Representative Jerrold Nadler of New York, the senior Democrat on the House Intelligence Committee, has written a six-page response to the FISA-abuse memo published Friday by the committee’s Republican staffers under the direction of Chairman Devin Nunes (R., Calif.).
I won’t get sidetracked by the fact that Nadler’s “Dear Democratic Colleague” letter has been “exclusively obtained” by NBC News — i.e., that it was leaked to the media, whereas the so-called Nunes memo was provided to committee Democrats before publication so they could seek changes. The Nunes memo had to be subjected to a rules-based process because of classified-information issues. The Nadler memo does not seem to contain classified information; it just responds to what the Republicans have produced, which is now public record.
I don’t agree with Jerry Nadler’s politics, but he is an able lawyer. What surprises me about his retort is how weak it is.
He posits four points, the last two of which are strictly political red meat. Of the other two, one provides an inaccurate explanation of the probable-cause standard in the Foreign Intelligence Surveillance Act (FISA); the other is an ill-conceived argument about Christopher Steele’s credibility. The latter provides a welcome opportunity to confront a wayward theory — which I’ll call “vicarious credibility” — that has been vigorously argued by apologists for the FBI and Justice Department’s handling of the Steele dossier.
Let’s take the easy stuff first — Nadler’s last two contentions.
A. The Nunes Memo and Deputy Attorney General Rod Rosenstein
Nadler’s point 3 insists that “the Nunes memo provides no credible basis whatsoever for removing Rod Rosenstein as Deputy Attorney General.” But the Nunes memo does not advocate Rosenstein’s ouster, nor does any credible commentator claim that the Nunes memo provides a basis to oust him. The memo merely notes that Rosenstein approved the last (third) application for reauthorization of the original FISA warrant. This is recited in the course of listing all the deputy attorneys general and FBI directors who approved applications (as the statute requires). Rosenstein’s approval is an unassailable fact. It is not a firing offense, nor is it portrayed as such.
To be sure, on the separate matter of the Mueller investigation, there is no shortage of reasons to be critical of Rosenstein’s passively erratic supervision. (See, e.g., here and here.) But that has nothing to do with the Nunes memo, or with the fact that Rosenstein happens to have signed one of the FISA applications — a fact that, on the woefully inadequate information we have, does not translate into knowingly submitting a misleading application to the court. For all we know, the last application — which would likely have relied on whatever intelligence the first three had yielded and may have had nothing to do with President Trump — could have been bulletproof.
B. The Nunes Memo and the Mueller Investigation
Nadler’s point 4 argues that “House Republicans are now part and parcel to [sic] an organized effort to obstruct the Special Counsel’s investigation.” Given the fantasy conception of obstruction Democrats have internalized since shortly after President Trump’s inauguration, it is little wonder that they now see felonies under every rock. The Nunes memo, however, has no bearing on the work of Special Counsel Mueller, which commenced seven months after the first FISA warrant on Page.
A cautionary note: Those Republicans trying to use the inquiry into FISA abuse as a Trojan horse for discrediting Mueller’s investigation are undermining both their own credibility and that of the FISA inquiry. Similarly, Democrats trying to make Mueller’s probe of Russia’s election-meddling synonymous with “Trump–Russia collusion” are doing themselves and Mueller no favors. But to investigate missteps and potential abuses in investigations relevant to the 2016 election does not impede Mueller in any way, much less corruptly obstruct him.
Now to the more serious contentions.
C. The “Vicarious Credibility” Theory
Nadler’s Point 1 (and to some extent, Nadler’s Point 2) is especially useful because it exemplifies a fatal flaw in the reasoning of defenders of the Justice Department and FBI’s use of the Steele dossier in a warrant application, including such commentators as the usually excellent Orin Kerr at Lawfare. It involves imputation to the informants of the investigator’s trustworthiness or biases — vicarious credibility.
Proceeding from the erroneous premise that former British spy Christopher Steele is the source of the dossier information, these analysts posit the inarguable proposition that the government is not required to provide the court with all of the potential credibility problems of an information source. Therefore, the theory goes, there is nothing untoward in the government’s failure to inform the FISA court that Steele’s information was bought and paid for by the Democratic presidential candidate, and then used in an application in which agencies run by the Democratic president sought court-authorized surveillance of the Republican candidate’s campaign.
Here’s the problem: Steele is not the source of the information. For purposes of the warrant application, he is the purveyor of information from other sources. The actual sources of the information are Steele’s informants — anonymous Russians providing accounts based on hearsay three- and four-times removed from people said to have observed the events alleged.
An example makes the point. Say I’m a prosecutor in a narcotics investigation. My DEA case agent is simply the best — many times rewarded by his agency and the Justice Department for his competence, diligence, and expert knowledge of how international drug cartels work. The agent brings me information that three of his sources have told him X is using a shipping container to import cocaine. I thus want to get a search warrant for the shipping container. When the court asks me what my probable cause is, I don’t get to say, “Gee, Judge, I have this fabulous investigative agent who’s got more performance awards than Tom Brady has touchdown passes. He tells me his informants are certain about the container.” If I try that, the judge — assuming he remains calm and doesn’t throw the warrant application in my face — either sends me back to law school or patiently explains that the issue is not the credibility of my investigative agent; it’s the credibility of my investigative agent’s informants.
Similarly, if the informants were shown in the application to be creditworthy, it would not matter that my case agent had been fired for dishonesty or incompetence.
This is the truly outrageous thing about the Steele dossier saga that I am surprised commentators don’t, or won’t, see. In the case of these FISA applications, the principal problem is not Steele himself but his information. We can never even get to the task of evaluating whether Steele’s anonymous, Russian, multiple-hearsay sources have some bias against Trump or Page. We don’t know who the sources are, and the FBI seems never to have corroborated them.
In the case of these FISA applications, the principal problem is not Steele himself but his information.
For a very long time, I confidently assured people that attacks on the FISA applications were sure to be futile. It did not matter that Steele had credibility issues, I explained; the critical thing was the credibility of Steele’s informants. The FBI and DOJ, I insisted, would never bring a court information based on factual allegations that the FBI had not corroborated, provided by sources that the FBI had not checked out. I intuited that Steele’s name probably did not even appear in the warrant application. That is because once the underlying sources are verified, the credibility of the person who tipped the investigator off to misconduct is beside the point. As a prosecutor, I took information from terrorists, mobsters, drug lords, scam artists — you name it. There is nothing wrong with taking information from a suspect source as long as the investigator then rolls up his sleeves and corroborates the information. That means establishing the credibility of the witnesses who claim to have seen or heard the sinister activities that are said to be the probable cause justifying the warrant.
Alas, it appears that I was wrong. From everything we have heard thus far, the FBI did not corroborate Steele’s informants. Their inflammatory allegations about Trump are acknowledged to be “salacious and unverified.” According to the Nunes memo, FBI corroboration efforts were only in their “infancy” at the time the first warrant was sought, and they never yielded anything but “minimal” verification (which may be a charitable way of putting it).
Now, that would not be a decisive infirmity if the warrant contained other reliable information that established probable cause, even if the Steele allegations were stripped out. But we’ve been told that did not happen. According to the Nunes memo, the FBI’s then–deputy director Andrew McCabe testified that there was not enough information outside the Steele allegations to show probable cause. (We’ll come back to that testimony, and the dispute over it, shortly.)
To justify a finding of probable cause, the government must satisfy the court as to the credibility of the informant who, it is claimed, witnessed the factual transactions described in the warrant. There is no vicarious credibility: The informant’s reliability cannot be shored up by the impeccable credentials of the investigative agent. The agent is not the witness; the informant is. The agent’s job is to satisfy the court that the informant was in a position to provide a reliable (generally, firsthand) account of what happened, and that the informant should be believed. Typically, the prosecutor and agent will give the court an assurance along the lines of “Confidential Informant No. 1 (CI-1) works at the bank, is intimately familiar with activity in the suspect’s account, and has described the following transactions . . . ”; and/or “CI-1 has provided similar information to the FBI on three other occasions and has been corroborated each time.”
There is nothing close to that in Steele’s dossier. His informants are anonymous. Some of them do not claim to have seen or heard the things they report — they’ve purportedly gotten the information from others. Meanwhile, Steele himself does not deal directly with the sources; he has a network of connections developed in his years as a British intelligence officer. In effect, A tells B something that is reported to C, who tells it to D, Steele’s guy in Russia, who passes it along to Steele. We have no idea whether A was in a position to observe what he told B, much less what credibility problems beset A, B, C, and the rest on down the chain. This is no doubt why Steele himself, in the libel cases against him, has taken the position that he cannot vouch for the truthfulness of the allegations in the dossier. Rather, he admits they were hearsay claims that he passed along because they seemed alarming, but that needed vigorous investigation by the FBI to determine their provenance and accuracy.
Observe the dynamic here: The closer a source of information is to firsthand knowledge, the more irrelevant are the merits or demerits of the ultimate purveyor of the information to the government and, ultimately, the court. Let’s say Steele or FBI agent Peter Strzok had a source who claimed that Trump conspired with Putin to hack Democratic email accounts. The notorious fact that Steele and Strzok harbor animosity against Trump would not matter as long as their source had firsthand information and a history of reliability.
By contrast, if their informants could not be checked out and their information could not be corroborated, the impressive professional credentials of these investigators would not validate their informants. Thus, the investigators’ animus against Trump would become a very alarming problem. As experienced agents, they are well aware that they must corroborate their sources before proceeding with something as serious as a FISA warrant application; if they failed in that rudimentary duty, it is perfectly reasonable to ask whether their personal feelings got the better of them — especially since we now have direct sources whose credibility cannot be evaluated at all, a purveyor of their multiple-hearsay allegations who is deeply biased against the target, and the overlay of a presidential election in which one candidate’s opposition research is being used to justify spying on the other campaign. As Professor Kerr correctly explains, the facts of every case are different and there are indeed times when credibility problems are so dire that the government is obliged to reveal them to the court.
Now, let’s return to Nadler’s letter.
Before elevating the vicarious-credibility theory to a new level of farce — by portraying Steele as an “expert” witness — the congressman argues that “the Nunes memo does not provide a single shred of evidence that any aspect of the Steele dossier is false or inaccurate in any way.” Again, this is backwards. It is the burden of the Justice Department and FBI to establish that the allegations they provided to the court were sufficiently corroborated to show probable cause. There is already a mountain of evidence (including Steele’s own admissions) that the dossier is deeply suspect. The Justice Department and FBI relied on it nonetheless. It is thus for them to satisfy the public that their warrant application was valid; it is not up to the critics to prove that the dossier is false.
So far, the FBI and Justice Department have provided only cause for grave concern that they gave a federal court unverified, highly unreliable information that was essential to the court’s probable-cause finding, and that they did so without being candid with the court about the biases of the information’s purveyor. That being so, the burden is on the FBI and the Justice Department to prove that they did not act improperly in seeking the FISA warrant — especially since they, rather than the rest of us, are in possession of the information that they insist would vindicate them.
Nadler is a shrewd lawyer, but he has spent his life in legislatures rather than courtrooms; he is perhaps to be forgiven for not understanding how expert testimony works. He asserts that “Christopher Steele is a recognized expert on Russia and organized crime.” Let’s stipulate that this is the case. Nevertheless, before you ever get to the point of having an expert explain factual transactions that are beyond the ken of the layman, the occurrence of these factual transactions has to be established by competent, reliable witnesses. Steele and the FBI point us to none.
Steele might know more about Russia and organized crime than Putin himself. Such expertise would still not endow his anonymous, hearsay sources with credibility, or fill in the chasm left by the lack of verified factual information for an expert to interpret.
D. The Nunes Memo and FISA Law
Nadler’s point 1 asserts that the Nunes memo does not “rule out the possibility that considerable evidence beyond the Steele dossier” supported the FISA court’s finding of probable cause to believe Carter Page was a Russian agent. This framing of the issue gets things backwards: We are supposed to have extreme confidence that the findings of a federal court are not based on suspect evidence. Nadler’s standard, in effect, is: It is not impossible that the court relied on solid evidence here. That is not what we aspire to.
In any event, the Nunes memo asserts that Andrew McCabe testified before the committee that “no surveillance warrant would have been sought from the FISC [Foreign Intelligence Surveillance Court, or ‘FISA Court’] without the Steele dossier information.” I hasten to add that I am quoting from the memo, not from McCabe’s testimony — of which the Nunes memo gives us a description, not a verbatim excerpt.
Some committee Democrats have suggested this is a misleading rendering of what McCabe said. Interestingly, Nadler makes no such claim. He appears to be taking it as a given that McCabe did opine that the warrant application would not have been sufficient absent the allegations in the dossier. That makes the credibility of the dossier critical. Yet, as we’ve seen, the credibility of the dossier has never been established and almost certainly cannot be.
Nadler takes three shots at repairing this defect, all unavailing.
First, there is some razzmatazz about the timing of the warrant application. Noting that the FBI opened an investigation of George Papadopoulos in late July 2016, Nadler leaps to the conclusion that “the FBI was actively investigating the Trump campaign months before they approached the court about Carter Page.” So what?
There was never any doubt that FBI agents were investigating long before they went to the FISA Court in October 2016. The Bureau’s opening of a Papadopoulos investigation on paper does not necessarily mean agents were acting on it meaningfully (there seems to have been no Papadopoulos FISA warrant, he was not interviewed until late January 2017, and he only pops up in the Nunes memo in connection with a FISA warrant application for Page). But no matter: We know the FBI was investigating Page in July — when he went to Russia and the Bureau started getting Steele’s reports. Nevertheless, how long the agents were investigating before applying for a FISA warrant has nothing to do with what they put in the warrant application. They could have been investigating Trump and Page for a hundred years before approaching the FISA court; warrants are issued based on the quality of the information proffered to the court, not the duration of the information-gathering process.
Second, Nadler suggests that FISA law, as he purports to describe it, strongly supports an inference that the FISA court’s probable-cause finding was sound — i.e., the finding that Page was a Russian agent. But, after labeling the Nunes memo “deeply wrong on the law,” Nadler proceeds to get the law wrong, and misses the point in any event.
According to the congressman, to justify a FISA warrant, the Justice Department had to show probable cause that “the target of the electronic surveillance is . . . an agent of a foreign power.” No, that’s only half of what must be shown. Nadler neglects to explain that, because Page is an American citizen, the Justice Department also had to establish that the clandestine activity in which Page supposedly engaged on Russia’s behalf may have involved federal crimes. (See sec. 1801(b) of Title 50, U.S. Code.)
This is a salient omission. The Steele dossier alleges that Page was implicated in crimes — potentially heinous ones. As I have pointed out previously, we don’t know everything the FBI knows, so we can’t discount the possibility that Page was not only involved in criminal activity but involved in it with Russians. That seems highly unlikely, though, given that Page has not been charged with any crimes, and that there are no obvious reasons for suspecting him of crimes besides what is alleged in the dossier (allegations he strenuously denies). That is, FISA law appears to make the dossier more relevant and problematic, not less.
On the question of criminality, I note that many commentators point to Page’s involvement in a prior FBI investigation involving Russia, and darkly observe that he “was on the Bureau’s radar screen for years” before the October 2016 FISA warrant application — as if this supposition were a substitute for FISA’s legal requirement of proving criminal activity. But the 2013 investigation into which Page stumbled was a case of Russian agents trying to recruit him as a source. Far from doing anything criminal, Page appears to have cooperated with the FBI and Justice Department to nail the Russian spies. (See the Justice Department’s complaint in United States v. Buryakov, at pp. 12–13 — Page is “Male-1,” whom the Russian spy Victor Podobnyy refers to as an “idiot,” and whose 2013 interview by the FBI is described in paragraph 34.) Again, we don’t know everything the FBI knows, but based on what we have been told, it appears that in the prior case, Page worked with the United States against Russia; that does not jibe with the allegation in the FISA warrant application that he worked with Russia against the United States.
It appears that in the prior case, Page worked with the United States against Russia; that does not jibe with the allegation in the FISA warrant application that he worked with Russia against the United States.
Putting aside Nadler’s glaring omission in describing FISA’s probable-cause requirements, his statement that the law requires a finding that Page was a foreign agent does nothing to establish that Page was, in fact, a foreign agent. The question is whether the alleged facts presented in the warrant application showed that he was a foreign agent involved in potential federal crimes. Nadler does not come close to addressing that.
Third and finally, Nadler offers more vicarious-credibility theory: We should believe there was probable cause in the FISA warrant application because “the Department of Justice thought so,” and “a federal judge agreed.” This tautological proposition is even more risible than the claim that Steele’s professional credibility and expertise are a substitute for corroborated factual assertions by reliable informants. The Justice Department’s job is to establish probable cause, and the court’s task is to find probable cause based on a sufficient factual showing. These burdens are not satisfied by the happenstance that a prosecutor handed an application to a judge, who then signed a warrant. If they were, we’d be living in a very different kind of country — one that would defy Jerry Nadler’s history of civil-liberties activism.
Congressman Nadler’s attack on the Nunes memo is wholly unpersuasive, and in several particulars frivolous.
— Andrew C. McCarthy is a senior fellow at the National Review Institute and a contributing editor of National Review.