Dantas’ Inferno: “Veja Vindicated!”

The image “https://i2.wp.com/i113.photobucket.com/albums/n216/cbrayton/brasil14.jpg” cannot be displayed, because it contains errors.
Snippet of the “Dantas-Holder” dossier as it ran in Veja.

Brazilian political and economic commentators perform their analyses before the fact. Before they know that it actually happened, they have an explanation for it. They present opinion divorced from information. –Ricardo Kaufmann (See O Globo: “Chávez Won the Referendum Because He Manipulated the System!”)

[Veja’s claim, that it could neither confirm nor deny the veracity of the claim] is not a joke, it is an epitaph. A journalistic death certificate. A negative conclusion would have made this a non-story, destined for the roundfile. But this is not even that. It is a mere suspicion raised by a highly suspect source which, despite “six months of investigation,” is just as flimsy as before the supposed investigation began. —Alberto Dines

Absence of evidence in not evidence of absence –Donald Rumsfeld, on Saddamist weapons of mass distraction

Extraordinary claims require extraordinary evidence –Carl Sagan

Reuters aims to report the facts, not rumours. Clients rely on us to differentiate between fact and rumour, and our reputation rests partly on that. –“A handbook of Reuters journalism”

J.J. Rendón, psychologist, publicist, with a degree in ontopsychology, who managed the presidential campaign of Carlos Andres Pérez, backed this theory by reaffirming that these are the two basic and necessary conditions that a rumor catches hold in the mind of the people: First, the subject of the story must have a certain level of importance so that it is repeated by he who hears it, and next, the facts must be dressed in a certain amount of ambiguity.”Revista Latinoamericana de Comunicación Social, April 2001

Sometimes it is reasonable to argue from a lack of evidence for a proposition to the falsity of that proposition, when there is a presumption that the proposition is false. For instance, in American criminal law there is a presumption of innocence, which means that the burden of proof is on the prosecution, and if the prosecution fails to provide evidence of guilt then the jury must conclude that the defendant is innocent. Similarly, the burden of proof is usually on a person making a new or improbable claim, and the presumption may be that such a claim is false. For instance, suppose that I claim that I was taken by flying saucer to another planet, but when challenged I can supply no evidence of this unusual trip. It would not be an Appeal to Ignorance for you to reason that, since there is no evidence that I visited another planet, therefore I probably didn’t do so. –The Fallacy Files, Argumentum ad ignorantiam

From the Ministério do Planejamento (Brazil) clippings page today: Veja magazine this week once again asks us to applaud its courageous, rigorous and exemplary reporting in running a May 17, 2006 story on a “dossier” purporting to show that senior government officials had offshore bank accounts, allegedly stuffed with bribe money.

It claims credit for helping the federal police break the case, and adds one piece of new information about its actions in the case:

  1. Dantas had asked for, and received, anonymity as the source of the dossier, in exchange for providing it.
  2. Veja then wisely and courageously, in a spirit of civic-mindedness, reneged on that promise by
  3. Giving (sworn?) statements to the federal police as to Dantas’ authorship of the dossier

On what theory was Dantas guaranteed anonymity in the first place, I wonder?

Reuters, for example, has very careful procedures for publishing information from single, anonymous sources.

Use named sources wherever possible because they are responsible for the
information they provide, even though we remain liable for accuracy, balance and
legal dangers. Press your sources to go on the record. … Reuters will publish news from a single, anonymous source in exceptional cases, when it is credible information from a trusted source with direct knowledge of the situation. Single-source stories are subject to a special authorisation procedure.


Anonymous sources are the weakest sources. … Stories based on anonymous sources require particularly rigorous cross-checking. We should normally have two or three sources for such information.

Veja cites one independent source for its conclusion that ” it cannot be ruled out conclusively that the dossier might contain accurate information.”

That source is anonymous, was hired — by Veja‘s own account — by Veja to do the examination, and is not quoted on the record. Nor are the results of the examination Veja commissioned published.

We are not allowed to assess the credibility of Veja‘s only corroborating source on the Dantas fairy tale or know the facts on which Veja based its interpretation of his or her findings.

Veja states that it finds the expert’s findings “inconclusive,” but does not produce the expert to confirm and validate its interpretation of his or her findings.

It puts words into that expert’s mouth but does not identify the expert, or produce him or her to verify that Veja is reporting the results of his or her analysis accurately.

As the Reuters manual points out:

We aim to show, not tell. The reader should judge people with the facts and quotes we supply, not our interpretations.

Note to persecuted whistleblowers of the future: Veja makes promises to anonymous sources that it does not keep. Veja will snitch you out in a New York minute. Unless it is paying you to tell it what it wants to hear.

See also

A subhead accompanying Diogo Mainardi’s interview with Dantas in the same issue, however, made the following statement: “This man could bring down the government if he told what he knows.”

It also stated, in its May 24, 2006 issue, that while it could not confirm the authenticity of the dossier, or the accuracy of the factual claims it embodied, it found Dantas’ story plausible. My translation:

Using all legal means, VEJA tried to confirm the veracity of the material handed over by Manzano [and compiled with Frank Holder, with whom they met in Zurich.] Submitted to examination by an expert hired by the magazine, the material presented numerous inconsistencies, but none of them sufficiently strong to completely eliminate the possibility that the papers contained true information. … The magazine made it clear that it could not prove the authenticity of these papers, which could all be a fraud. Even so, it is difficult to believe that the banker would have spent so much time and money to hire and equip international spies only to come away with a bunch of phantom documents.

Which is part of the reason why the latest revised version of Veja‘s apologia pro leak journalism suo is so laughable … and utterly contemptible.

Em maio de 2006, VEJA revelou que o banqueiro Daniel Dantas, dono do Opportunity, tinha em mãos uma lista com contas bancárias supostamente mantidas no exterior por figurões da República. Encomendada por Dantas ao americano Frank Holder, ex-espião da agência de investigações Kroll, ela seria uma evidência do enriquecimento ilícito das autoridades nela mencionadas – inclusive o próprio presidente Lula. A reportagem de VEJA teve acesso à lista, sob a condição de que o nome de Dantas não fosse divulgado. Uma perícia contratada pela revista revelou, no entanto, diversas inconsistências no material, que tornaram impossível comprovar cabalmente a inexistência das contas sem ao mesmo tempo desmentir sua existência.

In May 2006, VEJA revealed that banker Daniel Dantas, owner of Opportunity, had in his possession a list of bank accounts supposedly maintained outside Brazil by senior public officials. Commissioned by Dantas from the American Frank Holder, a former spy for the Kroll investigative agency, it supposedly showed that the authorities listed there were illegally enriching themselves — including President Lula.

Veja‘s Mainardi plugs his book to Globo’s Jô Soares

Both Kroll and Holder acknowledge, in their contract dispute in New York federal court, that Holder had a consulting contract with Kroll at the time, aftering terminating his employment with the firm as of January 31, 2006, I think it was.

At the time Dantas (reportedly) approached Veja, however — by Veja’s own account, on August 15, 2005 — he was still a Kroll employee. As he was when he had his phone conversations with Veja‘s Márcio Aith between August 24 and 28.

He is now employed by Ferrell Law, the firm that is defending him in his contract dispute with Kroll — and whose founding partner defended members of the Lucchese crime family some controversial legitimate businessmen from New Jersey who were cleared of all charges in the late 1980s.

Kroll claims Holder already had a parallel contract with Ferrell to service Dantas and Brasil Telecom at the time Veja ran the “dossier” story, but Holder claims he was only hired by the firm in 2006, to lead its investigative arm.

As to “spy,” Holder claims he is not a spy, but is on the hook for spying. Meanwhile, Kroll claims it did not hire him to spy, and describes itself as an investigative agency, a lot of whose work is rummaging through public archives and puzzling over spreadsheets. Boring beancounter stuff, not the stuff of 007. This sensationalist Homeric epithet is prejudicial and stupid.

VEJA had access to the list, on the condition that Dantas’ name would not be published. An expert analysis commissioned by the magazine revealed, however, various inconsistencies in the material which made it impossible to prove with certaint that the accounts did not exist, without, at the same time, being able to rule out that they did.

Why, in initially accepting this condition, is Veja not to be considered on the wrong side of the generally accepted principle, as the Reuters manual puts it, that

We can never allow our sources to make allegations, contentious statements or vituperative attacks behind a cloak of anonymity. It weakens our credibility and gives the sources an opportunity to benefit at our expense. It is fundamentally unfair to the other party and thus biased.

And why, in reneging on that promise, is Veja not to be condemned for burning its source? (Assuming that Dantas did not explicitly or tactitly agree to the tearing up of the previous bargain.)

There are two powerful reasons behind a reporter’s refusal to expose sources. To be trustworthy, one must keep one’s word. And, more specific to journalism, sources are the lifeblood of newsgathering. In a world of secrecy and spin, the person who gives good information is better than gold. From Watergate to the latest corporate whistleblowers, the benefits flowing to the public from this pact of confidentiality are invaluable. —“The Journalist and the Whistleblower,” NYT, Feb. 6, 2004

Personally, I happen to agree that once your anonymous source burns you — gets you to vouch for bullshit that does not pan out — then the deal is off.

But you still bear the responsibility for vouching for the bullshit in the first place. It is your job to, as Ronald Reagan said, “trust, but verify.” This is what you get paid to do.

Making a deal for anonymity without first (1) corroborating the claims made, and (2) confirming the credibility of the source is (1) stupid and (2) unbelievably stupid.

It is like buying a used car without test-driving it, and signing an “as is” liability waiver on the sale without first having your own mechanic look the thing over.

I would dearly love to read that report, by the way, know exactly who they commissioned it from, and read the expert’s conclusions for myself in the expert’s own words.

Did the person Veja hired to analyze the documents also demand and receive anonymity?

I myself am going to commission an independent expert to examine Veja‘s claim about this report, as to the question of its existence or nonexistence.

Can my expert state conclusively, from examining the issue of Veja that sets forth the claim, whether that inconclusive report actually exists or not?

I have no evidence at all for this claim, mind you, but I find it implausible that Veja would suddenly start telling me the fact-checked truth — Veja disbanded its fact-checking department a few years back — all of a sudden.

I therefore conclude that that the expert report on the dossier documents probably does not exist, or, if it does, that it supports the absolute ambiguity that Veja claims it does.

Let Veja produce a copy of it and show me how wrong my suspicions are.

The burden of proof is on those who want us to pay them money for their information products, with an implicit warranty as to their quality.

Otherwise, Veja is failing to provide any concrete evidence for its claim that there is no conclusive evidence — which seems to contradict the federal police claim that the lack of evidence for the existence of the accounts is conclusive evidence for their nonexistence.

Veja, in other words, is suppressing the evidence it says it possesses for the alleged lack of conclusive evidence.

It wants we should just take its word for it.


Diante de sinais de que Dantas usava a lista como elemento de chantagem em uma disputa empresarial com fundos de pensão de estatais, VEJA revelou sua existência e apontou o banqueiro como seu autor – fato, claro, negado por Dantas. Além disso, a revista enviou, na ocasião, toda a papelada que reuniu sobre as supostas contas ao procurador-geral da República, Antonio Fernando de Souza, que, efetivamente, investigou o caso.

Given signs that Dantas was using the list as part of a blackmail scheme in a dispute with the pension funds of state-owned firms, VEJA revealed its existence and pointed to the banker as its author — which he, of course, denied.

What signs were those? This whole “blackmail theory” is a loaded question, I think, because the concept of blackmail implies a desire to conceal guilt on the part of the blackmailed party.

Dantas is interviewed in the issue in which the dossier is published, an interview in which Dantas claims he was asked to pay a $50 million bribe to the PT to resolve his business problems. A claim which Mainardi and his editor make approving, supportive noises about.

Furthermore, the magazine sent the entire collection of papers on these alleged accounts to the federal attorney, de Souza, who in fact investigated the case.

And what did he conclude? The only credible reports on the investigation we have so far (from Leandro Fortes) concern the investigation conducted by Disney Rosseti of the federal police.

Na semana passada, ao concluir o inquérito sobre o episódio, a Polícia Federal confirmou integralmente a história de VEJA e indiciou Dantas e Holder por calúnia – crime que consiste em atribuir falsamente a alguém fato criminoso.

Last week, upon concluding an investigation of the episode, the federal police confirmed VEJA’s story in its entirety and indicted Dantas and Holder for slander — a crime that consists in falsely attributing a criminal act to someone.

Veja summarized its “story in its entirety” as follows:

Veja used all means at its disposal to prove the veracity of the information. It was not possible to arrive at any conclusion — positive or negative.

Veja boldy predicted that the claims underwritten by the Dantas-Holder dossier would turn out to be either true or false.

The police have now concluded they were false.

Veja was right all along!

I give you a hot tip: Denver will either win the Super Bowl next year, or else it will not.

I then ask you to pay me for this hot gambling tip.

The bet is a sure money-maker, provided you can find somone stupid enough to bet on the proposition that Denver will neither win nor lose the Super Bowl next year. (With the addition of sudden-death overtime, ties have all but been eliminated.)

You tell me to go chase myself. Because you are not an idiot. Hardly anyone is that big an idiot.

O indiciamento de Dantas só foi possível em razão de documentos fornecidos pela revista e de depoimentos dados por seus editores. Ainda que não tenha obtido a quebra internacional de sigilo para verificar a existência das supostas contas, a PF concluiu que elas são falsas. Baseou sua decisão em “conversas informais”.

The indictment of Dantas was only possibly because of documents this magazine furnished, and sworn statements given by its editors.

Under the prevailing Press Law, if I understand his correctly, Brazilian journalists cannot be compelled to reveal their sources.

The theory, familiar from the argument in favor of similar “shield laws” in the United States, is that the knowledge that journalists can be coerced into revealing sources results in a “chilling effect” on the Deep Throats of the world — legitimate whistleblowers who run risks by providing information of legitimate public interest.

Not having yet obtained access to international banking records to verify the existence of the supposed accounts, the PF concluded they were false. They based their decision on “informal conversations.”

Who says?

Elas teriam convencido os policiais de que tudo não passa de invenção do banqueiro e seu cúmplice americano. Com o indiciamento, Dantas agora pode ser denunciado pelo Ministério Público e responder a mais um processo criminal.

Those conversations purportedly convinced police that it was all a fabrication by the banker and his American accomplice. With the indictment, Dantas can now be formally charged by the prosecutor and face yet another criminal trial.

Leandro Fortes of CartaCapital published a play by play of the investigation last week (does not say what his source was, though) in which he describes the deliberative process as follows (my rough translation):

Rosseti then got into touch with the DRCI, the Ministry of Justice’s Department for Asset Recovery and International Cooperation, in order to make use of mutual cooperation agreements with Switzerland (where the Lacerda and Gushiken accounts supposedly existed) and the Cayman Islands (Dirceu’s alleged account.) He waited 60 days and received terse replies. The Swiss government alleged legal obstacles to responding to the request, because technically speaking it can only provide this type of information if there is evidence that the supposed accountholder is involved in crimes of corruption in his or her country or origin. That is to say, if the evidence of corruption is precisely the existence of offshore accounts, the suspicion dies stillborn. The same occurred in the Caymans. In those two cases, the PF was supported by COAF [the Brazilian FinCEN, sort of], the Treasury ministry’s intelligence bureau, and concluded that none of the parties had ever sent a single dime to the alleged accounts that Dantas took to Veja.

The existence of the accounts itself would have been the evidence of wrongdoing required to reveal information about whether the accounts existed or not.

The inference being that the Swiss authorities did not accept the “dossier” as legitimate evidence of a crime, triggering its duty to disclose information about the accounts.

The whole dispute is a textbook case in two logical fallacies: the ‘appeal to ignorance” (argumentum ad ignorantiam) and the “fallacy fallacy” (argumentum ad logicam).

In some cases, reasoning that takes either of the following two forms is fallacious:

There is no evidence against p.
Therefore, p.

There is no evidence for p.
Therefore, not-p.

When Joseph McCarthy promised to present 81 cases of persons with Communist ties infiltrated into the Truman administration, he used this argument in connection with Case 40:

Of Case 40, he said, “I do not have much information on this except the general statement of the agency … that there is nothing in the files to disprove his Communist connections.

On the other hand, it is not always a fallacy to accept “absence of evidence” as “evidence of absence.” The Fallacy Files again:

… when extensive investigation has been undertaken, it is often reasonable to infer that something is false based upon a lack of positive evidence for it. For instance, if a drug has been subjected to lengthy testing for harmful effects and none has been discovered, it is then reasonable to conclude that it is safe.

The legitimacy of the conclusion depends on the completeness and methodological soundness of the investigation done.

The fallacist’s fallacy — if an argument that P is fallacious, then not P — goes something like this:

Tom: “All cats are animals. Ginger is an animal. This means Ginger is a cat.”.
Bill: “Ah you just committed the affirming the consequent logical fallacy. Sorry, you are wrong, which means that Ginger is not a cat”.

Upon closer inspection by a zoologist, however, Ginger turns out to be a meowling, clawed furry creature who licks herself, has triangular ears, hunts birds and mice, and so forth.

Facts are the sword that cut the Gordian Knot of speculation, rumor and Rumsfeldian logic-chopping.

Which is essentially what Veja is engaged in here: It focuses on interpretations to be drawn from ambiguous evidence without producing evidence for the proposition that the evidence is ambiguous.

That is to say, it claims to have performed due diligence, but has not produced any real evidence that it did so — the supposed expert’s examination of the “dossier.”

The entire issue is a red herring.

The real issue is not the dossier, but the facts the dossier purports to refer to: the existence of the accounts, and whether those accounts actually belong to the persons they are attributed to.

Veja is changing the subject while filibustering — a primary, though not in itself conclusive, indicator of the banana-republican guilty plea.

The argumentum ad logicam can be seen as a variant of the ad hominem fallacy, because it relies on the opposing speaker’s seeming lack of credibility, a factor which is usually irrelevant to the actual correctness of a given theory (although it can help bolster the evidence in an inductive argument).

There is also the fallacious attribution of fallaciousness to an argument. You see this often on Wikipedia in the form of Neutral Point of View Disputes in which it is claimed that the article in question fails to take into account the possible validity of, say, the Hollow Earth or Flat Earth theory, or the luminiferous aether, or that some people say they sincerely believe that 2+2=5.

Veja, for example, seems to be implying that the conclusions of the police investigation are an example of a fallacious Argument from Ignorance:

There is no evidence for p.
Therefore, not-p.

That is to say, it implies that the conclusions of the police investigation are no less inconclusive than Veja’s “rigorous, six-month investigation.”

This is a clash of interpretations, not a factual dispute, carrried on in some parallel universe in which E.D. Hirsh never wrote Validity in Interpretation and Spock is an evil, bearded pirate.

The presupposition here, however — and of Brazilian lynch-mob journalism in general, and Senator Joseph McCarthy’s standard of evidence — is that the burden of proof is on the accused to provide evidence that not-P, not on the accuser to provide evidence that P.

Example: Veja provides no evidence that it had a document examiner prepare a report on the dossier.

Therefore, I reason, Veja is lying.

I conclude that it never had any such report prepared.

In the event that Veja produces such a report, my reasoning proves to have led me into error.

There is only one way to settle the question: Either Veja produces the report in question, and allows us to assess whether its interpretation of the purported findings of that report is valid, or it stops claiming to have such evidence. Put up or shut up.


Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

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

Google+ photo

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

Twitter picture

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

Facebook photo

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


Connecting to %s