How Did Christopher Steele Collect Information after Sources (Allegedly) Dried Up?

Sorry to those who think I’m overly focused on the Christopher Steele dossier, but I’m reading Luke Harding’s book on the Russian investigation, which uses the dossier as a centerpiece. I may do a longer post about what his overall narrative does, but for now there’s a weird paragraph that conveniently is in this long excerpt I want to focus on.

After introducing the first report of the dossier (the one that features the pee tape and dated, non-email kompromat), Harding writes,

The memo was sensational. There would be others, 16 in all, sent to Fusion between June and early November 2016. At first, obtaining intelligence from Moscow went well. For around six months – during the first half of the year – Steele was able to make inquiries in Russia with relative ease. It got harder from late July, as Trump’s ties to Russia came under scrutiny. Finally, the lights went out. Amid a Kremlin cover-up, the sources went silent and information channels shut down.

There are several details that conflict with known facts and/or claimed (in some cases, sworn) ones.

First, Harding suggests there were 16 reports in all. I’m not sure whether he’s suggesting the final total of reports written between June and early November was 16 or whether he’s suggesting there were 16 additional reports in all, for a total of 17. Either way the number works out (there were 17 total reports, one of which was written after November). But that makes the November reference weird. There was no report written in early November. The last known report before the election was dated October 20, and then there wasn’t another one until that December 13 one.

  • 080: June 20, 2016
  • 086: July 26, 2015 (citing events in 2016)
  • 095: not dated
  • 94: July 19, 2016
  • 097: July 30, 2016
  • 100: August 5, 2016
  • 101: August 10, 2016
  • 102: August 10, 2016
  • 136: October 20, 2016
  • 105: August 22, 2016
  • 111: September 14, 2016
  • 112: September 14, 2016
  • 113: September 14, 2016
  • 130: October 12, 2016
  • 134: October 18, 2016
  • 135: October 19, 2016
  • 166: December 13, 2016

In any case, Harding gets the December date sort of correct later in the passage. Except he describes Glenn Simpson giving John McCain the report, dated December 13, before McCain called Jim Comey about it on December 8.

Less than 24 hours later, Kramer returned to Washington. Glenn Simpson then shared a copy of the dossier confidentially with McCain, along with a final Steele memo on the Russian hacking operation, written in December.

McCain believed it was impossible to verify Steele’s claims without a proper investigation. He made a call and arranged a meeting with Comey. Their encounter on 8 December 2016 lasted five minutes. Not much was said. McCain gave Comey the dossier.

I explain the significance of these December dates in this post.

Things are even weirder with the third sentence in this passage.

For around six months – during the first half of the year – Steele was able to make inquiries in Russia with relative ease.

According to the public narrative, Steele wasn’t working for Fusion until the Democrats asked for a Russian focus in June. And the first of his released reports relies on reporting from June. But Harding here suggests Steele was working on it for the six months before that! I pointed to circumstantial evidence that Fusion paid Steele on March 22, April 6, and May 25, in payments they don’t associate with Perkins Coie, in addition to the payments that were probably to him on July 13, August 2, September 1, October 5, and November 1.

Now check out the following sentences. Starting in “late July … the lights went out and … the sources went silent and information channels shut down.”

As the timeline above makes clear, the numbering in the dossier gets funky almost immediately, but the most likely reading suggests after that first, June 20 report, there are 4 reports from late July, and the remaining 12 reports all postdate late July. Report 100, the first post-July one, is sourced to “early August 2016” (and dated August 5).

Now, maybe the paragraph is just totally screwy. But if there’s any basis in fact to it, it suggests the public timeline is wrong (something which may be backed by the payments). More importantly, it suggests Steele’s extensive (albeit very indirect) network of sources stopped providing intelligence not long after he allegedly started his inquiry.

16 replies
  1. brightdark says:

    Wasn’t there a report that Steele never went to Russia himself and depended on his sources for all the information? Unnamed sources that used sources that they never named to Steele. So they could have been talking directly to the FSB/GRU/Putin and nobody would have known it on the receiving end.

    I’ve always thought the trying to prove the dossier thing is like trying to nail jello to a wall.

  2. earlofhuntingdon says:

    One possibility is that Steele was constantly working his sources throughout the year, but that as the subject of his work became more obviously Trump, the powers that be in Moscow made it known that cooperation with Steele on that subject would be hazardous to one’s health.

    If so, that would be added support for the idea that the Russian government had its own op focused on Trump, that Trump was and presumably had been a special subject for some time, and that looking into it would be considered trespassing.  More smoke, not yet (publicly) fire, but encouragement for Mueller’s team to look harder.

    • milkshaken says:

      that would be my guess: As long as Trump was considered just an unimportant buffoon, there were enterprising insiders connected to Russian security services willing to trade gossip about Trump’s compromising history. Also I remember there were quite a few  “sudden natural deaths” and few arrests and “suicides” in the Russian intelligence and military community reported in the second half of 2016 – I wonder if some of them were in any way connected to this.

  3. zonefreezone says:

    If he was working sources the first half of the year, did he take 6 months to type and deliver them?

    Makes no sense. It looks like June thru Nov were fruitful months but not the beginning of the year supposedly prior to Steele’s engagement with Fusion.

    If the lights went out in July, yeah,no.


    Theory: Farfetched: Steele (or someone yet to be identified) compiled and was working for Brennan/Clapper/Clinton prior to being “engaged by PCoie” in April. After the emails were dumped they had to work up a bogus timeline to hide the true basis for and identity of Steele preliminary “clients” this earlier info gathered and they had to do it fast. How much of the dossier info  is contemporaneous – I haven’t rechecked but it’s probably none.

    Why Luke Harding would slip up and provide the true timing given his motives is hard to discern.


  4. calvin412snax says:

    I’m reading Luke Harding’s book on the Russian investigation, which uses the dossier as a centerpiece. I may do a longer post about what his overall narrative does….

    Definitely eager for this take, emptywheel!

    I was just listening to this Fresh Air interview with Harding in which he discusses the book. I think it’s interesting that Harding places an extraordinary amount of faith in the Steele dossier and its sources, whereas you are much more dubious on him and them. While I understand that you have largely been critiquing use of the dossier by Trump’s opponents from a strategic perspective, you also seem to question its reliability in general (specifically with regard to the hacks and leaks, correct?). Does Harding put so much stock in the dossier because he thinks the money-trail is sufficient to prove collusion, and that part of the dossier is more legit?

    • Avattoir says:

      Harding refers to the content of what Steele put in his reports to Fusion as “claims“.

      For some years, I was in a job that involved reading intel reports, fairly regularly in fairly large volume. “Claims” was not how I was taught or instructed to treat any report from a trained intel officer or agent, nor was it how I ever read one.

      Instead, I’d start out with looking for thoroughness, in reporting of everything that might be both at all relevant & conceivably true, or at least potentially valuable to know about, along with some sort of thought-out (preferably standardized to the agency) glossary or guide to the reader as to on what level of credibility the reporter assesses to each item that reported. In my experience, my approach was (and hopefully still is) is how government agency intel reports, whether foreign or domestic, should be written, be read & be properly appreciated – ideally, of course. And the standard I’m describing has in place for the better part of 4 decades before I ever had anything to do with any of them.

      It’s difficult to conceive of how Harding would not have been told this.

      But Harding also would know that it’s easier to sell a product by over-selling it: see e.g. SoS Powell at the U.N. in March 2003; see e.g. candidate speeches by the Trump campaign in 2016 & going back at least as far as 2012.

      That is, it’s easier to sell a publisher on such a book project as Harding was selling, by proposing to treat reporting of as many & much as possible of the most sensational stuff as “claims” – using the vernacular sense as his excuse, as opposed to reflecting on all the context & other such nous involved in interpreting the assessment of each item.


      I admit to being among those annoyed with Ms. W’s apparent fixation on the so-called Steele dossier. I’ve read her posts on this & have yet to see any objective basis justifying lavishing on it nearly that level of attention.

      But this post by Ms. W, and some of the comments below (a little it) have put an end to my annoyance. And this not merely b/c FCOL this is her blog so we must accept her, like we should everyone, as we find her.

      Without going on & on about this, I’ll just make this one observation in the hope it’s not too cryptic: that, on this subject & all that’s related to it, and at this time, I feel very few things in life would be capable of making Ms. W happier than for her to have established contact with Mr. S.

      Maybe they could do a podcast: pretty sure folks not smart or lucky to be regular readers here would listen. I doubt they’d have any serious communication issues, once it started.

      • bmaz says:

        Your description of how a prosecutor reads reports is exactly what cracks me up so much with people who ask “oh my, was it used to get FISA warrants??” Yeah, no, it was not. Even asking that belies a lack of understanding of how a Title III warrant is constructed and presented, much less the more fastidious FISA application process. Could info also in the dossier that was independently obtained and verified have been in an app? Sure, but that is not the same as it being “based on the dossier”. Not to mention the first FISA warrant we know of seems to have clearly predated the dossier, much less Steele’s time working on it. The dossier itself will never be evidence in and of itself, because that is not how evidence actually works, so the hype and furor over it kind of escapes me.

  5. earlofhuntingdon says:

    Steele is former MI6.  After being outed, he continued to work for the government, and then left for private business.  Given that he’s self-employed, he would be constantly marketing for clients.  He sells them information and insight into what it might mean.  To have that expertise to sell, he would be constantly working his sources, to stay current on who’s in and who’s out, what they’re doing, with whom, and occasionally why.

    Steele would have “typed up” his information when he had a customer to sell it to.  Typically, he would have presented it in a way that would keep the customer coming back for more.  He would also revise what and how he presented it in response to varying feedback received from his customer(s), as well as in response to broader events.

  6. Willis Warren says:

    Maybe he means the fiscal year?


    The easiest explanation is that the sources dried up in October, right before the election.  The July paragraph is just bad editing

  7. Willis Warren says:

    Looks like I was right about the manuscript containing errors. Harding admits he threw it together pretty quickly

    First of all, this book has been secret. It wasn’t announced. It wasn’t on Amazon. No one knew about it until two weeks ago. We – I have a wonderful publisher called Sonny Mater (ph) who brought everything forward when the Mueller indictments came out. I stayed up all night and I rewrote the epilogue, and the book has been kind of crashed out in record time.

  8. Kim says:

    Do we know for certain that Steele wasn’t working the Russia/Trump intel before being hired by Fusion?

    “late July … the lights went out and … the sources went silent and information channels shut down.”

    That would be about the time that Fusion’s client changed to HRC campaign, about the time Trump gets the nomination.

    Would that be about the time that Putin realizes Trump might be the next Prez?
    Suspect Putin would not be happy with Russian sources working for HRC campaign.

Comments are closed.