Threading the Needle: Structural Reform & Canada's Intelligence-to-Evidence Dilemma

Becuase I am a patriot, and wasn't available to sell my country out today in Finland, I have written yet another paper on intelligence-to-evidence. This one tries to straddle the distance between "accessible for non-lawyers" and "technical enough for lawyers". I try hard in this paper to lay out what intelligence-to-evidence is, in my view. Most importantly, I propose what I call "moneyball" solutions to this problem, expanding and refining those I have suggested elsewhere and supplementing the solutions that have been raised by others (which as mostly complementary). I have spent a lot of time talking to people about this, and nothing I have heard has persuaded me things can't be done better. It is not quite a Gordian a knot as many seem to assume. On the other hand, there is no "home run" solution. A lot of players will need to come to the table with renewed determination. The paper is intended as a draft working paper. I welcome comments and feedback. It may be downloaded here.

The paper's abstract is as follows:

This article canvasses the “intelligence-to-evidence” dilemma in Canadian anti-terrorism. It reviews the concept of “evidence”, “intelligence” and “intelligence-to-evidence” (I2E). It points to the legal context in which I2E arises in Canada. Specifically, it examines Canadian rules around disclosure to the defence: the Stinchcombe and O’Connor standards and the related issues of Garofoli challenges. With a focus on CSIS/police relations, the article discusses the consequences of an unwieldy I2E system, using the device of a hypothetical terrorism investigation. It concludes disclosure risk for CSIS in an anti-terrorism investigation can be managed, in a manner that threads the needle between fair trials, legitimate confidentiality concerns and public safety. This management system rests on three legs:

  • Manage the relevance “tear-line” so that crimes less intrusive on CSIS information holdings are preferred over ones that are more intrusive. This strategy requires applying a prosecutorial insight to those investigations and planning their conduct to not prejudice trials. I bundle this concept within the category of “collecting to evidential standards” and “managing witnesses”.
  • Legislate standards to create certainty from the murk of evidence law. Here, two innovations stand out: legislate O’Connor style third-party status for CSIS where: CSIS’s investigation is a bona fide security intelligence investigation; CSIS and police do not have full, unmediated access to each other’s files; and, CSIS does not take an active role in the police investigation. But do not build this legislated third-party status around rigid barriers on information-sharing. Second, legislate ex parte, in camera procedures for Garofoli challenges of CSIS warrants in which special advocates are substituted for public defence counsel.
  • Manage the public safety risk by creating a fusion centre able to receive investigative information from all-of-government and fully apprised of the public safety risks associated with an ongoing investigation (or parallel investigations). Ensure it includes representatives from all the services with legal powers to respond to threats. The fusion centre would not itself be an investigative body, and would have O’Connor-style third-party status, something that would not require legislation but which might benefit from it.

A Law for New Seasons: Bill C-59 from the "Big Picture" Perspective of National Security Reform

Over the next few months, I will try to post thoughts on Bill C-59, the government’s massive national security overhaul package. Kent Roach and I have posted two quick assessments: an oped in Maclean’s and a longer piece at the Institute for Research on Public Policy website. I also provided reactions to the media in various placed, including on The House here and Power & Politics here.

(We always worry about pushing out analyses of such complex legislation on an insta-response basis, and qualify what we say with an open invitation to point out errors and omissions. Like most people, I learn best when I write, reflect, discuss, revise.)

In this space, I want to meditate on two issues emerging in the discussion.  First, that C-59 is about correcting C-51, creating the impression (fanned by some politicians) that C-59 rolls back security powers.  Second, the resource and burden issue.


C-59: Reforming without subtracting

A word of warning: Kent and I always took the view that C-51 was dealing (mostly) with real problems, but the solutions were so festooned with their own shortcomings that they didn’t solve the problems, but did create a host of new ones.  (The speech crime was the exception: it was always a solution in search of an invented problem).

I won’t repeat our analysis here. (We set out our conclusions in the 600 pages of False Security.)

This is by way of saying: I was never in the “repeal and return to the prior status quo” camp.  Because that status quo meant returning to a security law system that creaked with age and inadequacy.


Fixing the Problematic Parts

If we expect the state to protect us, we need to give it tools.  In part, this is because I believe the civil liberties implications of the day after a security failure are always worse than the civil liberties challenges raised during a calm, premeditated effort to give security services reasonable tools to prevent that incident. (After some bomb goes off, everyone assumes that it stems from a failure of law, and that we need fewer rights.  Usually, the reason is more complex: sometimes it is operational. And sometimes it is simply a manifestation of the old IRA slogan about security services needing to successful all the time, and terrorists only once. Those impossible odds mean something will always happen.  And so you need social resilience, not a stampede to turn your society into North Korea.)

When we do security law and policy reform properly, the questions always are: which tools, are they proportional, and are they compatible with a liberal democracy (and avoid the “burning villages to save them” problem).  And for anti-terror tools, focused on a threat embedded in a civil population, “overclocking” on your tools may precipitate the very threat you intend to stave-off.  (Witness the nonsense discussion on the margins of the internet last month, after the UK incidents, raising the prospect of mass internment. Setting aside the egregious rights violations, this is out past Pluto in terms of security: people need to spend more time examining the blowback consequences of mass internment. It’s a pretty good way to turn a difficult security environment into a 100-year war.)

C-59 is about correcting C-51’s (unnecessary, probably-never-actually-wanted-by-the-security-services) excess, and I think it generally does a good job here (with the real remaining concern being the light-touch amendments to the Security of Canada Information Sharing Act, renamed and tempered, but still vast).  For instance, I doubt CSIS ever wanted to be in the detention and rendition business – so why create a law that made that a legal possibility?

For more on these fixes, see our IRPP piece, linked above.


Dealing with (Some) of the Puzzling Omissions

But C-59 is also about giving new powers to the security services.  Four things stand out. First, by placing CSIS threat disruption powers on a more plausible (although surely still novel) constitutional foundation, it makes those actually usable.  (CSIS has clearly not been prepared to use threat reduction that raised constitutional issues under C-51, probably appreciating that the C-51 formula was an invitation for controversy in the courts and out).

Not everyone will think we’ve hit the sweet spot.  See Michael Nesbitt’s excellent analysis.  But we are way closer than with C-51 – with that bill’s formula, it was really hard to find a constitutional lawyer (not taking instruction from government) who thought we were even in the ballmark.  And whatever we might conclude about how carefully drafted some of the new “closed list” powers are, I simply cannot think of any other way to square the constitution with some of the more potent threat reduction powers I believe are quite properly on the table (e.g., interfering with a suspected terrorist’s communications). 

Second, I had not quite appreciated the extent to which CSIS was on the cusp on being paralyzed by its old law. For one thing, the limitations in its Act on retaining information – most dramatically illustrated by the Fall 2016 Federal Court decision on the CSIS ODAC initiative (see a write up here) -- must be deeply constrictive of CSIS deploying big data analytics – or even basic Boolean searching – on information…that they cannot have.  There are, of course, all sorts of privacy concerns – which is where close study is required of both the revamped collection and retention rules and their checks and balances. But at some point, one must concede that if you are to have an intelligence service, it needs to be able to collect, retain and analyze intelligence. (Privacy protections have always has been about checks and balances, from their inception in the early common law through to the present day).

For another thing, I had not quite appreciated how dramatically changes in the concept of Crown immunity – and doubts about its application to CSIS operations – must be crimping operations. It may not be too much of an exaggeration to say, with all the new terrorism crimes introduced since 2001, that every CSIS officer and source covertly infiltrating a terror plot is at risk of prosecution. CSIS recruiting must go something like: “Thank you for your service. As soon as you participate with this group, you are a criminal. But we’ll put in a good word with the prosecutor – assuming we’re prepared to cough up our secret op details. Hopefully things will be ok.”  The response must be something like: “No way.” Or: “Ok, give me $8 million.”

I have no way to know if the problem is that dramatic. But legally, it may be. And if so, together the limit on CSIS data retention and the crimp on human source immunity is pretty serious.  It might mean that Canada risks not having a real security intelligence service. 

Unless you think the world is much safer than I think it is, that is an unhappy prospect.  It is actually astonishing that this was not fixed a long time ago.  So the issue is: are you happy with the C-59 solutions?  And in responding, the first thing I look for it: checks and balances.  So far as I work through the details, I think they measure up quite well – indeed, potentially very well, measured against international comparisons.

Third, the Communications Security Establishment has been burdened with too little law, and too narrow a mandate. On law, we have known since it was first given statutory footing in 2001 that the issue of Canadian-origin information intercepts raised constitutional issues. People have been writing about it for a long time. But it was one of those questions that were, um, academic, until Snowden.  After that, it became a matter of public controversy, and litigation.  Fixing this was never that hard – and I am very pleased to see that C-59 proposes what I think is a viable and even elegant approach.  (Although there is a bug in the drafting, I think, that may leave the problem unfixed.  That requires more explaining, and I will blog on that soon.)

On mandate, CSE’s cybersecurity mandate basically reaches: get into a defensive crouch, protecting your core and vital organs, while the North Koreans, Russians, Chinese, hackers etc pummel you. But the world has changed since 2001. The new “active” and “defensive” cyber operations powers, and the broadening of the traditional cybersecurity mandate make a lot of sense.  Again, that assumes you agree that the world presents real security challenges that require viable responses.  If you do, then the remaining question is: are you happy with the checks and balances?

Four, tempering C-51, and adding a whole host of checks and balances is actually security-affirming.  In a democracy, the activities of the security service depend on consent and cooperation. Security powers that validate a lot of conspiracy theories erode that “social license”. 

C-51 took a lot of conspiracy theories from “plausible only if you assume everyone is a legal rogue and ethically unhinged”, to “legally possible, even if still doubtful in practice because the people involved are not venal and unethical”.  (Our various commissions of inquiry criticized the services, but did not suggest wrongdoing was ill-intentioned – with the exception of the poisonous leaks someone released to smear Maher Arar.)  But as anyone who has spent more than 5 minutes working in a human institution knows, people and institutions make mistakes – sometimes enormous mistakes. Silos, group think, cognitive bias, habit, incompetence, laziness, inattention, petty jealousies.  All the vices of the human form. Law, guidelines, protocols, oversight, review and checks and balances are what we use to minimize the prospect of systems failing, especially where the consequences of failure are significant.

C-59 puts the law back in play as a code of conduct, in a way that C-51 relaxed too much.  I think that is important. One might expect this of a law professor. But I cannot really think of any examples of where “the gloves are coming off” approach to security law and policy in a democracy has worked well.  It tends to produce outcomes that some future political leader needs to apologize for, after a commission of inquiry, disastrous court losses, public acrimony and a general erosion of public trust.


Administrative Burden: Better than the alternative

And that brings me to the administrative burden conversation.  C-59 will amp up the checks and balances in national security law considerably.  So considerably that Canada may well be back to where it was in 1984: a leader in this area.  Predictably, there will be anxiety that this will shackle responses, drain resources and infuse lawyers and overseers into the nitty-gritty of security work.  C-59 is, in some respects, the judicialization of intelligence that former CSIS director Jim Judd disliked a decade or so ago.

It is also consistent with developments in other Five Eye states, and even the French have new law in the area of intelligence. (The French, famously, have had little).  It is inevitable: as soon as you focus on security threat emanating from your civil society, intelligence starts to drift closer to police work.  And so, it needs to abide by at least some of those standards that guard police work (many of which echo those announced by Robert Peel in establishing the first police force in the 19th century).

The new systems could be impossibly bureaucratic.  Or they could be elegant and effective.  Much will turn on design, resourcing, staffing. Inattention on these issues will produce disasters: impairing necessary security conduct, done by cautious, risk-adverse services; and/or overpromising on accountability without delivering.

But I will say this: they are the quid pro quo to accomplishing that security expansion noted in the first four points of this blog.  C-59 should establish a regularized, professionalized system of checks and balances.  And whatever burden they impose, that would be dwarfed by the burden imposed by a creaky, inadequately constructed security system that lurches from scandal to commission of inquiry to judicial slap-down; with powers uncertain, planning interrupted by public controversy and all your staff-time devoted to appeasing a disgruntled Parliament, judge or commissioner.  In other words: the 2000s. I don’t know anyone (in any walk of life) that wants to go back to the scandal/response system of national security policy-making. That would be bad for security and rights.



In sum, C-59 is probably in, or near, the Goldilocks space between too hot and too cold. Which is not to say it is perfect, or that it fixes everything, or will please everyone.  For instance, the SCISA is not falling. (The author chuckles to himself.) And it isn’t to say we won’t suddenly discover a new concern in the 150 page document.

But based on about 5 readings of the full text and some deep dives on some of the more complex parts, it appears to be more carefully crafted than anything we’ve seen in this area in a long time – probably the 1988 Emergencies Act, and before that the 1984 CSIS Act.  That’s a good place to be, going into the parliamentary process.

The Unraveling of Canada's Legal Justification for Force in Syria? The Trouble with "Unwilling and Unable"

Events yesterday in Iraq seem likely to complicate the legal basis for Canada’s participation in the anti-Daesh coalition. As summarized by Jennifer Daskal over at Just Security:

U.S.-led forces hit a convoy carrying pro-Syrian government forces advancing inside a deconfliction zone inside Syria. The convoy was reportedly traveling toward the al-Tanf military base used by U.S. coalition forces to train anti-ISIS fighters. U.S. and coalition officials assert that the Russians “apparently” attempted to dissuade the convoy from entering the area, that they first fired warning shots and deployed two US aircraft as a show of force, and only struck the convoy after it failed to heed the warning, as a means of protecting U.S. and coalition forces.

Jennifer Daskal also reports US official statements that the “strike was a proportionate response done for purposes of force protection—an act of self-defense in an effort to protect U.S. forces.” She observes: “This is, on its face, quite plausible. And, if accurate, lawful as a matter of both international and domestic law.”

The facts remain uncertain. But I shall assume for the sake of this analysis that the convoy comprised Syrian government controlled militia or military.  And I shall assume that Syria is now actually interested in fighting Daesh (whether or not this particular convoy was tied to that effort).

If so, I am less persuaded that the strike was lawful as a matter of international law. Or more accurately, I think it undercuts that entire legal basis for the presence of Coalition forces in Syria.

The Basics

Leah Sherriff and I examine the international law of using force against a non-state actor in the territory of another state in our article on targeted killing. Interested readers will find there the detailed footnotes for the sources and principles I cite below.

To summarize, a state may not use force against another state, or on that other state’s territory without the territorial state’s consent. There are two exceptional circumstances.

The first is UN Security Council authorization. That does not truly exist for the conflict against Daesh, notwithstanding the S/Res/2249 (2015). I agree with those analyses that see this resolution as an intentionally ambiguous resolution that falls short of the language associated with Security Council authorizations on use of force.

The second exception is self-defence against an armed attack, including collective self-defence done in conjunction with a state that has suffered this attack. Self-defence against Daesh is, in fact, the legal basis for Canada’s use of force in Syria. In late October 2014, when Canada joined the American-led coalition against Daesh in Iraq, it reported to the UN Security Council that Canada was invoking individual and collective self defence under article 51 of the UN Charter, explaining “States must be able to act in self-defence when the Government of the State where a threat is located is unwilling or unable to prevent attacks emanating from its territory”.[1]  On the specifics on the Daesh case, Canada is asserting collective self-defence with Iraq against Daesh, with self-defence extended to Syria (where Daesh also operates).

But Syria itself has not consented to the use of force on its territory (and even if it had, tacitly or otherwise, it could revoke that consent). And it itself is not the originator of the armed attack. Hence we face the common post-9/11 conundrum:  Directing military force against a non-state actor almost always requires use of force on the territory of another state, including some that do not consent.

There is serious incongruity in the idea that a non-state actor may use violence whose scope and effect rises to the level of armed attack, and then hide behind the territorial sovereignty of a state that, however unwillingly or unwittingly, serves as host. And as a practical matter, some states – including, especially, the United States – have rejected a formalistic approach that would allow this shelter. These states have instead pursued a doctrine of “unwilling or unable”. In Ashley Deeks’s words:

The “unwilling or unable” test requires a victim state to ascertain whether the territorial state is willing and able to address the threat posed by the nonstate group before using force in the territorial state’s territory without consent. If the territorial state is willing and able, the victim state may not use force in the territorial state, and the territorial state is expected to take the appropriate steps against the nonstate group. If the territorial state is unwilling or unable to take those steps, however, it is lawful for the victim state to use that level of force that is necessary (and proportional) to suppress the threat that the nonstate group poses.[2]

Whether “unwilling or unable” is truly part of customary international law is hotly debated. However, the international community has demonstrated more receptivity – although certainly not enthusiasm – for the doctrine since 9/11 and especially since 2014. The United States and Turkey have reaffirmed the existence of an unwilling or unable doctrine to justify self-defence against Daesh in Syria, and they are not alone. They have been joined by Australia, Belgium, Canada, Germany, and, implicitly at least, Denmark, Norway and the United Kingdom. In response to the US notification to the United Nations, then Secretary General Ban Ki-Moon reportedly stated: “I also note that the strikes took place in areas no longer under the effective control of that [the Syrian] government.” Other states, such as Jordan, Bahrain, Qatar and the United Arab Emirates have participated in air strikes in Syria without articulating legal justifications, leading at least one commentator to posit that they are “relying on the same legal theory as the United States and UK.”

Still other states, such as France, have embarked on a similar course under the shelter of a UN Security Council resolution 2249 (2015) that is (as suggested) creatively ambiguous about the legal authority for directing force at Daesh in Syria.

Collectively, this constitutes considerable state practice and – in the case of the United States, Turkey, Canada, Australia, Belgium and Germany – emphatic opinio juris supportive of the “unable or unwilling” doctrine as a basis for invoking UN Charter Article 51 self-defence.


Yesterday’s Events

State practice of what is less clear. It is one thing to intrude on a state’s territory to exercise self-defence strictly limited to the attacking non-state actor. It is quite another to stray beyond this terrorist-specific targeting and direct force against the territorial state’s own assets or infrastructure.  That is why the US missile strike against the Syrian airbase after the chemical weapons attacks in April cannot be justified on the same self-defence theory supporting use of force against Daesh. (See my own views on that incident here and here.)

Yesterday’s clash with Syrian forces raises new legal doubts. They boil down to this: how can the “unwilling and unable” justification for force in Syria be maintained once the Syrian government asserts itself on its own territory against Daesh? “Unwilling and unable” begins to look like transparent pretext. Moreover, it beggars imagination that it could be used as legal justification for a Coalition forces strike staving off the very government said to be unwilling and unable in the first place.

To conclude that Coalition use of force is lawful in these circumstances you would need to accept the following:

1. Coalition forces have been lawfully employing force against Daesh because Syria was unwilling and unable.

2. The unwilling and unable justification persists even after the territorial state demonstrates it is no longer unwilling and unable.

3. And more than that, the Coalition forces may lawfully use military force to suppress efforts by the once unwilling and unable state aimed at reasserting control over its territory.

The “force protection” argument for yesterday’s Coalition strike only gets you so far. “Force protection” is not lawful self-defence if the presence of the Coalition now itself constitutes an armed attack against Syria, because the original “unwilling and unable” basis for the Coalition presence has evaporated. (Imagine this outside of the Syria context: If one state invades another, and parks itself on the second state’s territory, an effort by the invaded state to resist the invasion is not an attack against which the invading state may claim self-defence.)


I acknowledge that these legal niceties will determine little of what happens in Syria. And I imagine others will counter with their own legal reasoning. That might be something like this: once lawfully engaged in self-defence under an unwilling and unable doctrine, the Coalition may use such proportionate force as is necessary to end the armed attack by Daesh, including against the Syrian government if it gets in the way. Put another way, the unwilling and unable justification is a one-way-ratchet that cannot be notched back because the delinquent territorial state has a change of heart or capacity.

If that is the legal argument, it amounts to: once a failed state whose sovereignty has been suspended, always a failed state with a suspended sovereignty. That does not seem the sort of position likely to increase international stability or avoid capricious misuse in the future.

At the very least, events on the ground in Syria demonstrate clearly the risk of predicating use of armed force on a fuzzy doctrine like “unwilling and unable”.  States do not always remain unwilling.  And to use force to ensure they remain unable would be the height of absurdity.

In sum: since Canada is participating in the Coalition, it will be drawn along into a difficult legal quagmire if the anti-Daesh campaign now bleeds into an armed conflict with the Syrian government.


[1] Letter dated 31 March 2015 from the Deputy Permanent Representative of Canada to the United Nations addressed to the President of the Security Council, UN Doc S/2015/221 (31 March 2015)

[2] Ashley S. Deeks, “‘Unwilling or Unable’: Toward a Normative Framework for Extraterritorial Self-Defense,” (2012) 52(3) Virginia Journal of International Law 483 at 487-88.

The Citizenship Revocation Zombie: The Endless Saga of Bill C-6

Bill C-6 constitutes the Liberal government’s efforts to reverse the Tory changes to the Citizenship Act.  Among other things, it restores fraud in acquisition as the sole basis of revoking Canadian nationality.  It repeals the Tory provisions permitting revocation for those dual nationals who commit terrorism offences.

Why this is a good thing

For reasons discussed here and here and here and here, I do not agree with revocation for dual national terrorists.  I understand why it superficially seems like a good idea.  I also understand the impulse that demands denunciation and exclusion.

But revocation does little to make us more secure, while at the same time throwing other values under the bus.

As I discuss here, the Tory regime suffers from obvious constitutional infirmities. And as I discuss here, it is a measure motivated by a sense of grievance, not a cold-eyed assessment of its usefulness as a counter-terrorism tool.  Indeed, in all the parliamentary proceedings, there has been no cogent evidence of which I am aware setting out how revocation makes us safer.  The security presumption amounts to a naïve “Not in My Backyard” supposition.  Pick at the logic of that assumption, and all you are left with is that it just feels good to denounce terrorists.

Which of course is important – and it is exactly what we do with the criminal law, a much more robust and meaningful tool. If we have a problem with anti-terrorism in Canada, it is that we resort too often to administrative measures – which would include revocation – and not enough to prosecutions. But that is another discussion.

Bill C-6’s tardy trajectory

Reverting to C-6, that bill has taken a dog’s age in an apparently very dilatory parliamentary process. Now it is at third reading in the Senate, where under Senate rules it can be further amended.  And proposed amendments there are, including on additional due process standards for the classic revocation for fraud.  That is a good thing, assessed by my lawyerly eye.

But there is also an amendment proposed by Conservative Senator Lang that would restore a species of revocation for a terrorism offence.  The cross-referencing in all these changes is difficult to follow, but as I understand the language: the amendment would permit a criminal court judge trying a terrorism offence to make a declaration as part of any sentencing.  This declaration would have the effect of stripping nationality, but only (in practical effect) for dual nationals.  (The amendment replicates the existing standard guarding against revocation that would violate Canada’s international obligations and produce statelessness).

Meanwhile, the defendant would (still) have the onus of proving that they are not a dual national to avoid this prospect.

Why this is a bad thing

I will give Senator Lang credit for a system more honest that the existing scheme.  The existing scheme introduced by the Harper government was all dressed up as an administrative proceeding, but was clearly punitive.  At least Senator Lang’s proposal embeds revocation in an unequivocal penal setting.  (Although, not having researched it, I cannot think of another instance where a sentencing element for a crime is external to the Criminal Code).

At the very least, the close link to a Canadian criminal proceeding obviates concern about revocation being used in response to some doubtful foreign prosecution, or used by a political minister (the current system) rather than a judge.

But, it will surprise no one to learn that I still think it is a bad idea, on stilts.  First, it does precisely nothing to address my concerns about the security downside of revocation, discussed here, or the practical incongruity of where it would be available (in practice, for terrorist plotters but not actual terroristic killers never charged with terrorism offences, for the sort of reasons discussed here).

But second, the Charter section 11 protections in penal matters clearly now attach.  And there will be no doubt about the applicability of section 7 (fundamental justice) or section 12 (cruel and unusual treatment).  As discussed here, in the United States, the US Supreme Court held long ago that punitive revocation was cruel and unusual – it does not take much imagination to contemplate Canada’s Supreme Court following this reasoning.

As for onus of proof, the jurisprudence on onus of proof under section 11(d) and section 7 could be clearer, but placing the onus on a person to prove that they are not in the category of persons who may be punished by revocation (a form of aggravated punishment) strikes me as, well, supremely problematic.

Take this passage from the Supreme Court's decision in Pearson, directly related to onus of proof in sentencing: "it is clear law that where the Crown advances aggravating facts in sentencing which are contested, the Crown must establish those facts beyond reasonable doubt ...While the presumption of innocence as specifically articulated in s. 11(d) may not cover the question of the standard of proof of contested aggravating facts at sentencing, the broader substantive principle in s. 7 almost certainly would." (The Court cited with approval the earlier case of Gardiner and the following passage from an article by JA Olah: "... because the sentencing process poses the ultimate jeopardy to an individual . . . in the criminal process, it is just and reasonable that he be granted the protection of the reasonable doubt rule at this vital juncture of the process.")

A majority of the Court again cited these authorities more recently in R v. DB to conclude: "the Crown is obliged to prove, beyond a reasonable doubt, any aggravating factors in sentencing on which it relies" (at para 78).

It is also notable that in the US, revocations (where available at all -- basically, fraud) attract robust due process protections to a criminal standard of proof standard.

A Dual National Canadian is Not-Quite Canadian Enough

All that should be enough to deter any reasonable parliamentarian from venturing down this impossible path. But that’s all nothing as compared to the nuclear issue: The Lang amendment creates a separate penalty regime for dual national and for single-nationality Canadians. For exactly the same conduct and for exactly the same offence, the dual national is exposed to the considerable consequences of revocation, on top of criminal sentence. The single-national is not. 

(Perhaps you could argue that this additional peril might be mitigated by a sentencing judge, for instance, tempering the prison sentence in favour of revocation.  But there is no obligation that they do so.  And the Supreme Court in Pham suggested that there are limits in terms of whether immigration consequences can figure into criminal sentencing.  Nor, as I have suggested, do I think preferring revocation over incarceration is a great idea from a security perspective. Nor do I think you can quantify the relative implications of revocation – leading ultimately to permanent banishment – as compared to a sentence of years. Citizenship, after all, has been properly called the “right to have rights” and the consequences stemming from its revocation may be both more serious and enduring than a criminal sentence. In this last respect, this story of what happened to two former Britons is sobering).

So bottom line: as with the Harper government scheme, we still have a massive section 15 equality problem. Dual nationals would not be equal before and under the law, based entirely on the fact that they have (and perhaps are often stuck with) a dual nationality – or because they can’t prove that they are not dual nationals.  I discuss the niceties of section 15 issues at length here.

The takeaway

Generally, lawyers and law professors are reluctant to speak definitively about how courts might decide questions of constitutional law.  That is because prediction in this area is more idiosyncratic art than true science, and it is embarrassing not to be omniscient. 

But I am going to go out on a limb here (and yes, if ever proved wrong, I will leave this post up): zombies shall walk the earth and winged pigs flock to the sky before a system as proposed in this amendment satisfies the Charter.

Revoking Citizenship for Terrorism Offences: Testimony to the Senate studying Bill C-6

The following are my speaking notes from my testimony before the Senate on bill C-6 and terrorism citizenship stripping:

Speaking Notes

Bill C-6

Standing Senate Committee on Social Affairs, Science and Technology

Feb 16, 2017

Good morning.  I’d like to express my thanks to the committee for inviting me to appear on bill C-6.  In my statement, I shall focus exclusively on the issue of citizenship revocation for terrorism.  These were powers enacted in 2014, and would be repealed by C-6. I support that repeal.

I will focus on two key points.  First, terrorist citizenship stripping provisions are different from other rules, such as revocation for fraud.  Second, they are problematic from a security perspective.

On the first point:  those defending the 2014 revocation law sometimes claim that war criminals have citizenship stripped, and that the 2014 law is, therefore, no different.

This analogy is misleading.  Nazi collaborators have lost citizenship -- because they lied about their conduct at the time they became Canadians.  The revocation is for fraud, not war crimes.  No Canadian has lost their citizenship for a war crime committed while a Canadian.

The 2014 law did something different: it used citizenship stripping as a supplemental punishment to penalize dual nationality Canadians for things done while a Canadian. In so doing, it effectively converted the many Canadians with dual citizenship into probationary Canadians.

Let me turn to why terrorism citizenship revocation is poor security policy.

First, there simply is no empirical basis to conclude that dual nationals pose a more serious security risk than single national Canadians.

Singling out dual nationals for the special risk of revocation is not, therefore, rationally connected to a security objective, a lethal shortcoming both from a security and constitutional perspective.

Second, Canada has worked arduously and deployed tools like no-fly lists, passport revocation, peace bonds and outright criminal prosecution to stop Canadians from traveling for the purpose of participating in terrorist activity. But the objective of citizenship stripping is, ultimately, to deport these people. If truly dangerous people are deported, the net effect may be to speed foreign fighters on their way.

Again, this raises questions of rationality.

But third, an irony of the structure of terrorism law in Canada is that citizenship revocation will not be imposed on the most dangerous of people. The manner in which terrorism offences are designed in the Criminal Code means they are most likely to be applied to plotters, not killers. Once someone becomes a killer, police and prosecutors are much more likely to charge the person with murder, not terrorism offences.  That is the reason why Justin Bourque, Richard Bains and most recently Alexandre Bissonette (at least so far) were not charged with terrorism offences.

Not a single person currently in prison in Canada for a terrorism offence enacted after 9/11 has actually committed an act of violence.  They were pre-empted.  And so as pre-empted plotters, they now face the risk of revocation, assuming they are dual nationals.  Killers would not.

Four, revocation would consume huge resources, best deployed for different objectives.  Setting aside the inevitable constitutional challenges to revocation itself, subsequent efforts to remove these former Canadians would be an arduous undertaking.  The risk of maltreatment in foreign countries may be high, placing Canada in the invidious position of trying to remove people to torture. 

We have been down this path before, with immigration security certificates.  At the end of the last decade, the government was spending multiple millions of dollars per year per security certificate, often in a vain attempt to defend the measures.  This was not money well spent.

Indeed, more money by a large margin was spent trying to remove people, than Canada has spent in total so far on counter-violent extremism and terrorist disengagement strategies.  While these programs are themselves unproven and untested, they are an area in which Canada clearly needs to improve its practices.  At the end of the day, even for terrorist convicts, rehabilitation, not displacement to other countries, is the most plausible security strategy.

That is because, fifth, displaced dangerous people remain dangerous – and indeed potentially more dangerous than if they remained in Canada. They will likely be impossible for Canadian security services to monitor, they may be free of strictures such as peace bonds, and history suggests that dangerous people outside of Canada can continue to do harm to Canadians and Canadian interests. 

Nor, sixth, would other countries readily welcome the arrival of a person radicalized to violence in Canada – and almost all of those serving terrorism sentences are made-in-Canada terrorists.

Canada’s international anti-terrorism objectives are poorly served if we become an exporter of instability, rather than invest in the hard work of terrorist disengagement.

In sum, it is my view that little recommends terrorism citizenship stripping, and there much that condemns it.

Thank you for your attention.  I welcome your questions.

Streamlined Anti-terror Investigations: Quick Notes on the UK Experience

In the spirit of constructive input, I have prepared two (admittedly lengthy) blog entries on how to make anti-terror investigations more seamless than they are at present in Canada. In this blog, I articulate the key challenges and then look at how the UK has addressed these matters. In a subsequent blog, I offer “lessons learned” for Canada and propose a path forward.

CSIS immunity from criminal culpability for acts done in foreign fighter investigations: Observations on SIRC report

In its annual report on CSIS activities released last week, the Security Intelligence Review Committee (SIRC) included some cryptic comments about CSIS overseas foreign terrorist fighter investigations.  Specifically, it stated: “CSIS should ensure its employees fully understand the extent to which certain activities present legal risks. To this end, SIRC recommended that CSIS seek legal clarification on whether CSIS employees and CSIS human sources are afforded protection under the Common Law rule of Crown Immunity in regards to the terrorism-related offences of the Criminal Code of Canada.”

It is impossible to draw definitive conclusions about what SIRC is talking about here, but a few inferences seem reasonable:

  • This comment seems to be directed at CSIS security intelligence investigations (that is, intelligence investigations) and not CSIS’s new (post bill C-51) “threat reduction” powers (discussed in my post immediately prior to this one).
  • In Bill C-44, CSIS was given a clearer legislative basis to conduct overseas security intelligence investigations, and one assumes that is exactly what it is doing.
  • And based on SIRC’s statement, CSIS has either employees or sources who are collecting information overseas on terrorism-related matters, but doing so in a manner that may expose them to Canadian Criminal Code culpability.
  • At a guess, there are two reasons why these employees/sources may have legal liability. 
  1. First, Canada’s terrorism offences are extraterritorial – they extend to conduct by Canadian citizens overseas that if done in Canada, would be a crime. The most likely candidate is “participation with a terrorist group”.  A CSIS source in a terrorist group is almost certainly someone who, along the way, has participated in a terrorist group.  Indeed, these kind of issues came up in the Toronto 18 case. And employees may also commit such crimes, while acting undercover with a terrorist group.
  2. Second, the entire Criminal Code applies to federal employees who commit an act that is crime in both Canada and the place overseas where it happens (s. 3.73(4)).  So there is a whole lot (conceivably most of the Criminal Code) that a CSIS employee might do that triggers potential criminal exposure, while they act undercover.
  • CSIS security intelligence operations include no express statutory carve-out for this criminal culpability. (CSIS’s new threat reduction powers post-C-51 do, but only if blessed by a secret federal court warrant). (I don't think s.20 of the CSIS Act provides enough cover, since it only extends peace officer protections, which I don't believe would permit illegal conduct, absent proper application of s.25.1 of the Criminal Code. And at any rate, that s.20 only applies to CSIS employees, not sources or agents. I take it SIRC has the same concerns, or it wouldn't have raises this issue at all).
  • And so if CSIS employees or sources are to be immune from criminal exposure for their conduct in security intelligence investigations, then it will be because of classic crown immunity rules.  Now, we know that the police do not possess crown immunity for illegal conduct, done as part of their peace officer duties.  That is precisely why Parliament added the controversial s.25.1, carving out such immunity, to the Criminal Code. (Shockingly, the RCMP appears not to have employed properly the s.25.1 carve out in the Nuttall case, where officers were found to be engineering a terrorism plot. They were, therefore, exposed to potential culpability).
  • What has not been decided, as best I know, is whether CSIS enjoys crown immunity in their operations. I think they are likely in a different position than RCMP: police enjoy “police independence” and thus are legally distinguishable from the Crown in their conduct of criminal investigations. CSIS does not enjoy this independence, and so can be conflated with the Crown. (Saying more than this is a research project for which I currently have no time.)
  • But even assuming CSIS enjoys such crown immunity, it seems like the sort of thing that would extend to employees.  At a guess, it is much less likely it extends also to sources.  The prospect that sources are also covered becomes more likely (I suspect) if they are more than information sources: crown immunity seems more likely if they are actual agents acting at the direction of CSIS.  An agent is just that: someone who is not an employee but who is acting under the command and control of the government.
  • But if CSIS is directing agents (or undercover employees) to do things that break Canadian anti-terrorism law, that looks like the kind of thing that probably should be done as threat reduction, and not security intelligence. And so if it is done without the federal court warrant required for those threat reduction activities that break Canadian law, we will have an issue of whether CSIS acted legally under its legislation. Indeed, the very failure to obtain a warrant here seems likely to be precisely the excess that would strip the crown immunity from the agent/employee anyway, because it does not comply with statutory immunity provision. (See para. 37 et seq).
  • And on top of that: if a person is a CSIS agent/employee, and if they are in fact committing a criminal offence -- perhaps in some sort of agent provocateur role – then even if they were immune under crown immunity, there are legal troubles. The fact of the illegality becomes critically material if criminal charges are later brought against one of the CSIS targets.  Specifically, we could have failed prosecutions against such targets when, a la Nuttall, a court concludes there has been entrapment or some other abuse of process associated with the CSIS conduct.

All of this to say that SIRC’s cryptic statement in its recent report raises a host of immediate legal issues – and we shall never know how they are resolved even though they are elemental to the rule of law, and possibly prejudicial to our ability to prosecute foreign fighters if the legal advice and subsequent CSIS conduct is amiss.

Canada's Security & Intelligence Community after 9/11: Key Challenges and Conundrums

I have posted my latest paper here. It (briefly) canvasses a range of outstanding issues that have arisen in the organization of Canadian security and intelligence since 9/11. The abstract reads as follows:

The Canadian security & intelligence community’s historical development and scope reflect the country’s relatively favourable geopolitical circumstances. Since 9/11, anti-terrorism has been the country’s clear security priority, possibly to the point of ignoring other critical issues. Because responses to terrorism involve both criminal law and intelligence-led preemptive activities, Canada’s chief police and intelligence agencies now overlap in their investigations to a considerable degree, creating conundrums for both operations and accountability. This article traces the impact of these developments on the Canadian management of national security, and the institutional design of Canada’s S&I community and accountability mechanisms. It concludes with a series of questions Canadian policy makers must ponder in deciding how best to address Canada’s operational and accountability national security challenges.

Yesterday's Law: Terrorist Group Listing in Canada

Kent Roach and I have posted our latest paper here. This paper reviews Canada's terrorism group listing laws. The abstract reads as follows:

Canada’s approach to proscription differs from that of other Westminster democracies. With a different constitutional tradition, Canada does not ban organizations; instead it penalizes certain forms of conduct, above mere membership, with terrorist groups. “Terrorist groups” include entities listed proactively by the executive, but also entities that meet a functional definition in Canadian criminal law. In practice, the latter, functionally-defined terrorist groups have figured in most terrorism prosecutions – few cases have involved listed groups. With the new focus on Daesh (a listed group), that may begin to change. However, executive listing raises unresolved constitutional doubts in Canada, prompting concerns that reliance on proscription may be more trouble than it is worth. In many respects, therefore, terrorist group listing is yesterday’s law, sparked by the events on 9/11, but of marginal utility thereafter. There may be reasons of administrative expediency to preserve listing, but the tool is most doubtful when used as a precursor to criminal prosecutions.

Completed Terrorism Prosecutions in Canada: Updated Table

I keep an informal, running table of completed terrorism prosecutions under the post-9/11 terrorism offences for my own reference purposes. I post the most recent iteration of that table on this blog, showing 26 completed prosecutions resulting in guilty verdicts as of last week (although in two of those cases, there was ultimately no conviction because the cases were tossed because of police entrapment). The table may be downloaded here. (If I have missed anything, I'd be grateful for an email).

Killing Citizens: Core Legal Dilemmas in Targeted Killing of Cdn Foreign Terrorist Fighters

My article with Capt. (Ret.) Leah West Sherriff on Canadian and international law and targeted killing is now accepted and forthcoming, Canadian Yearbook of International Law. We have posted the current version here. The abstract is as follows:

For the first time since the introduction of the Canadian Charter of Rights and Freedoms, Canada is an armed conflict with an insurgency that has actively recruited Canadians and directed them to use or promote violence against Canada. In the result, the Canadian government may ask its soldiers to target and kill fellow Canadians, or to assist allies in doing so. This situation raises a host of novel legal issues, including the question of “targeted killing” confronted by the United Kingdom in 2015 when it directed military force against several Britons believed to plotting a terrorist attack. That incident sparked a report from the British Parliament highlighting legal dilemmas. This article does the same for Canada by focusing on the legal implications surrounding a targeted killing by the Canadian government of a Canadian citizen. It examines how a Canadian policy of targeted killing would oblige Canada to make choices on many weighty legal matters. First, it discusses the Canadian public law rules that apply when the Canadian Armed Forces deploy in armed conflicts overseas. It then analyzes the international laws governing military force, scrutinized from the perspective of use of force (jus ad bellum) and the law of armed conflict (jus in bello). It also examines an alternative body of international law: that governing peacetime uses of lethal force. The article concludes by weaving together these areas of law into a single set of legal questions that would necessarily need to be addressed prior to a targeted killing of a Canadian.

Learning Lessons from the Driver Terror Matter

Between them, the Nuttall/Korody case and the Driver take-down raise tremendous questions about Canadian anti-terror strategies. And there is a tremendous temptation to rush to judgment, which can be perilous. On the other hand, there should be judgment and it should be expeditious.

A. Are we a learning society?

Canada is bad at drawing public lessons from past incidents. Lessons, when articulated and even when heeded (sometimes), usually follow time-consuming, expensive, protracted and often delayed (and in the Air India bombing, woefully delayed) commissions of inquiry. Even now, close to two years after the October 2014 attacks, we know very little. The focus of accounting there was on Parliament Hill's security structure, not the broader security and intelligence landscape.

All of this opaqueness can be contrasted with Australia’s detailed seventy-five-page report released shortly after a December 2014 terrorist attack in Sydney.

B. The Risk of Insta-Conclusions

An early judgment from the Driver case might be that we got lucky. The peace bond failed. The tip-off came from the FBI, not a domestic service.  And the intervention from authorities came when the bombing plot was on the cusp of execution. This is too Jack Bauer a plot line for comfort.

Another early judgment might be that authorities got it wrong. The court got it wrong. The people who concluded Driver was not violent got it wrong. The insta-judgment may be that Driver was always dangerous. The conditions should have been tightened, not relaxed. Surveillance should have been unrelenting.

Those are the narratives that risk becoming rooted, which is fine if they are true. But they may not be.

1. Was everyone wrong?

For instance, it could be the case that the authorities were right: Driver was not at high risk of violence when the peace bond was issued. Instead, Driver made the move from radicalized to radicalized-to-violence after the peace bond -- and perhaps (as colleagues Dawson and Amarasingam have suggested) because it unrooted him from the few stabilizing elements in his life, however unpalatable those were. Psychology is not physics. Things change, they are not immutable.

If this is true, it leads to very different policy conclusions than the "everyone was wrong" thesis. Not least, it reaffirms the need to get to grips with counter violent extremism programming and develop best practices in that area.

2. Surveillance on steroids?

There will be very close consideration also of whether the authorities should have maintained more intrusive surveillance. Those will be compelling calls. And more intrusive surveillance could have been facilitated by a more intrusive peace bond.

The conditions imposed, for instance, on Mr. Harkat in an (analogous) security certificate were, at one time: fitting of an electronic monitoring bracelet, to be worn at all times; a strict curfew, with solo departures from his house precluded and accompanied departures only on CBSA approval; strict geographic limits on his ventures outside his house; no visitors other than those listed in the order; no communication with anyone supporting “violent Jihad”; consent to interception by CBSA of all communications; consent to CBSA searches of his house and possessions; surrender of travel documents; and, a bar on possessing weapons.

These conditions seem of the sort that would have forestalled a Driver-type plot.

But there is the difficulty of tailoring conditions to actual risk -- with hindsight certainly, security certificate conditions were overkill, and also deeply controversial.

And more materially (literally), there is cost. By the end of the last decade, the government was budgeting $59 million for two-year periods on security certificates – over $11 million per person on a security certificate. This included more than surveillance costs -- it also included legal costs for the fraught legal contest. But even if you limit these costs to CBSA, CSIS and CIC funding, it still amounts to $3 million per person per year in support of the security certificate.

In February 2016, the government was aware of more than 180 individuals with Canadian connections who were abroad and suspected of engaging in terrorism related activities. One hundred of these were believed to be in Iraq and Syria. Another 60 had returned home. Earlier, in October 2014, the RCMP was reportedly tracking 90 individuals who intended to travel, or had returned from overseas, although it is not clear how many of these were affiliated with Daesh. And these numbers don't include the terror fans who are prepared to be think globally, act locally.

For the sake of argument, let us assume that we impose peace bonds on only 50% of 100 of these potentially dangerous people -- some proportion of the others are prosecuted (even more expensive), die overseas, never return or return and a decision is made that they are not dangerous. (And some of the latter would probably be surveilled anyway, even if no other legal step is taken against them).

Now, given Driver, we decide to apply security certificate-style surveillance to these 50 people. If we assume that the agency costs noted above cover surveillance, that works out to $150 million dollars a year. In 2016-2017, the RCMP's planned budget for all federal major criminal investigations into "serious and organized crime to reduce its impact on Canada and Canadian interests; financial crime to maintain the integrity of the Canadian economy, government and financial systems; and national security threats to support and protect Canada’s national security" is $294 million. So we are talking about adding the equivalent of 1/2 of that total budget in the form of terror-peace bond surveillance costs alone. 

I am not saying this is beyond Canada's means -- the RCMP has a huge budget and CSIS has a big one. And I am not disputing that the $3 million/year cost might be bloated and that policing can be very inefficient. But it is not so bloated and inefficient that there are no real costs at issue. And real costs require real prioritization. 

Keep in mind, meanwhile, that the government has pledged $35 million over five years on counter-violent extremism. This looks like chump change against the costs of persistent surveillance.

So what should our priorities be? I spend more time than I care to admit trying to understand this area of law and policy. But I am not comfortable arriving at a definitive conclusion on the data we have before us.

C. Learning to Learn Lessons Learned

So I circle back to my original point: If this were Australia, we might have a prompt, public lessons-learned report to help us decide and justify, based on facts and not rooted insta-judgment.

This is one practice well worth emulating.




Aaron Driver Matter: Questions awaiting Answers

[Upate: We have prepared two pieces outlining our post-RCMP press conference questions on this matter in the Globe and National Post.]

Yesterday, Aaron Driver was killed during a police anti-terror operation in southern Ontario. RCMP are expected to provide more details today, but early media reports suggest that bombs were involved. See Stewart Bell's Post reporting here. There are now also several overviews of Driver's history (see, e.g., the Globe story here) -- he was well known to police. Most notably, he was subject to an anti-terror peace bond, a mechanism that allows intermediate constraints on liberty on evidence short of what is required to prove an outright terrorism offence. That peace bond was sparked by Driver's online expression of pro-Daesh views and support.

It is a recurring truth that every terrorism incident tends to affirm each person's prior policy preferences. Those who wish tighter laws see them as justifications for tighter laws. (Or in Donald Trump's case, torture). Those who prefer softer forms of anti-terrorism see every incident as a failure to apply properly that approach. Evidence in support of such different positions is often a secondary consideration.

I am vulnerable to that tendency as well, and strive to counter it by self-imposed socratic probing of my own reactions. And so here are some of the questions to which I need answers in order to gauge the law and policy implications of this latest event.

As I write this, I am also cognizant that this is a very human case -- there was loss of life, and by all accounts, there could have been even graver loss of life. Both of those things needs to always be in our minds.

1. Did the peace bond work or did it fail?

For a more detailed account of peace bonds and how they work, see here.

A hasty assumption may be that the peace bond here failed. After all, if the reporting is correct, Driver was well on his way to building a bomb (and may have done so), despite being subject to peace bond limitations.

The peace bond "failed" argument may come in several forms.

a) Prison Uber Alles Theory

First, there will be those who think Driver should have been outright incarcerated a long time ago.  And indeed, when anti-terror peace bonds were created in 2001, they were derided as too weak for real terrorists, and critiqued as potentially too strong for false positives -- that is, innocent persons swept into their embrace by the very low burden of proof on government. Peace bonds are vulnerable to what Kent Roach and I describe as the "Goldilocks" problem.

Incarceration depends, however, on a crime. And our terrorism criminal law already sets the tripwire for terrorism crime very far from actual acts of violence. (None of the two dozen or so persons in prison for post-9/11 terrorism crimes got further than plotting before they were charged and convicted).

So does that mean that we should set the tripwire even further back? In essence, bill C-51 did that by creating a new speech crime -- had it been in force at the time, perhaps it might have reached Driver's speech conduct.

Of course, broadening the sweep of criminal law means that more people will go to jail for ideas that would never, left alone, have resulted in violence -- the terrorist fanclub is many multiples bigger than the terrorist cadre.

There will be those who take no discomfort in the idea of people imprisoned for their hateful thoughts. Personally, I can think of nothing more counterproductive from both a civil liberties and security perspective. The civil liberties implications of imprisoning people for opinions should be obvious. As for security: past practice suggests that putting a radicalized person in prison is a good way of ensuring that person emerge after there sentence radicalized and violent. That seems to be the pattern in recent French terrorist incidents, where imprisonment served as a radicalization-to-violence boot camp.

Canada does not have a sustained prison "deradicalization" program. Nor is it clear that deradicalization is even possible, although terrorist disengagement from violence may be.

So those advocating incarceration as a silver bullet should be careful what they wish for.

b) The peace bond conditions constraining liberty were not stringent enough

This is a more pressing issue. I do not, however, think we know enough to pass judgment. It may have been a mistake to relax the peace bond terms to eliminate (reportedly) an electronic bracelet requirement -- perhaps that was enough to allow Driver a greater range of action. On the other hand, maybe the peace bond conditions provided police with investigative access to Driver greater than would have been the case if reliant on regular surveillance powers. If so, then the peace bond might have facilitated disruption of the apparent plot.

c) The peace bond condition on deprogramming should have been imposed

Driver successfully challenged in court the constitutionality of part of the peace bond law (as amended by C-51) allowing the state to impose counselling -- billed by the court as "deprogramming". Was this decision an example of the constitution getting the way of good security?

Well, the government chose not to appeal the Manitoba lower court decision -- which was rather thin in its reasoning. It is difficult to see this decision from a single lower court as the last word on the matter.

But more than that, there is the question of whether a "treatment program" would have worked. Deradicalization is a novel area of the social sciences, and there is considerable doubt in the literature that I have read as to whether it can ever work. Moreover, forced participation in such a program may have the effect of simply entrenching views.

On the other hand, it may be possible to dissuade people from making the leap to violence. And we do need some sort of exit strategy from an endless cycle of disrupt or incarcerate. A UK parliamentary study (correctly) urged the need for an exit ramp from that country's equivalent to peace bonds, known originally as control orders: "a “graduated scheme, which commences concurrently with the [liberty limiting] measures, with the sole purpose of engagement and de-radicalisation”.

Parking someone on a peace bond is not a permanent solution.

It is possible to see, therefore, the Driver case as underscoring the need for serious investment in counter-violent extremism, so that such matters do not end with an explosion and preventive use of lethal force. Some will scoff at this conclusion, but I do not think the evidence exists one way or another as to whether we can develop an effective counter-violent extremism strategy to preempt these situations. And I don't think we have much choice but to try.

d) C-51 was good! No C-51 was bad!

Already commentators are approaching this case through the optic of the debate on 2015's anti-terrorism law. I am on record with my views about C-51. As careful readers will know, I am not among those who contested C-51's objectives. Rather, as Roach and I discussed in our book (still very bang-on-topic! still reasonably priced!), the law managed to muddle solutions to real problems. We believe it will have negative consequences for civil liberties and security if the powers accorded in C-51 are exercised to their legal limit. (And it is therefore those ill-conceived limits that demand reconsideration).

But in relation to Driver and peace bonds, C-51 so far is a minor cast figure. Peace bonds were not the key controversy in C-51. They have existed since 2001 to respond to feared terrorism offences. The government began its proceedings against Driver under this old law, only switching to the C-51 regime when that law came into force. C-51 did relax the standards for obtaining a peace bond -- they went from very low to very, very low. But I have little doubt that Driver would have been subjected to a peace bond under the original, very low standard -- certainly that was the Crown's belief.

I think that this is more generally true for other peace bonds since C-51 came into force in June 2015. The impediment to peace bonds prior to C-51 wasn't the law -- it was police resources, a different security environment and generally a more conservative approach to using this tool.

As noted, C-51 may have made Driver vulnerable to new speech crime charges, had it been in force. But as also noted, I do not believe incarcerating people who say the wrong things is a good idea.

C-51 also opened the door to CSIS threat reduction activities -- known popularly as "disruption". These include (most controversially) disruption activities that break the law and breach human rights in the Charter. We do not know if any of these new powers were material in this case, although we do know from Minister Goodale's statement that CSIS was involved in this case somehow (probably in its intelligence capacity). To date, there is no reason to believe that CSIS used its law breaking powers. (Nor, incidentally, does this case seem to have any of the controversial hallmarks of the Nuttall/Korody entrapment matter.)

e) CSIS and RCMP Coordinated

There seem to be early indications of police/CSIS coordination, and it is welcome. But the how of the coordination will be interesting, given our longstanding critiques that Canada does not manage inter-agency anti-terrorism investigations as seamlessly as we believe necessary in the current threat environment, or as seamlessly as the UK's MI5 and police do.

The system worked, sort of. And it is important to understand whether this was by design or luck. There will necessarily be questions about whether there was an intelligence/police failure. Here are some bottom-line questions:

  • How did someone known to authorities and subject to a peace bond get as far as posing a credible (and perhaps actual) suicide bomb threat?
  • Why was the police intervention 11th hour, in the nick of time (and perhaps after it, given that there was a reported explosion causing injury)?
  • Given the presence of a weapon and that there was apparently a computer video involved, Driver was in clear violation of his peace bond terms (and so basically automatically subject to imprisonment for up to 4 years). So why was there no arrest much earlier?
  • Why was there no public warning?
  • All of this is to say: Where was the early detection? Was this a resource issue? Did something go missing? Will be discussing a failure to "connect the dots"?

In truth, these are elemental, key questions, and I fear we will not learn the answers. Unlike other countries, we do not do thorough, public lessons-learned reports. We still, for instance, do not know all the details about what went wrong (or right) in the October 2014 terrorism events.

To be clear, I am not advocating gotcha inquiries. Just fact-based, clear-eyed assessments that can be evaluated outside of the hothouse of the security services themselves.

I am told the RCMP will be holding a press conference today at 1:30. In a perfect world, these questions will be asked and answered.

f) Civil liberties groups did their jobs

I add a commentary. I am seeing some suggestions on social media that the civil liberties group that contested Driver's peace bond was wrong to do so. But most of today's rights exist because someone -- often a civil liberties group -- defended them in relation to a deeply unpopular person or group. The core right at issue here was expression, something that enables me to write this blog and tweeps to tweet. 

Defending liberty while preserving security is like balancing an umbrella on the tip of one's finger: it is never in perfect equilibrium. Instead it is always a dynamic exercise in balance involving constant reaction and correction. Without an active espousal of civil liberties values, we risk the umbrella falling forward in constant reaction, and never any correction. Civil liberties groups do their jobs, just as security services do. We need them both to bring their A games.


Imminent release of book on Canada's 2015 security laws

After working flat out for now more than six months, Kent Roach and I are on the brink of releasing our full book on Bill C-51 (and the other recent security laws).  This is a big book that goes well beyond and does much more than recast the backgrounders on www.antiterrorlaw.ca.  It is as much about how Canada is getting security wrong as it is about how recent laws are jeopardizing rights. Some may regard the title (over which we mulled for much time) as provocative.  We regard it as measured, honest and accurate.   Our readers can be the judge.

The book is available for advance purchase from the publisher and on Indigo-Chapters.  As a free extract, we have also posted the Preface and Chapter 1 on line here and here or through the publisher’s website.

Despite its length (much longer than we originally planned), we worked with the publisher to price the book at as accessible a price point as possible — much lower than for a conventional work of this nature.  And we worked hard to meet rigorous academic and legal analysis standards.  But we set out to write a book that it is also a rewarding read for readers without a background in law or security but who are interested in really getting into the details behind all the smoke and rhetoric over anti-terrorism in Canada.  Put another way, we hope that people will read this book — and no, we don’t expect to make money, or really care if we make a dime.  (In any event, royalties for this kind of book usually are enough to pay for a nice gourmet pizza for our families.)

This has always been a project in active citizenship and we believe that we are duty-bound as handsomely paid law professors to contribute to an informed citizenry, able to parse complex issues.  An open society needs an open law, and national security in a democracy depends on informed consent.

We have tried to frame the basis for our conclusions as transparently as possible — so people can trace our sources to test why we take the views that we do.  See our Preface, linked above, for what we have tried to do with this book, who we are, the perspectives we bring, and how we have found our facts.

Of course, there will be people who will disagree with us.  We do not have all the answers.  So let us start the polite, informed and thorough discussion between people of good faith that we never had during the C-51 debates; a discussion about how we can achieve REAL security in a rights-observing country.  This is as grave an issue as any a society might encounter, and it should never be fodder for slogans and partisan positioning.

A shout out to our publisher who has, from day 1, also joined us in this vision of civic duty.  More than anything else, we hope that this book sells well so that publishers willing to support these kind of efforts are able to make ends meet, and continue to play their vital role.

Thank you to the many people who have written to us over the months with their questions, comments and insights.  We have not been able to respond to all, but have responded to many.

As we did throughout this project, we always welcome polite, dignified and substantive comments, with apologies again if we cannot respond to all such missives.  After all: classes start this week!

Antiterrorism Law and "Cultural" Baggage: How Terrorist Activity is Defined in Law

This weekend witnessed a small contretemps over Justice Minister MacKay's comments about the foiled mass atrocity attack in Halifax (currently an alleged attack because matters remain to be decided in criminal court).  Minister MacKay was quoted as saying "What we know of these alleged plans for a mass attack against our friends and our neighbours in the province, is that the attack does not appear to have been culturally motivated, therefore not linked to terrorism." 

This statement -- seemingly asserting that terrorism in Canadian criminal law is predicated on a "cultural" motivation -- struck many observers (myself included) as odd. My initial assumption was that the Minister was seeking to signal, however inelegantly, that this recent event was not Al-Qaeda or ISIS inspired violence.  This assumption seems belied by the Minister's spokesperson's statement, as reported in Global news: "MacKay spokesperson Clarissa Lamb, responding to a query from Global News as to why the minister had referred to cultural motivations, said MacKay 'was simply relating his understanding of what law authorities have concluded at this point.'"

The Minister, in other words, may actually believe that somewhere in Canadian law or jurisprudence, terrorism  is predicated on a "cultural" motivation.  For reasons I outline below, this misunderstanding would be of considerable concern.  But first, the law: Terrorism offences in Canadian law are almost universally predicated on a defined concept of "terrorist activity".  You may review a video "explainer" in which I lay out the elements of this definition here.

There is a "motive" aspect to the definition.  The listed motives are "political, religious or ideological purpose, objective or cause".  Culture is not mentioned. 

I am left puzzled, therefore, by the assertion made by the Minister's spokesperson.  I shall assume, therefore, that Minister MacKay was using "culture" as a shorthand for "religious".  But even then, by being underinclusive in his understanding of the true reach of terrorist activity, he would be implying that Canadian law is directed at one (and only one) motive for politicized violence: religion. 

If the press reporting is any indication, some Muslim Canadians are taking the Minister's statement as coded language for "religious violence is terrorism and nothing else".  The obvious inference, given the heated government rhetoric supporting its C-51, Antiterrorism Act, project is that religious violence means the Al Qaeda and ISIS inspired brand.

There is no doubt that AQ/ISIS inspired violence is a challenge for our democracies, as events in Europe even this past weekend clearly demonstrate.  But it is not the only form of politicized violence we should be concerned about, as Europe's largest recent mass casuality terrorist experience unequivocally demonstrate (Andres Breivik's murderous attacks in Norway).

The drafters of the 2001 Antiterrorsim Act who created the concept of "terrorist activity" were wise, because they were familiar with the many guises in which terrorism can arise.  They did not single out one type of cause and focus the law on it alone. 

The fact is, however, that since then, prosecutorial attention has been focused almost exclusively on AQ/ISIS inspired forms of terrorism.  In course of preparing our forthcoming book on Canadian antiterror law, and what the government's bill C-51 will do to it, Professor Kent Roach and I are reviewing all the reported cases in terrorism trials to date. If one excludes a single Tamil Tiger terrorism financing trial and the remnants of the Air India case, all of the terrorism trials since 9/11 we have reviewed have related to AQ (or ISIS) inspired terrorist activity.

At the same time, there have been events — such as the Bourque shooting and bombings incidents or attempts in Ottawa (Royal Bank) and Alberta (a plot against a Veterans Affairs office) that might, plausibly, meet the "motive" element of terrorist activity.  These were “political” and “ideological”, not religious act.  Yet, as best as I have been able to gather, they were not prosecuted as terrorism offences.  

In the present Halifax case, what differentiates the described events from "terrorist activity" of the sort at issue in cases like the Toronto 18 may not be the motive -- even a Columbine like motive might meet the definition or political or ideological.  Rather, it may another aspect of the definition of "terrorist activity": the so-called purpose element.  To be terrorist activity, the underlying wrongful act must be done "in whole or in part with the intention of intimidating the public, or a segment of the public, with regard to its security, including its economic security, or compelling a person, a government or a domestic or an international organization to do or to refrain from doing any act, whether the public or the person, government or organization is inside or outside Canada".  Simply killing people because your ideology demands death and mayhem may not meet this purpose requirement.  Maybe the evidence showed that Bourque simply wanted to kill because of his beliefs, not because he wished anything to come of his actions?  Maybe the Halifax plotters were of the same frame of mind?

This fine parsing of the law matters, and deserves serious treatment. We must be clear in our messaging about what terrorism is or is not under our law, and when it is not judged terrorism, why that is. We are at risk that our antiterror criminal law will be perceived, not as criminalizing politicized violence regardless of the cause, but aimed in practice at one particularly concerning and notorious form of politicized violence — the religious AQ/ISIS inspired brand.  We are not well served in this country by a practice or impression in which Muslim politicized violence (and only Muslim politicized violence) is terrorism, and everyone else's is something else.

It would be useful, therefore, for the political executive not to facilitate this impression. It would be better to underscore the reach of terrorism offences to the full range of motivations actually named in the Criminal Code.

Understanding How Much Terrorist Propaganda is already criminal, even without C-51

In reviewing the commentary on the new speech crime in Bill C-51, there appears be to an underappreciation of how much speech is already criminalized in Canada's current anti-terrorism law. We invite readers to consider the actual record.  For instance, in 2010, Mr. Namouh was charged and successfully prosecuted for (among other things) "enthusiastically participat[ing] in most of [a terrorist groups' propaganda activities".  Among other things, the accused participated in conveying "a message to Austria and Germany threatening terrorist action if their soldiers are not withdrawn from Afghanistan".  The accused also participated in most of the groups more clearly propagandistic activities, including (as described by the court):

  • analyzing the speeches of Al Qaeda leaders
  • inciting violent jihad
  • calling for support for jihadist groups
  • redistributing Al Qaeda materials
  • acting as a spokesperson for captured jihadists
  • singing the praises of jihadist leaders who died for the cause 
  • ensuring the security of online communications between jihadists
  • taking part in psychological warfare
  • providing military training with the purpose of implementing violent jihad 
  • producing a series of videos called the “Califate Voice Channel,” with the aim of transmitting news from the jihadist front
  • publishing jihadist magazines online
  • acting as an official media outlet for two groups taking part in terrorism. 

The accused was deeply invested in his cause and was not an idle apologist of things terroristic.  This undoubtedly contributed to the ultimate outcome.  But still, the behaviour cited by the Court in support of the participation and facilitation convictions ranges from outright threats to propaganda more distantly linked to violence.  Nevertheless, this propaganda style speech contributed to the convictions.

If you wish to review a digest analyzing the reach of the current law (and listing the practical reasons why we should be very cautious in going further in criminalizing speech), see the law review article Professor Roach and I prepared, issued as a working paper for TSAS, and now forthcoming, Alberta Law Review.

To sum up: this is not, and never has been and never should be, a debate about "is free speech absolute".  In our law, it clearly is not, and never has been.

And so the debate about the proposed new speech crime -- which Professor Roach and I assess as extremely vague and potentially vast -- is not (and never has been) "should terrorist incitement, threats, recruiting, instruction, or similar sorts of propaganda be outlawed".  It already is.  Instead it is about "what else should be considered 'terrorist propaganda'" -- that is, how much speech should be swept into that orbit. 

Professor Roach and I have argued in both the media and in our more detailed analysis that too much is caught up in the new offence -- we have no difficulty naming situations in which the law will penalize unpopular speech very far removed from violence.  We have no difficulty naming situations in which the law could penalize popular speech very far removed from violence.  We have also been underscoring our concerns about what speech chill may mean for the most important tools available to the state in dealing with radicalization to violence -- counter violent extremism programs and open source intelligence. 

We fear that the offence overreaches and violates rights while at the same time diminishing security.  If we are right, this would be an accomplishment few would be content to claim.  It takes some doing to offend civil liberties while at the same time making it harder for the RCMP to keep us safer.  Whatever this offence's constitutionality, it certainly pays little heed to the "law of unforeseen (or even foreseeable) consequences".

This concern with forseeable consequences and unforeseeable second order effects will animate our critiques of the areas of this bill, which we are working on now.

And so we hope that these are exactly the issues before every lawmaker contemplating this law. 

As lawyers, we can set out in firm detail our opinion on the speech provision's reach and its constitutionality, and are happy to have the usual lawyerly debate. 

But there are also operational issues in play, and lawmakers and the public deserve a response.  Can we expect an unfettered and unmanaged explanation of how the new offence would dovetail with the RCMP's important counter violent extremism program, one that appeals to our logic and does not simply reflect "marching orders" from political masters?

We live in this society too, and we want an effective response to radicalized violence, even if we do not believe the facts (as they exist at present) require us to go as far as the prime minister in seeing it as an existential crisis.

New Ottawa Terrorism Arrests: The Role of the Internet

Unusually for this blog, I am posting in response to police arrests that have just occurred.  I do so, only because these arrests come at a time when we are debating a new antiterror bill (Bill C-51) with quite concerning -- in some respects, quite radical -- legal content, including in the area of free speech.  The new arrests risk becoming fodder in the political debate.

I leave it to others to discuss the difficulties of law enforcement, administration of justice, presumption of innocence and the like in a highly politicized environment.

I make here only two points: My own view is that the new arrests are not an argument for extending those laws.  Quite the contrary -- they demonstrate the long reach of those laws.  Anyone who points to these arrests and says they justify new laws asks too much of our ability to suppress logic.

Second, I voice a note of caution about the apparent emerging narrative concering "radicalization by internet".  The Ottawa Citizen reports today: the "eventual trial could offer some critical insights into how young jihadis are radicalized, recruited and sent overseas. The RCMP said Project Servant collected evidence through witness statements, social media postings, and intercepted communications, including 'different snippets from Skype'. The RCMP said the radicalization of the three men who were charged primarily took place on the Internet."

The role of the internet in radicalization (and more particularly radicalization to violence) is a topic of both active discussion (and assumptions) and academic inquiry.  It would be foolish to suggest that it never happens.  Moreover, the past is little guide to the present and future in the cyberworld.  But the available research tends to support the view that the internet is enabler, but not cause.  See the paper Kent Roach and I prepared that deals, in part, with this topic.

It is important, also, to distinguish between the internet as broadcaster (that is, the archiving of information) and the internet as communications tool.  Does the RCMP mean that radicalization took place because the men were exposed to terrorist propaganda on the internet, or does it mean (as the examples above suggest) that the internet was a communicaiton tool?

This matters: radicalization to violence caused by internet propaganda is an argument for internet censorship.  Internet use as a tool of communication is an argument for suveillance powers.

Since both of these matters raise different issues (free speech in one, privacy in another), a careful disaggregation of the facts is necessary.  There is risk in simply saying "the internet caused it" and then using that to justify the full range of censorship, surveillance, regulation etc.  The facts actually matter.

We need that criminal trial to understand what is at issue.

Backgrounder #1: Assesing the Antiterrorism Bill's New Speech Crime

Kent Roach (University of Toronto) and I have now gone "live" on our separate website, established by our book publisher Irwin Law.  This website hosts our detailed analysis of the Bill C-51, the government's antiterrorism bill.  That website is: www.antiterrorlaw.ca.  We have now posted our first detailed "backgrounder", examining in exhaustive detail the new "advocate and promote terrorism offences" speech crime.  You may link to the website and then to the full analysis by clicking here.

Terrorist Words & The Long Arm of the Law: The Existing Record

As already noted this morning, Kent Roach and I have posted our working paper on terrorism glorification offences.  In it, we emphasize that the existing anti-terrorism offences (those enacted post-9/11 in Bill C-36) already criminalize a substantial amount of speech.  Put another way, we already have terrorist speech crimes.  We then posit that going one step further and criminalizing (in addition) "glorification" or "apologie" of terrorism (aka, "Hurray the terrorists!") would be constitutionally suspect, and likely counterproductive from a counter-radicalization and security investigation perspective.

Kent and I are now working on a "report card" on the 20+ terrorism trials we have had in Canada since 9/11, in an effort to place current debates into an empirical context.

In this post, I anticipate that report by returning to the point made above on reach of the current laws.  We have had a terrorism trial in Canada in which a person was convicted of, among other things, terrorist propaganda: R. v. Namouh, 2010 QCCQ 943.

This case involved (in part) a transnational bombing plot in association with the Global Islamic Media Front, which the court found to be a terrorist group under the Criminal Code because of its activities in Europe. The accused was convicted of conspiracy to detonate explosive device (s.465(1)(c)); participating in terrorist group activities (s.83.18(1)); facilitating terrorist activity (s.83.19(1)); extortion in association with a terrorist group (ss.83.2, 346).  He was sentenced to life for the bombing plot.  But his participation and facilitation activities also attracted sentences of 4 and 8 years respectively.

What were these activities?  The accused "enthusiastically participated in most of the GIMF’s propaganda activities".  Among other things, the accused participated in conveying "a message to Austria and Germany threatening terrorist action if their soldiers are not withdrawn from Afghanistan".  The accused also participated in most of GIMF more clearly propagandistic activities, including (as described by the court):

  • analyzing the speeches of Al Qaeda leaders
  • inciting violent jihad
  • calling for support for jihadist groups
  • redistributing Al Qaeda materials
  • acting as a spokesperson for captured jihadists
  • singing the praises of jihadist leaders who died for the cause 
  • ensuring the security of online communications between jihadists
  • taking part in psychological warfare
  • providing military training with the purpose of implementing violent jihad 
  • producing a series of videos called the “Califate Voice Channel,” with the aim of transmitting news from the jihadist front
  • publishing jihadist magazines online
  • acting as an official media outlet for two groups taking part in terrorism. 

The accused was deeply invested in his cause and was not an idle apologist of things terroristic.  This undoubtedly contributed to the ultimate outcome.  But still, the behaviour cited by the Court in support of the participation and facilitation convictions ranges from outright threats to propaganda more distantly linked to violence.  Nevertheless, this propaganda style speech contributed to the convictions. 

All of this is to say that our current laws are very far reaching: terrorist propagandists may be prosecuted and convictions obtained in Canada, assuming you can find and arrest the propagandists in the first place.

Terrorist Babble & the Limits of Law: Assessing a Prospective Canadian Terrorism Glorification Offence

Kent Roach and I have posted our working paper issued by the Canadian Network for Research on Terrorism, Security and Society and dealing with terrorism glorification offences.  The article is currently in peer review.  It can be downloaded from SSRN.  The abstract is as follows:

Since 2007, the Canadian government has repeatedly expressed interest in a terrorism ‘glorification’ offence, responding to internet materials regarded by officials as terrorist propaganda and as promoting ‘radicalization’. In the wake of the October 2014 attacks, this idea clearly remains on the government’s shortlist of responses. This article addresses the merits of such a criminal offence. It include analyses of: the sociological data concerning ‘radicalization’ and ‘radicalization to violence’; existing offences that apply to speech associated with terrorism; comparative experience with glorification crimes; and, the restraints that the Charter would place on any similar Canadian law. We conclude that a glorification offence would be ill-suited to Canada’s social and legal environment. This is especially true for Charter purposes, given the less restrictive alternative of applying existing terrorism and other criminal offences to hate speech and speech that incites, threatens or facilitates terrorism. We are also concerned that new glorification offences could have counter-productive practical public safety effects. Instead, we recommend modest amendments to the existing criminal law allowing the government to respond effectively to speech that is already criminal under existing Canadian terrorism or other criminal offences. Specifically, we favour a carefully constructed means of deleting (or at least ‘hiding’) the most dangerous forms of already criminal internet speech.