In July, ODNI declassified and released 2017 HPSCI(Permanent Select Committee on Intelligence) Majority staff report regarding "Russia's Influence Campaign Targeting the 2016 US Presidential Election". Below is the link to ODNI press release:
...
President Obama directed the creation of this January 2017 Intelligence Community Assessment after President Trump defeated Hillary Clinton in the 2016 election, and it served as the basis for what was essentially a years-long coup against the duly elected President of the United States, subverting the will of the American people and attempting to delegitimize Donald Trump’s presidency.
...
The declassified 46-page report in PDF format can be downloaded via the link below:
I'm archiving its text to make it searchable here. This is part 4.
The ICA Mischaracterized the Fragment As Supporting "High Confidence" Judgments. To judge with "high confidence" - as the ICA does - that Putin's goal was to help Trump win would, per ICD 203 guidance, require "clear and reliable reports from multiple sources." Moreover, these reports would need to be of sufficient quality to confidently explain Russian actions that were not consistent with helping Trump win, and to also rule out alternative explanations for the Russian influence operations. The ICA failed to meet these standards, and the highly compartmented nature of the raw reporting made it difficult or impossible far most readers to see the foundational sources (see box "Confidence Definitions").
What the ICA Says: Confidence Definitions
High confidence generally indicates that judgments are based on high-quality information from multiple sources.
Moderate confidence generally means that the information is credibly sourced and plausible but not of sufficient quality or corroborated sufficiently to warrant a higher level of confidence.
Low confidence generally means that the information's credibility or plausibility is uncertain, that the information is too fragmented or poorly corroborated to make solid analytic inferences, or that reliability of the sources is questionable. [ICA p 13]
The NSA justification for not signing on with the CIA-FBI "high confidence" judgment on Putin's intentions to help candidate Trump (NSA preferred "medium") highlights the weakness of the evidence for this major judgment:(redacted)
(redacted) NSA has moderate confidence in this assessment given a limited source base, lack of
corroborating intelligence, and the possibility for an alternative judgment.(redacted)(redacted) The Director of NSA, Admiral Rogers, testified: "It ultimately boils down to a HUMINT source that did not have direct access ... I didn't find the level of corroboration that led me to high [confidence]... I didn't see multiple sources."
The ICA includes a "Scope and Sourcing" statement describing the reliability of the key CIA HUMINT reporting that is misleading (see box "Scope and Sourcing").
What the ICA Says: "Scope and Sourcing"
The ICA notes that key judgments are based on a single "well established" source.
(redacted) We make some judgments based on the reporting of an established clandestine source with secondhand access through identified subsources. The source is well established, and other examples of (redacted) reporting have been corroborated through other streams of human and signals intelligence. [ICA p.i]
(redacted) The established source with secondhand access provided us our only specific information on President Vladimir Putin's order to pass collected material to WikiLeaks; the timing of the formal influence campaign; the existence of specific, planned Russian Foreign Intelligence Service (SVR) efforts; some specific details of Putin's views of Secretary Clinton; and the reported role of the Federal Security Service (FSB) hacking operations related to the US election. [ICA p.ii]
(redacted) Although the ICA correctly describes the primary source's reporting history, it does not explain that information a primary source passes on from a subsource is only as good as the subsource's access, knowledge, and bias. If those factors are unknown, then confidence in the information is affected accordingly.
When a source sends a report that is unclear, the utility of the information is limited until it can be clarified, and particularly for a major judgment.
(redacted) In the case of the ICA's foundational reports from the established source, it was not possible (redacted) to clarify the meaning of report language or identify how it was obtained by the subsource.(redacted)
DCIA Ordered the Publication of a Second Substandard Report, From an Unknown Subsource, Cited by the ICA to Allege That Putin Preferred Trump. This information was both unverified and implausible and, like the unclear fragment, CIA professionals originally declined to publish it when it was first collected. It was only published on DCIA's orders after the election on 16 December 2016 and was subsequently used, without caveats, to source the first bullet of evidence for the judgment that Putin "developed a clear preference" for candidate Trump (see box "First Bullet").(redacted)
__What the ICA Says: The First Bullet of Evidence on Putin's "clear preference" for Trump
- (redacted) "As early as February 2016, a Russian political expert possessed a plan that recommended engagement with [Trump's] team because of the prospects for improved US-Russian relations, according to reporting from (redacted) government service." [ICA p 2]
The ICA bullet text is alarming, implying the existence of a Russian plan for engagement with the Trump campaign that most readers would see as strong evidence of President Putin showing a "clear preference" for candidate Trump. But the ICA omits critical report context which, had it been made available to the reader, would show the report to be implausible - if not ridiculous - and missing so many key details as to be unusable.
(redacted) CIA operations officers declined to publish the report when it was acquired (redacted) in February 2016, considering it "odd" and "_lacking authoritativeness."(redacted)
(redacted It was only disseminated in December 2016, on DCIA's post-election "full review" order to put out previously unpublished information, and experienced CIA officers said that it "would not have met the threshold" for dissemination otherwise.(redacted)
The ICA fails to clarify that "the plan" was just an email with no date, no identified sender, no clear recipient, and no classification. CIA could not vouch for the ultimate source's vetting, validation, or access.(redacted)
(redacted) Station officers were unable to obtain further clarification of this report from the (redacted) service (see box "Second Report's Context Warning").
What the Raw Intelligence Says: Second Report's Context Warning
(redacted) CONTEXT: "the CIA can neither independently vouch for (redacted) vetting or validation of the ultimate source nor the ultimate source's access to the reported information. The document contains no classification. The document did not carry a specific date or identify the originator." (redacted)