Do the Lava Jato Leaks Show Illegal or Unethical Behavior? A Debate Between Brazilian Legal Experts

As readers of this blog are likely well aware, last month The Intercept published a series of articles, in both Portuguese and English, that called into question the fairness, legitimacy, and motivations of the Lava Jato (or “Car Wash”) anticorruption operation in Brazil. These articles were based on private text messages between prosecutors and then-Judge Sergio Moro (and among members of the prosecution team) that The Intercept obtained from an anonymous source (widely suspected to be an outside party who hacked prosecutors’ cell phones). The revelations raise a number of questions about the Lava Jato operation, including whether the leaked text messages demonstrate that Judge Moro violated Brazilian law and/or ethical codes, and if so whether these breaches would invalidate the convictions of at least some of the Lava Jato defendants, most notably former President Luiz Inácio Lula da Silva (known as Lula).

Shortly after the first set of Intercept stories came out, I offered my own perspective on the implication of the leaked text messages (see here and here). But on the specific question of whether these text messages were unlawful or unethical, I was and remain uncertain, not least because evaluating this particular question requires expertise in Brazilian law. To help shed further light on this topic, and to assist others in understanding the complex legal and ethical questions at stake, today’s blog post features a point-counterpoint debate between two Brazilian legal experts with opposing perspectives on this question:

  • First, Ademar Borges de Sousa Filho (a Professor of Law at IDB-Brasilia and a practicing defense attorney) makes the case that the text messages disclosed by The Intercept demonstrate that Judge Moro behaved unethically and unlawfully, and that his lack of impartiality requires the nullification of the conviction of Lula (and possibly other Lava Jato defendants, though any such decisions would need to be made on a case-by-case basis).
  • Next, Luciano Benetti Timm (the National Consumer Protection Secretary at the Brazilian Ministry of Justice and Professor of Law at FGV São Paulo) presents a rebuttal, arguing, first, that the unauthenticated text messages obtained by The Intercept are not legally admissible, and that even if they were, they do not demonstrate any illegal partiality, or unethical behavior, by Judge Moro, and therefore do not provide grounds for questioning the convictions of Lula (or any other Lava Jato defendant).

Before proceeding, I should note that there are a number of other legal and political issues that are being hotly debated inside and outside of Brazil related to the Lava Jato case, Lula’s conviction, and related matters. The pieces below do not address these other issues, because I specifically requested Professor Borges and Professor Timm to focus narrowly on the question of the legality/ethics of the communications between Judge Moro and the Lava Jato prosecutors. I hope that readers find the debate below useful and enlightening on this issue.

POINT (Professor Ademar Borges): The Lava Jato Leaks Demonstrate that Former Judge Sergio Moro Lacked Impartiality, Meaning that Lula’s Conviction Must be Vacated

The exposés published by The Intercept Brazil have disclosed the content of messages exchanged between former federal judge Sergio Moro and Lava Jato prosecutors, particularly lead prosecutor Deltan Dallagnol. The large number of messages exchanged between the prosecutor who was in charge of the Lava Jato task force and the federal judge responsible for adjudicating the charges brought by that same prosecutor call into question the impartiality of that judge, and require nullification of the case or cases in which that lack of impartiality has been revealed.

The Accusatory System and the Impartiality of the Criminal Judge in Brazil

To understand and evaluate the impropriety of the text messages between the presiding judge and the prosecutor, it is first necessary to understand that Brazilian law rigidly separates the functions of the judge and the prosecutor in criminal proceedings. Indeed, this is no longer the subject of significant disagreement in national case-law. In 2014, the Supreme Federal Court (the Supremo Tribunal Federal, or STF) set an important precedent (ADI 5104 MC, Rapporteur Justice Luís Roberto Barroso, Full Bench (Plenary)) that affirmed the absolute separation between the functions of accusing and of judging in the criminal process, declaring that “the 1988 Constitution made an unequivocal choice for the accusatory system—not the inquisitorial system—creating the basis for a profound change in the conduct of criminal investigations and prosecution of criminal cases in Brazil.” The most important feature of this accusatory system, as Justice Barroso’s majority opinion stressed, is the establishment of “a rigid separation between prosecuting and judging.” This rigid separation, the STF continued, has two main consequences for Brazilian criminal procedure: First, the judge ceases to play an active role in the investigation and prosecution stage, in order to preserve his or her neutrality at the trial and judgment stage of the case. Thus the judge must remain passive, acting only in response to the parties. As STF opinion put it, Brazilian criminal procedure “segregates the investigation, prosecution, and trial phases, and it is not permissible for the judicial authority to determine the course of the investigations, to the detriment of its own neutrality.” Second, and relatedly, the judge must treat the prosecution and defense equally, remaining equidistant from both parties. These principles, though articulated especially clearly and forcefully in Justice Barosso’s 2014 decision, were hardly novel, as the same basic understanding had already been asserted on other occasions. For example, in a 2008 decision (HC 95009, Rapporteur Justice Eros Grau, Full Bench), the STF declared that Brazil’s accusatory system prohibits a judge in a criminal case from becoming directly and personally involved with the police and prosecutors in the planning of the investigations, because a judge who does so can no longer be considered independent and impartial. The requirement of judicial impartiality, the Court held, requires the criminal judge’s “complete disengagement from investigative activity and active combat of crime.”

Moving from these general principles to the specific statutes on point, the Brazilian Criminal Procedure Code guarantees the impartiality of the judge in Article 252 (which establishes rules of “impediment,” prohibiting a judge from presiding over a case when the judge has a family relationship with the parties or has acted as a lawyer or witness in the case) and in Article 254 (which establishes what in Brazil are called rules of “suspicion,” allowing either party to demand recusal of a judge who has certain other relationships with the parties to the case, relationships that the judge must disclose). As most relevant here, Article 254, paragraph IV requires disclosure and on-demand recusal if the judge has “advised” either party. Furthermore, the general “suspicion” clause of the Civil Procedure Code (Article 145, paragraph IV), which is applicable to criminal proceedings, says that a judge is “suspected” whenever he or she is “interested in the judgment of the case in favor of either party.”

In addition to these statutory provisions, Brazilian courts have applied analogical and teleological reasoning to elaborate and refine the rules on judicial impartiality (see, for example, here). As the Superior Court of Justice (the Superior Tribunal de Justiça, or STJ) has explained, there are a multitude of other connections between judges and parties that can corrupt the impartiality of the judge, in addition to those expressly provided by the Criminal Procedure Code (see, for example, HC 331.527, Rapporteur Judge Jorge Mussi, Fifth Panel, 10/21/2015). When deciding whether there is sufficient “suspicion” of the judge, Brazilian courts assess whether the personal attitude of the judge in the concrete case demonstrates a lack of equidistance in relation to the parties. In making this assessment, Brazilian courts take an objective approach, asking whether there are concrete facts sufficient to call into question the impartiality of the judge, who is obliged to offer sufficient guarantees to eliminate any doubts about his or her impartiality. As most relevant here, the STJ has held that a judge may exhibit lack of impartiality by conduct that demonstrates a preference or predisposition toward any of the parties. For example, the STJ has held a criminal judge who speaks about a case with a higher court judge who would participate in hearing the appeal of that case has impaired impartiality (HC 172.819, Rapporteur Judge Jorge Mussi, Fifth Panel, 02/28/2012). And perhaps the most common reason that a criminal judge is deemed to have lost his or her objective impartiality is when that judge participates actively in obtaining evidence in preliminary proceedings (see, for example, STF, HC 94641, Rapporteur Justice Joaquim Barbosa, Second Panel, 11/11/2008).

To be clear, none of this is to deny that in Brazil, both the prosecution and the defense may hold private meetings (what in some legal systems are called ex parte meetings) with the judge. Indeed, the STJ has ruled that “the conversation between magistrate and prosecutor on cases in progress does not, in itself, reveal suspicion of the mentioned authorities…. Lawyers are received very often by … judges without any kind of nullity.” That said, informal private conversations between a prosecutor and a judge may result in the loss of the judge’s impartiality when:

  1. the content of the communications demonstrate that judge has “advised” the prosecutor (within the meaning of Article 254, paragraph IV of the Criminal Procedure Code); or
  2. when the judge’s conduct in the proceedings demonstrates partial behavior; or
  3. when the judge performs functions typically performed by the Public Prosecutor or the police, such as participating in the planning of the investigation or the prosecution’s strategy (see STF, HC 95009, Rapporteur Justice Eros Grau, Full Bench, 11/06/2008).

Finally, in addition to these legal requirements, Brazil’s Code of Judicial Ethics also requires (in Article 8) that the judge must remain equidistant from the parties throughout the process and avoid any kind of behavior that may reflect favoritism, predisposition, or prejudice. Furthermore, Article 9 of the Ethics Code states that the judge must provide the parties with equal treatment, and that although this requirement does not prohibit the judge from holding a private hearing with lawyers for one party, this is only permitted if ”the other party is guaranteed the same right if requested.”

The Informal Conversations Between Judge Moro and the Lava Jato Prosecutors Regarding Pending Cases Violated the Requirement of Judicial Impartiality

In light of the rules and principles just discussed, it is clear that the messages exchanged between former Judge Moro and prosecutor Deltan Dallagnol indicate a lack of judicial impartiality.

First of all, although (as noted above) Brazilian law authorizes the judges to receive lawyers and members of the public prosecutor’s office in their chambers to discuss ongoing proceedings, the extensive exchange of private electronic messages with only one side (here the prosecutors) to discuss matters relating to those proceedings violates both the law and the Code of Judicial Ethics, both of which require the judge to remain equidistant between the parties, so as to maintain judicial impartiality.

Some messages exchanged between Judge Moro and the Lava Jato prosecutors are particularly problematic:

  • In one of them, Judge Moro advised Mr. Dellagnol to alter the order of the phases of the Car Wash Operation, which would imply accelerating certain arrest and search and seizure requests and delaying others.
  • In another message, Judge Moro suggested that the prosecutors issue a press release clarifying alleged contradictions in Lula’s testimony.
  • In yet another text exchange, Judge Moro informed Mr. Dallagnol that he had obtained information about the alleged transfer of ownership of property of one of Lula’s sons, and Judge Moro suggested that the prosecutor speak with a particular person to confirm the information.

These messages reveal the judge’s partial behavior—a clear predisposition towards the prosecution, as well as involvement in strategic planning and the timing of the investigation itself. Such messages show not only that Judge Moro was not impartial in a general sense, but that he was “advising” the prosecution, within the meaning of Article 254, paragraph IV of the Criminal Procedure Code. And the abnormality of this interaction between the judge and the prosecutor—and the lack of equidistance between the prosecution and defense—becomes all the clearer when one observes that Judge Moro did not adopt the same attitude of “collaboration” with the defense attorneys.

Now, to be clear, the Lava Jato Operation involves hundreds of cases and criminal defendants, and it is not possible to declare with confidence that these breaches of the judicial impartiality principle affects all of these cases. Whether the judge’s partial behavior requires the nullification of a conviction depends on whether such lack of impartiality can be demonstrated in the individual case. But while it’s not possible to make a blanket statement about other Lava Jato cases, there is no doubt, based on the text messages that have been reported so far, that Lula’s conviction must be vacated due to Judge Moro’s demonstrated lack of impartiality in that case.

Now, on that point, some have argued that the fact that these text messages may have been obtained illegally (possibly by hackers, though this has not yet been definitively established) means that the messages may not be used by a defendant, such as Lula, to challenge his conviction on the basis of judicial partiality. This is incorrect. The alleged unlawfulness in obtaining the messages does not prevent them from being used in favor of the defendant to prove a lack of judicial impartiality (see, for example, STF, HC 74678, Rapporteur Justice Moreira Alves, First Group, 06/10/1997). The principal reason, as the STF case law has explained, is that the greater constitutional relevance of individual freedom in the face of the prohibition to use illegal evidence justifies the admissibility of unlawfully obtained evidence in favor of the defendant (see STF, HC 75261, Rapporteur Justice Octavio Gallotti, First Panel, 06/24/1997, and STF, RE 212081, Rapporteur Justice Octavio Gallotti, First Panel, 12/05/1997).

To sum up, while nobody disputes the importance of fighting corruption in Brazil, the ends do not justify the means. Nobody should be convicted except by an impartial judge. And this requirement of judicial impartiality requires the judge to adopt a position of equidistance in relation to the defense and the prosecution. A judge who exchanges hundreds of private messages with the prosecutor about the case will not be able to eliminate doubts about his impartiality—especially if these messages reveal that the judge was deeply committed to the prosecution’s point of view from the beginning. A judge who engages in planning an investigation and prosecution violates the ideal of ​​a fair trial and erodes the credibility of the judiciary. Former Judge Sergio Moro’s failure to adhere to this fundamental principle of judicial impartiality violated both the Judicial Code of Ethics and the law of criminal procedure. As a result, his rulings on those cases that he discussed informally and systematically with prosecutors—including that of former President Lula—must be vacated.

COUNTERPOINT (Professor Luciano Timm): The Lava Jato Leaks Do Not Demonstrate Partiality, Illegality, or Impropriety, and Do Not Justify Vacating Any of the Lava Jato Convictions

Professor Borges, like many other commentators inside and outside of Brazil, argues that the text messages between Judge Moro and the Lava Jato prosecutors demonstrate that then-Judge Moro acted unethically and illegally, and that in light of this new evidence, Lula’s conviction (and perhaps the convictions of other Lava Jato defendants) must be nullified. Professor Borges has laid out his case ably and clearly, but his analysis is incorrect. First, unauthenticated text messages obtained by illegal means are inadmissible as evidence, and thus could not be the basis for vacating the conviction of Lula or any other defendant. Second, even if one ignores that fundamental problem and assumes, for the sake of argument, that these text messages are authentic and legally admissible, none of the messages disclosed so far shows behavior that violates Brazilian law or ethical codes.

Illegally Obtained Evidence Is Inadmissible in Court

I shall begin where Professor Borges ends, with the question of whether the text messages disclosed by The Intercept could be used in a legal proceeding challenging the conviction of any of the Lava Jato defendants, such as Lula. The answer is no, for two reasons.

  • First, the text messages have not been authenticated. Under Article 439 of the Brazilian Civil Procedural Code (which applies to criminal proceedings when Criminal Procedure Code doesn’t have a specific provision on the same point), for an electronic document to be admissible in court, it must be printed and subjected to authenticity check by an independent expert. This has not been done. Unless and until it is, these alleged conversations mean nothing from a legal point of view. And this is the right approach, given that, without authentication, we do not know if the content of these messages is accurate. Indeed, the editor of The Intercept, Mr. Glenn Greenwald, has already acknowledged that he edited a few of the messages. While he explains his alterations as “corrections,” this fact highlights how easy it would be for the original source to doctor the messages.
  • Second, even if the messages were authenticated, they would be inadmissible in court because these texts were obtained by illegal means. Brazil’s Internet Law, in Article 7, paragraph III, guarantees the inviolability and confidentiality of private communications stored online, except when obtained pursuant to a judicial order. And Brazilian law expressly prohibits the use of illegally obtained evidence. Not only is this prohibition found in the Brazilian Constitution (Article 5, paragraph LVI), but Article 157 of the Brazilian Criminal Procedure Code also explicitly states that “unlawful evidence, … obtained in violation of constitutional or legal norms, is inadmissible and must be disregarded”). Professor Borges argues (as do a few other commentators) that even unlawfully obtained evidence can be used by a criminal defendant to challenge the impartiality of the judge. But this is incorrect, and the cases Professor Borges cites are inapposite. (An aside here, but an important one: unlike the United States, there is no stare decisis doctrine in Brazil, and courts can and do frequently change their opinions.) First of all, none of the cases cited by Professor Borges declare that illegally obtained evidence can be used to challenge a conviction on the grounds that the evidence in question shows the partiality of a judge. Rather, the cases that he cites deal with distinguishable situations, mainly involving the recording of telephone conversations in which only one side of the conversation knew the conversation was being recorded. In one of the cases (HC 75261), the recording was made by the police officers that negotiated the release of a victim, while another (HC 74678) is about the use of a phone recording made by a third party with the authorization of one person and without the authorization of the other. Thus, these cases involve materially different facts; they do not involve the illegal hacking of law enforcers fighting corruption. Additionally, more recent developments in the STF’s case law point in the opposite direction: Just last week, on June 28, 2019, the STF ruled that unlawfully obtained WhatsApp messages cannot be used as evidence in a court of law (HC 168052. Rapporteur Judge Gilmar Mendes. First decision made on 11/06/2019, other votes still pending). And it makes good sense for Brazilian law to prohibit the introduction of evidence obtained through illegal hacking, as allowing such evidence would create incentives for criminals to hack smart phones and other electronic devices of public authorities.

The Text Messages, Even if Authentic, Do Not Reveal Illegal or Unethical Behavior

Even if one ignored the above issue and treated these unauthenticated, illegally obtained text messages as legitimate evidence, this evidence would not suffice to show illegal or unethical behavior by Judge Moro or the Lava Jato prosecutors.

Before proceeding to the analysis of these specific messages, it’s necessary first to clear up some confusion about the background principles and norms of the Brazilian criminal justice system. Professor Borges asserts that Brazil has an “accusatory system” of criminal justice (in contrast to the “inquisitorial system” used elsewhere), and that this accusatory system “rigidly separates the functions of the judge and the prosecutor in criminal proceedings.” But this is at best an exaggerated and incomplete statement of how the Brazilian criminal justice system works. Although the Brazilian system contains elements of the accusatory model, it is not a “pure” accusatory system, and it retains important elements of the inquisitorial model. For example, Article 156 of the Criminal Procedure Code states that the judge may have a more participatory role during the trial, and may seek evidence, according to the principle of “seeking the real truth” (“princípio da verdade real”). More specifically, according to the Criminal Procedure Code a judge can interrogate the defendant, pose as many questions as he or she wants during the hearing, ask for technical expert evidence, and so on. The fact that the judge engages in these truth-seeking activities does not mean that the judge is biased. Quite the opposite. The judicial impartiality principle means that the judge’s rulings must be based on the evidence presented to the court, not that the judge must be a wholly passive participant following the model of a U.S. judge. Also relevant here is the fact that prosecutors in Brazil have a special status, different from lawyers for other parties. Article 128 of the Brazilian Constitution declares that prosecutors are “defenders of legal order, the democratic regime, and social rights” and have the same constitutional guarantees like judges to protect their independence and autonomy. In hearings, the prosecutor sits beside the judge at the bench, in contrast to lawyers for other parties, who stay beside the client and away from the judge’s bench. This is yet another important contrast with pure accusatory systems of justice. It is true that a handful of STF opinions (including ADI 5104 MC, the case mentioned by Professor Borges) use more sweeping language endorsing the accusatory model. But these decisions—which, again, are not binding precedents—should not be understood to radically change long-established Brazilian legal culture and practice regarding the more active role of the criminal judge and the special status of the prosecutor.

Turning to the more specific question of whether or under what conditions it is appropriate for a judge to engage in ex parte communications with prosecutors: First of all, as Professor Borges acknowledges, not only does the Brazilian legal system allow ex parte contacts between judges and lawyers for one side or the other, but in fact it is customary for lawyers to have private communications with a judge without the presence of the other parties’ lawyers. Professor Borges is right to acknowledge that the STJ has explicitly ruled that “the conversation between magistrate and prosecutor on cases in progress does not, in itself, reveal suspicion of the mentioned authorities…. Lawyers are received very often by … judges without any kind of nullity.” Nevertheless, Professor Borges and others argue that these private communications between Judge Moro and the Lava Jato prosecutors violate the principle of judicial impartiality, as enshrined in Articles 252 and 254 of the Criminal Procedure Code, Article 145 of the Civil Procedure Code, and Articles 8 and 9 of the Code of Judicial Ethics. But these assertions are mistaken. None of the cited provisions, or the more general Brazilian case law, prohibits the sorts of communications between the judge and the prosecutors that The Intercept has disclosed.

To begin with, we need to clear away some misleading suggestions regarding the applicability of certain legal or ethical rules to this case:

  • The idea that there’s a possible violation of Article 252 of the Criminal Procedure Code can be dismissed out of hand, as this provision is limited to the specific relationships (mainly family relationships) between the judge and a party. There is no serious argument that any of these specific relationships applies in the Lava Jato cases heard by Judge Moro.
  • Although Professor Borges also involves a general provision in Article 145, paragraph IV of the Civil Procedure Code (which says that a judge may be deemed to lack impartiality if the judge is “interested in the judgment of the case in favor of either party”), the Civil Procedure Code is not applicable here, because there are specific rules of the Criminal Procedural Code on the matter. (The case cited by Professor Borges to suggest that the Civil Procedure Code could be applicable deals with a very different situation, where the criminal defendant was the presiding judge’s attorney in a separate civil case. Judge Moro was a not a client of any Lava Jato defendant.)
  • Next, while Professor Borges and others have suggested that there’s a distinct problem in this case because the judge and the prosecutor were communicating via text message, rather than holding their ex parte conversations in the judge’s chambers, in fact neither Article 9 of the Judicial Ethics Code nor any other provision prohibits otherwise lawful ex parte contacts between judges and lawyers to take place by text message rather than in chambers. Smartphones can be and constantly are used for work, and there is no reason why communications through texts should be treated differently from in-person conversations, so long as the nature and purpose of the communications are equivalent. Indeed, Brazil’s National Justice Council officially decided to allow notifications via WhatsApp for both civil and criminal proceedings.

With that out of the way, we can proceed to the more significant objection to the text exchanges between Judge Moro and the Lava Jato prosecutors: that they demonstrate partiality (or, to use the Brazilian terminology, “suspicion”) within the meaning of Article 254, paragraph IV or the Criminal Procedure Code (or, perhaps, some more general principle of impartiality recognized in the case law). This claim is not as frivolous as the others just discussed, but it is still incorrect.

  • First, as emphasized above, the exchange of information between prosecutors and judges (the latter of whom apply the fundamental principle of “seeking the real truth”) are in fact quite common in Brazilian criminal justice. There would be a violation of the judicial impartiality principle only if Judge Moro had prejudged the merits of the case. But an examination of the text messages that have been disclosed (again, assuming for the moment that they are authentic) reveals only contacts concerning procedural aspects and measures related to the investigation, or general communications. At no time did Judge Moro address the merits or indicate that he had prejudged the legal or evidentiary issues. Nor do any of these texts indicate that Judge Moro planned to convict the defendants regardless of the evidence, or that he had any sort of political bias. For example, Professor Borges and others play up the fact that Judge Moro suggested that the prosecutors alter the order of the phases of the investigation. But according to the STF (AS 89 AgR. Rapporteur Justice Edson Fachin, Full Bench, 13/09/2017), more is required in order for there to be “suspicion” of a judge within the meaning of Article 254 paragraph IV. Moreover, in making his argument that the communications at issue here show lack of impartiality under Brazilian law, Professor Borges relies prior cases that are not on point. For example, the STJ’s decision in HC 172.819 involved a trial judge who had called the appellate court judges who were going to hear the appeal of the trial judge’s case, and who in these calls exhibited an interest in having his decision upheld on appeal. There is nothing similar in any of the text messages published by The Intercept.
  • Second, even if there were some question about the propriety of the text messages, the evidence is nowhere near clear enough to support a nullification of Judge Moro’s rulings. In advancing a challenge to a judge’s impartiality, the burden of proof is on the defendant to show “concrete and objective” evidence that of “partial behavior of the judge in the proceeding, incompatible with its functional competence,” in order to invalidate the judge’s exercise of jurisdiction over the case. (Though the case quoted was in the civil context, the same principle applies in criminal matters.) Indeed, Brazilian jurisprudence on suspicion of judges in the criminal area quite restrictive, in order to avoid opportunistic behavior. The texts disclosed in Judge Moro’s case are, at the very least, sufficiently ambiguous that neither Lula nor any other Lava Jato defendant could meet this demanding standard.
  • Third, even if it were true that Judge Moro exhibited partiality in the first instance trial, that failure did not prejudice Lula (or any other defendant), given that the verdicts were reviewed and validated by the Regional Federal Tribunal. The Regional Federal Tribunal not only confirmed the verdicts in the cases at issue, but increased the penalty of the defendants. Later on, STJ also affirmed the conviction. (Furthermore, the fact that the STF affirmed more than 80% of Judge Moro’s preliminary decisions in the Lava Jato cases casts doubt on any claim that he was systematically biased in his procedural or other pre-trial rulings.)

As for the separate but related assertion that Judge Moro breached his ethical duties as a judge by showing favoritism toward the prosecution, the response is generally similar to the response to claims that Judge Moro acted illegally. Although Brazil’s Judicial Code of Ethics—as well as the Federal Prosecutors’ Code of Ethics—list a variety of duties related to independence, impartiality, symmetry, and the like, there is no breach of any of these ethical duties if there is a conversation ex parte or if the judge seeks evidence in the name of the principle of the search for truth or makes suggestions concerning the general organization of an investigation.  And even if there were a breach of an ethical duty, unless there is also a violation some legal provision, this would not affect the finality of any verdict. Rather, an alleged ethical breach would be handled separately by the ethical board of the entity involved (judges or prosecutors), with a view to imposing an appropriate sanction on the individual who behaved wrongfully (warning, admonishment, censure, among others), without any implication for the cases already judged. As it happens, the Brazilian National Prosecutor Council already resolved the ethical complaint brought against Mr. Dallagnol in relation to the text messages disclosed by The Intercept, and concluded that not only the messages were illegally obtained and may not be authentic, but also that Mr. Dallagnol’s texts did not breach his ethical duties. There is little reason to believe that Judge Moro’s case will be decided differently.

In sum, the text conversations published by The Intercept are legally inadmissible and in any event do not show any abnormality when evaluated in the context of Brazilian law and practice, which allow private communications between judges and prosecutors regarding the sorts of matters discussed in these texts.

Before concluding, it’s important to step back and recognize the larger context here. In the past, attempts to prosecute high-level corruption networks in Brazil have been stymied by courts on a variety of formalistic grounds raised by powerful politicians and wealthy entrepreneurs. Lava Jato has been revolutionary because, for the first time, prosecutors have breached the seemingly impenetrable wall of impunity, and successfully convicted scores of powerful people from different political parties. All of these defendants were guaranteed due process of law and convicted only when their corruption was proven beyond any reasonable doubt. But the efforts to undermine Lava Jato continue. This does not mean that the normal legal and ethical rules do not apply to the Lava Jato operation. But it does mean that, when assessing the brief excerpts from past conversations, as well as the recent commentary on The Intercept’s reports, one must be appropriately sensitive to this broader context. Lava Jato is unprecedented, and everything about this operation and the actors involved should be evaluated bearing in mind Ortega Y Gasset’s wise observation: “I am I and my circumstances.”

68 thoughts on “Do the Lava Jato Leaks Show Illegal or Unethical Behavior? A Debate Between Brazilian Legal Experts

  1. First of all, I want to thank both of you for this thoughtful, enlightening, and respectful debate. This is a hugely important and complex question, and one that it’s hard for outsiders like me to comprehend (a fact that those who dislike what I’ve written are constantly reminding me). The topic is understandably one that arouses a lot of passion, and sometimes anger, on both sides, but sometimes the arguments generate more heat than light. The two of you have helped me to understand the strongest legal arguments on both sides of the question, and while I’m still not in a position to form a confident opinion, at least now I have a better comprehension of the key points in dispute. I hope other readers, both inside and outside Brazil, find the exchange similarly useful.

    I do have a number of remaining questions about the legal and ethical issues raised by the text messages published by The Intercept. I’ll pose these follow-up questions in separate comments, so as to start separate comment threads on each of them. I hope that other readers with views on these or other questions will post comments as well.


    One of the text exchanges highlighted by the Intercept involved a situation where Judge Moro had been told (by a third party) that there was a witness who had evidence that might be helpful to the prosecutor. Judge Moro relayed this information to Mr. Dallagnol. My question is whether this act itself–passing on information about a possible witness to the prosecutor–is itself illegal/unethical under Brazilian law. More generally, if a witness approaches a judge (directly or through an intermediary) and says, “I have evidence relevant to the case you’re hearing,” may the judge permissibly inform the lawyer for the party whom the evidence helps, or must the judge tell the potential witness to contact the lawyer for the relevant side directly? Are there any Brazilian laws or judicial decisions on that point?

    • No, the act itself of passing on information to the prosecutor is not illegal or unethical under Brazilian law. On the contrary: it is admitted by the article 40 of the Criminal Procedure Code. This article states that a judge, when confronted with evidence of a crime, has to send the documents to the prosecution. As you can see, this article is not about a witness, but I would say it’s fairly accepted that if a witness approaches a brazilian judge directly, the judge has to send the witness to the prosecution (as he would do with a document).
      Since the act of informing the prosecution is provided by law, a formal procedure would have to be done. In other words, it is expected that the judge sends a formal notification to the prosecution.
      The alleged messages, however, shows that this notification occurred in an app. For those questionning Moro’s acts, this mistake is more than a mistake: is a cause of nullification.
      In my opinion, it isn’t. In Brazil there is a widely accepted latin principle: “ne pas nullité sans grief”, which means that there is not a nullification if there is no prejudice.
      In this episode, there was no prejudice, since the witness was not questioned or heard.

      • If I may, that interpretation is not correct. Article 40 is very clear: “when, in the case file (autos) or papers that have become known, the judges or courts verify the existence of a crime of public action (in which charges are not required to be filed by the offended), will refer to the Ministério Público (state or federal attorney office) the copies and documents needed to the offerring of the indictment”. Let it be noted that I’m translating in a way to try and approximate the terms in a way that would make more sense to english speakers.

        In any case, this was *not* the case. Moro did not refer documents or written proof of a “crime of public action” to the MP. On the contrary, he tipped the prosecutor about a possible witness that could strenghten his case.

        Not only this referral *not* registered in any formal document, without entry log withing the MP, it was not used to offer any indicment regarding any crime of “public action”.

        A judge using of “hearsay” to tip the prosecution regarding a possible crime (not discovered in the case file or written documents) is NOT, in any shape or form, allowed by article 40 od the Criminal Procedure Code.

      • I do agree. The judge might even listen the witness as a “Court witness”, according to article 209 of the Brazilian Criminal Procedure Code.

      • I totally agree with you.
        This behavior is allowed by the Criminal Procedure Code (article 40).
        Despite it is prescribed to be registered, it is admitted thought app, since it didn’t reveal any prejudice. This only concerns about format, not material content.

    • No, it isn’t if this happens inside the official procedures where the defence can see the request. Not in the dark by telegram.
      You should read the last leaked messages today, it is decisive. The prosecutor even sends an unfinished piece to him to advance his ruling. I agree with the theory, LJ is important and a game changer in Brazil, but Mr. Moro is not LJ and LJ is not him. To question his conduct is not an attack to the operation, but many do the attacks and many see this way.
      Unfortunately, he inspired himself too much in the Manu Pulite and forgot that Brazil doesn’t have a judge for instruction. He completely lost himself and the media was too blind in the eager to put Lula in jail.
      If they would have wait until the country house, they would have a stronger case, with more proofs (after a not guilty in the apartment), but they had the election timing and decided to force a conviction. Using the media to break the presumption of innocence towards the general public was the nail in the coffin.
      As a comparison, give a look at the power point presentation from Delagnol and the presentation of the accusation of Making a murder. You can relate. They went to the media and sold horrendous crimes (the godfather of all Brazilian corruption), but when came to the lawsuit, none of the accusations had a base or was mentioned. The accusation made on the media by them doesn’t match the real facts. You need to understand. Politicians had bags of money tracked and taps saying that they would kill the person transporting before they say anything for the police. He is still free and recently elected. You had wiretaps with politicians conspiring against the presidency, detailing all their plans, they became ministers and all they said eventually start to become true.
      Lula is not innocent. Is impossible for any Brazilian president to be in our system, but he is far away from being what the media and the right wing want to portrait him.

    • I believe this is, of course, the most problematic part of the conversations. My understanding is that this is inadmissible under Brazilians laws, not only because of the Criminal Procedure Rules but also because of the Magistrates’ Ethics Code that requires all judges to keep “an equal distance from both parties avoiding all kinds of behavior that could reflect favoritism, predisposition of prejudice”. Laws governing criminal trials in Brazil are much of puzzle but there is an evolving understanding that under the 1988 all laws should be interpreted in a way that maximizes the judge’s neutrality (I’ll be back to this point in the next follow-up) questions. There’s also an equally problematic aspect of it: a source seeing the judge as the one in charge for the investigation, feeling free to reach to him tipping about the investigation, which means somehow the larger audience saw him as the investigator himself. The conversations suggest some sort of negotiation on how the information is going to be introduced to the case. My reading of current laws is that a judge approached by a source in that manner should limit him/herself to inform he is not the authority that should receive such information.

      Having said that I believe this is not, unfortunately, an exceptional case in criminal courts in Brazil. This transition towards a more liberal/progressive approach to criminal justice, made by the Constitution as pointed by Professor Sousa, is far from being well-reflected in daily activities of all courts.

      • Interesting. Perhaps the fact that the understanding of the law in this area has been evolving, and the fact that there seems to be quite a large gap between what the law’s requirements and routine practice, may explain why there seems to be such disagreement even among experts about how Judge Moro ought to have handled this situation.

        In your view, is this exchange sufficient to taint the trial? Or does the fact that, in the end, nothing seems to have come of this mean that, without more, this conduct might merit a reprimand of Judge Moro but not a vacation of the judgment?

        • I think this is a fair reading. Also, when it comes to corruption or violent crime it seems that there is a tendency to have the public opinion more tolerant with judges breaking rules of impartiality because these goals are perceived as crucial to “protect society”.

          I think that the exchange tipping on a potential witness and discussing ways to robust the case are enough to taint the trial. I think there is a less clear part of it which is whether the articles are enough to do that, or it’d require accessing the raw data to confirm the messages are authentic. I think the messages couldn’t be used to take any actions against Moro and the prosecutors because of the rules preventing illegally obtained evidence to be use against defendants (even in administrative/ethical proceedings). Vacating Moro’s decision here would be address the concerns with ensuring the right to a fair trial without necessarily digging into whether Moro and the prosecutors willfully committed any violation.

          We certainly will see some attempts to hold them accountable for that but I have trouble to see any of these attempts flying.

          There’s an interpretation under which the higher courts decisions remedy any procedural mistakes made by the lower court (here Moro’s). Here we are in unexplored waters, but I resist to that reading first because there’s no statutory provision pointing to that and the statute shouldn’t be read in a way that weakens defendants’ rights; and secondly because as evidence gathering is made before the trial court, higher courts’ decision are based on evidence collected by a judge that couldn’t do so and therefore there’s no valid decision for them to confirm.

          • Something that I am still unclear on is whether these messages enjoy the same level of privacy that personal messages between two citizens do. It seems to me that these communications between judge and prosecutor, if they were done in the traditional manner, would be recorded for the court’s record even if they weren’t generally available to the public. Is this true?

            The illegality of how they were obtained is problematic in using anything revealed to bring criminal charges but I don’t see why the authenticity debate is still ongoing if what I mention in the first paragraph is true.

  3. If this text is to portray a debate, shouldn’t the first to state his point be given the opportunity to at least react to the (lengthy) counterpoint?
    Shouldn’t mr. Borges also be offered the opportunity to counter?
    Have you given him the opportunity to do so, Mr. Stephenson?
    I for one would be very curious about Mr. Borges opinion on Mr. Timm’s statement.
    And (for the sake of impartiality) it seems a bit unfair to simply give Mr. Timm the last word. ( just before Mr. Moro is to appear in front the chamber of deputies.) Let’s see if this blog will be quoted (and cherry-picked from) once again, in a few hours…

    • The three of us discussed this, and I originally offered exactly the structure you seem to want (point, counterpoint, rebuttal), but in further discussions both Professor Timm and Professor Borges agreed that the format used above (point-counterpoint) would be best. And both of them saw the other’s text before giving final approval to publish the debate.

      But thanks for immediately assuming bad faith and manipulation. That’s really helpful.


    In that same exchange regarding a potential witness, after Mr. Dallagnol attempts to contact the witness and the witness says he/she doesn’t want to talk, Mr. Dallagnol says to Judge Moro that he (Dallagnol) might attempt to subpoena the witness based on “apocryphal news.” (I believe the Portuguese phrase is “notícia apócrifa.” I’ve heard conflicting things about what this phrase (which doesn’t have a clearly comprehensible English translation) means in this context. One interpretation is that it means “anonymous tip,” and that Mr. Dallagnol is proposing to apply for a subpoena based on the (false) claim that he heard the witness had evidence from an unknown source. A second interpretation is that “notícia apócrifa” doens’t mean “anonymous tip,” but rather “confidential source”–in which case Mr. Dallagnol was planning to apply for a subpoena based on the (true) statement that he’d heard from a source that wished to remain anonymous. A third interpretation, similar to but different from the second, is that “notícia apócrifa” means something closer to “unsubstantiated report”–that the subpoena was based on a tip that the witness had information but that didn’t provide any corroboration or explanation of the basis of the claim. In that case Mr. Dallagnol’s proposal would also seem unproblematic, at least on this point. Is there consensus among practicing Brazilian lawyers about what the phrase “notícia apócrifa” means in this context? Are there other published examples or usages in standard sources that one can point to?

    • There is even a fourth interpretation (also suggested in, for example, [1]), which I think is the correct one.

      Since a Judge is not allowed to informally suggest a witness to the Prosecution, the idea was to leak the information to the press, and then, based on a news story, subpoena the witness.

      So it is a sort of “tip laundering”, which makes use of the right to source protection accorded to journalists.

      Even before the “Intercept” leaks, there was wide suspicion that the Lava Jato task force resorted to this method.

      [1] (Portuguese)

    • No, there is not consensus about the phrase “notícia apócrifa”. And this is the most important message of all, in my opinion, because on a first read it may give the idea of doing something against the law.
      First of all, let me say that “apócrifa”, on the portuguese dictionary, refers primarily to a religious meaning: the Gospels that are not in the Bible. In other words, unofficial Gospels. However, in latin, the word comes from “apocryphum” or “apocrypha”, which means hidden or secret. On the “law language”, usually “apócrifa” is used to refer to something that doesn’t have the formal requirements. Example: a document without the signature (it exist, but has no probatory value; being a “apócrifo” document).
      That being said, I respectfully disagree with the fourth interpretation given by Mr. Scarparo below. Apparently he took the phrase “notícia apócrifa” and used the first word (‘notícia’) as news (news given by the press). However, there is no indication of that.
      In my opinion, when Dallagnol allegedly used this “notícia apócrifa”, he meant anonymus information or confidential source. I’ve spoken to a prosecutor about this message and he has the same opinion. I can tell you that is very common in Brazil that a witness doesn’t want to be questioned or even have his name on the records, so the prosecution has to say that it is an anonymus information or confidential source.
      Even if was this the case, I can see that this would be wrong if went ahead, because the name of the witness was brought to prosecution by the judge (something that is not wrong as I have mentioned on the other comment, but definitely was not an anonymus person).
      However, it is very important to see all the dialogue, as it goes on. And seeing that, as Dallagnol states that the witness doesn’t want to be questioned, Moro seems in doubt of what has to be done. He allegedly says “Now I’m in doubt” which I understand as doubt of this being a legal procedure.
      At the end of this dialogue, Dallagnol says he will call the person that refered the name of the witness.
      All being said, I don’t see this dialogue as a cause of nullification. Simply because the subpoena did not occur. Something wrong was about to be done, but the prosecution didn’t go further. The two names allegedly mentioned were not questioned or heard.

      • Respectfully, the principle of “pas de nullité sans grief” is not applied in regards to the partiality of a judge.

        “He tried to help, but didn’t succeed” does not invalidate the fact that he *tried* to help. That, in and of itself, is enough to question his impartiality and, therefore, to void his sentence.

    • It usually means “anonymous notice” for investigation purposes; according to Brazilian case law, “notícias apócrifas” authorize the beginning of investigations.

      • Given that understanding, is what Mr. Dallagnol proposed appropriate or not? Would it be considered a misrepresentation because the source, even if not disclosed in the filings, was known to Mr. Dallagnol (and to Judge Moro)?

        And would the exchange call the fairness of the overall trial into question, or would the fact that nothing apparently came of this exchange (the subpoena was never filed, and the witness never gave evidence, so far as I can tell) mean that it would be deemed a form of “harmless error”?

    • “Notícia apócrifa” is an anonymous tip. The most significant difference here is that in the U.S. this could mean basically a police officer saying he heard someone saying it and in Brazil is this more likely to be something documented but not attributable to a specific person, such as an e-mail sent by a fake e-mail account or an anonymous letter. Besides the protection to witness program, Brazil does not have protection for confidential sources to a criminal case, so the second interpretation does not make sense to me. Third does seem accurate as well given that (i) if that’s the case it wouldn’t be necessary to hide the actual source of information and (ii) an “unsustained report” is not enough to grant a subpoena precisely because there isn’t enough base to the story being told. I struggle to read this as anything but a made-up anonymous tip. The need to “launder” the source of the information signals the problem in revealing the actual source.

      • OK, if that’s correct, then this exchange does seem problematic. Though the inability of the prosecutor to submit a subpoena request based on information from a confidential (but not anonymous) source doesn’t make a lot of sense to me. Consider the following scenarios:

        Scenario A: I’m working for an organization (say a firm or a government ministry) and I observe evidence of serious criminal wrongdoing. I send a sufficiently detailed description of the evidence to a prosecutor from a fake email account (or by anonymous letter). The prosecutor (who doesn’t know who i am) can apply for a subpoena based on “apocryphal news” (an anonymous source) and get the evidence–and I preserve my anonymity and thus my safety from retaliation.

        Scenario B: Exactly the same, except that I send the email from my actual account, along with a plea to the prosecutor to keep my name out of any public court filings, because I’m worried about retaliation. Now, if your description of Brazilian law is correct, the prosecutor is in a bit of a bind: She can’t (legally) request a subpoena on the basis of an anonymous tip (“apocryphal news”) because she knows who I am, even though she could have applied for a subpoena, on exactly the same facts, if only I’d concealed my identity in the first place.

        Is that actually Brazilian law on this point? Doesn’t it seem kind of absurd? And if it is indeed the law, I can see why prosecutors might (perhaps routinely) fib about their knowledge of the source’s identity, to as to avoid unwanted (and, for evidentiary purposes, unnecessary) disclosure.

        But perhaps I’m missing something?

        • Let me try to navigate through the nuances of it.

          Prosecutors can use anonymous information to base their investigations. Courts however have ruled that anonymous tips are not enough to base some investigatory steps, as issuing warrants to get data protected by banking secrecy, tax returns, mail and phone records. A predecessor of Lava Jato, Investigation ‘Castelo de Areia’ (Sand Castle) was entirely overthrown because of that.

          So, in the Scenario A, prosecutors can start an investigation but couldn’t immediately file requests seeking the sort of data I just mentioned. Other lines of investigation, including reaching out to potential witnesses would be ok. Moreover, subpoenas in pre-trial investigations do not require judicial orders. If the prosecutors simply want to hear someone’s testimony, they don’t need a judicial order to reach to that person. The same applies to getting data that do not enjoy constitutional protection against warrantless invasion (those mentioned above), such as public records, information on business activities, etc.

          In Scenario B, the textbook answer is that the prosecutor cannot hide the identity of the source. On the other hand, she can apply to whatever kind of evidence she believes is necessary before courts. The identity of the source, on the other hand, will have to be disclosed to the judge and will be reflected in the records. There are some protections to witnesses in criminal cases, including giving them new identities, helping with costs to move to a different city (if that’s the case), protecting family members etc. The prosecutor is enabled to proactively request protection for witnesses in that fashion and these are probably the best safeguards a prosecutor could offer to the source in our example. If the records are declared “confidential” (which is pretty much the case in big investigations), the identity of the source is kept out of the reach of the public. Defendants on the other hand will have access to it once they get access to the records.

          I agree that by offering low protections, especially for whistleblowers as in your example, the law does not incentive people to report misconduct. In 2018, Brazil passed a new statute on whistleblowing but it did not advance substantially in whistleblowers’ protection.

          I hope it’s clearer now.

          I agree that the law has problems here in granting an efficient investigation but I’d say that the problem in the conversations goes beyond simply making up an anonymous tip and lays on doing that to hide the fact that the source is in fact the judge, who is prevented to act in that fashion.


    I’m still struggling to understand the meaning of the term “advised” in section 254, IV of the Criminal Procedure Code (the Portuguese word is “aconselhado”). Would this be understood, either generally or in context, to mean giving _any_ sort of advice, or specifically _legal_ advice, or something else? Are there Brazilian sources (treatises or prior cases or something similar) that take a definitive position on this? Or is the meaning of “advise” in this context still somewhat unsettled? (This obviously matters a great deal, because many of the interactions between Judge Moro and the prosecutors could be construed as giving advice on matters that were not strictly legal, such as whether to issue a press release.)

    • Advice about the case. In the new messages, for example, he warns a prosecutor that they forgot to add a document as proof, before the ruling. The prosecution adds in the next day.
      In another, he says (after had heard rumours) that he is against a possible plea bargain from E. Cunha.

      • That’s ridiculous! Those messages occurred in July of 2017. The impeachment was in August of 2016 and Eduardo Cunha is in jail since October of 2016.

        Eduardo Cunha’s supposed ‘plea bargain’ was just a rumour and it would never be in Sergio Moro’s jurisdiction.

    • It means a “specifical legal advice”. According to Renato Brasileiro de Lima (author of Código Processo Penal Comentado. Salvador: JusPodivm, 2017, p. 713), “if the judge legally advises one of the litigants, his position of neutrality will be compromised, because he may have anticipated his possible decision. There is no compromise of this neutrality if the judge merely tells the party that he should hire a lawyer, or if he provides procedural clarifications, without revealing his position on the subject of the claim”.

      • If you’re right about this, then it would seem to have potentially important implications for the way at least some of the media outlets are reporting on these texts. But I guess I’m still a bit confused about what counts as “legal advice.” The original Intercept reporting said that Moro did things like suggested prosecutors reverse the order of certain searches, and cautioned them against filing certain motions unless the supporting evidence was very strong. And I gather that the magazine Veja reported that Judge Moro did things like inform the prosecutors that they’d neglected to file certain evidence (and told them to file it right away, before the deadline passed), Would these sorts of “advice” count as _legal_ advice within the meaning of Section 254? Or would these be the sorts of “procedural clarifications” that Professor Brasileiro de Lima says are permissible?


    This question is somewhat related to the previous question, but it’s a bit more general. In my prior post, I noted that many of the allegedly problematic communications between Judge Moro and Mr. Dallagnol occurred in the context of managing the investigation. In that context, it seems natural (though perhaps not strictly speaking unavoidable) for the judge to give some guidance to the prosecutors, especially given the judge’s more active role in managing a criminal investigation than one finds in a pure accusatory system (though I know that latter issue is a point of disagreement between the two of you). At the same time, I recognize and I’m sympathetic to the point that these regular interactions may make the judge feel too connected to or sympathetic with the prosecution’s side. My question here isn’t really about this specific case, but rather a more general question: Is this a structural problem with the Brazilian criminal justice system, one that arises from what seems to an outsider like an awkward combination of the inquisitorial model (where the judge is playing a much more active role in, among other things, overseeing the investigation) and the accusatory model (which expects the judge to be more passive and uninvolved, so as to better preserve neutrality)? And do the sorts of problems that allegedly arose in this case in fact arise all the time, at least in complex operations where the judge is overseeing a very complicated investigation of multiple defendants, or is the sort of communication disclosed in this case (aside from the fact that the conversations involved text messages) genuinely unusual?

    • There is certainly a structural problem. Brazilian Criminal Code and Criminal Procedure Code were both enacted during a dictatorial period on the 1940s and consequentially reflect a more inquisitorial model. The Constitution, on the other hand, was enacted in late 80s, after the end another dictatorial regime and carries some of its traumas, in the sense that it envisions a criminal justice system where arbitrary decisions couldn’t take place, matching with a more liberal/progressive criminal procedure legal scholarship. Constitutional provisions weren’t followed by a large reform in legal rules governing criminal justice, resulting in multiple controversies in how to interpret the old rules or even if these rules stand before the Constitution. Moreover, the idea that Brazil in lives an accusatory criminal system is highly controversial among courts. Many judges believe and advocate for a more proactive role in fighting crime while others do believe this is simply not their job. Personally, I’ve seen judges that refuse to directly question witnesses even though the Criminal Procedure Code says they can.

      The fact the judge overseeing the investigation could be the same ruling the case is claimed by many to be problematic because the judge who granted subpoenas, dawn raids etc is more likely to have prejudice towards the defendant. This is however not strictly required by Brazilian laws. State Courts in Sao Paulo, for instance, have specific courts to rule on pretrial investigations: the Department of Police Inquiries. These courts have been affirmed by Sao Paulo’s Court of Appeals to be the ones with authority to validate plea agreements executed before charges are pressed (as in most of Car Wash cases). I believe this approach is much more promising solution to the structural problems of Brazilian system before a greater reform takes place.

      Finally, communications ex parte are not really unusual. Conversations about the merits of case or instructions on how to succeed in a case are wrong, even if they’re somehow frequent. This case has the particularity that we actually know word by word some of the conversations, which is somehow unprecedented. Cases discussing suspicion are usually based on much weaker evidence of “suspicion”.

      • Thanks for this detailed and historically informed clarification.

        I think you’ve put your finger on some of the reasons why this case is so difficult to evaluate (at least for people who are not already certain, ahead of time, that they support or oppose Judge Moro or former President Lula):

        First, as you cogently explain, there’s tension or inconsistency in how Brazilian law addresses the role of the judge in a criminal case. It seems like some Brazilian courts, as in your Sao Paolo example, have adapted in appropriate ways that perhaps should be emulated elsewhere, but I take it that this is not strictly required by the law (or at least that hasn’t been the understanding), and as a result in many cases we have the phenomenon that seems to have occurred here: A judge, acting in the more inquisitorial mode, is deeply enmeshed in supervising and directing a prosecution, but then that same judge is supposed to shift to a more detached posture, consistent with the adversarial system, with equidistance from both parties. Even if that’s conceptually possible, that system seems to invite trouble.

        Second, it sounds like there’s a major gap between the rules on the books and what happens every day in Brazilian courtrooms–a gap between (as Roscoe Pound put it) the “law in books” and the “law in action.” That doesn’t excuse what Judge Moro did, if he indeed engaged in illegal or unethical conduct. But it also makes it a bit hard to condemn him too stridently. Perhaps this case will be a “moment of truth” for Brazil, not just on Lava Jato specifically, but on the need to address this gap and reform not just the criminal procedure law, but everyday criminal procedure practice.

        • On point. Court structures’ are not legally required. In fact, in criminal law we have the “natural judge” doctrine which roughly speaking means that the judge having the first contact with a case should be the one presiding the trial. This is subject of an extensive debate on when and which under circumstances the doctrine applies. I mention that because this doctrine is often cited as the reason why the judge overseeing the pretrial investigation (granting warrants etc) should be the one ruling the case in the end. I also agree that this is a system that invites trouble in many ways.

          I also see that gap. It is not rare to hear some frustration on practitioners on how our criminal system actually works as opposed to what we learn from the books. I’d risk to say that even for an audience used to daily criminal courtroom activities, some parts of the conversations are a bit shocking, especially the parts involving tipping prosecutors on witnesses, how to robust a case etc.

          I hope this open the door for a “moment of truth” but I am skeptical about that due to the polarization on how this topic is being perceived. As we have a huge battle over the “narrative” here, it seems that we have limited space for a more nuanced perspective as “this is judicial misconduct but unfortunately it’s not as exceptional as we’d like to think”. Another aspects that makes it hard for people to really think about this as an indication of the need to rethink criminal justice system is that Car Wash is so much bigger and so much more complex than all other cases pending before criminal courts that people are less likely to relate with this context, making it hard to make this episode the basis for a much broader discussion.


    On the question of the admissibility of unlawfully obtained evidence, I wanted to ask you about another possibility that seems a bit different from the views that either of you expressed. (I came across this view in a piece by another Brazilian lawyer, though I’m having trouble re-locating the original source.) That view runs like this: A court may consider unlawfully obtained evidence if but only if that evidence shows that a criminal defendant is _actually innocent_; unlawfully obtained evidence cannot be introduced, even by the defendant, to show procedural violations or other defects with the trial. (US courts use something like an “actual innocence” standard in certain somewhat different contexts, mainly relating to re-opening convictions that have been finalized.) That view seems different from Professor Timm’s view, because if I understand Professor Timm correctly, Brazilian law doesn’t allow the introduction of unlawfully obtained evidence at all (at least if the evidence was obtained by something like computer hacking). That view also seems different from Professor Ademar’s view, as Professor Ademar seems to say that _criminal defendants_ can generally introduce evidence in their defense (or to contest the legality of their convictions) even if that evidence was illegally obtained. My question is whether that alternative understanding on the restriction on using unlawfully obtained evidence is supported by Brazilian law, or whether whatever commentator I was reading was incorrect in suggesting that this is the proper understanding of the rule here.

    • “Professor Borges argues (as do a few other commentators) that even unlawfully obtained evidence can be used by a criminal defendant to challenge the impartiality of the judge. But this is incorrect, and the cases Professor Borges cites are inapposite.”

      To me, that is a central ethical problem. I understand that there should be quite a lot of limits on the admissibility of illegally obtained information. It remains that not being able to provide illegally obtained information in any way doesn’t seem ethically justified when you deal with high-level officials, or other people of serious social significance. It reinforces their status as unchallengeable authorities. We’re indeed not talking about protecting the widow and the orphan, here, but we’re instead not willing to shed light on factual matters of social significance. Ethically, that’s quite different in my view.

      Moreover, following this line of thought, there also could ethically be a difference between 1. not tasking into legal account illegally obtained information, and 2. refusing to acknowledge or assess whether illegally obtained information are factual or not. 1 does not imply 2. And 2 generates a “chilling effect” that is detrimental to discussing matters of truth in public, as facts can be dismissed as mere opinion who are at odds with judicial truth, for instance. A concept which in some country can trigger legal action and stifle free speech.

      “For to distrust the judgement and the honesty of one who hath but a common repute in learning and never yet offended, as not to count him fit to print his mind without a tutor or examiner, lest he should drop a schism or something of corruption, is the greatest displeasure and indignity to a free and knowing spirit that can be put upon him.” — John Milton, Areopagitica.

    • I have a question both to you, Professor, as well as to the great commenters in this space: if, as noted by others, Judge Moro was simply using a atypical method to officialy communicate with the prosecution, then wouldn’t those messages me considered public records and, as such, not be protected by personal privacy?

      How would that situation be viewed in the US legal system?

      • Sorry for the delayed reply, I missed this comment earlier.

        I don’t know the answer to this question under Brazilian law, so I won’t hazard a guess.

        Under US law, assuming that an ex parte communication is legal, its contents would not necessarily need to be disclosed, at they would not typically be considered “public records” as a general matter. It all depends on the nature of the communication and what it was used for, as a legal matter. There’s not a general rule on this.

        And rules against illegally obtaining the information would, I think, apply. Suppose a prosecutor applies for a warrant over the phone. If a private party intercepts and records that phone call without authorization, that’s clearly illegal.

    • In regarding to the “unlawfully obtainted evidence”, as I understand it, they could be used in favor of the defendant. I believe there are problems regarding if he obtained those if good or bad faith (for instance, if Lula’s defense attorneys had, themselves, hacked those communications as opposed to coming to know them through unrelated public disclosure by journalists).

      In my contention, they should be admissible in any case, as they concern one of the “conditions” of the process, namely, the constitutional guarantee to “due process” that contains, in itself, the right to an impartial judge.

      We could delve into the debate regarding there even *existing* a process that is conducted by a partial judge, although I would contend that it is a case of it being “absolutly null”, as a sentence of merit by a “corrupted” judge can be rescinded within two years of the discovery of such act, as opposed to “at any time” in cases of a “non existing” process (e.g. if the defendant is not regularly summoned to take part of the process).

    • In general, a defendant may use unlawfully obtained evidence in his defense. Many scholars support this position (e.g. Vicente Greco Filho, Eugenio Pacelli, Fernando Capez).

      • Interesting. There seems to be quite a debate over this point. A couple of additional follow-up questions:

        * Are the rules on the admissibility of such evidence different for _post-conviction relief_ than at trial?

        * Are the rules on the admissibility of such evidence different if the evidence is being used to show “actual innocence”, rather than to establish a procedural violation or some other point?

        I ask because in US law these distinctions are sometimes important.

    • The typical discussion of unlawfully obtained evidence favoring the defense is indeed dedicated to cases of actual innocence. Even one of the Supreme Court Justices when commenting the case to the press said something like “Imagine someone convicted of murder and then we get evidence that that person is innocent, the evidence, even if unlawfully obtained, would be admissible”. This the kind of commentary you would get in a criminal procedure class.

      Having said that, I believe the conversations should be admitted in a possible claim to vacate Moro’s decision because they show a possible violation of the defendant’s right to a fair trial. Neither the Constitution, nor the Criminal Procedure Code condition the consequences of the violation of this right to any sort of concrete consequences for the defendant, nor to any actual defense standard. A restrictive interpretation in that manner, in my opinion, is an undue limitation to defendants’ rights. This is not to say that the court hearing this claim cannot inquire the credibility/authenticity of these messages. As for the moment we only have messages reproduced by the press, a court could conclude that access to the “raw materials” is necessary in order to decide.


    Even if Professor Timm is right that communication between the judge and the prosecutors is allowed, that it’s fine to do this by text message, and that there wasn’t anything inherently problematic about the content of these text messages, is it still a problem that Judge Moro did not make it clear to defense counsel that they could also communicate with him via text message? Would failure to do so violate the “symmetry” principle that Professor Ademar emphasizes?


    When the first appellate court (here the Regional Federal Tribunal) reviewed Judge Moro’s decision, did the court review only the legal questions or also the factual/evidentiary questions? And on those questions, did the court apply a de novo standard of review, or did the appellate court defer to Judge Moro’s findings (for example, reviewing them only for “clear error” or similar)? I ask because, the broader and less deferential the review on appeal, the stronger is Professor Timm’s argument that the affirmance on appeal undercuts Lula’s claim that Judge Moro’s alleged bias resulted in actual prejudice. But if review on appeal was highly deferential and/or limited to legal conclusions, the fact that the verdict was affirmed doesn’t really do much to show that the original verdict was unaffected by (alleged) partiality.

    • The crux of the matter, in my opinion, rests in the fact that the appellate court relies strictly on the findings produced under Judge Moro’s oversight.

      In other words, all the evidences the defense tried to have admited but were denied by Judge Moro’s (whose impartiality is under question) would not be admitted by the appelate court.

      Futhermore, in the newly released messages (by Veja magazine, a traditional right-wing publication), Judge Moro alerts Dallagnol that they did not file a certain evidence, and that he would wait until the next morning to rule on the case, so they could bring it to the case file.

      If he did not warn them about it, then the appelate court would not know of it when reviewing the case. That changes the content of the findings that they could rule on, in favor of the prosecution and against the defense.

    • Apellate courts in Brazil (Federal Regional Courts, in Federal Courts; and Courts of Justice, for State Courts) decide all the issue again. There is a general principle in Brazil called “devolutividade”: the matter is “fully given back” to the Appelate Court so that it may decide it all again on fact or legal basis.

    • The appeal courts review all facts. The point to me is that suspicion rules a protection to the right to a far trial in the following way. Knowing that judge and prosecutors colluded to produce one piece of evidence gives room to questioning whether they did the same in other episodes during the case. See that suspicion rules pointed by Professor Ademar are not limited to the extension of the advice given but encompass the whole case precisely because the defendant has the right to have his case being ruled by a knowingly impartial judge. If “fair play” guarantees were somehow violated during the case how can a defendant be sure that the same didn’t happen in other episodes?

      Evidence gathering involves a lot of decision from the judge’s side: dawn raids, admitting witnesses and documents, etc. The case brought to appeal are the reflection of the trial presided by the judge (they basically go through the case records and read the appeal and the opposition). If everything on the records is the result of the work of a biased judge, the court of appeals is seeing a biased product and that’s why I believe their affirmation of the decision cannot defeat a suspicion claim.

      • Adding: the Court of Appeals reviews all facts and evidence. The Superior Court of Justice and and Supreme Court limit their analysis to statutory and constitutional analysis.

      • That makes a lot of sense to me. It’s a bit difficult, as someone trained in the US, to know how to think about a system like Brazil’s, since in the US typically the appellate court reviews all the legal issues de novo, but reviews factual determinations with extreme deference to the trial court (reversing only for “clear error”). If it’s really true that on the first appeal the appellate court reviews _everything_ de novo (facts and law), why even have the first trial at all? I presume it’s because the conduct of the first trial shapes that record in ways that must (and should) influence how the first appellate court assesses the factual disputes. So my instinct is that the fact that the first instance appellate court reviews questions of both fact and law shouldn’t mean that serious errors, including a lack of judicial impartiality, at the first-instance trial court simply don’t matter so long as the appellate court affirms. But I asked the question because at least some Brazilian commentators have suggested that the affirmance of Lula’s conviction means that any alleged improprieties by Judge Moro at the first trial can’t call the conviction into question. That just doesn’t seem right to me, but I wanted to ask for others to explicitly weigh in on this point.

        • Let me try to put this way. When reviewing the facts, higher courts are limited to what is reflected on the records. In practice, it makes them to be somehow deferential to the facts as put together by the trial court. Notwithstanding, if the Court of Appeals finds that Witness A and B’s testimony is not enough to reach the conclusion that someone’s committed the crime he/she’s being accused of, it can reverse the decision.

          It is worth to mention that we only have jury trials for murder cases in Brazil. Therefore, questions like “Is there evidence beyond reasonable doubt that Mayor Z received bribes from the Drugstore W?” are answered by the judge in her decision in light of the evidence produced during trial. The Court of Appeals, can reach a different conclusion when faced to the same pieces of evidence because it interprets their strength or weakness in a different way.

          However, the evidence analyzed by the Court of Appeals is the evidence admitted to trial by the lower court and that’s why I don’t buy the interpretation that the affirmation of the conviction remedy any improprieties committed by the trial judge.

  10. There should be a disclaimer that Professor Timm has been appointed as National Consumer Protection Secretary at the Brazilian Ministry of Justice by former Judge Moro, who is the current Minister of Justice.

      • It’s difficult to do so when one of the parts is clearly politically biased in favour of Sergio Moro. More leaks today, showing Moro wasn’t particularly hurried to hear Eduardo Cunha’s ‘plea bargain’. Wonder why? Because they needed him to being the impeachment process against Dilma Rousseff. Do you understand now why I told you before that you’re on the wrong side of History here? There’s no real debate here, a judge cannot instruct prosecutors in a due process, and this has nothing to do with the specificities of Brazilian law.

        • That’s ridiculous! Those messages occurred in July of 2017. The impeachment was in August of 2016 and Eduardo Cunha is in jail since October of 2016.

          Eduardo Cunha’s supposed ‘plea bargain’ was just a rumour and it would never be in Sergio Moro’s jurisdiction.

        • Respectfully, the question whether the judge and/or prosecutors in this case acted illegally has _everything_ to do with the “specificities of Brazilian law.”

          And the specificities of the factual allegations matter quite a bit too, as the previous commentator pointed out. In his piece above Professor Borges ably argues for the position you hold. Professor Timm offers plausible counter-arguments that should be considered carefully and respectfully, even by people who disagree–as professor Borges does.

          I may or may not be on the wrong side of History-with-a-capital-H. On the particular issue being debated above, I actually don’t have a “side”–I’m still trying to make up my mind, which is why I invited to distinguished experts with opposing views to lay out their best arguments. I think we would all do better to examine the facts and arguments carefully. The stakes are too high to do otherwise.

  11. Professor Stephenson,

    I have been following your publications and the ensuing comments and really appreciate the legal breakdowns of this matter. Most journalistic publications that report on this issue cannot go into such detail and with limited Portuguese, I can’t read the analysis of the majority of the Brazilian sources. That being said, my girlfriend is Brazilian and is closely following this as it unfolds.

    Concerning the authenticity of the messages, she told me today that Moro and Dallagnol have both refused to provide their phones to authorities / sign some document that would enable authorities to recover the Telegram messages that Moro said he deleted, which would be an easy way to verify the authenticity of the allegations, no? Are there other pitfalls that would provide reason for them to not actively prove that the messages are either true or false?

    • What your girlfriend tells you is at least partially factually incorrect.

      1. Sergio moro has provided his cellphone to the federal police and they are doing their analysis. I am not sure about Dallagnol however.

      2. Regarding signing of papers so that telegram can release the information. This is politically motivated, asked by the politicians that are neck deep in trouble with the Lava Jato. Moreover, what kind of precedent would that set where to prove ones innocence on the back of unconfirmed, unverified, illegally obtained messages that can be easily edited (after the admission by the intercept that they had made an “edit error”, which included messages dated in the future and assigning the wrong names to the supposed text exchanges) would require one to give up his/her right to privacy?

      • Concerning the political bias behind those asking for Moro and Dallagnol to turn over their cellphones, this is definitely something to be considered and a good point.

        However, I’m not sure about the right to privacy as it pertains to the messages specifically sent between Moro and Dallagnol as well as the group text used by Moro and his aides. Considering that the argument that legitimizes these messages posits that these sorts of communications are just a shortcut for the standard procedures of communications between the prosecution and the judge during the phases leading up to the trial, which, I imagine, would be documented and recorded, why would these messages be considered any different? They aren’t personal communications between two private individuals but rather a more efficient and timely government channel by which to discuss public matters. At the least, shouldn’t they be recorded in the same fashion as other communications that take place between public officials concerning public events?

        • 1. It is known that personal conversations took place in the chat/chat rooms.

          2. Following your line of thought implies that if one were to fabricate a conversation about two people that work together, that would be a sufficient condition to release and display all work conversations between the two? I think any reasonable person would have concerns with the argument you present.

          • Anything mentioned in the chats that do not pertain to the matter at hand should not be revealed publicly, of course. The relevant authorities would redact anything that does not merit public disclosure. I don’t think I am claiming that personal conversations between two work colleagues should be disclosed. If that’s how it came off then apologies for the confusion. But since you mention how improper that would be, it should be noted that Moro did exactly that with the conversations that took place between Lula and Dilma.

            Not to mention that the recordings released by Moro were done illegally, which brings us to your second point. I agree that it cannot be known whether the recordings presented by The Intercept are factual or not. As far as I know, and barring the self-admitted editing errors by The Intercept, it has not been proven one way or another as to the veracity of the messages as a whole so I find it interesting that you are so sure that they are completely fabricated.

            All this being said, I don’t think you really answered my question – or sorry – my argument. If these specific chats were used to discuss Lula’s case and if the reasoning is that these sorts of chats are deemed suitable substitutes for more antiquated and inefficient forms of pre-trial communication, shouldn’t they be recorded? I legitimately don’t know if that is what is mandated in these sorts of pre-trial meetings between the judge and prosecution/defense but if the rule is that they should be transcribed for the court’s records, it seems as if these Telegram chats between Moro and the prosecution should also be on record for the court to review. I don’t think that is so far-fetched but if it is, I’d love to hear why and if you know for a fact that these sorts of pre-trial discussions are not on record, then I’d love confirmation on this point as it would provide a good counterpoint to what I have laid out.

            Personally, I don’t see a knight in shining armor on either side of this argument. It is near impossible to believe that Lula is untouched by the massive amount of corruption that happened under his watch and if I’m not mistaken, he has like five or six other cases pending against him. But if this one case in particular is compromised, then it is compromised. On side other side of the equation, Moro has done some very questionable things, such as the release of the Lula-Dilma recordings. Not to mention he now works for Bolsonaro, who most reasonable people agree is quite a debacle, to put it lightly, and most likely very corrupt himself. This makes me hesitant to take Moro at his word when it comes to potential impropriety before or during Lula’s trial. Do I identify more with the policies that Lula enacted? Yes. Does that mean that I am blind to his probable corruption? No. To me, that is not what is being debated here. Lula has many other cases against him that can put him right back in jail if he is deemed guilty.

  12. In capitalist civilization, or in so called ‘Democratic Civilization’ judiciary is the major corrupt house as of the “Sealdah Hwaker Market” (Kolkata, West Bengal, India). If you want to do something against this profession, as well as traditional judiciary corruption; then the court will hang you with or without evidence to eliminate you from this corrupt civilization.
    The present Brazilian judiciary is the prime or the confirmatory test off this acknowledged systemic corruption that imprisoned former President Lula, a strong Leftist reformer. The same events are abundant in India, Argentina like developing countries with US and the West as a whole, as well.

  13. It is not that hard to realize the messages’ ideological bias, including some serious indications that the people involved were consciously interfering in Brazilian presidential election. Not only against Lula, when he was leading the presidential race, but also against his substitute, Fernando Haddad.

    And it is very easy to acknowledge that Moro “advised” his friends (helping them to achieve their goals), which is clearly prohibited under Brazilian criminal law. No, this is not an “ethical” issue. It is a legal problem, and a fairly simple one, to be honest. In fact, there is not so much controversy about this point among Law professionals in Brazil, including those who disaprove Lula and his Labor Party.

    Moro’s behavior could be considered “normal” in Brazilian Justice system, but Brazilian Justice system is not “normal” at all. Acceptance does not make it legal.

    Again, the messages are not just “conversations” between judge and persecutors: there’s a blatant hierarchical relationship there. Moro is dictating strategic orders to better justify Lula’s conviction (less than two years before the presidential race). People being tolerant of those illegalities is something that speaks much about how legal issues were handled in the country during Lava Jato operation, especially regarding Lula’s convictions.

    The “illegal evidence” problem is almost irrelevant, because it really doesn’t affect the content of the conversations. A judge and some prosecutors combining efforts to prevent their foes from winning the presidential race? To put a rival candidate in prision? Wouldn’t it be more relevant to find out if the messages are real, in the fisrt place? Lava Jato is that necessary?

    The authenticity of the messages would possbily be confirmed only by the total disclosure of the cellphones’ data. The other way would imply a violation of Intercept’s constitutional rights. We must talk about “authenticty”, but we must keep in mind the only way by which it could possibly (legally) be tested, without turning the country into a dictatorship.

    • Respectfully, I think the illegality is not quite as clear as you suggest, if (as several Brazilian legal experts have suggested), the term “advised” in the Criminal Procedure Code is understood as limited to specific _legal_ advice. See my “follow-up question #3 above.

  14. For an update in English on how this is playing out in Brazil, the Guardian just published a very relevant report:

    “…efforts to disqualify the revelations were undermined on Friday when Brazil’s most influential conservative magazine, Veja, published a front-page report featuring damaging new disclosures about Bolsonaro’s most famous minister.

    Veja – long a cheerleader for Moro’s anti-corruption crusade – said its journalists had spent a fortnight poring over nearly 650,000 leaked messages between officials involved in the investigation, and concluded the former judge was guilty of serious “irregularities”.

    They included claims that – despite being a supposedly impartial judge in the ‘Car Wash’ inquiry – Moro had “illegally” steered prosecutors as they worked to convict Brazilian politicians…”

    with regards – Jonathan Fox

    • Thanks! Unfortunately the Guardian story doesn’t provide much detail, and the original Veja story is in Portuguese, but I’ll see if I can get someone to help me with the translation. One of the problems of covering this affair is the frequency with which new stories are coming out! That’s one of the reasons I’m trying not to get too locked in to any particular view.

    • A Brazilian friend was kind enough to supply a translation of the Veja story (at least the first one–maybe there are others out by now?). I’ve given it a quick read, but now I’m trying to go back through it more carefully to figure out what additional problems they might have unearthed. The cumulative weight of all the exchanges, coupled with the indisputably chummy and collaborative tone of the conversations between the judge and the prosecutor, certainly paint a troubling picture. I’m tempted to say that at least some degree of impropriety is now evident. I hesitate to do so mainly because I jumped the gun in my first commentary on the intercept stories, declaring that the exchanges were clearly improper, before I’d studied them carefully, tried to figure out the context, and asked Brazilian experts about the relevant law. But for what it’s worth, here’s my preliminary reaction to the main text exchanges that the Veja story highlights:

      * One exchange involves Judge Moro telling the prosecutor that they neglected to include a piece of evidence in the materials they filed–he noted that it seemed to be missing (apparently the prosecutors’ filings referenced it but it wasn’t in the submitted files). Moro alerted the prosecutors and told them to file it before the deadline. With the all-important caveat that I don’t know the relevant Brazilian law, this seemed harmless, and not like “legal advice.” Veja says (I’m paraphrasing), “Can you imagine if the Judge had given such advice to the defense??” My answer is, well, yeah–I could see that happening, and it wouldn’t be problematic. This seemed like nothing.

      * Then there’s an exchange where Judge Moro tells Mr. Dallagnol responds that he “needs” a petition from the prosecutor in the case of the temporary arrest of a defendant, apparently to ensure that he’s not released. Here Mr. Dallagnol says it will be filed by the deadline, and provides some “good decisions” (apparently prior court cases that could be cited?) for this situation. Judge Moro’s communication with Mr. Dallagnol here strikes me as ambiguous. He could be telling the prosecutors they need to take action to stop this person from being released, which seems to me highly inappropriate. But he could just be reminding the prosecutor of the deadline (“tomorrow at noon”) for a petition that the Judge has already been informed is in the works. Reminding a lawyer for either side of a deadline isn’t problematic, I don’t think. But I’m troubled by the fact that Mr. Dallagnol suggests cases that Judge Moro might invoke in ruling on the motion. I hope the Brazilian experts out there can weigh in on whether that’s the sort of advocacy that would be permissible in an ex parte communication with a judge, or whether that would need to be only in public filings. From a US perspective, that seems extremely inappropriate, but as I noted above I want to be cautious about leaping to conclusions about what’s acceptable in Brazil.

      * Then there’s a message where Judge Moro expresses his home that the prosecutors aren’t going to reach a plea agreement with Eduardo Cunha. The Veja story says that it is illegal for judges to comment on negotiations with collaborating defendants, and if that’s so, this seems clearly inappropriate (though it’s not clear it would taint any actual trial result). But Veja doesn’t cite any provision of the law or ethical code on this, so I’d be curious whether there’s widespread consensus that this is inappropriate.

      * There’s a somewhat confusing (at least for me) exchange about filing some document in the case of a document seized in a raid. It appears that Judge Moro informed a prosecutor (not by text) to file this one document separately from other documents (though the reason is not clear in the text exchanges, and Judge Moro also said “not to hurry” to upload the file in the e-processing system (again, for reasons that are unclear). I can’t figure this one out. Veja says (without citation) that this is a very serious problem because a judge can’t ask a party to “hurry up” (or not to do so). Is that right? It doesn’t seem to flow from sec. 254 of the criminal procedure code. It’s really hard for me to understand the context here, or to see why advice on whether hurrying up with the e-filing of a document (that apparently was going to be filed anyway) is a serious breach of law or ethics, as opposed to a fairly neutral issue of case management.

      * In another exchange, Judge Moro asks Mr. Dallagnol when the prosecution is going to file a petition in the Odebrecht case. Mr. Dallagnol replies that he needs one more day to ensure it’s high-quality, but offers to send along a draft first (apparently so Judge Moro can start working on his ruling). This one is definitely troubling to me. I don’t like the idea of a prosecutor submitting an advance copy of an official filing so that the judge can start working on a ruling–that certainly suggests that the final document is more for public consumption (with all the details ironed out and the legal arguments tightened up), and that this extra work isn’t as necessary for the judge. I certainly don’t think it’s standard practice in the US, or most other jurisdictions in the developed world, for prosecutors to send rough drafts of their motions to judges, so that the judges can start drafting their orders even before the official filings are in. Then again, the actual communication from Judge Moro could conceivably be characterized as yet another deadline/timing related issue, with no indication (at least not explicitly) that he’s prejudged the ruling on the merits. After all, he seems to want to get the filing precisely because he can’t start working on his order before then. But this one definitely smells funny, and I’d love to hear from anyone out there who thinks this sort of exchange is defensible. I suppose one other possible defense here: Judge Moro seems to decline Mr. Dallagnol’s suggestion–he replies that he can send the motion the next day. That doesn’t eliminate the fact that this exchange suggests an overly friendly, collaborative relationship, but it does seem to indicate that the possibly unethical proposal Mr. Dallagnol floated was never acted on, which may cut against the idea that this exchange taints the ultimate verdict in the case.

      * Next, there’s a chat where prosecutors discuss the fact that Judge Moro expressed a preference for an operation to take place early enough that the case can be presented before the holiday break. I won’t say much about this one because it seems to again involve issues of timing rather than substance. As in similar exchanges, the chummy collaborative tone is reasonable grounds for concern, but the actual suggestion by the judge seems mostly benign, going to timing of the operation and filings rather than any substantive matters.

      * The last text exchange again involves timing, and this one (like several of the above texts) also doesn’t have Judge Moro as a direct participant in the text messages. Rather, one prosecutor tells another that judge Moro suggested a particular date for a particular operation. I feel like I’m repeating myself here, but I’m not sure there’s any necessary impropriety in a judge (in what seems to be still a kind of quasi-inquisitorial system, or at least one where judges routinely have a supervisory role in investigations) making suggestions with respect to timing. And perhaps it’s worth noting here that for this and other exchanges, the worry that the judge was communicating with the prosecutors by text message is not obviously present. The prosecutors’ discussion could well be referring to something Judge Moro said to them in chambers.

      So, that’s my preliminary attempt to parse the specific text exchanges highlighted by Veja. But as I’ve said repeatedly (and as my critics keep reminding me) I don’t really know the Brazilian legal system, so I’d love it if anyone else out there who has more expertise could react to this and say a bit more about which of these exchanges are problematic, and why.

      • I’ll try to comment on each point.

        *It’s hard to know precisely how Moro knew about the evidence missing. But the way the conversation flows makes me read it as if the judge knew about the evidence because he is familiar with everything collected by the investigations and understands that prosecutors failed to mention and attach a strong piece of evidence. I don’t think that this was simply the case of not presenting a document already mentioned in the accusation piece.

        *Good decisions here seems to means other court decisions and opinions with language or argumentation that would be helpful when trying to put together future decisions. I think reminding a party of the deadline in an ex parte conversation isn’t problematic in general. Here we seem to be in a scenario where the judge went beyond simply telling what the deadline is, to actually pushing the party to file a document. I think that, coupled with the other dialogues, it increases the “collusive” tone of the conversations.

        *Law 12,850 says (Article 4th, 6th paragraph): “the judge won’t participate of the negotiations of the plea agreement, that shall take place between the police chief, the defendant and his or her attorney, with an assessment from the prosecutor or, if the case, between prosecutors and the defendant and his or her attorney”. It also says “the judge can refuse to validate an agreement that fails to meet the legal requirements or adapt it to the concrete case”. The Supreme Court faced to the question of the extension of judicial review over plea agreements. The ruling is a bit complicated to be honest, but I think it is fair to say that the Supreme Court was significantly deferential to the role of prosecutors in negotiating and deciding the terms of the agreement, limiting the review to the fulfillment of legal requirements or possibly manifestly absurd terms. Here, the judge seems to comment on the convenience or not to enter into an agreement with a certain defendant (regardless of the terms of the agreement and the value of his collaboration). In my opinion, this sort of behavior is outside what’s permitted by the law controlling plea agreements.

        *The point here is that the document seemed to be crucial to determine the involvement of Congress members in the crimes being investigated. If Congress members were involved, the case would’ve to be heard by the Supreme Court instead of Moro’s Court. The reportage implies that by delaying the filing of that document, Moro made an attempt to unlawfully keep the case in his Court instead of sending it to the due Court (the Supreme Court in this particular case).

        *I think sending drafts is not common practice in Brazil. I agree that it suggests an undue collaborative environment between the judge and prosecution. I honestly can’t see this as not being an anticipation of the decision an and indication of coordination.

        **I believe that suggestions on timing are not inherently problematic. The thing that they’re discussing pressing charges even before the evidence is gathered seems a bit more problematic. They also seem to coordinate timing so the case wouldn’t be time-barred, which is also a bit troublesome. Even if we’re talking about a quasi-inquisitorial system, we shouldn’t read it as if the judge was committed to the success of the investigation. My reading of the judge’s role in pretrial investigation is that s/he’s supposed to ensure the investigation will take place in accordance to all constitutional and legal rules.

Leave a Reply to Malte Huthoff Cancel reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.