FBI Thwarted an Investigation into How Russians Injected Disinformation in the Steele Dossier

Judging from the chronology of his interview transcripts, John Durham first started chasing his conspiracy theories about Hillary Clinton by September 17, 2019 (a detail I’ll return to). This list shows all the interviews cited in both Durham’s unclassified report and the annex; the italicized ones appear only in his classified annex (meaning they likely focus on the SVR intelligence at the core of that material).

In early September, Bill Barr’s office was micromanaging what Durham should investigate, including feeding him a binder of material. On September 16, he met with a partisan Cyber Agent named Nate Batty; he’s the guy who bolloxed the investigation of the Alfa Bank allegations, knowing that they came from Democrats. Then, on September 17, he met with “HQ Analyst-3” and she explained the nature of the SVR collection.

That September 17 interview is the only one exclusively listed in the annex. But it’s not the only interview Durham did with her. Between the unclassified report and the annex, Durham cited five interviews with this analyst, starting a month earlier, on August 14, 2019. In that interview, she described checking the SVR materials for information on the people prosecuted by Robert Mueller, a question he returned to twice more, in December 2019 and February 2020.

Where this analyst played the most significant role in his report, though, was in finding — in “significant intelligence information that first became available for the FBI to review in 2018” (perhaps not coincidentally after the DOJ IG investigation into the SVR material raised concerns about whether it had been sufficiently consulted during the Hillary email investigation) — “that as a result of [Russia’s access to sensitive U.S. government information”], Steele’s subsources could have been compromised by the Russians at a point in time prior to the date of the first Steele dossier report.”

It’s possible this analyst is Brittany Herzog, who testified about Steele’s subsources at the Igor Danchenko trial, though she left the FBI in 2019 to start grad school.

The bulk of what Durham included from this analyst pertained to how, in fall 2018, after she found evidence that Steele’s subsources had been identified before the first report in 2016, senior officials at FBI told her to stop documenting her work. She escalated the problem, ultimately to David Bowdich. Durham doesn’t discuss what happened then, even though his investigation continued past the time Bowdich departed.

We’ve never heard the results of that — except, perhaps, in questions by DOJ IG why the FBI didn’t unpack the possibility that Oleg Deripaska had injected disinformation in the dossier.

Which is why Durham’s own disinformation problem (well, one of them) is so interesting.

Having interviewed Analyst-3 about what was in the SVR files, he cited the DOJ IG report (by way of the SSCI Report) to affirm that Oleg Deripaska knew of the Steele project by July 2016. But then in a totally separate section, he casually asserted (citing NYT) that Steele worked for Deripaska (something he could, and should, have cited to Bruce Ohr’s 302s).

The FBI would have multiple reasons not to want to chase the disinformation in the Steele dossier, first in 2018, and then — after Mueller had established that Manafort was trying to get debt forgiven by him when he shared how the campaign planned to win and then discussed how to carve up Ukraine, another reason when Bowdich got that briefing. Over and over again, however, people serving Trump’s disinformation purposes never seem to want to pull the threads of Deripaska’s relationship with Steele and the possibility that Russia was sending disinformation coming and going.

Incidentally, Analyst-3 was not among the people who backed Durham’s theory that the Leonard Benardo emails were authentic.

Cited testimony

August 14, 2019: SVR didn’t have anything regarding regarding any Trump election campaign conspiracy with the Russians, nor did she see anything in FBI holdings regarding Carter Page, Michael Flynn, George Papadopoulos, or Paul Manafort, though there was material on Manafort that was not connected to the election or the presidential campaign.

September 17, 2019: Timing of SVR hacks. Victims targeted. 

December 10, 2019: Timing of SVR hacks. Victims targeted. The three things obtained: emails about hacking, analysis of hacked documents, and the stolen emails hacked. Hypothesis that the reference to “special services” in the SVR report was a reference to Christopher Steele. Details of the SVR report. Probable description of the compromise of Carnegie Endowment for International Peace. SVR didn’t have anything regarding regarding any Trump election campaign conspiracy with the Russians, nor did she see anything in FBI holdings regarding Carter Page, Michael Flynn, George Papadopoulos, or Paul Manafort.

February 19, 2020: The review team initially briefed Counterintelligence and Cyber executive management about their findings on the compromise of Steele’s sources during a conference call. Following the call, while driving home, Headquarters Analyst-3 was called by Acting Section Chief-2. Acting Section Chief-2 told Headquarters Analyst-3 that they appreciated the team’s work, but no more memorandums were to be written. A meeting was then held with Assistant Director Priestap and others. During that meeting, the review team was told to be careful about what they were writing down because issues relating to Steele were under intense scrutiny. SVR didn’t have anything regarding regarding any Trump election campaign conspiracy with the Russians, nor did she see anything in FBI holdings regarding Carter Page, Michael Flynn, George Papadopoulos, or Paul Manafort.

December 2, 2021: Moreover, significant intelligence information that first became available for the FBI to review in 2018 showed that the Russians had access to sensitive U.S. government information years earlier that would have allowed them to identify Steele’s subsources. Indeed, an experienced FBI analyst assessed that as a result of their access to the information, Steele’s subsources could have been compromised by the Russians at a point in time prior to the date of the first Steele dossier report. Two weeks later, the Deputy Assistant Director for Counterintelligence, Dina Corsi, met with the review team and directed them not to document any recommendations, context, or analysis in the memorandum they were preparing. The instructions, which Headquarters Analyst-3 described as “highly unusual,” concerned the team because analysis is what analysts do. Headquarters Analyst-3 was so concerned about the failure to fully exploit the materials involving Steele subsource information (and the possible need to bring information already exploited to the attention of the FISC) that she raised her concerns about the FBI’s lack of action in an email to her supervisor in the hope of having the issues explored further. See FBI-0009265 (Email from Headquarters Analyst-3 to FBI employees dated 10/17/2018). Although the team did not fully adhere to that instruction because of the need to provide context to the team’s findings, they did tone down their conclusions in the final memorandum. Headquarters Analyst-3 recalled that a separate briefing on the review was eventually provided by the team in the Deputy Director’s conference room, although Headquarters Analyst-3 could not recall if Deputy Director David Bowdich attended the briefing. Headquarters Analyst-3 did know that Bowdich was aware of the review itself. [T]here is reason to believe that even earlier in time [Russia] had access to other highly sensitive information from which the identities of Steele’s sources could have been compromised.