The Book





This blog comments on Canadian (and occasionally comparative) national security law to update my National Security Law textbook and now also my 2015 book, False Security: The Radicalization of Anti-terrorism, co-authored with Kent Roach.

Please also see for Bill C-51-related analyses by Craig Forcese and Kent Roach.

For narrated lectures on various topics in national security law, please visit my 2017 "national security nutshell" series, available through iTunes.

Please also visit my archive of "secret law" in the security area.

By Craig Forcese

Full Professor
Faculty of Law

Email: cforcese[at]

Twitter: @cforcese


Subscribe to National Security Law Blog
National Security Law Blog Search

Best Law School/
Law Professor Blog Award


Most Recent Blog Postings

A Feast of Commons Reports: National Security Studies by ETHI and SECU Released

This week, two standing committees released their reports on national security law and policy related matters. The Commons Standing Committee on Public Safety and National Security (SECU) released the product of its cross-country Fall 2016 hearings on national security writ large.  The Commons Standing Committee on Access to Information, Privacy and Ethics (ETHI) released its more focused assessment of one aspect of the last Parliament's infamous bill C-51, the Security of Canada Information Sharing Act.

Both reports make for interesting reading. And I am generally sympathetic to many of their recommendations.  A few quick words on each.  But before the dissecting, a general word: MPs should be congratulated in getting into the weeds on many difficult issues.  This is a hard area, and these reports join the government's own Green Paper in putting into circulation thinking on national security law and policy.

SECU Study

The SECU report is the shorter of the two studies, notwithstanding its broader coverage.  It has a fairly succinct summary of issues and evidence, and then a lengthy list of recommendations.  Not all the recommendations are closely anchored in supporting discussion, and so in some cases they are a bit unclear.  Generally, the recommendations range from very specific to very open ended.  I shan't assess on a recommendation-by-recommendation basis, but will address matters that caught my eye.

National Security Accountability

Readers of this blog will know that shortcomings in our national security review system is a recurring preoccupation in Canada. SECU highlights these difficulties in detail, and its report constitutes a welcome addition to the now vast literature on problems with Canadian national security accountability systems.

Its recommendations risk, however, a repeat of the Arar commission of inquiry error: two much bureaucratic complexity. There are basically two solutions to our institutional shortcomings in this area: a move to an all-of-government expert reviewer (sometimes dubbed a "super-SIRC); or the Arar commission proposal of "statutory gateways" between existing bodies, coupled with an expanded remit for some of them to capture agencies not currently subject to review.  The SECU report seems uncertain on which course of action to prefer.  There is some language favourable to a super-SIRC and a lot of language favouring statutory gateways, and then a complex-sounding means of coordination (including possibly the creation of still new bodies).

Too many moving parts in any policy proposal in Ottawa means certain death.  So while the spirit of the SECU recommendations is bang-on, I suspect the detailed recommendations will have limited impact.

CSIS Threat Reduction

SECU proposes abolishing statutory language suggesting CSIS can breach any and all Charter rights in engaging in "threat reduction".  This is welcome, although I think in practice there will still be a need to draft affirmatively what things we wish CSIS to do as threat reduction. Rather than a carte blanche, security legislation should lay out a detailed menu.  For reasons Roach and I discuss in various places (including some linked below), this enumeration approach would help put the regime on a more conventional constitutional footing.

SECU also recognizes the risk that CSIS and RCMP may be driving in the same lane post-C-51, and has some open textured language about, in essence, deconfliction.

This is a nod to the broader issue of intelligence to evidence.  But that is a complex matter deserving some very delicate law-making, and this report probably doesn't have the granularity to move the ball further down the field on that question.  That said, the related recommendation proposing the elimination of the bifurcated court process under the Canada Evidence Act is welcome.

Peace Bonds

SECU proposes a move to a "balance of probabilities" evidentiary standard for peace bonds -- the recommendation is unclear, but I assume this means abandoning the current "fear on reasonable grounds" (a low threshold).  I personally wouldn't support this. Balance of probabilities is a considerable burden of proof.  It will pretty much end the utility of peace bonds as a preemptive tool.  I think there are other safeguards that should be part of the peace bond regime, but this is not one of them. The state needs to have tools -- and this one has the virtue of requiring a proceeding in an open court (as opposed to all the secret administrative things like no fly lists).

Speech Crime

The SECU recommendations limiting the scope of Bill C-51's speech crime are important. That is a horribly overbroad crime at present.  It simply does not need to be so broad -- and reach so much speech only distantly linked to violence -- in order to meet the government's stated objectives. 

(I note that the dissenting Tory report expresses particular enthusiasm for the speech offence. I also note that a number of very serious lawyers had to debate at length in 2015 whether some Tory fundraising letters reproducing terrorism propaganda violated the speech crime. This is not a good offence if some very bright legal minds think there might be a real issue there. 

The Tories also argue that the offence has not injured free speech. Still, it has meant less privacy: RCMP documents suggest that the police are using the offence to seed investigations, including obtaining wiretap warrants. So to be clear: the police can wiretap speech in an effort to find the wrong kind of speech. This is not a happy situation).

The SECU recommendations cut away much of the excess of the speech crime, without undermining the stated purpose of this offence (which I happen to believe is already meet by existing crimes, but that is another debate. Personally, on this specific issue, I am with the NDP "supplementary opinion" to the report: the speech crime is unnecessary; repeal.)

No Fly List etc

There are many recommendations on improving the no fly list.  A redress system for false positives would be very welcome -- and we await action on this by government.  And use of special advocates during the secret appeal hearings would also be welcome.  Generally, there are a number of quite sensible recommendations on this topic. 

Likewise, the oblique reference to rolling back constraints on special advocate access to information in immigration security certificate matters created by C-51 is also welcome.

Lawful Access

SECU basically ducks the issue of lawful access, encryption and CSE collection of metadata and private communication.  The report does propose an appropriate warrant where CSE shares with other agencies -- a very vague recommendation.  But interpolating from what is intended, this is actually an issue, since I understand that some sharing may be occurring without warrant.  That is, CSE may be sharing information (incidentally) collected under its Mandate A with CSIS and police. In some instances, this may be information that the latter agencies could only collect had they obtained a warrant.  Since administrative end-runs around Charter rights are unconstitutional (says the Supreme Court), it would be nice to have a handle on this.  Likewise, however incidentally collected, the fact that CSE is acquiring Canadian personal information without judicial authorization raises its own constitutional problems.  I have discussed these issues before.  They are still unresolved.

Final observations

I suppose, not surprisingly, the report continues to be coloured by C-51 and the (disproportionate) focus on terrorism.  There are allusions to other national security threats.  But not a lot of detail on espionage, insider-threats, influence activities, cyber-security, etc.  Hopefully, there will be future opportunity to forward think on these and other national security challenges.

Overall, the report is a welcome benchmark on many of today's key (primarily anti-terror) issues.  Worth a read.

ETHI Report

ETHI's study is a deep dive into the obscure Security of Canada Information Sharing Act (SCISA).  I have less to say on this because the study is particularly comprehensive in its narrative. Covered both the criticism and defence of SCISA fairly, I thought.  And informative on government positions. 

SCISA's problems have always, in my view, been ones of drafting, and less conception.  I think the most of the ETHI report SCISA recommendations are sensible, justifiable and sound.  For the most part, implementing them (or reasonable variants on them) would not degrade the government's objectives with SCISA.  (I think there are technical issues involved in getting SCISA and the Privacy Act to work properly together, but I would be less concerned about subordinating the Privacy Act to SCISA if SCISA was more reasonable.  On the other hand, I still think the best solution would actually be a careful process of amendment and pruning of the multitudinous laws on security and information sharing to make them more coherent, adding where necessary; clarifying in other instances. SCISA was wallpaper over a fissured wall.)

ETHI also has recommendations on reformed accountability review. ETHI doesn't opt for either a super-SIRC or statutory gateways but does press for action of some sort.

And ETHI too raises the intelligence-to-evidence issue (I am so happy!) by recycling the Air India commission recommendations.  I confess that my own thinking on how to address intelligence to evidence has moved beyond the Air India recommendations, but keeping this issue alive in the public policy mind is critical.

Overall, a thorough study on highly technical issues.

If you have a couple of hours, these two studies are worth a read.

If you have even more time: Kent Roach and I discuss many of the issues at play in more detail here and here and here and here.  For the long version, still available, still current, still affordable, consider picking up our book False Security: The Radicalization of Canadian Anti-terrorism, written with verve and élan.







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.


Use of Force against Syria: The law is an ass so what to do about Pandora's Box?

In this entry, I follow the thread on some points made in my oped in the Globe this morning.  Readers may wish to consult that first.

The Hard Facts of this Case

The United States has bombed Syria. In it itself, there is nothing remarkable about this statement. After all, the United States and its allies have been using force in Syria against Daesh for years. The difference last night was, however, one of target and also, therefore, one of law. The sum consequence is: one more explosion for a smart bomb, and potentially one giant leap into the unknown for international law and relations.

The missile strikes last night targeted the Syrian military, not non-state insurgencies or terror groups in Syria. The use of armed force against Daesh is (controversially) justified as an exercise of collective self-defence, undertaken on the territory of a state (Syria) unwilling or unable to suppress armed attacks done by Daesh.

Whatever the disputed status of this “unwilling and unable” concept, self-defence is a legitimate exception to the legally-robust prohibition on one state’s use of force against the “territorial integrity or political independence” of another state (codified as Art. 2(4) of the UN Charter).  The other exception is authorization on use of force from the UN Security Council.

Largely because of Russian obstruction, there is no Security Council resolution authorizing force against the Syrian state or military.  Nor is there any plausible self-defence argument for use of force against the Syrian state itself (as opposed to against non-state actors mounting threats from its territories).  Syria has not attacked the US, or its allies. 

But Syria is massacring its own people, horrifically, in violation of most international law of any relevance.  Most recently, it is alleged to have (again) used chemical weapons.  (An obvious fear is that the US missile attack was provoked by bad intelligence – and the attribution of chemical weapon use to Assad’s government will prove incorrect.  Then we have an Iraq War redux. But Assad has used chemical weapons in the past.)


The Law is an Ass

Still, even with these Syrian atrocities: there just is no persuasive legal argument that these Syrian breaches justify use of force in response.  There is no other exception to the prohibition on use of force adding to those discussed above.  This is true whether one wants to call any additional exceptions “humanitarian intervention” or the “responsibility to protect”.  See also other discussions: here and here

People do, of course, urge the existence of these other carve-outs.  See most notably Harold Koh, here.  But unpack those discussions, and they turn on normative justifications – the should of international law, not the is. The is comes from basically two things: treaties and customary international law.  There is no treaty basis for either humanitarian intervention or R2P.  All that is left, then, is customary international law. And one would need to give supreme primacy to the state conduct (with opinio juris) of a handful of the world’s states to conclude an R2P or humanitarian exception exists.  And along the way, you’d need to ignore a lot of state expressions of dissent, not least dissent generated by the justifications offered by the United States in the 2003 Iraq War (where at least one line of justification focused on humanitarian impulses).

Nor is there any international law norm that permits use of force by one state against another to punish use of weapons of mass destruction.

And so we are left with a “wicked problem”: surely, we could prosecute the Syrian leadership for war crimes and crimes against humanity, if we had them.  But we don’t.  And so Syria acts with impunity, shielded by the well-meaning, international strictures on use of force.  And these well-meaning limitations on the use of force were intended by the framers of the UN Charter to operate in a world in which the Security Council worked – and where it would exercise collective defence.

Those framers were not pacifists.  But having lived through the Second World War, they wished to shackle as much as possible aggressive war – war undertaken for reasons other than self-defence.  The Security Council was supposed to deal with the circumstances (other than a state’s own immediate right to collective or individual self-defence) where threats to international peace and security obliged a forcible response.  And it was supposed balance great power rivalry.  But balance has meant, for most of its history, Security Council gridlock.

And so we are still left with the impasse: the law appears to be an ass, confronted with Assad’s atrocities.  And so perhaps the law should not speak in times of humanitarian crisis.


Pandora’s Box

The risk of that, of course, is both immediate and long-term.  The immediate risk is that a recycling of the “illegal but legitimate” claim from the Kosovo conflict in the Syrian context brings us a lot closer to a large inter-state war between the US and Russia than it ever did in Kosovo.

And the bigger problem: what is good for today’s hegemonic great power will be equally good for tomorrow’s, or indeed for today’s aspirants to great power status.  Whatever else can be said about the Article 2(4) framework: it has stabilized international relations between states.  Not perfectly, but nevertheless, to the point where use of force between states without lawful justification is anathema.  That is the explanation for the international community’s response to Iraq’s conduct in the Gulf War 1990.  And it is the reason why much of the world has sanctioned Russia for its Ukraine adventure.  (Both of those two instances involved a use of force culminating in de facto conquest.  But the illegality of the conquest is a by-product of the underlying prohibition on use of force.)

All of this is to say: Unravelling limitations on use of force in today’s uncertain political environment opens Pandora’s Box.


Making the Best of an Impossible Impasse

And so the question is: can we net whatever comes out of that box?

Or more concretely, if we are now committed to a path where force can be used for a noble cause outside traditional rules, can we confine it to those noble causes? Harold Koh proposes a list of sensible indicators designed to do just that.  And although I quibble with the any claim that they are currently part of international law, they certainly could be.  Better they than the alternative: expedient uses of force ungoverned by classic rules, used as much by the goose as the gander and for the wrong reasons.

So what policy prescription does this present. Many.  But one that comes immediately to mind this morning: No state should blankly endorse the US action. That is too much indefinite state practice that opens the door to competing and overboard narratives about “emerging customary international law”. 

Rather, any support should confine claims of legitimacy to the narrowest of facts: here, the need to act in the face of war crimes involving, not just the targeting of civilians, but the (so far, unfortunately, only alleged) use of an inherently illegal weapon of mass destruction by a state.  And we should borrow from self-defence rules and graft in some very strict proportionality rules. But stricter. (Many uses of force seem now to gravitate to regime change, which is the kind of thing that precipitates bigger wars and violations of the "Pottery Barn" credo: you break a country, you pay for it. Usually, we just walk away.)

In sum: if the use of force rules are to be sliced and diced, then the slices should be most minute and thinnest possible. Because a thick slice of baloney today is the kind of thing that can be used to justify the crassest form of aggression tomorrow.

Page 1 ... 2 3 4 5 6 ... 82 Next 3 Entries »