Yesterday The Wall Street Journal published reporting that was sourced to “an undisclosed U.S. intelligence report” about the origins of the coronavirus known as SARS-COV2, or COVID-19. This is misinformation and agitprop.
WASHINGTON—Three researchers from China’s Wuhan Institute of Virology became sick enough in November 2019 that they sought hospital care, according to a previously undisclosed U.S. intelligence report that could add weight to growing calls for a fuller probe of whether the Covid-19 virus may have escaped from the laboratory.
The details of the reporting go beyond a State Department fact sheet, issued during the final days of the Trump administration, which said that several researchers at the lab, a center for the study of coronaviruses and other pathogens, became sick in autumn 2019 “with symptoms consistent with both COVID-19 and common seasonal illness.”
There’s only one problem with the above excerpt: the sourcing does not actually seem to be to a previously undisclosed U.S. intelligence report and is misinformation and agitprop. We know this for a few reasons. The first of which is the reporting tells us this:
Current and former officials familiar with the intelligence about the lab researchers expressed differing views about the strength of the supporting evidence for the assessment. One person said that it was provided by an international partner and was potentially significant but still in need of further investigation and additional corroboration.
The second reason we know this is that the one person who does express support for the assessment uses language that no one who does any form of intelligence assessments would actually use:
Another person described the intelligence as stronger. “The information that we had coming from the various sources was of exquisite quality. It was very precise.
“Exquisite quality” and “very precise” are not terms used in intelligence assessments by the U.S. intelligence community. Rather, U.S. intelligence analysts, as well as their team and section and division leaders, and their colleagues who brief their work to senior leaders and decision-makers use the terms high, moderate, and low confidence, as well as highly likely or probable, likely or probable, or not likely or probable.
The Intelligence Community Directive Number 203, which is published by the Office of the Director of National Intelligence, delineates the analytic standards for U.S. intelligence analysis. It includes instructions on how to express likelihood in regard to confidence. None of the approved terms include “exquisite quality” or “very precise”. You’ll find the actual terminology, as well as the deconfliction guidance regarding assessing and stating confidence on page three of the directive.
The requirements for using analytic confidence indicators of high, moderate, and low come from the Intelligence Reform and Terrorism Prevention Act of 2004. A description of how all of this estimative language is used – from likelihood/probability of an event happening or not happening to the confidence in any specific part of a given assessment or the assessment overall – can be found on page four of the 2007 National Intelligence Estimate regarding Iran’s nuclear program. It is standard language — found in all National Intelligence Estimates.
- High confidence generally indicates judgments based on high-quality information, and/or the nature of the issue makes it possible to render a solid judgment. A “high confidence” judgment is not a fact or a certainty, however, and still carries a risk of being wrong.
- Moderate confidence generally means credibly sourced and plausible information, but not of sufficient quality or corroboration to warrant a higher level of confidence.
- Low confidence generally means questionable or implausible information was used, the information is too fragmented or poorly corroborated to make solid analytic inferences, or significant concerns or problems with sources existed.
Again, you will not find terms like “exquisite quality” and “very precise” anywhere in a U.S. intelligence assessment unless the author or authors of that assessment are quoting someone else directly. They are not terms used by U.S. intelligence personnel in their analytical products.
There is one additional important point demonstrating that this is not from a U.S. intelligence report: every single source claiming this is from an undisclosed U.S. intelligence report is anonymous. Not just anonymous, but glaringly anonymous.
The terms “current and former officials familiar with” and “one person said” and “another person described” does not tell us anything. Are these current and former intelligence officials? Are they current and former CDC officials? Are they current and former Department of Health and Human Service officials?
There is no way to know based on the way this reporting is sourced. We cannot know from the reporting who these officials are, what level of seniority they have or had, and which agencies or departments they worked for. That, itself, is an indicator that something ‘hinky‘ is going on.
The third reason we know that this information is not from a U.S intelligence report is because this story has been previously shopped and published, as well as debunked and walked back. Matt Tait, a former information security specialist at Britain’s equivalent to the National Security Agency, the Government Communications Headquarters (GCHQ), and who tweets as “Pwn” All the Things, helpfully breaks it all down for us by providing links to all the previous reporting sourced to or about this same assessment. For instance, this article from May 4, 2020 at The Daily Telegraph in Australia, followed by reporting in the Guardian from later that same day makes it very clear that the Five Eyes intelligence network (the intelligence-sharing network of the U.S., Canada, Britain, Australia, and New Zealand) are not responsible for this information. Here’s an earlier report from NBC on April 9, 2020 that also makes the same claims. Notice the similarity in the sourcing of the reporting to what The Wall Street Journal reported yesterday. Here are the first two paragraphs of that reporting (emphasis mine):
U.S. spy agencies collected raw intelligence hinting at a public health crisis in Wuhan, China, in November, two current and one former U.S. official told NBC News, but the information was not understood as the first warning signs of an impending global pandemic.
The intelligence came in the form of communications intercepts and overhead images showing increased activity at health facilities, the officials said. The intelligence was distributed to some federal public health officials in the form of a “situation report” in late November, a former official briefed on the matter said. But there was no assessment that a lethal global outbreak was brewing at that time, a defense official said.
The same ambiguous descriptions of U.S. officials also appear, as well as Department of Defense refuted the reporting, which NBC put in the third and fourth paragraphs:
On Wednesday night, the Defense Department disputed an ABC News report that an “intelligence report” had warned about the coronavirus in November.
“We can confirm that media reporting about the existence/release of a National Center for Medical Intelligence (NCMI) Coronavirus-related product/assessment in November of 2019 is incorrect,” said a statement by Air Force Colonel Dr. R. Shane Day, who is director of the National Center for Medical Intelligence, a unit of the Pentagon’s Defense Intelligence Agency. “No such NCMI product exists.”
Notice the difference in sourcing here. We have the name; rank; three titles: doctor (Dr.), colonel, and director; the name of the agency that Col. Day is a director of, and where that intelligence agency is located in the Department of Defense. While the allegations are ambiguous with no clear documentation and sourcing, the refutation here is clear, with proper documentation and sourcing.
Tait also provides us with a link to BBC reporting from June 15, 2020 that walks back these same claims in its earlier reporting. He also provides a link to Daily Beast reporting from June 23, 2020 that also debunks what The Wall Street Journal has resurfaced in their reporting yesterday. Finally, Tait provides us with links to both the Harvard study that serves as the basis for the geospatial assertions in all of this reporting – not the National Geospatial Intelligence Agency (NGA) – and the analysis debunking it by an open-source geospatial analyst.
Analysis from @harvard with imagery of hospital parking in Wuhan suggests early outbreak of COVID-19. They use maps from Tianyou Hospital (Oct 2018/19) to show difference, but car park is often full on @googleearth images 🧐https://t.co/PwaEIVwrPP pic.twitter.com/RfBtpmtu2N
— Benjamin Strick (@BenDoBrown) June 11, 2020
There are eight tweets in that thread with lots more imagery comparison (as well as a link to further debunking of the Harvard study) that seems to serve as the basis for all of the reporting that there was a disease outbreak in Wuhan, China in November of 2019. This apparently serves as the basis for the assertion that there was a major COVID-19 outbreak earlier than previously reported.
The official position of almost all of the top infectious disease and virology researchers, including those within the U.S. government, is that all evidence presented so far seems to indicate that SARS-COV2 and COVID-19 developed as a result of zoonotic – natural animal to human – transmission. Basically, this coronavirus mutated as a result of natural transmission within animal species until it was able to infect humans and then it did, somewhere near or within Wuhan, China. However, none of these researchers are able to definitively rule out that it might have been the result of an accidental leak from the virology lab in Wuhan. As a result, more inquiry and study are needed.
This is not indecision. Nor is it an attempt to have it both ways with the analysis. This is a result of how the language of scientific inquiry seems ambiguous to most people not trained as scientists.
As a social scientist – my doctorate is joint in criminology and political science – I am trained and educated in the same basic concepts of scientific inquiry as are the infectious disease and virology researchers that have been working to ensure that we all survive the pandemic. Terms like query, hypothesis, theory, methodology, operations, findings, and conclusions, as well as deductive versus inductive reasoning, p values, significance, quantitative, qualitative, or mixed methodology are part of the stock in trade of scientific inquiry, as well as its jargon. What science, scientists, and the technical language and jargon used by scientists are not good at is providing complete certainty.
The beauty of science, which is also its underlying dynamic, is that as new information and data become available and more refined and powerful methods and techniques for analysis are developed, conclusions change. Scientific explanations must be falsifiable, otherwise, they are not scientific. Scientists, because of their education and training, intuitively understand the ambiguity of technical language they use professionally is not always good at making their findings clear and plain for non-professionals.
The same holds true for those who work in national security, producing or consuming intelligence. Here too there is a very specific type of professional language. Those terms are not necessarily clear to non-professionals. Whether journalists being pitched information by a source or by the people reading their reporting.
It is very clear that once again, someone is trying to promote the idea that the U.S. and/or one of its Five Eyes intelligence partners had intelligence that COVID-19 was produced in the virology lab in Wuhan. This, despite there not being any clear evidence to support this.
It is possible that someone in the Wuhan virology lab was either exposed naturally to the virus as it first began to circulate or someone at that lab got wind of something strange happening in terms of disesase in the area, collected some samples, brought them to the lab, and then accidentally wound up releasing them. The real question is: are either of those scenarios probable and, if so, is it high, moderate, or low probability?
Neither The Wall Street Journal‘s reporting yesterday, nor none of the similar reporting that began in April 2020, provides any real indication that what is being alleged is at all probable. All of this reporting, including yesterday’s, as well as the assertions/accusations leveled at congressional hearings on this topic and convoluted questions by Fox News’ (which, like The Wall Street Journal is owned by Rupert Murdoch) White House correspondent, are improbable. Meaning, as in somewhere between unlikely and highly unlikely.
What is clear however, is that someone wants this long-debunked misinformation and agitprop brought back into the discussion at this time and has sought to launder it through The Wall Street Journal to do so.
Edited to Add by Adam L. Silverman on 30 MAY 2021: The tweet that was appearing below the final paragraph of this column has now been relocated back to where it was supposed to be. I have also corrected some mild stylistic errors, which has not affected the point of the column.