, 2010) and to enhance their own career, eg by obtaining a new mandate, or being appointed to a higher court.

51  Moreover, judges are surely not indifferent to the existence of a status differential among them. Like any other individuals, the presidents and vice-presidents of courts, and the presidents of chambers when they exist can exercise their function by using power ‘over’ and/or ‘with’ their colleagues. Authoritative legitimacy results not only from hierarchy but also from specific expertise or from particularly long experience. Thus, new members may be deferential to their more senior colleagues (Hettinger and others, 2003). Social psychology studies demonstrate that new members are far more anxious about social pressure. In contrast, group members who are familiar with one another are much more likely to express views inconsistent with those already expressed in the group. Therefore it is not surprising that the attention paid to the assessments of other members of the body, especially—but not only—among new members, explains the possibility of conformity effects (Martinek, 2010, 83).

52  From a power perspective, coalitions of judges can operate for a short time for a single case or become established. In the case of a five-member chamber, for instance, a systematic team of three judges can persist if they are constantly opposed to the two other colleagues. Interestingly, structural factors may also affect the power of référendaires or law clerks when they exist vis-à-vis judges. At first glance, there is a clear power asymmetry involving those who attend the conference, the délibéré, and who can obviously decide to dismiss the others. However, the balance of power is far more complex. In the case of the CJEU, for instance, when new members have no experience in judging, length of service gives undeniable power to experienced référendaires who know both EU Law and the institution (Deliberation and Drafting: Court of Justice of the European Union (CJEU)).

53  Thus, in comparing power asymmetries at the US Supreme Court and at the CJEU, Kenney explains:

Asymmetries in technical legal expertise are more likely at the ECJ, where turnover is higher and where the judges may not be knowledgeable about EC law or may have specialized knowledge in only one area of that law (eg competition law) and know little about milk quotas, pensions, or other technical matters. Justices may not really care about the intricacies of the legal reasoning or even have particularly strong feelings about the outcome, in which case, a référendaire or clerk who felt passionately could arguably have a great deal of influence if he or she managed to persuade the boss (Kenney, 2000, 615).

The same comment can be made regarding linguistic skills. If the clerk is more fluent in French than the judge, his or her influence can be considerable (Kenney, 2000, 615).

54  Considering the influence of personalities, strategies, processes, and power will give us a clearer picture of most negotiations. However, some fundamental aspects related to the ‘system of meanings’ of any social environment are overlooked by these four approaches. Hence a fifth and final approach underscores the importance of cultural variables such as value systems, codes, rituals, or even memories shared by the parties present.

55  National cultures differ in many ways. Notions of time, compromise, risk or power can be understood and used in various and sometimes diametrically opposing manners. Thus Western negotiators do not evaluate the passing of time in the same way as their Eastern counterparts. All cultures value time, but they do so differently. If we take the example of power, cultures where the ‘distance from power’ is greater tend to accept unequal relationships where political or religious leaders have ‘power over’ others, while cultures with a lower power distance favour more democratic decision-making based on the notion of ‘power with’ (Hofstede, 2010). Such differences are not only related to the degree of equality in structures, but also to varying understandings of expertise and gender.

56  In the same vein, some scholars and practitioners claim that negotiating behaviours vary across cultures (Faure and Rubin, 1991). To them, US, Russian, Brazilian, or Chinese negotiators—to name only a few—use language, space, and time in a distinct way. American negotiators, for instance, ‘tend to be blunt and legalistic while employing a conceptual vocabulary drawn from such diverse fields as labor relations, Christian theology, and sport. They are uncomfortable with silence and ignore body language. They enter a negotiation with their own timeframe and usually press for an early agreement, especially if the issue at stake has political significance at home’ (Quinney, 2002, 1). In contrast, Chinese negotiators are often presented as being distrustful of legalistic negotiations (Salomon and Freeman, 1999; Faure, 1999), whereas Russian negotiators are, in the view of these authors, associated with a specific sense of authority and relationship (Schecter, 1998). The list of national characteristics is endless: Japanese negotiators are supposed to be particularly concerned with face-saving and not to disclose much about themselves and their goals, while French negotiators are often characterized as ‘aggressive’ and experts in threats and warnings (Adair and others, 2004).

57  Most realist scholars criticize this approach, accusing it of overgeneralizations based on national stereotypes. Although these scholars admit that awareness of local norms and practices is key, especially during the preparation phase and the preliminary contacts, they still disregard this approach as superficial and not helpful in explaining negotiation outcomes.

58  Moving beyond these cultural differences, it is useful to remember that culture is not only defined as a set of beliefs, values, and behavioural expectations, but that these elements are learned and passed on to new generations. This intergenerational dimension allows us to consider the influence of collective memory on international relations (Rosoux, 2018). From this perspective, culture is not restricted to various norms regarding non-verbal communication, power and time, to name only a few elements. It is also related to the significance and uses made of the past. There is much at stake here, since representations of the past are associated with representations of the other and ultimately of one’s own group. As the French novelist Georges Bernanos wrote in Les Enfants Humiliés, ‘the future does not belong to the dead, but to those who speak for them, who explain why they are dead’ (Bernanos, 1949, 29; Uses of History and Collective Memories by International Courts and Tribunals).

59  The way in which a society deals with its past through official discussions, commemorations, and monuments has direct consequences not only for nation-building, but also for international negotiation (Rosoux, 2004). The peace process in the Middle East is symptomatic in this regard. The Israeli and Palestinian narratives of the 1948 war and the origins of the Palestinian refugee question demonstrate the extreme significance of memory issues (Aggestam and others, 2015). The intractable character of the negotiations between Turkey and Armenia, or the highly emotional Polish-German negotiations about the legacy of the Second World War, remind us that the appropriate unit of measurement in peace studies is probably neither years, nor decades, but generations. So, besides the evolution of the parties’ interests, which remains the main variable considered in the field, the ways in which the past is interpreted, misinterpreted, or even manipulated undeniably impact international negotiations.

60  At first glance, we could argue that international adjudicators share the same culture. Most of them come from Western law schools and were progressively socialized within particular institutions. Moreover, they are all influenced by a common ‘subculture of justice’ that leads them to adhere to predominant norms and practices (Spohn, 1990, 1212). The CJEU, for instance, is sometimes described as a ‘homogenous group endowed with a unified and atemporal rationality’ that explains ‘the maintenance of a transnational esprit de corps’ (Vauchez, 2010, 247). The homogeneity of the group can easily be illustrated by the frequent promotion of référendaires to judicial functions. This common experience, though, does not fully even out the diversity resulting from a variety of nationalities, legal traditions, and political cultures. On the contrary, both dimensions seem to be intimately linked (Representation of Cultural Diversity in International Adjudicatory Bodies; International Adjudication and Cultural Diversity).

61  In the case of the CJEU, an analysis of the speeches made to judges who arrive and leave the Court indicates a systematic emphasis on striking a balance between, on the one hand, the representativeness of a specific national legal culture and, on the other, the eagerness to promote a resolute European horizon (Vauchez, 2010).

62  The plurality that seems inherent to an institution such as the CJEU is particularly palpable when we consider the institutional energy devoted to translation (McAuliffe, 2009; Translation: Court of Justice of the European Union (CJEU)). Far from being a purely procedural stage, this linguistic dimension sometimes has direct consequences on the negotiation process within chambers. The noticeable differences between the translations of the same judgment are telling in this regard. They show that judges reached a solution which was mutually acceptable to the majority of, if not all, members of the chamber, thanks to the use of concepts that can be understood in different ways.

63  Beside a plurality of languages and conceptions, the judges of the CJEU also differ in terms of socio-professional background. This divergence is particularly salient if we compare, on the one hand, British and Irish judges—who are closely related to judicial networks such as the magistracy or the Bar—and, on the other, Belgian, German, or Luxembourgish judges, who mostly belong to academic networks (see Vauchez, 2010, and Cohen, 2010). These professional and cultural origins explain to some extent the variety of attitudes towards cases. Far beyond the superficial aspect of a cross-cultural dimension emphasizing national stereotypes, it is striking to observe the coexistence of different legal traditions. Scholars and practitioners frequently depict the court as a ‘legal melting pot’ or a ‘laboratory for legal pluralism’, where Scandinavian, Romano-Germanic, and Anglo-Saxon traditions coexist, without forgetting the specificities of former socialist countries in central Europe (Nicola, 2016).

64  To give a more specific example, the role of a chamber’s president reflects his/her conception of authority: he/she may be eager to either initiate and encourage discussion among the members of the chamber, or to suggest a particular position to be followed by them. The variety of understandings of power gives rise to further issues related to the impact of gender (Gender Balance in International Adjudicatory Bodies). Do women judges bring a different voice to the bench? If so, do they reveal a particular political culture? At this stage, there is no consensual answer to these questions. Scholars who have explored this issue at the national level are divided. Some have found significant differences between men and women in several legal areas (Songer and Crews-Meyer, 2000), while others have detected no discernible differences between them (Segal, 1997). This pending question indicates one of the limitations of the cultural approach. It is not sufficient to predict future processes, but is complementary to other approaches. It emphasizes a crucial dimension, but, taken alone, cannot explain why protagonists have reached a particular outcome.

65  The starting point of this study was the lack of bridges between the field of international adjudication and negotiation theories. Major differences remain between these spheres. As has already been mentioned, adjudicators have to reach an agreement while most international negotiators fail to do so. Moreover, notions such as ‘collegiality’ or ‘esprit de corps’ are rarely appropriate in international politics, especially in post-war settings. Broadly speaking, approaches to international negotiation were not developed to fit with the nature of judicial decision-making. The dissimilarities between these spheres, however, should not result in them being considered as distant islands.

66  The understanding of adjudicative deliberation is key. In the eyes of most practitioners, it is the ‘moment of truth’ in the sense that it reveals the extent of each adjudicator’s independence and impartiality (Bernardini, 2014, 16). It is therefore critical to use as many analytical tools as possible to do justice to the complexity of the process. Our consideration of negotiation theories confirms that one single perspective is hardly sufficient to crack this judicial enigma. The range of possible approaches is an invitation to continue with our research.

67  If we accept that adjudicative deliberations cannot be reduced to mere processes of coordination, since the interests and the perceptions of adjudicators do not coincide, it is worth exploring how well each of these approaches fits with the particularities of negotiation. Personalities and instincts do matter. Strategies and tactics are inescapable. Choices of particular procedures, the search for prestige, and cultural variables do impact the process. Nevertheless, none of these variables determines the outcome in a systematic way.

68  Although the notion of power is and remains central to understanding any negotiation, it does not make for a full understanding of the multi-layered nature of the judicial process. Despite numerous studies using statistics and robust methodologies, the adjudicative deliberation cannot merely be seen as a competitive game between self-interested pursuers of prestige and influence (Cooter, 1983; Rosenberg, 1999). As Judge Harry Edwards explained, a ‘model that takes each appellate judge as an atomized individual casting a purely individual vote in any given case will not produce a good explanation of how judges decide cases’ (Edwards, 2003, 1656).

69  The limitations of this paper did not allow us to reflect on the richness of all international adjudications. Its objective was a modest one. By tracing links between adjudicative deliberations and negotiations, it suggested new ‘lenses’ through which we can examine judicial decision-making. At this stage, many opportunities for further research are conceivable. Analysing a varied range of cases from different adjudicative bodies and differing time periods could provide insights. Although current lacunae in the literature ‘bridging’ international law and political science can be seen as impediments to our research, they also provide a fascinating challenge. As the Belgian novelist Simon Leys wrote in other circumstances, ‘the truths that we are looking for are like butterflies: in fixing them, we kill them’ (Leys, 1983, ii).

Valérie Rosoux