Power before the mirror

There are moments when a case ceases to be merely a case. Documents begin to connect, personal relationships intersect with the structures of government, institutions begin to appear within the very material they are supposed to examine, and an investigation into specific facts begins to reveal a much larger question: what happens when power has to investigate power itself?

That is where the Banco Master case appears to have arrived.

Daniel Vorcaro’s phone became an archive of this proximity. It contains messages, contracts, records of meetings, contacts, and references to people who occupy very different positions within the Republic. Based on this material, the Federal Police produced a 218-page report. The document brought together material involving Alexandre de Moraes, Paulo Gonet, Andrei Rodrigues, and André Mendonça, along with other figures connected to the banker. The investigation began to reveal a network of relationships.

Alexandre de Moraes appears in messages attributed to the justice and in records of contact with Vorcaro. The Federal Police identified references to meetings between the two. In one conversation, shortly before his arrest, Vorcaro asks whether he should leave the country. Elsewhere in the material, there are references to requests made by the banker for authorities to be contacted on his behalf. The extraction also revealed an important limitation: in some conversations, Vorcaro used disappearing messages sent as screenshots created through the Notes app, and the other party’s responses were not recovered.

The material also includes a contract worth approximately R$131 million between Banco Master and the law firm of Viviane Barci de Moraes, the justice’s wife. It was in this document that the Federal Police found one of the clearest records of Alexandre de Moraes’s involvement.

The discovery came through forensic analysis of the files found on Daniel Vorcaro’s phone. By examining different versions of the draft, investigators analyzed metadata and the digital records preserved within the documents. Those records indicated that the file had been opened and modified on a computer at the Supreme Federal Court used by Alexandre de Moraes. The user registered for the modification appeared as “Ministro Alexandre de Moraes.” In one version, the modification was recorded on January 15, 2024, at 11:04 p.m., after Vorcaro had returned the draft with his own changes. The same record appeared in another version of the contract.

It was through this analysis of the metadata and the file’s own digital records that the Federal Police discovered that Moraes had edited the contract. Alexandre de Moraes later acknowledged that he had edited it. Viviane Barci de Moraes’s law firm also confirmed that he had access to the document and explained that the justice had been consulted to assess potential conflicts involving the engagement.

The contract called for 36 monthly payments of R$3 million net, reaching approximately R$131.2 million gross. Some of the payments were subsequently identified in the financial records examined in the case.

The file preserved more than the text of the contract. It also preserved part of the history of how the contract itself was edited.

And the file continued to reveal other connections.

Paulo Gonet and members of his family appear at least 33 times in the material gathered by the investigation. The attorney general appears in messages involving Ciro Soares and Daniel Vorcaro. In one of them, Ciro sends the banker a photograph of himself with Gonet and tells Vorcaro that Gonet wanted to speak with him. There are also references to Gonet’s son and to a trip to London connected to Vorcaro.

Gonet appears inside the archive.

Then he appears before the archive.

He is the person who asks for the nullification of the investigations conducted by André Mendonça.

That circumstance makes it even more important that the material be known, preserved, and examined by the appropriate authorities.

Ciro Soares occupies a recurring position in this network. A lawyer with a strong professional presence in Bahia, he appears as one of the bridges between Vorcaro and different authorities. Federal Police material records his proximity to Gonet and Mendonça and his role as an intermediary in contacts involving the banker.

In February 2025, Ciro sent Vorcaro a photograph of himself with André Mendonça and wrote that he was “working for you.” He also mentioned having taken the justice to the Vasco Vasconcellos tailor shop to have a suit made. Mendonça confirmed that he had been at the tailor shop, denied being taken there by Ciro, and denied that third parties had paid his expenses.

The suit is one detail within a much larger story. It therefore needs to occupy the proportion it deserves, without disappearing and without carrying a meaning that the larger body of evidence has yet to establish.

In March 2025, Vorcaro met with Mendonça again in São Paulo, outside the justice’s official schedule. Mendonça confirmed the meeting and said that the banker had sought to discuss matters involving court-ordered debts that were of interest to Banco Master. Mendonça stated that he merely listened and later ruled against Vorcaro’s interests.

These are different points of contact, at different moments, involving different people. When they appear within the same archive, they begin to form a map of relationships.

And that map reaches the Federal Police. Andrei Rodrigues, the Federal Police director-general, also appears in the material. Vorcaro referred to him in conversations concerning his interests and sought access to officials who could influence the institutional environment surrounding Banco Master.

At the same time, the relationship between Mendonça and the Federal Police was already under strain.

In August, the 27 regional superintendents of the Federal Police issued a public statement defending the agency’s director-general, Andrei Rodrigues, and the investigative independence of the institution. The statement said that investigative activity should remain guided by facts, evidence, and the oversight mechanisms established by law.

It was against this backdrop that the August 24 meeting took place.

Mendonça summoned members of the Federal Police and ordered an analysis of messages found on Vorcaro’s phone, including the identification of interlocutors and the preparation of a report. The deadline was 72 hours. Among the names identified were Moraes, Gonet, and Andrei Rodrigues.

The Federal Police complied with the order.

It produced the report.

And the report brought information about members of the Supreme Federal Court into the Supreme Federal Court itself.

At this point, the institutional dimension of the case becomes impossible to ignore.

Mendonça ordered the investigative step. The Federal Police conducted the analysis. Moraes appears in the material. Gonet appears in the material. Andrei Rodrigues appears in the material. Mendonça then oversees the proceeding that produced part of this information. And the debate over the validity of the investigation begins to involve people who appear in its results.

A few days later, Daniel Vorcaro was questioned at the Papudinha by a judge serving as an aide to André Mendonça, with the participation of a representative of the Public Prosecutor’s Office and without the presence of the Federal Police. His defense had alleged intimidation and requested an urgent hearing. Vorcaro made allegations involving members of the Federal Police, including Andrei Rodrigues.

From that point on, the case ceases to be a straight line. It becomes an overlapping structure of documents, people, institutions, and competing accounts.

This is precisely the kind of situation in which legal procedure matters. Rules governing jurisdiction and authority exist to prevent power from being exercised arbitrarily. Procedures determine who may investigate, who may order investigative measures, who supervises an investigation, and who must decide when a case reaches officials protected by special jurisdiction.

Those rules must be respected.

But there is another limit that must also be respected: a procedural issue cannot automatically become a mechanism for making the facts disappear.

That is precisely where the current controversy lies. Paulo Gonet argues that Mendonça exceeded his authority by directly ordering the Federal Police to identify specific individuals and deepen the investigation into them, particularly because another Supreme Court justice appeared among the names uncovered. Under this interpretation, the matter should have gone before the appropriate body of the Court.

There is, however, a legal question that prevents such a simple reading.

The Supreme Court itself has already held that there is no general rule requiring every investigation involving an official with special jurisdiction to receive prior authorization from the full Court. In 2024, in its ruling on ADI 7.496, the Court recognized the authority of a reporting justice to exercise judicial oversight over investigative acts involving officials with special jurisdiction and rejected a requirement of prior collegial approval established by a state rule.

That does not automatically resolve Mendonça’s situation. The issue must be assessed in light of the specific nature of his actions, the origin of the order, the scope of the investigative measures, and the particular circumstances of this case.

It does, however, prevent an absolute rule from being imposed when the Supreme Court itself has not established such a rule in its jurisprudence.

And this is the point I consider essential: the procedural question can be examined while the facts continue to be investigated. A possible procedural irregularity may have consequences for a particular decision, investigative measure, or act. That does not necessarily mean that every piece of information uncovered must disappear.

If a measure was improper, correct it. If authority was exceeded, determine the legal consequence. If an investigative step needs to be repeated by another authority, repeat it. If the full Court needs to rule on a particular stage, let it rule. The investigation continues.

Because what exists in this archive is not a single piece of information. There are messages, documents, contracts, metadata, editing records, meetings, relationships, payments, allegations, and competing accounts. This is the body of material that needs to be understood.

The case cannot be reduced to a dispute over who has the formal authority to press a particular button. The question of authority must be resolved. The underlying material must be investigated. Both questions can exist simultaneously.

That possibility is precisely what seems so important to me.

Because what now exists before the country is a structural problem.

The Supreme Federal Court has a singular role in the Republic. It is the guardian of the Constitution and exercises constitutional review over laws and government acts. Its authority is tied to its ability to control the exercise of power.

But what happens when the power that needs to be examined exists inside the Court itself?

That question changes the nature of the problem.

We are dealing with an institution that has to examine material involving one of its own members. This is precisely the kind of situation in which institutional architecture needs to be stronger than personal relationships.

A democracy should not depend on the personality of whoever occupies a seat. It should not depend on the confidence a particular justice inspires. It should not depend on the reputation of a particular prosecutor. It should not depend on how close or distant individual officials are from one another.

It needs mechanisms capable of functioning when the people involved are powerful.

This is where the word corporatism stops being merely a political accusation and becomes an institutional risk that needs to be confronted.

Every institution tends to protect its own stability. That can be healthy when it means preserving institutional independence. It becomes dangerous when preserving the institution begins to be confused with protecting its members.

The Supreme Court needs to be preserved. Its justices need to be accountable for their own actions. Those things are compatible.

Public trust also needs to be preserved. And public trust does not come from the absence of questions. It comes from the ability to face them.

For that reason, the Gonet issue needs to be viewed in its entirety. The fact that his name and the names of family members appear repeatedly in the material does not determine the outcome of his legal position. But the context exists.

The same applies to Mendonça. His conduct needs to be examined.

The same applies to Moraes. The messages, meetings, contract, and digital records need to be examined.

The same applies to Andrei Rodrigues. The references made to him need to be examined.

An investigation exists precisely to turn this collection of questions into possible answers.

There is no reason to anticipate a conviction. There is also no reason to anticipate immunity. There is reason to investigate. And there is an even greater reason to investigate when the names uncovered belong to the most powerful core of the institutions responsible for the investigation itself.

What concerns me is the possibility that the procedural issue could be turned into a pretext for preventing the facts from being investigated to the end and, where crimes and individual responsibility are ultimately established, preventing those responsible from being punished.

If a decision was unlawful, that decision can be challenged. If an act was invalid, it can be invalidated. If an investigative measure was carried out by the wrong authority, it can be repeated.

But the facts are still waiting for an answer.

That distinction is fundamental.

Procedural law does not exist to shield facts from examination. It exists to organize the path through which facts are examined. When procedure becomes a door that prevents an investigation from entering, the purpose of the system itself begins to lose meaning.

Brazil needs to discuss this. It needs to discuss how investigations involving Supreme Court justices are initiated, who supervises them, conflicts of interest, institutional oversight, transparency, and mechanisms that reduce the possibility of an institution being asked to decide alone on facts involving its own members.

This does not mean weakening the Supreme Court. It means creating conditions under which its authority can endure.

A constitutionally powerful institution needs equally powerful mechanisms of self-restraint.

The greater the power, the greater the need for transparency. The greater the authority, the greater the need for oversight. The greater the ability to decide other people’s lives, the greater the ability must be to withstand scrutiny of itself.

The Banco Master case exposes this question in a particularly uncomfortable way.

A banker’s phone revealed relationships. A contract revealed an edit recorded in its own metadata. The metadata led investigators to Moraes’s involvement in the alteration. Messages revealed connections. Meetings added context. Gonet and members of his family appeared. Mendonça appeared. Andrei Rodrigues appeared. The Federal Police moved to the center of the dispute. The Public Prosecutor’s Office began challenging the way the investigation itself had been conducted. And at the center of it all stands the Supreme Court.

It is a kind of archive within the archive.

The more one looks, the more relationships appear. And the more relationships appear, the greater the obligation to investigate.

Not to produce a predetermined conviction.

To find out what happened.

That should be the simplest position of a republic.

The government investigates. The investigation uncovers evidence. The evidence is examined. Responsibility is determined afterward.

When power appears within the material under investigation, the obligation to pursue those answers becomes even greater.

This is where the mirror becomes unavoidable.

For decades, the Supreme Court has looked outward. It has reviewed acts of the executive branch. It has reviewed legislation passed by Congress. It has established limits for public officials. It has protected fundamental rights.

Now there is a situation in which investigative material reaches into the institution itself.

That is the institutional test.

A constitutional court must be able to look at itself. It must be able to recognize that its members can become subjects of investigation. It must be able to distinguish defending the institution from defending its members. It must be able to preserve its rules without allowing those rules to be used to interrupt the search for facts. It must be able to correct a possible procedural failure without erasing what the investigation has revealed.

And if the existing institutional architecture does not provide enough security for that, there must be the courage to change it.

Brazil cannot build a republic in which ordinary citizens are subjected to a degree of scrutiny that those in power cannot withstand when the scrutiny turns inward toward the institutions themselves.

There cannot be a zone in which institutional authority is absolute while the possibility of examining that authority is relative.

There cannot be a point at which the rule of law ends precisely when an investigation reaches those who possess the greatest power.

That is why the discussion surrounding this case needs to move beyond the names involved.

Moraes is one name. Mendonça is one name. Gonet is one name. Andrei Rodrigues is one name. Vorcaro is one name.

The larger problem lies above them.

It lies in the structure that must produce an answer.

An answer that is legally valid.

An answer that is institutionally legitimate.

An answer capable of surviving public scrutiny.

Because the country does not have to choose between justice and rules. It needs both.

It does not have to choose between investigation and due process. It needs both.

It does not have to choose between preserving the Supreme Court and allowing its members to be investigated. It needs both.

What cannot happen is for legal procedure to become a mechanism for protecting power.

If there is an error, correct it. If there is responsibility, investigate it. If there is a suspicion supported by concrete evidence, investigate it. If there is no basis, let the investigation establish that.

The outcome must come from examination.

Not immunity.

That is how an institution demonstrates strength.

Not by preventing questions from being asked.

By being able to answer them.

A mirror does not exist to produce a beautiful image.

It exists to show what is there.

And perhaps that is precisely what the Supreme Court needs to do now.

Look at its own image.

Endure the discomfort.

Allow the documents to be examined.

Allow the metadata to be examined.

Allow the contracts to be examined.

Allow the relationships to be examined.

Allow each official to answer for what they actually did.

Allow the law to apply even when power exists inside the institution itself.

Because a court that can look at every other branch of government but struggles when the gaze turns back toward itself reveals a weakness that no constitutional authority can conceal.

Power needs limits.

And it also needs mirrors.

In the end, perhaps the most important question is not who will be investigated. It is whether Brazil has built institutions capable of investigating anyone when the evidence warrants an investigation, including when that person is sitting inside the very room where the decision will be made.

That is the question before which power must recognize itself.

And answer.