Jessica Davis: I'm an intel expert. Here's what I see at the POEC
Some of the police intelligence reports are very good. Some are very bad. This should worry us.
By: Jessica Davis
Over the last few weeks, we’ve been treated to an inside view of the intelligence function in the Ottawa Police Service (OPS) and the Ontario Provincial Police (OPP). It’s been a mixed bag. Some of the assessments released to the Public Order Emergency Commission so far are good, professional, and accurate. Others are far less so. The variation in the quality of these reports points to some serious issues around threat assessment in the OPS and the OPP. As the people of Ottawa can no doubt attest, the police response was not informed by accurate intelligence. Even when it was available, it’s not clear that Ottawa police leadership were reading it, something that no doubt contributed to the invocation of the Emergencies Act and the duration of the occupation.
As someone who has worked in intelligence in the Canadian government (and now outside it) for over 20 years, first in the Canadian Armed Forces, and then at FINTRAC (Canada’s financial intelligence unit) and at the Canadian Security Intelligence Service (CSIS), I’m intimately familiar with analytic best practices and common pitfalls of evaluating all kinds of intelligence and information. I’ve had the benefit of extensive training both in Canada and in the U.S. (including at the Sherman Kent School at the CIA), and have been training intelligence analysts (and intelligence consumers) for the better part of a decade. I write my own newsletter where I analyze current security and intelligence-related matters. (Link below.)
With all this experience guiding me, what I see in these intelligence reports concerns me.
Let’s begin with a quick primer on police (or criminal) intelligence. Criminal intelligence, as it’s often referred to, is intelligence produced in support of police forces. Many, if not most, major police forces in Canada have some sort of intelligence function. There are also intelligence functions housed within provincial forces, such as Criminal Intelligence Service Alberta, and of course the intelligence function (Provincial Operations Intelligence Bureau) within the Ontario Provincial Police. In Ottawa, that intelligence function is housed in the OPS Intelligence Directorate. While criminal intelligence is quite different from security intelligence (such as that produced by CSIS), common standards apply, and there are more commonalities than differences in how intelligence should be produced and evaluated.
Intelligence assessments should draw on a variety of sources. For criminal intelligence, this will largely be things like open source reporting (which can include Twitter feeds, public Facebook posts, and other information shared freely), as well as reports from sources developed for the police force or information from partner agencies. Good intelligence reports often refer to those sources either directly or obliquely: references to “open source research” or information from partner agencies or informants are common.
Assessments should use plain language, assess threats in terms of probabilities (in a way that the consumer can understand), avoid logical fallacies, and strive for unbiased analysis. It’s unrealistic to expect intelligence analysts to remove all trace of bias from their assessments: what’s more important is having a robust review process that challenges assumptions and biases.
A good example of an intelligence assessment can be found in the Public Order Emergency Commission documents. On January 20th, 2022, the OPP produced a strategic intelligence overview assessing the likely outcome of the convoy and protests. The document suggests that thousands of vehicles would participate in the convoy, and that there was “no exit strategy for departing Ottawa: the intent appears to be to remain in Ottawa until all COVID 19-related mandates and restrictions are lifted.” The author further notes that while the intent of the participants to remain on Parliament Hill until the mandates are lifted is likely unrealistic in the long term, “even a small number of tractor trailers parked” would “almost certainly” be disruptive in the short term. While this language doesn’t go so far as to suggest that an occupation would take place, it gets most of the way there. And to be fair: an intelligence analyst suggesting that a thus-far peaceful movement of people would occupy a city would not be seen as credible without multiple, independent, corroborating sources.
By comparison, the assessments produced by the Ottawa Police Service a few days later (25-29 January) stand in stark contrast. The January 29th report not only quotes Rex Murphy as a reputable source of information, it plagiarizes him. Take this excerpt from Murphy’s column published on January 25th in the National Post:
“Most protests are way past tired. Same core folks; same lame chants. Ugly signs. And what they are pleased to call tactics, same there. Block a city intersection close to downtown media outlets. Have two or three glue themselves to something. Go “dead” when the police have to drag you away. This was boring even in the ’60s.”
And compare it to this language from the Ottawa Police Service Intelligence Assessment from 29 January 2022:
“Most protests are repetitive. Same players, same chants. And what they are pleased to call tactics, same there. Block a city intersection close to downtown media outlets. Have two or three glue themselves to something, waiting for the same old supper—hour news shows and write-ups in hard-left handbills, blogs, and undernourished Twitter feeds. The vast majority has long ago stopped giving them any attention.”
(Late in the paragraph, the author mis-attributes more Rex Murphy content to National Post reporter Catherine Lévesque.)
There are so many problems here that I don’t really know where to begin. While intelligence assessments are not held to the same standard as academic articles, plagiarism is still a big deal. Citing an opinion columnist as a source of information is also not acceptable, particularly when it’s being used as a thinly veiled appeal to authority, meant to influence the consumer of the intelligence product, in this case, OPS leadership.
Again, quoting Murphy, the author also goes on to argue that politicians are hypocritical, and that this is one of the drivers of the protest. The inclusion of this quote demonstrates a clear political bias. Finally, the description of the convoy is at odds with the reality on the ground — the author described it as a ”grass roots” and “organic” movement. By January 29th, it was clear that there was significant organization behind the movement, as evidenced by, among other things, the hotel room bookings and logistical plan. Further information suggests that some support for the protest was inorganic. So not only is this assessment full of common problems, it’s also dead wrong.
The problems don’t stop there. On February 16th, 2022, the OPP assessed that “foreign adversaries may have attempted to leverage the freedom movement blockades and protests to protect or enhance their own strategic economic and political interests. Previous reporting noted controversial political figures in the U.S. voicing support for the blockades in Ottawa and Windsor, and foreign funding of the Ottawa blockade, much of it from the U.S.” It’s unclear where this information/assessment is drawn from, and is in stark contrast to the Director of CSIS’s statement on February 6th, 2022 that there were “no foreign actors at this point supporting or financing the convoy.”
Because the source of the OPP assessment isn’t clear, and the terms “actor” and “adversary” are undefined, it leads the reader to question the validity of both of these assessments.
It is within CSIS’s mandate to assess this type of threat, as they focus on espionage and foreign interference. OPP’s mandate on this issue is less clear, and it’s also not clear where they would have drawn information from to come to this conclusion. What we do know is that some money was being donated to the crowdfunding campaigns, and some foreign individuals, most notably a former U.S. president, were supportive of the convoy. However, Canadian officials do not commonly refer to the United States as an “adversary.”
There are also references in the intelligence reports to the financing of the convoy, but almost all refer solely to the crowdfunding campaigns. These campaigns were a major media preoccupation for most of the convoy, but the operational financing of the protest likely occurred primarily through cash donations and email money transfers. This is the type of intelligence that on-the-ground police forces would be expected to collect and analyze, given the importance of financing for sustaining the convoy.
This inside view of the intelligence function within two major police forces does not inspire confidence. While these documents and related testimony likely represent just a snapshot of the material being produced during the convoy, they raise serious questions about the professionalism of intelligence assessment within the OPP and OPS. While some lapses in professionalism can be excused during a crisis, some of the most concerning reports were written before the convoy had become a crisis in Ottawa. This is both a problem of analysis and of effective supervision. Accurate, reliable intelligence that clearly expresses uncertainty should inform operational plans for this type of event. Yet, as we’re learning, there appears to have been a breakdown in both the production of professional intelligence, and how, or if, the recipients consumed it. This goes some way towards explaining the lack of effective policing that Ottawa saw during the convoy.
The Line is Canada’s last, best hope for irreverent commentary. We reject bullshit. We love lively writing. Please consider supporting us by subscribing. Follow us on Twitter @the_lineca. Fight with us on Facebook. Pitch us something: firstname.lastname@example.org