Author(s): Virgílio Afonso da Silva
Source: European Journal of Legal Studies, Volume Issue ()
Do We Deliberate? If So, How?
Virgílio Afonso da Silva*
Academic studies usually present the points of view of their authors. In the specific field of deliberation in constitutional and supreme courts, although the dynamic, quality and results of the deliberations are analysed from various points of view, that perspective almost always comes from outside the court. What judges think of their deliberative performance or what they think of the deliberative model in the court to which they belong is rarely known. This article aims to address this issue by presenting the thoughts of justices on a certain supreme court regarding the deliberations in which they participate. Its goal is thus not to formulate general hypotheses about deliberation in constitutional and supreme courts or even specific hypotheses about a particular court. It presents some of the results of a broad study on the deliberative practices of the Brazilian Supreme Court. This research was based on interviews with the justices of the Court as well as other sources. These interviews sought to understand what the Supreme Court justices think—or at least what they say they think—about the deliberative process in which they participate, especially their views on how the deliberation and judgement sessions are organized, as well as on the value of consensus and collegiality.
Keywords: supreme courts, constitutional courts, deliberation, judicial behavior, judicial review
Academic studies generally present the perspectives of their authors regarding a given subject. In the specific field of deliberation in constitutional and supreme courts, although the dynamic, quality and results of the deliberations are analysed from various points of view, that perspective almost always comes from outside the court. Thus, it is interesting to explore judges' opinions of their deliberative performance or the deliberative model of the court to which they belong, as these opinions are seldom publicly known or taken into account.
This article aims to address this issue by presenting the thoughts of justices on a certain supreme court regarding the deliberations in which they participate. Therefore, it is not a study that defends a certain deliberative model nor is its goal to formulate general hypotheses about deliberation in constitutional and supreme courts or even specific hypotheses about a particular court. In a certain manner, this article aims to contrast some of the assumptions and hypotheses I formulated elsewhere about judicial deliberation in general, and about the deliberative practices of the Brazilian Supreme Court in particular, with the opinions of the judges of this court.In that article, I pointed out countless deliberative shortcomings in the Brazilian Supreme Court and concluded that in fact, there is no deliberation among the judges on this court.
As will be shown throughout this text, in general, the Supreme Court judges disagree with this assessment. It is possible to argue that in fact, they are satisfied with the deliberative model adopted in this Court. This article presents some of the results of a broad study on the deliberative practices of the Brazilian Supreme Court. This research was based on interviews with the justices of the Court as well as other sources. Contrary to the objectives of similar studies conducted regarding other courts,the main goal of these interviews was not to reveal what takes place in the deliberation and judgement room because Brazilian Supreme Court sessions are public and their plenary sessions are broadcast live on television. Instead, the interviews sought to understand what the Supreme Court justices think — or at least what they say they think— aboutthe deliberative process in which they participate.
To achieve this objective, this article is divided into five main sections. The first section presents the general objectives and methodology of the research.Section two briefly presents the core concept underlying the interviews: deliberation. The third section systematizes the justices' general opinions concerning the organization of the Brazilian Supreme Court's deliberation and judgement sessions. In this section, my more general hypothesis stating that the judgement session is not a deliberation sessionis questioned. Section four focuses on the value of consensus in the deliberative process. Finally, the fifth section presents the justices' thoughts on the importance of collegiality in the deliberative process and their opinions about its existence in the Brazilian Supreme Court. In the conclusion, in addition to systematizing the results of the case study, I also present an exercise called 'institutional creativity'. Conducted at the end of each of the interviews: each justice was asked to define what, in his opinion,would be the best deliberative model for the Supreme Court. To a certain extent, this institutional creativity exercise encapsulates the view that justices have of their roles as deliberators within a collegial body such as the Brazilian Supreme Court.
Since the deliberation and decision-making process adopted by the Brazilian Supreme Court has been the same for many decades, each new justice must follow the practices dictated by tradition and the Court's rules of procedure. However, that does not mean that all justices share the same view of their role on the Supreme Court as a collective institution. In other words, despite the fact that an increasing amount of information is accessible — from online databases, through TV Justiça (a TV channel operated by the judiciary that, among other things, broadcasts the plenary sessions of the Brazilian Supreme Court live), on YouTube or even on Twitter — we still cannot assess the role that the justices themselves want to play or what they think of the current decision-making process of the Court. It would not be sound to assume that all justices share the same views on the value of collegiality or dissenting opinions, the role of the justice rapporteur, or the effects of the extreme publicity surrounding the Court's deliberation and decision-making process. The interviews were intended to provide this input to better understand the Brazilian Supreme Court's deliberation practices from material hitherto unavailable.
Between September 2011 and August 2013, seventeen justices (incumbent and retired) were interviewed. The interviews were structured (i.e., the same for all justices) and consisted of 36 questions, some with sub-questions, on the following subjects: the role of the justice rapporteur, concurrent and dissenting opinions, deliberation dynamics, deliberation and the legitimacy of judicial review, agenda setting and deliberation, methods of constitutional interpretation, the value of consensus, interruptions during the deliberation process,collegiality, publicity and TV broadcasting, deliberation and binding precedents, and deliberation and public opinion. Each interview lasted an average of 1 hour and 15 minutes. The longest interview took 2 hours and 45 minutes and the shortest, 45 minutes. The questions had not been revealed in advance and all interviews were conducted face-to-face. Each interview was recorded and subsequently transcribed.
To ensure confidentiality, the names of the justices were replaced by letters. Although there is no recognizable order to these letters, a clear division was made: letters A through I represent the justices who were incumbent at the time of the interview, and letters N through U represent those who were already retired at the time of the interview. In the text, I do not distinguish between incumbent justices and retired justices, except in those cases in which this distinction would be helpful to clarify contrasts between their views.
Despite their busy schedules, the justices were generally welcoming to the goals of the research. In many cases, they were willing to schedule more than one appointment to ensure that the interviews would be conducted at the ideal pace. Since only a few justices refused to be interviewed, it can be assumed that the results have a robust explanatory power regarding the deliberative practices of the Brazilian Supreme Court.
This article — as well as others presenting the results of my research — does not have the typical structure of a law journal article.As stated above, it does not aim to defend a thesis on the Brazilian Supreme Court's deliberative process or to describe this process from a purely external perspective, much less to offer a comprehensive review of literature on the matter at hand. The goal is to deliver something that could be called an internal description. Just as the Supreme Court's decisions are the result of 11 different individual opinions that somehow have to fit into a final document, this internal description of the Supreme Court's deliberative practices also tries to compose a picture of that institutional practice from myriad individual points of view of its members. The only difference is that in this research, it is 17 justices that comprise this picture rather than 11.
However, it is important to stress that although this article focuses mainly on presenting the opinions of the Brazilian Supreme Court justices on issues related to the deliberative practices of this Court, it is not merely a collage of points of view. On the one hand, because those points of view have been sorted out and systematized; on the other hand, because although I do not intend to take sides on the issues addressed, it was occasionally necessary to note some contradictions in the justices' statements or highlight some factual inconsistencies related to them.
A final clarification concerning the goals of the research and interviews is that their main topic was not the justices'attitudes on the tens of thousands of decisions made every year by the Court. Special focus was placed on the most important, politically and morally controversial decisions because many statements about, for example, the role of the justice rapporteur, the number of dissenting and concurring opinions, or the dynamics of the deliberation process only apply to those controversial cases.
In other words, a strictly quantitative study might show a different scenario than that which served as the backdrop for my research. However, I think that the choice to focus on a rather small set of decisions is justified. If one seeks to analyse the Brazilian Supreme Court as a constitutional court, then it does not make sense to take into account the deliberations of the justices when deciding the tens of thousands of interlocutory appeals they decide every year. What really matters here is the justices' attitudes in their decisions on those politically and morally charged cases that constitutional courts typically decide, such as those involving political reform, campaign financing, abortion, stem cell research, same-sex marriage, affirmative action, drugs, and so forth.
The interviews took at least one assumption for granted: the better the deliberative performance of a court exercising judicial review, the better the court itself is. The positive value of deliberation as such was therefore not in question. The theoretical framework that underpins this assumption has already been developed elsewhere and need not be fully analysed here.
It may nevertheless be summed up as follows. Justices in supreme and constitutional courts are not directly elected. Therefore, the legitimacy of these courts in reviewing the legislation passed by democratically elected parliaments must have different foundations. The most frequently mentioned source of their legitimacy is surely the fact that they can protect the constitutional rights of minorities. However, there are other sources, and the quality of the deliberation is one of them. When courts do not deliberate (or when they deliberate poorly), they add nothing (or very little) to the work already done by the legislature. In other words, if court decisions consist of pure head counting, then Waldron is right to challenge their legitimacy. Thus, the assumption that courts are (or at least can be) institutions with a distinct deliberative potential is paramount to their legitimacy.
The assumption that deliberation is a central feature in the decision-making process of a supreme or constitutional court defined the organization of the interviews. The concept of deliberation underlying the questions is that of internal deliberation: 'the effort to use persuasion and reasoning to get the group to decide on some common course of action', which involves 'giving and listening to reasons from others inside the group'.Most questions were related to what I consider the 'conditions under which the full deliberative potential of an institution can be attained'.The most important of these is collegiality. The value of collegiality, however, was not taken for granted, at least not to the degree that the value of deliberation was. The justices could deny — and some of them indeed did — the positive value of collegiality. The concept of collegiality underlying the interviews implied qualities such as the disposition to work as a team, the willingness to listen to arguments advanced by other justices (i.e., being open to being convinced by their good arguments), cooperativeness in the decision-making process, and mutual respect among judges.An entire section is also dedicated to a final feature of my concept of collegiality, namely, 'the disposition to speak, whenever possible, not as a sum of individuals but as an institution (consensus-seeking deliberation)'.
In the discussion regarding whether there is truly deliberation on the Brazilian Supreme Court, a preliminary question was whether the judgement session should be characterized as a deliberation session. This question addresses the more general thesis I advanced elsewhere, according to which the judgement session is not a deliberation session.The question posed to the justices was rather straightforward: 'Do you think that a judgement session in the Brazilian Supreme Court is also a deliberation session?' The concept of deliberation underlying the entire interview was expressed at the outset: 'During this interview, deliberation should be understood in its widest sense, as the exchange of arguments within a collegial institution with the goal of persuasion and decision making'.
Although the opinions of the justices have seldom been unanimous, those who said that the judgement session is not a deliberation session frequently clarified that this was the case only because of the workload and not necessarily due to the way the session was organized.More than a few justices mentioned Article 135 of the Court's rules of procedure, which defines the voting order in the Supreme Court. This article suggests that before presenting the position of each justice, there will be an oral debate:
Art. 135. Once the oral debate is finished, the Chief Justice will hear the reading of the written opinion of the rapporteur [...] and of the other justices, in reverse order of seniority.
This oral debate is regulated by Article 133 of the Court's rules of procedure:
Art. 133. Each justice may speak twice about the issue under discussion; he or she may also speak one additional time, if necessary, to explain a modification in his or her opinion. No one will speak without authorization from the Chief Justice, and no one will interrupt the justice who is speaking without asking him or her to briefly cede the floor and receiving that permission.
Clearly, these are not the characteristics of an open debate. Of course, in practice, on the few occasions during which there is an exchange of arguments before each justice's opinion is presented, this rigidity in the Court's rules of procedure is usually not followed. However, in the most important cases before the Brazilian Supreme Court, the practice has been to move directly to the reading of the written opinion of each justice without the preliminary oral debate mentioned in Article 133. Nevertheless, several justices still consider the judgement session to be a deliberation session as well. This appears so clear to some justices that they do not even bother to provide much justification.
Others justify this understanding through more arguments, as did Justice B:
I have the impression that it is a deliberation session. I think that a decision can be constructed that may not have even been in the written opinion of the justice rapporteur.
This outcome, mentioned by Justice B, corresponds substantially to one of the objectives that Fearon attributed to deliberation, which is the construction of a result that may not have been the initial idea of any participant in the debate.Similarly, other justices also appear to believe that it is a deliberation session because although the opinions may be written before the session begins, there is an opportunity for change and adaptation, particularly when such change is possibly based on comments from other justices:
Yes [it is a deliberation session], because you start your reasoning from the beginning, you explain its path, and in the course of this presentation, you may be interrupted by a colleague who tries to establish different premises. I think that this is true deliberation despite the fact that you had your ideas previously organized in a written format.It is a deliberation session in terms of the conclusion of the process. We meet together to discuss and exchange ideas, and we mutually complement each other [to] reach what we understand to be the best solution for each distinct case.
Even those who understand that the model does not encourage debate may argue that the judgement session is also a deliberation session:
I believe [that it is a deliberation session] in the sense of an exchange of ideas and an eventual readjustment of positions. Certainly yes, although the model does not particularly favour this.
In a similar sense, but with a bit more of a critical tone, Justice H argued:
It is a mixture of voting and debate. You are debating, and at the same time, you are also voting. This hampers deepening the discussion because if you could do it as the Court's rules of procedure dictate and simply participate in a debate, you [would think] 'I am going to participate in the debate like an intellectual sharpshooter and later I will reformulate my written opinion'. It would be better.
The most clearly contrary opinion arguing that the judgement session is not a deliberation session was that of Justice C:
I don't think so. It is a session for the presentation of points of view that are either already defined or that are reinforced during the discussion but are not brought about by it. The discussion itself is only an opportunity for the affirmation of points of view that are already in some way in the heads of the justices of the Supreme Court. The system does not frame this as a deliberative process, and it cannot be because to do so, it would have to be another environment, an environment for discussion, not one for defending points of view. What would work would be this: 'look, I am thinking of this, I am thinking out loud, but I want to listen, let's see, I am proposing this, what do you think? This or that could be considered...'. That is different, but the system does not foster it.
Finally, only one justice distinguished between the sessions of the panels and the plenary sessions in relation to the deliberative potential of both:
In the panel, [there is] greater [possibility for debate], perhaps because it does not have the solemnity of the plenary session. I think that the panel has a more deliberative profile. In most cases, the plenary session is nothing more than the presentation of stagnant, individual opinions.
The determination of who has the floor in a debate is rarely completely open. There must be some form of organization. As mentioned above, debate on the Brazilian Supreme Court is regulated by Articles 133 and 135 of the Court's rules of procedure. The former regulates the oral debate before the votes are taken; the latter defines how the voting is conducted. Since there are rarely oral debates before the voting, the only moment for deliberation is during the voting. The rules are quite rigid for this process: the Chief Justice will take the votes of the justice rapporteur and the other justices in the reverse order of seniority.
As mentioned, this speaking order, which is always the same, does not appear to encourage debate.Even if there are other courts — especially in common law countries — that also use the criterion of reverse order of seniority, there are some peculiarities in the Brazilian court that make this fixed order potentially more problematic. In the supreme courts of both the US and the UK, the reverse order of seniority is used in a preliminary meeting at which the justices give their views on the case being decided. After the preliminary meeting, the opinion of the Court and, if any, the concurring or dissenting opinions (in the US court) or the draft opinions of the individual justices (in the UK court) are shared with all the justices before the very final decision is taken.
Nevertheless, some justices have no problem with the fixed order of opinion reading. Some simply stated: 'I do not see this as relevant',or 'I think that this [...] is not essential'without additional comment. Others argue that this rule is not important because it is not followed to the letter: 'This does not seem too important to me because it is not rigidly observed in reality'.However, most of the justices believe that the rule is positive. There are a wide variety of reasons for this, some of which are conflicting. Some see the requirement that the newest members of the Court vote first as a form of protection. Within this form of protection, the idea of protecting justices from mutual influences appears.
I think this is good. I think it is good because each one has their own moment to speak. And to begin with the freshman is good because—I was once a freshman — the freshman enters the Court with great respect for the senior [members] and can have a tendency to vote along with them; therefore, he should be the first to vote.
I think it is interesting because it allows the youngest to bring new ideas. For example, if he were shy, he would not feel at ease to dissent after the opinions of the more senior members, so it has this positive aspect.
I have never considered this issue. However, there is a reason for it. [It is] always the newest member [who presents his opinion first] so that he is not influenced by the more senior members. I think that this is an intelligent rule, a golden rule.
Other justices emphasize a positive effect on the more senior members rather than the newer ones:
At first, I thought this was horrible. Then, I realized that the political advantage of this is the following: if there is disagreement, it is the more senior members who decide; it is a form of permanence to have the final opinions come from the most senior members who are the last ones to vote.
In contrast, Justice G argued that it is good because the younger justices can bring new ideas to the decisions:
I confess that I never considered this. To begin with the newest [is] more thought-provoking; it brings new values, and the debate becomes richer.
Those who are opposed to the model usually suggest the rule adopted in other Brazilian courts in which the order of speaking varies according to who the rapporteur is:
I think it would be much more democratic if it varied according to the distribution of the case. It would be much more democratic and perhaps create greater security in voting by not always leaving the younger members to be cannon fodder.
Other courts follow the order of seniority after the rapporteur. It is the rapporteur and then goes by seniority. I think that this would be better because the criteria of the order would vary, but there would still be an order.
In the already mentioned article, I argued that a fixed order of reading the written opinions could lead to a difference in the weight of each justice, especially for the most senior justicesbecause the case may already be mathematically decided by the time it reaches the most senior associate justice in the voting process. As discussed above, there are several courts that adopt the same order (inverse seniority). Once more, the contrast with the US and UK supreme courts may be illustrative. The inverse seniority criteria are used in these courts in preliminary (and secret) sessions.In the Brazilian Supreme Court, after the most senior associate justice and the Chief Justice cast their votes — i.e., share their opinions for the first time with all other justices — the judgement session is finished.
Nevertheless, the justices have a quite a different perspective. Although they are aware that the risk exists of the most senior associate justice becoming of little relevance, almost all of them point to the fact that it is not the procedural rule that is relevant in these situations but the personality of the justices, especially the more senior associate justices and particularly the most senior. Thus, if there is a risk that the opinion of the most senior associate justice becomes of little relevance, it would be due to his lack of initiative, not a procedural rule.
In the justices' responses, the most cited names were Justices Moreira Alves and Sepúlveda Pertence, as examples of the more active senior justices, who did not wait their turn to express their opinions — while the example of the most passive justice was always Justice Celso de Mello (the current most senior justice). In the following paragraphs, I transcribe these responses without additional commentary because they appear to me to be illustrative of the role that the personality of the most senior associate justice (or of the other rather senior justices) can have in the deliberation.
This will depend on the style of each justice. For example, Marco Aurélio Mello makes use of a provision of the Court's rules of procedure that Moreira Alves and Pertence often used: 'I am not voting before my colleagues, but, given the written opinion of the rapporteur, I would like to debate'. Moreira Alves did this constantly. Marco Aurélio Mello did it quite a bit, but less often, and Pertence also did it a lot, a bit less than Moreira Alves. In contrast, Celso de Mello, who [currently] is the most senior associate justice, does it much less.
The Court's rules of procedure allow each justice to intervene in the discussion, so Marco Aurélio Mello uses this a lot; he immediately asks to speak. In contrast, Justice Celso de Mello is more reserved; he winds up giving a lecture at the end of the session. He is the last to vote and has very little influence. He makes a speech that he thinks is for history books.
Moreira Alves would not let the chance go by: when he was the senior associate justice and in certain cases realized that the Court was leaning against his position, he would ask to speak and even ask to interrupt the session in order to view the case files.Not just anyone is able to fight in the plenary session because you will debate there with people with tremendous experience. At times I joked 'you have to hold the reins firmly or you may fall off the horse'.
He did not do this [speak before his turn]; Celso de Mello is much more evasive whenever possible.
Moreira Alves was like that; you began to vote and he would interrupt. He opened the debate at the beginning. Who had this personality? Moreira Alves, Pertence. Now, Celso de Mello does not. I think that he isn't even listening to what the others are saying.
Thus, because of this tendency to attribute the weight of each justice in the deliberation to personality issues rather than to procedural rules, there are few who believe that the voting order creates an imbalance in the debate. In addition to Justices A and D, who were mentioned in the previous section as being opposed to the current fixed voting order, only retired Justice T mentioned some concern with a possible imbalance.
The fact that so many justices highlighted the personalities of their colleagues as pivotal, and provided clear examples thereof, should be taken into account. In light of this point, my original arguments, presented in the article mentioned above, should be refined to include the 'personality' variable. However, this does not mean that my arguments are rebutted.The fixed order still potentially decreases the influence of the most senior justices because they read their opinions last. The fact that some senior justices were and still are able to overcome the burden of 'jumping the queue' by expressing their points of view before their turn does not mean that the burden does not exist; it only means that the burden is not insurmountable. The case of Justice Celso de Mello, who has been the most senior associate justice on the Brazilian Supreme Court since 2007,clearly shows that not every senior associate justice is able or wants to be as active as they would need to be to have some degree of influence.
Related to the previous issue, another question also arises addressing the strategy of how to proceed when a case is mathematically decided, i.e., when there are already six opinions in favour of a given position. In these moments, the influence of the justices who have still not voted is reduced given that it would be necessary for one of the justices who had already read their written opinion to change his or her position to alter the course of the judgement, which is not a simple task.
One of the interview questions was raised precisely to determine how the justices act or would act in a plenary session in a situation such as this, i.e., if he had still not read his opinion, which is contrary to the already consolidated majority. In this situation, the basic options would be (i) to simply read the written opinion as it was written, (ii) to reconsider the written opinion (during the plenary session or after requesting time view the case files) to try to counter the arguments of the majority, (iii) to try to join the majority but offer some different arguments that could perhaps be accepted; or (iv) to simply join the majority.
Most of the justices argued that their response would depend on the case being decided. Many of them said that all the strategies are possible: 'you could take any one of these four strategies, any one'.Justices E and R had similar responses:
All the responses would be possible depending on the case, but I would say that if it is an issue that I consider to be important, and the Court is heading in a direction diametrically opposed to that which I believe correct, I would ask to be allowed to intervene before my turn and present my argument, even if only briefly. If I find that my argument would not be wholly accepted, I would try my second best option, which would be to try to neutralize the extreme that seemed to me to be unfavourable. And, in certain cases, if I believe the issue unimportant and the majority was already determined, I may just let it go.
There is no rule. For example, if I understand that the consequences of a given decision would be disastrous, I would ask for time to view the case files; if I think that the consequences of the decision would not be disastrous but nevertheless have a dissenting argument, I would maintain my position, but not read the written opinion to avoid disturbing the others. [I would say] 'I dissent in the terms of the written opinion, which I will file later...' since I knew that I would not change the opinions of the other justices. If I had a point of view that could be accepted by the majority, I would [present it].
Here the variable 'workload' was also mentioned as a factor that could influence the behaviour of the justices. In this sense, Justice Q stated:
The tendency is towards the latter [to simply join the majority]. In second place, the next to last [try to join the majority but offer some different arguments that could perhaps be accepted]. And this has a lot to do with the workload.
However, there are justices who emphasize that it is not possible to compromise, especially when handling the most important cases. For example, Justice B argued:
I never thought of joining the majority, no. If I had a differing position, I wouldn't do that. I think that I would have to maintain the dissenting position depending on the importance of the issue.
The position of Justice G is a bit more incisive, but takes a similar direction: 'No, I think that in the most polemical cases, I would offer my own contribution as I would have taken it to the judgement session'.
The position of Justice F was different from all the others. He stated that the justices simply cannot vote in another way that is not faithful to their conviction with the exception of the Chief Justice:
The ten associate justices of the Brazilian Supreme Court have the sole and exclusive commitment to vote along with their own conviction and must vote as such. The possibility of a Supreme Court justice voting against his conviction is exclusive to the Chief Justice. The associate justices on the Supreme Court do not have the right to do this and cannot do this without running the risk of subverting their consciences.
This point of view is especially opposed to that expressed by Justice E, who explicitly mentioned the possibility of voting according to a second preference to try to neutralize the extreme that would appear unfavourable. It seems that Justice F would understand this strategic behaviour as a 'subversion of the conscience' of the judge.
In many constitutional and supreme courts, during the oral arguments justices may ask questions to the lawyers.Although this moment precedes the deliberation itself,it can have an important effect on the debate among the justices.
In the Brazilian Supreme Court, this practice does not exist. At most, the justices may ask questions to clarify a factual issue. When questioned about the importance of debate with lawyers, several justices emphasized the value of the oral argumentsand many stated that they favour the possibility of debating with the lawyers. Nevertheless, many justices pointed to a certain lack of experience among many of the attorneys who present oral arguments at the Supreme Court.
Among those favourable to greater interaction between justices and attorneys, some mentioned that it should not simply follow the current model. Some changes were suggested:
I think that this interaction should take place. And certain formalities that create a distance between the judge and the lawyer should be eliminated.
Maybe the most important change would be the realization of oral arguments (with or without debate) in a distinct (previous) session, as usually occurs in other constitutional courts and supreme courts:
I think that the oral argument as it is conducted in Brazil has limited value because it is a very unilateral process. The rapporteur generally already has his opinion written, which means that for him, the oral arguments rarely have any consequence. I would favour a model in which the oral debate were made in a session prior to the judgement session.
Although favouring a debate between justices and attorneys, Justice B mentions that with the Supreme Court's current workload, this interaction would not be possible or even productive:
An oral argument in the plenary session, for example, may have a very strong influence — not necessarily on the rapporteur, but on those who still have to vote — and in some cases certainly influence the opinions of the other justices. But the workload is a pivotal issue here. [A debate with the attorneys] would perhaps require having only that judgement [on that day].And the presentation of oral arguments could not be limited to fifteen minutes [as it is today]. The system would have to be completely remodelled.
Finally, although several justices might be favourable to having greater interaction with attorneys, they point to a certain lack of experience of those who make oral arguments in the Supreme Court, which can undermine the potential of this interaction:
On one occasion, I began to ask the attorney a few questions, but I forgot that he was quite young, and he became embarrassed, so a colleague nudged me — 'why are you causing trouble?' I apologized, [and said] 'No further questions'. So there is this issue as well. Since it is not usual [to ask questions], the attorney can understand that this is being done as a form of pressure.
Justices C and I were more explicit about the lack of experience of the attorneys:
What happens here in the Supreme Court? The person who comes to argue here at times is quite young and does not even fully understand the case, so it is impossible to engage in a dialogue. This can be detrimental. I think it is not very productive.
The attorneys are extremely unprepared! What is happening today? The person graduates, [...] takes the bar exam, and the next day is presenting oral arguments before the Supreme Court. One day, during a break in the session, a young man, who was young enough to be my son, and a young woman came to me. She sat here and he sat there. He [...] looked at me and asked: 'did you receive my brief?'. I looked at him and said: 'young man, look, according to the Court's rules of procedure, you should call me Your Honour, but you can call me Sir and start over'... [In other countries,] there is a long way before being admitted to the Supreme Court's bar. This rule would facilitate a better dialogue because it would be a dialogue among equals. But [here], you are clearly not dialoguing among equals.
On the Brazilian Supreme Court, as on almost all similar courts around the world, the decisions are made by the majority. That is, six votes are enough for a case to be decided in a certain direction. Even so, there are studies that indicate that some courts make an effort to decide cases by broader majorities or even by consensus.
Questioned about a similar tendency on the Brazilian Supreme Court, only one of the incumbent justices and two of the retired justices were able to identify a consensus-seeking tendency. Justice G stated:
There is an effort in this direction because, one way or another, the justices, when they eventually meet outside the sessions, debate. During the interval of a session, they debate the issues. Thus today, there is an effort to reach an institutional decision.
No other justice expressed a similar perception.In fact, some believed that this effort to reach a consensus should not even exist:
If I could give weight to a unanimous decision and to a majority decision, I would give a higher weight to a majority decision because it is the unequivocal understanding that the issue was discussed. In the plenary session, I very much resent the fact that at times, for one reason or another, an issue is decided without greater discussion.
Although they understand that this search for consensus should exist, others do not see an environment that is propitious to this on the Brazilian Supreme Court:
There is none. It's each one for himself and God for all. As soon as I entered in the Court, I asked, '[Justice] Jobim, listen, don't you discuss things here?'; [he answered] 'Ah, no, I tried to have discussions, but it didn't work'. [...] I think that [...] when you have a score higher than this minimum of six to five, the legitimacy of the decision is more robust. You will always have a justice like Marco Aurélio Mello, who will always dissent, but even this is good. [For example], the decision on same-sex civil union: I strongly insisted that we discuss this issue beforehand so that we could establish some premises and reject others to avoid each justice making his own speech. That is not good.
An attempt to explain the absence of a consensus-seeking orientation was given by Justice F, who attributed this role to the Chief Justice and to the rapporteur of each case. When the Chief Justice or the rapporteur do not take on consensus (or broad majorities) as a goal, it is very difficult for this to occur naturally. His explanation deserves a longer transcription:
This is the role of the Chief Justice; it is the role of the rapporteur, who has to take the initiative. The other associate justices here are quite passive. You can ask 'should it be like this?' My answer is no. And you ask, 'So, why is it like this? Is there an explanation?' There is. The number of cases that we have to decide is so cumbersome that we think, 'ah, I am not the rapporteur of the case, I am not the chief of the panel, I am not the Chief Justice, I vote how I want to vote; here is my vote, it's done. Why would I have to try to find a compromise solution?'. This is the role of the rapporteur, of the justice who first dissented, or of the Chief Justice.
Following this idea that it is up to the Chief Justice (or to the rapporteur) to try to foster consensus or broader majorities, Justice I mentioned that he acted this way when he was on the Electoral Court:
We have to sit down together, and it's easy: 'look, people, let's have coffee and do this and that; let's establish some shared premises and let's go into the session in agreement'. I have done this. One day I had a big problem [on the Electoral Court]. There was a delay, a justice was absent, and in the other room I had a good bottle of whiskey. I sent the waiter to bring some cheese and served a few glasses of whiskey to my colleagues and [said] 'take a look at this here' and we began to talk and we went to the session... An informal conversation can work miracles. There is no demerit in this at all. Because this is not academia, do you agree? We are one of the political branches, which must take decisions. And the more consistent, coherent, firm, and univocal the decisions, the better for the country, the better for the citizens.
However, the absence of a search for consensus does not appear to be peculiar to the most recent benches of the Court. The large majority of the retired justices stated that they also did not see a trend towards consensus.In fact, in the case of retired justices, almost none appeared to see a problem in lacking a consensus-seeking orientation.To the contrary, it is identified as an attitude that demonstrates respect for individual positions:
No, no, this doesn't exist. On the Supreme Court, the individual's point of view is strongly respected.
Never, never. And in fact, there were many decisions made by tight majorities [...] this is inevitable, there is not a minimal chance on the Supreme Court, above all at a time like this, to achieve anything different, [such as] 'let's try to make a bit stronger majority'. This possibility does not exist. 
No, no one was ever concerned with this. [Because a search for consensus] would imply coercing people to adopt a line of thinking when in reality, they are not prone to this.
These answers — especially the latter — seem to perfectly express the fear that consensus-oriented deliberation could suggest: 'that a justice is open to compromise her own view of the underlying legal merits of an appeal in order to achieve some extraneous, distinctly non-legal or policy goal'.
However, some retired justices did mention that even without a general trend towards consensus-oriented interactions, the older benches of the Court sometimes identified cases that deserved special attention and that, if possible, should be decided by a broader majority or by consensus. In these situations, a 'session of the council' or even an 'administrative session', both held behind closed doors, could be used for a preliminary debate:
You had cases in which it was convenient for a decision to be taken by a truly substantial majority. These were cases for a session of the council. Precisely to know the different points of view, to not have a decision with too much dissent and to not have unnecessary debates during the session. The Supreme Court was very careful about this issue.
At the time when there were administrative sessions, these issues were discussed. 'Should this decision be unanimous?', 'Even if there are dissenting opinions, should it appear unanimous?'. And this was a political decision that was related to some high public interest of the country.
However, these 'sessions of the council' have been virtually discontinued. One of the justices explained the decline in the importance of the sessions of the council as follows:
It was pretty much dissolved as Justice Marco Aurélio Mello joined the Court. He was adamantly against this preliminary conversation. [...] Sometimes he didn't show up or showed up to protest, etc. Since then, the meetings have very rarely been held.
The Brazilian Supreme Court is a collegial body. This does not mean that there is collegiality on the Court. The definition of collegiality that was presented for the justices was quite broad. It should be understood not in the sense that there necessarily be friendship or that dissent among the justices should be always avoided, but in the sense that there be a willingness to work as a team, to listen to the arguments of colleagues and to be open to them, and when possible, to try to speak as a group and not as an individual. Given this definition, the questions presented to the justices were clear and direct: Is there (or, in the cases of the retired justices, was there) collegiality on the Supreme Court?
The division between incumbent and retired justices was probably the clearest in this area. Practically all the incumbent justices stated that the environment was not at all collegial, while almost all of the retired justices, especially those who left the Court long ago, said that it was. And more than one justice mentioned the expression supposedly coined by Justice Sepúlveda Pertence, who once asserted that the Brazilian Supreme Court is composed of 11 islands:
Collegiality is the ideal. It does not work like that, but it is the ideal. No, it does not exist here. There is even a saying that you can put in quotes: Sepúlveda Pertence said that the Supreme Court is constituted of 'eleven islands'! He always said this, and I think this is very true. After a few years, I said to him one day 'and they don't even form an archipelago'.
Look, it doesn't exist [collegiality]. There are eleven islands, and it absolutely does not exist. At times you may try to convince a certain group so that you have a pre-majority, so that your position will prevail. But collegiality does not exist on the Supreme Court. And it will take a long time to achieve it!
Among the incumbent justices, only Justice G adamantly declared that 'there is an environment of collegiality'. All the others clearly said there was not, or mentioned conditional factors, which were often not satisfied, for this collegial environment to be attained. Moreover, although many linked the lack of collegiality to the current deliberative model, the perception that there is no collegiality is not directly related to their opinion of about this model. In other words, even the justices who favour the current model, or are at least satisfied with it, still attribute the lack of collegiality to it, at least to some degree.
I think that [collegiality has] great importance, although at least in the more complex cases, it is usually absent.
I think there is no collegiality. I think that this sense of collegiality does not exist. I think that the system does not favour it and there is resistance to it by some.
This 'resistance by some', mentioned by Justice D, can be deliberate resistance, i.e., a justice who believes that being open to being convinced by others or to trying to speak as a group and not as an individual is harmful to the work of the Court. But it can also be the result of personal difficulty, at least in the view of Justice F:
It depends on each person's profile. We had here a fellow justice, and I admire him a lot, Eros Grau. [As] an only child, he had difficulty with this collegial body [...] I feel happy in the plenary session, I feel fortunate. I would be sad if I was alone. Thus, this is not only related to being a judge, but it also has to do with being human.
As mentioned above, the view of some of the retired justices is quite different, especially those who left the Court long ago.
It did exist. It existed [a collegial environment]. [Justice] Brossard once wrote an article, and it was quite fortunate: he said that 'among the justices of the Supreme Court, there is a lot of cordiality, there is a lot of friendship, without any intimacy'. As an outside observer, I think that this has changed. I don't know if my observation is pertinent, but you see how the Barbosa Court was... This is one of the reasons that leads me to think that it's changed.
Not today. There is no doubt that there is none. Yes, there was more collegiality at that time.
When I entered the Court, there was an environment of collegiality. It was something curious because there were brutal debates, but they were debates about reasons, such as the debates between justices Néri da Silveira and Moreira Alves, Pertence and Velloso, and especially when Pertence and Moreira Alves got sharp. But no one left offended. In recent years, they began to strike at the person. That is not right. Instead of attacking the reasoning, one attacks the person. This began to create discord. And now there is hate there, and the environment is quite heavy. I was there recently, one speaks about another, complains, 'I was treated poorly, this way and that.
Retired justices certainly do not always see their time on the Court positively in relation to collegiality. Some of them also mentioned the argument of the eleven islands:
Complete collegiality, no. I insist on the argument of the archipelago. Once in a while, let's say, some islands had a greater tendency to come together. But this was by chance.
However, when faced with the question of whether there had been some change in the environment of collegiality from their time to the present, all the retired justices stated that they identified a clear change, which always tended towards less collegiality.
Although nearly all the justices believe that there should be more collegiality on the Supreme Court, this opinion has not always been compatible with their own attitudes. As observed above, collegiality refers to the disposition to work as a team, to listen to the arguments of colleagues and be open to being convinced by them, and to try speaking as a group rather than as an individual whenever possible. Nevertheless, when asked about the frequency with which they were convinced by their colleagues, the responses indicated a lack of collegiality. There are those who said they never changed their position, those who asserted they had but not in important cases, and those who said that this happens frequently but were unable to give an example:
Look, if you asked me to name a case, I can't remember precisely. But I have many times, many times [been convinced by a colleague's argument].
Yes, although it was a small change.
Sincerely, perhaps one time or another when there was a shift in the jurisprudence that I may not have known, but very rarely, very rarely.On the Supreme Court, never! On the truly polemical cases I have never voted without being sure which of the two points of view in confrontation was mine. So it was never necessary to change.
There was no case [in which it was necessary to change my opinion].
Only one justice declared a willingness to vote in line with the reasoning of his colleagues when they were better than his, offering examples and justifications:
I had some cases, mainly in criminal issues. I was always more severe than my colleagues. But, at times, they advanced some arguments that I considered irrefutable. In such cases, I had no doubt. When I was not able to win over my conscience...
As stated at the beginning of this article, the justices of the Brazilian Supreme Court are, to a large degree, satisfied with its deliberative practices. In an exercise conducted at the end of the interviews called 'institutional creativity' in which each justice could define what, in his opinion, would be the best deliberative model for the Supreme Court,there were few variations in relation to the current model. In regard to deliberating in public, only 2 of the 17 justices were opposed to the practice. Some retired justices defended the so-called 'sessions of the council', which were sessions prior to the judgement session — without public — to discuss procedural issues. However, even these justices are not opposed to the public plenary session.
However, a very common practice, analysed throughout this article, appears to trigger a certain rejection among the justices: the custom of bringing written opinions, at times lengthy ones, to the deliberation and judgement sessions. Although virtually all justices follow this practice, which is not established by the Court's rules of procedure, all of them stated that if they could define the rules and practices of the Court on their own, they would prefer that only the justice rapporteur present his written opinion, which would be debated freely by the other justices.
These two positions — to favour deliberating in public and to oppose taking written opinions to the deliberation session — may be considered compatible. Nevertheless, the practice of taking written opinions to the session has been consolidated precisely as the publicity of the deliberations increased, that is, when the plenary session came to be broadcast live on TV. In other words, according to this view, the justices only take written opinions (and at times lengthy ones) to the plenary session because a large public is watching. However, the justices do not appear to have identified this pattern.
In any case, the division between retired and incumbent justices, which in part coincides with the division between justices with experience deliberating in front of the cameras (incumbent justices) and those without this experience (retired justices), appears to point to an important change associated with the deliberative practice in the Brazilian Supreme Court. Not only are the deliberations in the most polemical cases rather a sequential and very formal reading of opinions, very far from an open debate; collegiality, essential to sincere deliberation, also appears to have drastically decreased in recent years.
* Professor of Law, University of São Paulo. This article presents some of the results of a broad study about the deliberative practices of the Brazilian Supreme Court. This research was supported by a grant from the São Paulo Research Foundation - FAPESP (grant 2011/01066-0). An early draft of this article was presented at seminars of the research group Constitution, Politics and Institutions, at the Law School of the University of São Paulo, at the Getulio Vargas Foundation Law School in Rio de Janeiro (FGV Direito Rio) and at the Inter-Institutional Research Group of the Political Science and Public Law Departments of the University of São Paulo, the Getulio Vargas Foundation Law School in São Paulo (FGV Direito SP), and the UFABC. I would like to thank the organizers and participants of these three events for their valuable comments, especially Diego Werneck Arguelhes, Ivar A. Hartmann, Fernando Leal, Thomaz Pereira and Pedro Cantisano (Rio de Janeiro) and Rogério Arantes, Luciana Gross Cunha, Vanessa Elias de Oliveira and Conrado H. Mendes (São Paulo). I would also like to thank Guilherme Benages Alcantara and Tatiana Alvim for their hospitality in Brasília during my visits to the Brazilian Supreme Court. Finally, I also would like to thank all the justices that agreed to be interviewed for my project (see note 12).
 Of course, this does not mean that judges' perspectives are never taken into consideration in academic studies. There are enough studies, especially on judicial behaviour, that include interviews with judges as part of their dataset. However, they rarely involve interviews with supreme court or constitutional court judges and even more rarely within the specific field of judicial deliberation in such courts.
 Virgílio Afonso da Silva, 'Deciding Without Deliberating' (2013) 11 International Journal of Constitutional Law 557.
 For example, the main objective of the interviews conducted by Kranenpohl with the justices of the German Constitutional Court was to better understand the dynamics of the deliberation sessions of this court, which are completely secret. See Uwe Kranenpohl, Hinter dem Schleier des Beratungsgeheimnisses (VS Verlag für Sozialwissenschaften 2010).
 It is certainly impossible to identify whether justices are offering their honest opinions or the answers that they deem most compatible with their position or public image. This article should be read with this limitation in mind. However, this is not a shortcoming; instead, it is part of a methodological option. The goal of this article is not to investigate, for example, whether there is actually a consensus-seeking tendency in the Court or how often justices change their opinions. In fact, to achieve these goals, quantitative research would have been the best method. However, if the goal is to uncover how the justices see themselves as deliberators, the best way to accomplish this is to interview them, even if there is a risk of insincere answers. To minimize that chance, the confidentiality of the answers was emphasized before each interview. For more details, see section I (Methodology).
 The text in the first section, which summarizes the methodology and goals of the research on the deliberations of the Brazilian Supreme Court, is repeated in all the articles that present the results of this research.
 See da Silva (n 2) 570.
 The use of the pronouns 'he', 'his', and 'him' does not imply a gender-based option for the masculine over the feminine in this article. It simply results from the fact that none of the women who are or were justices in the Brazilian Supreme Court chose to participate in this research (see note 13).
 After this research was concluded, the Getulio Vargas Foundation Law School in Rio de Janeiro began a project called 'História Oral do Supremo Tribunal Federal' (Oral History of the Brazilian Supreme Court), which has interviewed several justices of the Brazilian Supreme Court. Although thematically wider in scope, these interviews also contain questions related to the deliberative process in the Court.
 See note 47 below.
 Only four incumbent justices refused to be interviewed despite many attempts to gain their interest: Celso de Mello, Joaquim Barbosa, Cármen Lúcia Antunes Rocha and Rosa Weber. Since these two latter justices refused to talk and retired Justice Ellen Gracie Northfleet never answered several invitations sent to her, unfortunately, no women were interviewed for this research.
 Even articles that include interviews do not usually have the structure of this article. Again, Kranenpohl's research is an exception. In addition to the above mentioned book (Kranenpohl, Hinter dem Schleier des Beratungsgeheimnisses (n 3)), see also Uwe Kranenpohl, 'Herr des Verfahrens oder nur Einer unter Acht? Der Einfluss des Berichterstatters in der Rechtsprechungspraxis des Bundesverfassungsgerichts' (2009) 30 Zeitschrift für Rechtssoziologie 135; Uwe Kranenpohl, 'Die gesellschaftlichen Legitimationsgrundlagen der Verfassungsrechtsprechung oder: Darum lieben die Deutschen Karlsruhe' (2009) 56 Zeitschrift für Politik 436; and Uwe Kranenpohl, 'Die Bedeutung von Interpretationsmethoden und Dogmatik in der Entscheidungspraxis des Bundesverfassungsgerichts' (2009) 48 Der Staat 387.
 The following incumbent (at the time of the interview) and retired justices were interviewed. Incumbents: Ayres Britto, Cézar Peluso, Dias Toffoli, Enrique Lewandowski, Gilmar Mendes, Luiz Fux, Marco Aurélio Mello, Luís Roberto Barroso and Teori Zavascki. Retired: Carlos Velloso, Eros Grau, Francisco Rezek, Ilmar Galvão, Moreira Alves, Nelson Jobim, Sepúlveda Pertence and Sydney Sanches.
 An example related to the role of the justice rapporteur may illustrate the importance of this clarification. While it is true that in the vast majority of decisions, the justices tend to vote along with the justice rapporteur without further inquiry, this is not the case in those more politically and morally controversial decisions, which are also the decisions that draw greater public attention outside the Court. The same applies to the practice of bringing lengthy written opinions to the plenary sessions and reading them; this usually happens only in those major decisions.
 The definition of a controversial case is far from clear-cut. For example, it is not possible to state that all plenary decisions (as opposed to panel decisions) or all non-unanimous decisions are controversial. There are both panel decisions and unanimous decisions that may be considered controversial. Maybe the best example of the latter is the decision on same-sex civil unions, from 2011 (ADI 4277). Although it was a unanimous decision, its subject-matter is quite controversial. This is the reason why, instead of trying to provide a clear concept of a controversial decision, I decided to deliver many examples of recent decisions that should be, at least for the goals of this research, considered controversial. Not coincidentally, the unanimous decisions used as examples often have many concurring opinions.
 For more details on these theoretical discussions, see da Silva (n 2). For other defences of judicial review grounded in the deliberative attributes of supreme or constitutional courts, see, for instance, John Rawls, Political Liberalism (Columbia University Press 1993) 231; Conrado Hübner Mendes, Constitutional Courts and Deliberative Democracy (Oxford University Press 2013); Ronald Dworkin, Freedom's Law: The Moral Reading of the American Constitution (Oxford University Press 1996) 1–38; Christopher L Eisgruber, Constitutional Self-Government (Harvard University Press 2001).
 See Jeremy Waldron, 'The Core of the Case Against Judicial Review' (2006) 115 Yale Law Journal 1346, 1391.
 John Ferejohn and Pasquale Pasquino, 'Constitutional Adjudication: Lessons from Europe' (2004) 82 Texas Law Review 1671, 1692. In opposition to the concept of internal deliberation, Ferejohn and Pasquino define external deliberation as 'the effort to use persuasion and reasoning to affect actions taken outside the group', which involves 'the group, or its members, giving and listening to reasons coming from outside the group'.
 da Silva (n 2) 562–67.
 ibid 562–63.
 ibid 563.
 See ibid 570: 'the plenary session means 'opinion-reading session' rather than 'deliberation session'.'
 Similarly, see Mathilde Cohen, 'Ex Ante versus Ex Post Deliberations: Two Models of Judicial Deliberations in Courts of Last Resort' (2014) 62 Am J Comp L 951, 999: 'the deliberative ideal lacks verisimilitude for certain courts by virtue of their skyrocketing dockets.'
 Emphasis added. Art. 134, § 2 of the Court's rules of procedure also mentions oral debates.
 For example, Justice I ('I think so, it certainly is'), Justice Q ('I think so, I think so') and Justice O ('There is, there is [deliberation]').
 See James D Fearon, 'Deliberation as Discussion' in Jon Elster (ed), Deliberative Democracy (Cambridge University Press 1998) 44.
 Justice G.
 Justice A.
 Justice E.
 Justices R, P and S are also sceptical about the possibility of having deliberations in the judgement session. Justice R: 'It is rare, rare, it's rare [to have deliberation]'; Justice P: 'At times'; Justice S: 'It is more for [reaching] a decision than for deliberating'.
 Justice D. For a comparison of the level of cohesion in the panels and the plenary session, see Evan Rosevear, Ivar A. Hartmann, and Diego Werneck Arguelhes, 'Disagreement on the Brazilian Supreme Court: An Exploratory Analysis' (31 October 2015). Available at SSRN: https://ssrn.com/abstract=2629329, 18.
 Similarly, see Mendes (n 16) 167: 'As the interaction becomes more rigid and codified (like the ritual in which the order of individual votes follow a criterion of seniority), deliberation naturally loses spontaneity. And although deliberation cannot be seen as mere 'spontaneous conversation', hard rules of interaction may turn it artificial'.
 Justice B.
 Justice T.
 Justice C.
 Justice O.
 Justice S.
 Justice I.
 Justice R.
 Justice A.
 Justice D.
 See da Silva (n 2) 570–72.
 This may explain why there is almost no study addressing the issue at stake here, namely, the difference in the weight of each justice in the deliberation process. Leflar, however, argued that even in the preliminary conference, the order should not be fixed. See Robert A Leflar, 'The multi-judge decisional process' (1983) 42 Maryland Law Review 722, 726: '[t]he order in which judges state their views on cases during a conference has bearing on a judge's opportunity to influence the decision'.
 Even though any justice is allowed to change his or her mind before the judgment session is concluded, i.e., even after the most senior Associate Justice read his opinion, this rarely — if ever — occurs.
 Justice F.
 Justice I.
 The possibility of interrupting the judgement session and requesting to view the case files ('pedir vista') is established by the Brazilian Civil Procedure Code (Article 555, § 2): 'Every judge may interrupt a judgement session if he considers himself unable to reach a decision at the given moment'. These requests are sometimes also used to postpone the decision.
 Justice O.
 Justice Q.
 Justice R.
 Justice T: 'I think that a variation in the voting order would be very significant precisely for [avoiding differences in the weight of the vote between the justices who vote earlier and those who vote later]'.
 I would like to thank Diego Werneck Arguelhes for pointing this out to me.
 The justices mentioned above as having more active personalities, Moreira Alves and Sepúlveda Pertence, were the most senior associate justices from 1986 to 2003 and from 2003 to 2007, respectively.
 Justice H. This justice added a fifth strategy, which was analysed in the previous section: 'There is still a fifth [option]. You can say the following; 'look, the opinion of the rapporteur appears to me to be incorrect, and I would like to anticipate my opinion' to try to influence, in the best sense, the other justices who will vote afterwards. I will not let it reach me with the fact consummated; I will anticipate my dissent right away'.
 Justice E.
 Justice R.
 'Advocate', 'lawyer' and 'attorney' are terms used here in the broad sense to encompass all those who present oral arguments in the plenary session of the Supreme Court.
 See Mendes (n 15) 162.
 Justice F: 'My experience shows that the oral arguments can lead to a change in the outcome of the judgement'.
 Justice G.
 Justice E.
 Unlike what occurs in many courts, in the Brazilian Supreme Court, many cases are often judged in a single day (or, to be more precise, in a single afternoon).
 Justice O.
 Justice C.
 Justice I.
 See Kranenpohl, Hinter dem Schleier des Beratungsgeheimnisses (n 3) 181.
 Justice G.
 Justice B: 'No, I do not believe that this exists, I don't see it '; Justice D: 'No, it doesn't exist, it doesn't exist'.
 Justice A.
 Justice I.
 Only Justice R stated that 'There was [a tendency to seek consensus]. Then it disappeared. But at first, it was there, there was an effort. [...] When things got very complicated [...] one would ask to view the case file in order to try to reach consensus later. At times, the request to view the files was merely instrumental only to try later to discuss the issue over a coffee'.
 With the exception of Justice T: 'I would even like if it were like that, but I knew it wouldn't be because unity never prevailed'.
 The answers of the justices of the Brazilian Supreme Court in this regard are very similar to the most current justification of British judges for maintaining the seriatim system: (Andrew Le Sueur, 'A Report on Six Seminars About the UK Supreme Court' (2008) 1 Queen Mary University of London, School of Law Legal Studies Research Paper, 32).
 Justice O.
 Justice Q.
 Justice N.
 Benjamin Alarie and Andrew Green, 'Should They All Just Get Along? - Judicial Ideology, Collegiality, and the Appointments to the Supreme Court of Canada' (2008) 58 University of New Brunswick Law Journal 73, 82.
 Article 151 of court's rules of procedure establishes that secret sessions might take place: (I) when one of the justices provides relevant reasons for it or (II) by request of the Chief Justice to discuss administrative matters. The first are sessions of the council and the latter are the so-called 'administrative sessions'. The sole paragraph of Article 152 explicitly provides that in the first case, the judgement session that follows the session of the council must be public.
 Justice O.
 Justice S.
 Justice P. The administrative sessions still take place several times a year. However, in these sessions, the justices do not discuss cases that are pending in the Court; they only address administrative matters. In some exceptional cases, the administrative sessions may still be used to define some procedural details of the decision-making process of pending cases, but never the merits of a case.
 Justice C. On the 11 island metaphor, see, for instance, Guilherme Forma Klafke and Bruna Romano Pretzel, 'Processo Decisório no Supremo Tribunal Federal: aprofundando o diagnóstico das onze ilhas' (2014) 1 Revista de Estudos Empíricos em Direito; Diana Kapiszewski, 'How Courts Work: Institutions, Culture, and the Brazilian Supremo Tribunal Federal' in Javier A Couso and others (eds), Cultures of Legality (Cambridge University Press 2010) 62.
 Justice I.
 Justice E.
 Justice D.
 Justice O. The Barbosa Court [2012-2014] was marked by open conflicts and fights—sometimes very harsh and impolite — between the Chief Justice Joaquim Barbosa and some of the associate justices, and sometimes between Barbosa and attorneys. Since the plenary sessions of the Brazilian Supreme Court are broadcast live on TV, these fights could also be followed live on TV.
 Justice Q.
 Justice R.
 Justice P.
 For a different perception, see Kapiszewski (n 81) 63–65. It is important to bear in mind that the retired justices may unconsciously romanticize their time on the Court, especially those who left the Court long ago. Additionally, when someone is asked about his or her perception concerning events, persons and situations from the past, their memory is likely to be inaccurate or biased. For a good account of several cognitive biases (including confirmation bias, rewriting of memory and others), see Anthony G Greenwald, 'The totalitarian ego: Fabrication and revision of personal history.' (1980) 35 American Psychologist 603.
 Justice H.
 Justice E.
 Justice I.
 Justice Q.
 Justice U.
 Justice S.
 In the definition of this 'ideal model', the justices were not bound by any constitutional, legal or regimental actual provision. In other words, they were at liberty to create the model they desired.