Posts

Think of the HPSCI Report as a Time Machine to Launder Donald Trump’s Russia Russia Russia Claims

On July 11, 2017, I noted that we had proof that Trump’s claims he had been wiretapped were false. That’s because, if the Intelligence Community had found an exchange like the one Don Jr released that day — in which someone working for Russian oligarch Aras Agalarov offered Don Jr, “very high level and sensitive information but is part of Russia and its government’s support for Mr. Trump” and he responded, “If it’s what you say I love it especially later in the summer,” then the confidence level for the Intelligence Community Assessment that Russia had affirmatively tried to help Donald Trump get elected would have been high, even for the NSA (which said it only had moderate confidence).

And yet we’re still arguing over whether that judgement was fair eight years later.

The HPSCI report released the other day — which is dated September 18, 2020, but which right wing HPSCI Chair Rick Crawford misleadingly released alongside a statement pretending it was done in January 2017, and which the architect of the case for the Iraq War, Fred Fleitz, says was done in 2018 —  seems to have been a response to a more rigorous SSCI Report released weeks earlier that confirmed Putin did want Trump to win. (Indeed, Fleitz offers a garbage explanation to claim this HPSCI Report is more credible than the SSCI one.)

The HPSCI report manages to challenge the SSCI conclusion by revisiting a different question: Not, did Russia take actions to help Trump win the election that created a grave counterintelligence threat (the SSCI report included, but went far beyond, the evidence released in the Mueller Report) but, did the IC claim that Putin did want Trump to win, made in a rushed report published in the first days of 2017, have enough evidence behind it at the time?

The report accuses John Brennan of having made up his mind to that question, yes, Putin aspired to help Trump, and thereby influenced the result. Fair enough (though as I read it I wondered whether Brennan knew of the advance notice of the email leak that George Papadopoulos got, and so had a source of confidence not reflected in the report; and the nature of the claim in the HPSCI report differs from the nature of the claim in a report John Ratcliffe released last week). But the HPSCI report does the exact same thing, delivering up the result that Donald Trump wanted, and it did so, in part, by intentionally remaining dumb to many, but not all, of the things that had been learned since.

Just as one example, the report asserts,

[T]he available intelligence showed: No Putin orders directing or suggesting operations intended to elect Trump (by contrast, Intelligence on Russian operations on German elections specifically mentioned Putin’s goal of defeating Chancellor Merkel).

This refers to this ICA judgment predicting that Germany would be the next country in whose elections Russia would interfere. Only, that judgement turned out to be wrong; Russia conducted the same kind of hack-and-leak campaign targeting Emmanuel Macron in May 2017 (with the help of pro-Trump influencers), so while that might be a good argument in January 2017, it fails after May 2017, to say nothing of September 2020.

More importantly, it only references the intelligence available through December 29, 2016, so wouldn’t include the damning email to Don Jr disclosed in July 2017. Because the ICA didn’t include ongoing FBI investigations, it wouldn’t include Papadopoulos’ brags about Russian interference that the FBI knew about, but didn’t, couldn’t, include in the report. It wouldn’t include the intercepts between Mike Flynn and Sergei Kislyak discovered in that very period as the IC sought to explain why Putin decided not to retaliate against US sanctions.

The HPSCI report concluded there was not enough intelligence to back a high confidence conclusion that Putin wanted to help Trump win by focusing only on the time before FBI started looking in earnest.

In any case, as I’ll show, in several places, the report breaks the conceit that they’re evaluating only the information available to the IC on December 29, 2016.

The HPSCI Report, then, is not so much a useful piece of analysis, but a time machine, an attempt, weeks before the 2020 election, to set the clock back for candidate Donald Trump to the time before it became clear he really did benefit from help from Russia.

And it doesn’t even do that very well.

The report structure

As laid out below, the report doesn’t hide that its sole goal is to erase the judgment that Putin wanted to elect Trump. Its first finding is that everything else in the ICA is sound, followed by seven poorly-organized findings ostensibly explaining why the assessment that Putin aspired to help Trump was unsound.

  • Finding 1: The bulk of the ICA judgements were sound.
  • Finding 2: Significant tradecraft failures cast doubt on the ICA judgments of Putin’s intentions, claiming that only on the judgment that Putin aspired to help Trump, the ICA tradecraft failed.
  • Finding 3: The ICA failed to acknowledge that key judgements were based on raw intelligence that did not meet tradecraft standards: This claims that of 15 sources behind the judgment, 12 were unremarkable, but three “contained flawed information and these became foundational” to the claim Putin aspired to help Trump win.
  • Finding 4: The ICA excluded significant intelligence that contradicted its judgment. This section discusses Russia explaining why there were downsides to both candidates.
  • Finding 5: The ICA disregarded Russian behavior that undermined its judgment that Putin aspired to help Trump win. This section, which notes that Moscow was receiving reports on US polling but doesn’t mention (!!!) that some came from Trump’s campaign manager, argues that since the election got close in its final weeks, you would have expected Putin to dump all the other derogatory intelligence he had on Hillary. That argument provides an opportunity to parrot the SVR documents discussed here. It also looked at what it claimed were Russian leaks (really, just one) that it said made Trump look bad. Finally, it ludicrously suggested that the disparate hacking of Hillary is just attributed to her being an easier target.
  • Finding 6: The ICA misrepresented documents on Putin’s intention. This pertains significantly to advice Putin got (it may also rely on the SVR documents). A 10-page attack on the dossier is put in here for contrived reasons, which I’ll return to in another post.
  • Finding 7: The ICA failed to consider alternate explanations. This section significantly revisits the SVR documents.
  • Finding 8: The ICA was unnecessarily rushed.

Some of this is quite reasonable. For example, Finding 4 notes that Russia was going to be unhappy with either candidate; I think Trump opponents often forget that Putin didn’t want a strong Trump, he wanted a Trump whose narcissism would create more problems than Hillary.

In Finding 8, sure this report was rushed. It had to be.

Some of the criticism of Brennan — if accurate, but as noted the complaint here is different from the complaint in more recent reports — seems fair.

The flawed reports (Finding 4)

In several other areas, the analysis only survives by relying on that time machine effect.

The report claims that of 15 pieces of intelligence to back the Putin finding, three were not just substandard, but were pushed through by Brennan.

The first I address here. Brennan pushed to include a report that Putin approved the DNC leak because he was “counting on” a Trump victory. The CIA, years later, stands by the quality of the source and the fact that the interpretation in the ICA, “was most consistent with the raw intelligence.”

The second of these three reports is far more interesting. It describes a report from Kyiv (the Republicans spell it Kiev) laying out a plan, starting in February 2016, to place someone pro-Russian on Trump’s campaign team. The analysis of this lead focuses on questionable sourcing and potential Ukrainian bias.

But the time machine effect of this report frees Republicans from accounting for the fact that Trump, starting in February 2016, in fact did place a pro-Kremlin official on his “election team,” Paul Manafort, and Manafort sought to monetize his role there by getting pro-Russian Ukrainians and a Russian oligarch to pay him.

And Konstantin Kilimnik, in Ukraine, seemed to know of that plan before Manafort was installed. This report may have looked problematic for inclusion in December 2016 (though by that point Manafort’s cover-up of his Russian ties was public). But it looked prescient by 2020.

The third report is similar. HPSCI’s response begins a long focus of the report attempting to debunk the underlying intelligence — a claim that Russia perceived Republicans to be less supportive of human rights — with a bunch of whataboutism. How dare you call the party of torture less supportive of human rights, the right wingers wail. Did you know that Reagan said, “tear down this wall”? All the while ignoring that Trump ran on an affirmatively pro-torture platform.

“They’re both poison” (Finding 5, 7)

In other words, a central pillar of the report is to complain that intelligence analysts didn’t consider alternative explanations for the intelligence they were looking at.

This was about the stage of reading this report where I could get not get the scene from Princess Bride where Vizzini attempts to outsmart the Dread Pirate Roberts out of my head.

 

That’s true because some of the arguments — and they go on for pages and pages — sound just as stupid as Vizzini does. Republicans tie themselves in knots trying to come up with alternative explanations. Republicans refuse to consider that the SVR hacks, which I wrote about here, were meant for intelligence collection but the foot-stomping GRU ones were not. They treat all the SVR reports — including the ones that, FBI had decided years before 2020 were objectively false or the one that Ratcliffe released days before this report warning it “may reflect exaggeration or fabrication” — as true and damning. They obsess about the derogatory claims about Hillary’s health and mental fitness without even considering the report Hillary released in real time after her pneumonia scare. They actually believe a claim that European leaders doubted Hillary could lead. These reports obviously play on right wing biases, and sure enough HPSCI’s Republicans cling to those Russian spy claims in the report, just as they have since Tulsi emphasized them. In a report that wails mightily (and correctly in at least one case, cited the parallel CIA report) about leaving out contrary information, HPSCI simply leaves out the Jim Comey allegation in one of the SVR reports, which if true, would explain why Putin wouldn’t have to (and didn’t) dump damning intel close to the election: Because Putin believed that “Comey is leaning more to the [R]epublicans, and most likely he will be dragging this investigation until the presidential elections,” something that turned out to be true. In other words, they cherry pick which Russian spy products they choose to parrot, one of the sins they accuse the ICA team of, but they do so with years of hindsight that made clear how foolish that was.

This report has an entire section on how Putin would have tracked polling and so knew he could get Trump over the line if he dumped opposition late in the campaign (which, of course, he did), without blushing about one source for that polling: Manafort’s regular provision of it via Kilimnik, something that became public between the ICA release in 2017 and this HPSCI report in September 2020.

Their claims get more ridiculous from there. Even in the face of the non-stop flood of Hillary emails released in 2016, right wingers cling to the single report from Colin Powell calling Trump “a national disgrace” as proof Putin doesn’t love Trump.

I mean, it’s pretty funny to me. But then ultimately it gets back to what Wesley, the Dread Pirate Roberts, said when Buttercup concluded that Wesley had put the poison in his own cup. He didn’t. He put poison in both glasses.

All the ICA did — and it’s worth reading how the “aspired to” section includes a lot of explanation as to why Putin would prefer Trump to which right wingers didn’t and don’t object, such as Trump’s willingness to partner with Russia on terrorism or make deals — all the ICA did was say that Putin wanted Trump to win. And right wingers have gone to all lengths, up to and including parroting Russian spies in the White House, to degrade the strength of that claim as it was made from high to moderate confidence years after it became clear the judgment was correct.

Ultimately the effort was intended and bound to drive more polarization. Which point the right wingers make but — oh my goodness look how they do it?!?! They point to the IRA’s activities after the election that claimed to oppose Trump’s election.

But they source that to the Robert Mueller IRA indictment, dated February 16, 2018, over a year after the ICA was completed.

By February 2018, there was abundant public evidence that Putin preferred Trump, including that letter to Don Jr as well as the guilty pleas of George Papadopoulos and Mike Flynn. By 2020, the date of this report, court filings were public describing Manafort’s lies that, Amy Berman Jackson judged, he told to cover up what happened at an August 2, 2016 meeting with alleged Russian spy where they discussed how Trump planned to win, how Manafort would get paid by pro-Russian Ukrainians and debt relief from Oleg Deripaska, and a plan to carve up Ukraine. But only here, only amid their desperate attempt to find proof that Vladimir Putin does too hate Donald Trump, do they confess they’ve read any of the charging documents from Mueller.

That is, the time machine was fake, just an attempt to make all the evidence laid out in the SSCI report go away.

As I’ll show, what HPSCI did with the dossier was even worse — so much worse I had to break it out as its own post. There, they don’t even try to maintain the illusion they were dumb to everything they learned since the ICA.

But as to their main report, claiming to assess the treatment of the intelligence in 2017 — a feigned ignorance that is central to their rebuttal of one of three “substandard” intelligence reports — they can’t even maintain that ploy as they attempt to whatabout proof that Putin wanted to help Trump win.

Share this entry

Tulsi Gabbard and John Ratcliffe Reveal Putin “Was Counting on” a Trump Win

It’s funny, reading the two rehashes of the 2017 ICA that John Ratcliffe and Tulsi Gabbard released in the last weeks.

There are parallels and common judgments between them (probably in part because the CIA one was limited to “CIA materials provided to congressional oversight investigations”). Both say the confidence level for the judgment that Putin “aspired” to help then-candidate Donald Trump win the election was too high. Both say John Brennan big-footed the process in a problematic way. Both complain about the short timeline. Both complain that “the highest classified version of the ICA had been shared with more than 200 US officials;” neither acknowledge that that was neither anticipated nor, presumably, the fault of Obama appointees, who were long-gone by the time Trump’s appointees disseminated it that broadly (and in fact other documents Tulsi released suggest that ICA drafters intentionally planned a less-classified version to be disseminated at that level, to avoid the problem Trump’s appointees complain about). Both complain about how the Steele dossier was added as an appendix, though (as I’ll show in a follow-up) they’re inconsistent about how they claim it was.

But there are differences. the document from Ratcliffe — who released the first of the SRV documents contemporaneously with the HPSCI report that obsessed about them — doesn’t appear to mention them at all.

The two reports treat three pieces of intelligence on which the “aspired” judgment was based differently (the CIA one may not treat one of the HPSCI complaints at all). As I’ll note in my main post on the HPSCI report, CIA treats one document that HPSCI considers problematic as reliable but compartmented in a way that made inclusion problematic.

Perhaps the most interesting detail you get from reading both in tandem pertains to one phrase in a document about which “a senior CIA operations officer observed, ‘We don’t know what was meant by that’ and ‘five people read it five ways,'” basically, about whether that phrase hade been read the correct way. As of a few weeks ago, in Ratcliffe’s report, the CIA was still trying to protect this intelligence, but not Tulsi. She declassified most of four pages of discussion about the phrase, with information about the access — the source was well-established, had authoritative access to something but second-hand access to this information, but for some reason the CIA was not able to clarify what the source meant by the phrase. The HPSCI Report complains that the ICA didn’t note that this person had an “anti-Trump bias” (emphasis original).

And Tulsi declassified what the intelligence said (even though she hadn’t in the less classified version of the ICA she had released a day earlier).

Putin had made this decision [to leak DNC emails in July] after he had come to believe that the Democratic nominee had better odds of winning the U.S. presidential election, and that [Trump], whose victory Putin was counting on, most likely would not be able to pull off a convincing victory.

The HPSCI memo goes on to complain that Brennan included this. It invents a number of other readings this could have meant, besides that Putin wanted to help Trump win. Maybe Putin expected Trump to win, in July 2016 when no one else did? Maybe Putin counted on a Trump win at the RNC? They even tried to undermine the intelligence by claiming that all the things Putin did to tamper in the election could have served the other goals he also had.

None of the confirmed activities — leaks, public statements, social media messaging, and traditional propaganda — corroborate the ICA interpretation of the fragment, because these activities were all consistent with Putin’s objectives to undermine faith in US democracy, without regard for candidate Trump’s fate.

Putin approved the DNC leak because he was counting on Trump to win, the fragment said, and HPSCI Republicans want to believe that maybe Putin just wanted to undermine faith in democracy.

Well, anyway, as I said, Ratcliffe didn’t declassify any of that. He did send analysts back to review the underlying intelligence, and here’s what they said.

The DA Review examined the underlying raw intelligence and confirmed that the clause was accurately represented in the serialized report, and that the ICA authors’ interpretation of its meaning was most consistent with the raw intelligence.

And Ratcliffe also backs the quality of the source behind this claim.

The DA Review does not dispute the quality and credibility of the highly classified CIA serialized report that the ICA authors relied on to drive the “aspired” judgment.

So between them, Tulsi and Ratcliffe provided us something genuinely new. According to a reliable but ambiguous intelligence fragment, CIA got intelligence that said Putin approved the DNC leak  “because he was counting on” Trump’s victory.

Update: I’ve fixed the quotation mark in the title: just the “counting on” is a direct quote.

Share this entry

Tom Cotton Does Nothing as OPM Hack Equivalent Happens in Plain Sight

Both WaPo and MuskWatch have written about the declaration that former acting Chief of Staff to the then-Acting Social Security Commissioner, Tiffany Flick, submitted in a union lawsuit against the Social Security Agency on Friday. To support a bid for a Temporary Restraining Order arguing, in part, that the way DOGE has handled Social Security data exposes the unions’ members to fraud, Flick described how DOGE boys were given rushed access to the most sensitive kind of Social Security data, including:

The Enterprise Data Warehouse, which houses SSA’s master files and includes extensive information about anyone with a social security number (including names, names of spouses and dependents, work history, financial and banking information, immigration or citizenship status, and marital status);

The Numident file, which contains information about the assignment of social security numbers; and

The Master Beneficiary Record and SSI Record files, which contain detailed information (including medical data) about anyone who applies for or receives Social Security or SSI benefits

While WaPo’s Lisa Rein (who has been covering this particular takeover closely and was cited in the filing) ends her piece quoting Flick saying, “the risk of data leaking into the wrong hands is significant,” neither Rein nor MuskWatch considers the full implications of this. (And to be fair, the union’s lawsuit, which represents general government employees, doesn’t either.)

Though this complaint includes a FISMA component, meaning the unions are arguing, in part, that the government is violating its own cybersecurity rules, it does not and cannot make a national security argument: That treatment of the entire country’s data in this fashion presents enormous national security risks.

As Flick describes, Elon’s DOGE boys came into the Social Security Agency harboring and clinging to conspiracy theories about fraud, even when offered explanations to debunk them.

20. [snip] We proposed briefings to help Mr. Russo and Mr. Bobba understand the many measures the agency takes to help ensure the accuracy of benefit payments, including those measures that help ensure we are not paying benefits to deceased individuals. However, Mr. Russo seemed completely focused on questions from DOGE officials based on the general myth of supposed widespread Social Security fraud, rather than facts.

[snip]

51. Additionally, even with only read access DOGE can, and has already, used SSA data to spread mis/disinformation about the amount of fraud in Social Security benefit programs. The agency can always do more to ensure accurate and timely benefits payments, and it continues to pursue improvements. However, fraud is rare, and the agency has numerous measures in place to detect and correct fraud.

Having nothing more than conspiracy theories, DOGE demanded — and got (partly by replacing the Commissioner with a staffer who had worked with DOGE in advance) — that Akash Bobba be granted access to virtually all of Social Security Agency’s data, immediately. Bobba appears, with description of his access at GSA, in this Wired profile. Bobba got access to that data via a telework option, meaning he was located with a bunch of other people not cleared into this data itself.

22. Throughout this time, Acting Commissioner King requested that Mr. Russo report to her, as the CIO normally would, but he consistently gave evasive answers about his work. It appeared to me that he was actually reporting to DOGE.

23. During the week of February 10, with daily pressure from Mr. Russo, the CIO’s office tried to rapidly train Mr. Bobba to get him access to SSA data systems so he could work on a special project for Mr. Russo at DOGE’s request and so that he could “audit” any of the work of SSA experts.

24. We worked to provide Mr. Bobba with the necessary information and information security training but had to do so in a truncated manner and outside normal processes.

25. Given that, I do not believe Mr. Bobba had a sufficient understanding of the sensitive nature of SSA data or the ways to ensure such data’s confidentiality. These are complicated systems with complex policies governing very large programs, and it simply is not possible to become proficient within a matter of days.

[snip]

28. [snip] I understood that Mr. Bobba was working off-site at OPM while he was analyzing the SSA data. I also understood that other, non-SSA people were with him and may have also had access to the protected information. My understanding is that Mr. Russo approved a telework agreement for Mr. Bobba (while at the same time directing CIO management to work onsite full-time) to allow him to work out of OPM. But our standard telework agreements state that employees need to work in a private location and should be careful to protect systems and data from unauthorized access. Mr. Bobba’s work didn’t seem to align with those requirements.

[snip]

36. It was never entirely clear what systems Mr. Russo wanted Mr. Bobba to have access to, but Mr. Russo reportedly stated that Mr. Bobba needed access to “everything, including source code.”

[snip]

43. But the request to give Mr. Bobba full access to these databases without justifying the “need to know” this information was contrary to SSA’s longstanding privacy protection policies and regulations, and none of these individuals could articulate why Mr. Bobba needed such expansive access. I also understood that Mr. Bobba would not view the data in a secure environment because he was living and working at the Office of Personnel Management around other DOGE, White House, and/or OPM employees.

Even if we could assume these DOGE boys — at least three of whom (Edward “Big Balls” Coristine, Branden Spikes, and Sam Corcos) have been shown to have suspect ties — have no other motive than to spin false claims of fraud, this would still be a massive security risk. But as Flick repeats over and over, these DOGE boys were always evasive about what they were really up to. And as she describes, these boys are working off site, without the kind of confidentiality protections that would apply within SSA.

By handling the data like this, they make it child’s play for adversaries to help themselves as well.

It’s not just that DOGE has found almost nothing while compromising the most sensitive datasets in government. It’s also that the way they’re doing so, driven in significant part by this haste, has made it exceedingly more likely someone else will compromise the data.

The risk is not just fraud (the harm laid out in the lawsuit). It’s spying, on an even greater scale than China achieved with the OPM hack.

And the members of Congress who’re supposed to oversee such issues have done nothing — at least nothing public.

I’ve included contact numbers for the Senate Intelligence Committee (which is the most likely to give a shit about possible compromise like this), as well as the Chair and Ranking members of other committees with jurisdiction. If one of them is your Member of Congress, call and ask why they’re abdicating their duty to protect the country from obvious compromise.

Senate Intelligence Committee

GOP

Tom Cotton (202) 224-2353

Jim Risch (202) 224-2752

Susan Collins (202) 224-2523

John Cornyn (202) 224-2934

Jerry Moran (202) 224-6521

James Lankford (202) 224-5754

Mike Rounds (202) 224-5842

Todd Young (202) 224-5623

Ted Budd (202) 224-3154

Dems

Mark Warner (202) 224-2023

Ron Wyden (202) 224-5244

Martin Heinrich (202) 224-5521

Angus King (202) 224-5344

Michael Bennett (202) 224-5852

Kirsten Gillibrand (202) 224-4451

Jon Ossoff (202) 224-3521

Mark Kelly (202) 224-2235

Senate Homeland Security Committee

Rand Paul (202) 224-4343

Gary Peters (202) 224-6221

House Intelligence Committee

Rick Crawford (202) 225-4076

Jim Himes (202) 225-5541

House Homeland Security Committee

Mark Green (202) 225-2811

Bennie Thompson (202) 225-5876

Share this entry