About this Project

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 www.antiterrorlaw.ca 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.


For a continuing conversation on Canadian national security law and policy, please join Stephanie Carvin and me at A Podcast Called INTREPID.


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

By Craig Forcese

Full Professor
Faculty of Law

Email: cforcese[at]uottawa.ca

Twitter: @cforcese


National Security Law Blog Search
Subscribe to National Security Law Blog

Best Law School/
Law Professor Blog Award


Most Recent Blog Postings

Latest Book: Available from Irwin Law in April 2018.


Bill C-51: Still Baking With the Wrong Recipe on the Charter of Rights & Freedoms

First my compliments to the Chef...

I have watched with interest the senate proceedings on bill C-51 and generally congratulate the senate standing committee on national security and defence for a much more substantive and meaningful deliberation on bill C-51 than we witnessed in the sorry spectacle that was the Commons committee process.

The senate committee has invited many government witnesses and security service watchdogs -- all notably absent at the Commons level -- and all of these witnesses have presented more detailed and substantive positions that elevate the debate over this bill.

But I am afraid there is still a fly in my soup...

This is not to say, however, that they have cured concerns that Kent Roach and I have raised about the bill.  I would categorize these concerns into three categories. 

  • First, concerns that the government seeks to achieve mostly legitimate security objectives with a bill that overreaches unnecessarily with measures that exceed what is needed to meet these objectives, and thus cause avertable injury to civil liberties while risking serious counter-productive side-effects on the security side (e.g., the new speech crime and the overly broad authorization for CSIS to conduct kinetic threat reduction steps).  Kent and I have proposed amendments that would, in our view, enable the security objectives to be achieved, without the collateral damage.  But apparently with little effect.
  • Second, concerns that the bill underreaches in failing to deal with serious shortcomings in Canadian national security law (e.g., the failure to oblige information sharing by CSIS of the sort that the Air India inquiry adjuged in 2010 necessary both to cure historical problems and also on-going coordination problems between RCMP and CSIS; the failure to cure the stove-piping problem with existing review bodies and extend the remit of review powers to the many government agencies engaged in national security, but subject to no review for lawfulness and compliance with directives).
  • Third, concerns that the bill violates the Charter of Rights and Freedoms (e.g., again the speech crime and also the puzzling new warrant regime for Charter breaches by CSIS).

Debating the latter, estoric Charter issues has proven fruitless -- these are too arcane a subject for a political process.  Generally speaking the government (official) position is "Yes C-51 is Charter compliant" (although I am not sure, personally, how many first born children the government would bet on the question). 

As for outside government: they certainly exist, but it is pretty hard to find a public or constitutional lawyer outside of government who agrees with the government position.  But such is life, and the recurring pattern these days involves vain efforts to cure constitutional issues at the parliamentary level, followed by lengthy, expensive litigation to defend the constitution in the courts.

And I need to see the manager about the cake we had for dessert...

All that said, I cannot help but comment on a new version of an old government line on this Charter issue.  The government persists in articulating the jawdropping position that because court warrants exist for searches (and also detentions), then it follows a court can, by warrant, permit the override of any and all Charter rights.  I've lost count of the number of times that I or Kent Roach have outlined how this is a fundamental category error

I won't repeat in full that analysis here.  Instead let me simply offer up this analogy: Compare the Charter to the aisle in your local grocery store marked "baking goods".  There you will find a whole host of ingredients, including such things as flour, baking soda, sugar etc.  Some of these things look a lot alike -- for instance, many are white and granular.  But of course, they are very different both in their taste and in the properties they bring to any baking project.

In the government worldview, all are, however, exactly the same -- they all can be treated the same way.  And so where a recipe calls for flour, another ingredient such as baking soda or sugar may be substituted. 

The reality, of course, is that not every ingredient -- and not every Charter right -- is the same. 

The section 8 protection against "unreasonable" searches and seizures and the section 9 protection against "arbitrary" detention have built in qualifiers ("unreasonable" and "arbitrary"), and have been understood for an age (in the Charter, the US Bill of Rights and in their more ancient common law analogues) as permitting an independent authority (usually a judge) to pre-authorize searches and detentions. 

These pre-authorizations prevent a breach of these rights because it is the preauthorizaiton (among other things) that makes the search "reasonable" or the detention "non-arbtirary". 

And so these are rights that in their fabric are associated with warrants -- and keep in mind what warrants are.  They are one-sided, secret judicial proceedings done with only the government present and only with government information provided and done in advance of government action, without its full contours on the ground yet established.

It is striking that some government senators are now pointing to the Spencer decision -- which was, in effect, a classic s.8 search and seizure case -- and suggesting that in some manner the Supreme Court blessed C-51.  In Spencer, the Supreme Court said, in essence, warrants are required for identifying information associated with IP addresses. (For a description of Spencer and why it is a notable s.8 search and seizure case, see here.) 

But again, this was a case about s.8 search and seizure, and not every other right in the Charter. It is a case, in other words, about the traditional area in which warrants are part and parcel of our legal tradition -- all it does is extend s.8 with its tradition of warrants to a novel new technology which raises new sorts of privacy concerns.

So the best way to analogize the government's use of Spencer as justification for C-51 is: "Look! The Supreme Court said that flour really is included in the recipe!  So that means we really can use baking soda!" 

The cake was very chalky, strangely fluffy and tasted awful...

Let me expand on the baking soda: rights like the Charter s.12 bar on "cruel and unusual treatment" and basically every other Charter right are different from s.8 and s.9, and there is no such "reasonableness" or "arbitrary" qualifier on the right and no such tradition of warrants designed to forestall the violation of the right.  (I acknowledge that s.7 has more internal pliability than say s.12, at least for its procedural component.  Less so, it seems to me, when the court establishes a substantive element of fundamental justice).

Bottom line: We don't have, in Canada, "cruel treatment warrants" or "denying reentry to the country warrants" or "speech repression warrants" or "denial of habeas corpus warrants".  That is, courts don't decide that rights can be abrogated in advance is a one-sided, secret judicial proceedings done with only the government present and only with government information provided and done before the government action, without its full contours on the ground yet established.  We have repeatedly asked for accurate and compelling analogues for what the government proposes in C-51, and we are still waiting.

It is certainly true that in our Charter "recipe" no right is truly absolute.  But again, in the recipe that is the Charter, the nature of limitations matter.  Limitations are permissible only through advance use by Parliament of the notwithstanding clause (s.33) or through the complicated, back-end, open court adjudication on a full factual record of a limitation "prescribed by law" (which almost always means legislation that identifies intelligibly the nature and contours of the rights constraint) under s.1.  Neither of these forms of limitation are found (or satisfied) in C-51.

Seriously, Mr Manager?  Just because the Cordon Bleu school uses baking soda for fluffy bread doesn't mean they'll let you take flour out of cakes

Yet, under the government's theory a judge may, through warrant, now pre-authorize a breach of any right.  Again, I won't repeat the full objection to this, which is rehearsed elsewhere (including here).  But just take one example. 

It is a violation of the Charter to commit torture (s.7 and s.12 of the Charter would be violated).  Under the government legal theory, however, it is now perfectly feasible that such a breach would be exonerated if Parliament were to pass a law saying "police can do anything so long as first blessed by a court".  In the government's radical legal theory, the existence of the Charter bar on torture can be negated by this sort of generalized ex ante court blessing.  Anything done under the shelter of a court warrant cannot, in the government's reasoning, be unconstitutional.  It is, quite literally in some cases, a "get out of jail free" card. 

Beyond being a puzzling position for a government whose political executive rails against judicial overreach, it is simply untrue that "whatever a judge says, goes and is inherently constitutional".   Judges don't decide the persistence of a Charter right.  The existence of the Charter does not depend on judicial sufferance.  The government's theory now presents what I suppose we could call the "judicial notwithstanding clause": rights can be abrogated in advance at judicial whim.

But as I say, since the government seems committed to this constitutional adventure, the legislative process is apparently not a place where they can be persuaded to take a more conventional, and more defensible approach. 

The more conventional approach would be: enumerate specific powers that you wish CSIS to have, in the legislation, to reduce threats to the security of Canada.  In this manner, you "prescribe by law" powers that may (by implication) constrain a Charter right.  If it is challenged, the legitimacy of this new power then becomes fodder for a conventional inquiry for a court under s.1 in the regular, open-court, full evidentiary record manner. (This would also have the necessary effect of allowing us to debate in a transparent manner exactly what sort of powers we are giving CSIS.  The present bill amounts to "anything other than bodily harm, obstruction of justice or violation of sexual integrity", with the government then saying "but don't worry, it won't actually be used in a scary manner").

This conventional approach would avoid what will happen after C-51 is passed: a court challenge to the new warrant provisions that can reasonably raise the sort of objections noted above, accompanying objections that the peculiar warrant proceeding happens also to violate separation of powers principles by dragooning the courts into a Charter-limiting role and blessing unconstitutional state conduct, and a lot of very interesting arguments that will be able to deploy the Supreme Court's blessing of "reasonable hypotheticals" to explore exactly how bad the new C-51 powers could be.

All of this would have been totally avertable by a little more care and attention to the Charter's fundamental recipe.  But no, "Let them eat our very chalky, strangely fluffy and awful tasting cake!"



Bill C-51: Catching up on the "Catching Up with Our Allies" Justification for new CSIS powers

One of the recurring government talking points in the C-51 debate has been, in relation to the new CSIS powers, "we are only catching up with our allies".  This statement has puzzled and perplexed and I have raised preliminary concerns about its accuracy before. 

We have repeatedly suggested that the government release details supporting this claim, but (par for the course in this debate) it has so far ignored such requests (despite the fact that any memorandum on such a topic would be on foreign and comparative law, would not be Canadian legal advice and thus presumptively not solicitor-client privileged). 

The closest we have to details appeared in a recent and rather misleading one page document on C-51 that some government MPs have posted on their website (more on that below).

Neither Kent Roach or I purport to be specialists in foreign law, and so we have done our best to share what we know about the government claims with those who are (that is, lawyers, professors and sometimes security service members in the countries named by the Canadian government).  This blog entry represents a running tally of what has been said on this topic and what we have discovered. 

Again, we invite amplifications or corrections from those who actually know something about the laws of the countries listed.  We are doing our best, but appreciate any corrections.

Setting the Ground Rules

First, some basic methodological rules. 

1. What is the CSIS Power in Question?

Bill C-51 provides that CSIS may take “measure, within or outside Canada” to "reduce" threats to the security of Canada (broadly defined).  These measures may include things that would violate Canadian law or our constitutionalized bill of rights -- if such conduct is first blessed by a court in the form of a warrant, issued in a secret proceeding ex parte, and without appeal.  In issuing such a warrant, the judge may authorize CSIS "to do any other thing that is reasonably necessary to take those [threat reduction] measures".
The only constraint is that the CSIS threat reduction measure must inflict no bodily harm, no obstruction of justice and no violation of sexual integrity. The measures must also be "reasonable and proportional in the circumstances, having regard to the nature of the threat, the nature of the measures and the reasonable
availability of other means to reduce the threat."
These powers are not tied to investigating crimes or collecting intelligence — they are about threat reduction, tied to the vast concept of “threats to the security of Canada” in the CSIS Act (basically, anti-terrorism, sabotage/espionage, “foreign influence activities” and subversion).  And, as the language above suggests, CSIS can take these measures either in Canada, or abroad (and despite any foreign law).

2. So What Does this Mean for Foreign Comparisons

Given all this, it is not methodologically or logically sound to justify this new CSIS power by committing any of the following errors:

Category 1. the power by foreign services to gather intelligence.  CSIS already has powers to collect intelligence that include the power to conduct search and seizures for this purpose under warrant.That is not what C-51 is about -- it is about powers to "reduce" (aka disrupt) threats, using physical/kinetic means and not by spying.  And so pointing to special powers that foreign agencies have for intelligence collection purposes does not justify bill C-51.

Category 2. foreign peace office/law enforcement powers to detect or prevent crimes. 

The Canadian police already have the power, under the Criminal Code s.25.1, to violate criminal laws where engaged in a criminal investigation, subject to the requirement not to inflict bodily harm, obstruct justice or violate sexual integrity. (This power is the obvious model for the new C-51 CSIS powers, but it is much narrower and subject to back-end public reporting obligations that do not exist for the CSIS power.  See our discussion here.)  The RCMP also have the responsibility under the Security Offences Act to apprehend the commission of national security offences. 

With these legal rules, the highly preemptive nature of most terrorism offences, and the concept of intelligence led policing, the RCMP is in the business of disrupting/preventing feared terrorist acts, in the broadest sense of the term. 

And so any comparison to a foreign country that has police services that perform this same function is logically not relevant in terms of justifying giving disruption powers to CSIS, a classic security intelligence organization.  (This kind of argument could also demonstrate a lack of familiarity with existing Canadian police powers.)

Category 3. powers a foreign intelligence service may exercise in its international operations.  Foreign intelligence services do have the power to violate foreign laws in their foreign operations (CIA, MI6). But they won't have the same power to violate domestic laws or to operate domestically. 

A person who points to these foreign powers to justify similar powers, to be exercised domestically, by CSIS misunderstands the reason why most states separate their foreign and domestic intelligence operations.  And, more than that, what an agency can do abroad is simply no precedent for what CSIS should be able to do domestically.  In other words, a precedent for CSIS's domestic law breaking, threat reduction powers must itself involve domestic, law breaking, threat reduction powers.  Anything else compares apples to oranges.

Category 4. misinterpreting foreign law or failing to define adequately what is meant by terms "disruption" or "threat reduction". It is important to understand that different states apply different understandings to similar terms.  When some states speak of "prevention" they may mean something very different from what is actually on the table with C-51 and CSIS's proposed threat reduction powers.

Category 5. comparing a limited domestic power to be conducted in full compliance with law/human rights norms with a power by CSIS to break the law and violate the Charter (with warrant).  It is important to understand the scope of any disruption power a foreign service may have.  Even if it can perform a threat reduction activity analogous to what is at issue in C-51, a modest power that does not not allow the service to exceed the law or human rights is no justification for bill C-51, which proposes exactly that.

Category 6. comparing the proposed CSIS power to a country whose security services operate without a legal superstructure and whose conduct is a bit of a mystery.

What has the government said?

So far, this is what the government has said about foreign states:

Minister Blaney said this during is opening statement to the Commons committee:

Currently CSIS can detect security threats but is unable to take action unlike most allies are doing. With the new threat disruption mandate, CSIS would be authorized to take direct action to disrupt threats to the security of Canada at home and abroad like most of our allies, such as Sweden, Norway, Finland, Denmark, France, United States, United Kingdom, and Australia. It's about time, Mr. Chair.

He said this at the Senate committee:

Senator White: Thanks to all of you for being here today.

Minister Blaney, the U.S., U.K., Finland and Norway, for example, allow their respective security agencies to disrupt, in one form or another, criminal and/or terrorist activities.

Mr. Blaney: That’s correct.

On clause-by-clause in the Commons committee, government officials were asked this question by Parliamentarians:

Hon. Wayne Easter: Thank you, Mr. Chair.
Then my question would be do any of our Five Eye partners, United States, Australia, New Zealand or U.K. give these kinds of disruptive powers to their equivalents of CSIS and do they allow their agencies to supercede either domestic law or constitutional rights?
Mr. John Davies: Yes—not New Zealand, but the others, Australia, U.K., U.S. absolutely have equivalent powers. The minister mentioned that in his remarks when he was at the committee.

And some government Members of Parliament have now posted the following on their websites:

Some have alleged that the Conservative Government is not correct in stating that other allies allow their national security agencies to disrupt threats. What allies can do this work?
In the US the Central Intelligence Agency can, pursuant to the National Security Act, conduct domestic threat disruption with an executive order. In the United Kingdom, MI5 can, pursuant to section 1 of the Security Service Act conduct any activity to protect national security. The Norwegian Police Security Service has a mandate to prevent and investigate any crime against the state, including terrorism. The Finnish Security Intelligence Service is mandated to prevent crimes that may endanger the governmental or political system, and internal or external security, pursuant to section 10 of the Act on Police Administration. We must ensure that CSIS has the same tools to keep Canadians safe. [emphasis in original]

What have we learned?

So far, we have spoken to one former security service employee and four security or law professors in the United Kingdom (who in some cases have in turn spoken to other security service personnel).  We have spoken to three law professors in the Australia, or members of their research teams.  We received input from one French law professor.  We have received information from two law professors in the United States.  And we have exchanged notes with one law professor in Sweden, with expertise in Nordic law. These are people specializing in security or terrorism law.  Helene Buzzetti, a journalist with Le Devoir, has spoken to embassy officials from many of the countries named in the government statements above.  We have also done our earnest best to read foreign laws. This is what we have have learned so far in table form.  More detailed discussions follow. 

To summarize: we have found precisely no instance of a true precedent for the C-51 CSIS powers.  In every instance, the Canadian government's comparison with a foreign agency constitutes one or more of the methodological errors noted above.  We repeat our initial concern: Canada is not "catching up to allies".  It appears to be on an adventure of its own.  Again, we welcome corrections if we are misunderstanding foreign law.

Summary Table

State (agency)

Threat Reduction Analogy?

Power to Violate Laws/Human Rights?

Category Error Committed by Canadian Government

Australia (ASIO)

No. Powers tied to intelligence collection.


Category 1 Error: Justifying CSIS threat reduction powers by pointing to foreign intelligence collection powers


No. Powers ties to intelligence collection


Category 1 Error: Justifying CSIS threat reduction powers by pointing to foreign intelligence collection powers


Yes, to some extent in the exercise of a law enforcement function analogous to what our own Canadian police do anyway


Category 2 Error: Justifying CSIS threat reduction powers by pointing to foreign police powers analogous to what our own police do anyway


Category 5. Comparing a limited domestic power to be conducted in full compliance with law/human rights norms with a power by CSIS to break the law and violate the Charter (with warrant).


N/A (because powers not prescribed by law)

N/A (because powers not prescribed by law)

Category 6. Comparing the proposed CSIS powers to a country whose security services operate without a legal superstructure and whose conduct is a bit of a mystery.


Yes, to some extent in the exercise of a law enforcement function analogous to what our own Canadian police do anyway


Category 2 Error: Justifying CSIS threat reduction powers by pointing to foreign police powers analogous to what our own police do anyway


Category 5. comparing a limited domestic power to be conducted in full compliance with law/human rights norms with a power by CSIS to break the law and violate the Charter (with warrant).


Yes, to some extent in the exercise of a law enforcement function analogous to what our own Canadian police do anyway


Category 2 Error: Justifying CSIS threat reduction powers by pointing to foreign police powers analogous to what our own police do anyway


Category 5. Comparing a limited domestic power to be conducted in full compliance with law/human rights norms with a power by CSIS to break the law and violate the Charter (with warrant).

United Kingdom (MI5)

Not truly. It plays a more active role that CSIS but does not have legal powers of the scope at issue in C-51


Category 4. Misinterpreting foreign law or failing to define adequately what is meant by terms "disruption" or "threat reduction"

Category 5. Comparing a limited domestic power to be conducted in full compliance with law/human rights norms with a power by CSIS to break the law and violate the Charter (with warrant).

United Kingdom (MI6)


Yes, in limited circumstances, but may only do so internationally

Category 3. Justifying CSIS law breaking powers to those a foreign intelligence service may only exercise in its international operations. 

United States (US)



Category 3. Justifying CSIS law breaking powers to those a foreign intelligence service may only exercise in its international operations.

Category 5. Comparing a limited domestic power to be conducted in full compliance with law/constitutional norms with a power by CSIS to break the law and violate the Charter (with warrant).



So far, Australia comes closest to having a power equivalent to the CSIS C-51 proposal.  Except it's not really close.  Australia's CSIS equivalent, ASIO, has new powers to conduct "Special Intelligence Operations" (SIOs).  In conducting these operations, ASIO is immunized for law breaking, subject to a requirement it not induce a crime or cause serious injury, torture, commit a sexual offence or cause significant lor or serious damage to property.  So far, this sounds like the C-51 CSIS proposal.

Except for this: these SIOs may be authorized by executive approval where an operation will assist ASIO "in the performance of one or more special intelligence functions".  The latter are, in turn, listed in paragraphs 17(1)(a), (b), (e) or (f) of the ASIO Act: 

(a)  to obtain, correlate and evaluate intelligence relevant to security;

(b)  for purposes relevant to security, to communicate any such intelligence to such persons, and in such manner, as are appropriate to those purposes;

(e)  to obtain within Australia foreign intelligence pursuant to section 27A or 27B of this Act or section 11A, 11B or 11C of the Telecommunications (Interception and Access) Act 1979, and to communicate any such intelligence in accordance with this Act or the Telecommunications (Interception and Access) Act 1979; and

(f)  to co‑operate with and assist bodies referred to in section 19A in accordance with that section [including a law enforcement agency].

As confirmed by the law professors to whom we have spoken, this means that SIOs are aimed at intelligence gathering.  This is not threat reduction.  Put another way, Australia has amped up the intelligence function of its domestic intelligence service.  It has not converted its domestic intelligence service into a threat disruption agency. 

ASIO also has a new power to conduct, under ministerial warrant, an invasion of computer systems that might amount to disruption of computer systems: "if there is reasonable cause to believe that data relevant to the prejudicial activities of the identified person may be accessible by using a computer or other electronic equipment, or a data storage device, brought to or found on the subject premises—use the computer, equipment or device for the purpose of obtaining access to any such data and, if necessary to achieve that purpose, add, copy, delete or alter other data in the computer, equipment or device". 

Again, this is no anology to the C-51 CSIS proposal.  For one thing, it is (supremely) more limited, and carefully codified than the almost endlessly open textured proposal in bill C-51.  For another, it too is tied to an intelligence purpose: the computer access warrant can only be issued where it will "substantially assist the collection of intelligence".  So again, this is not threat disruption per se.  It is about intelligence collection.

Similar observations might be made about other ASIO powers to, e.g., detain.  They are tied to intelligence investigations, not threat reduction per se, and indeed limited to intelligence investigations in relation to terrorism.  The closest analogy in Canadian law would be the investigative hearings provisions in the Criminal Code, involving police.

All of this is to say that the Canadian government's comparison to Australia seems to constitute a Category 1 error, per the methodological guidelines above.

This is not to say that the Australian powers are benign.  They are very controversial.  But if the government does want to give CSIS the sort of powers that ASIO has, then let's have that debate.  Because what it's proposing goes well beyond the ASIO powers, and therefore, Australia is no precedent.


In her Le Devoir article here, behind a paywall at this time, Ms Buzzetti contacted the Danish embassy and spoke to the political attache.  On that basis, she reports: PET, the domestic Danish intelligence service has no disruption powers.

PET's website contains a framework statutory document listing the agency's functions.  Those functions include a reference to "preventive measures".  But in the detailed description ("F. preventive security"), none of the activities sound anything like disruption or threat reduction "measures" of the sort at issue for CSIS in C-51.  That is, it does not look like PET can act physically, unilaterally to prevent security risks, and instead gathers intelligence and helps coordinate activities by other agencies.


As noted above, Canadian government MPs are now claiming "The Finnish Security Intelligence Service is mandated to prevent crimes that may endanger the governmental or political system, and internal or external security, pursuant to section 10 of the Act on Police Administration."

The Finnish embassy did not respond to Ms. Buzzetti.  We have conducted our own research and consulted with others.  A complicating issue for Norway, Sweden and Finland is the degree of overlap between police and security intelligence.  The two functions tend to be blended -- and the Finnish service was until 2010, the "Security Police".  These Nordic service appear, therefore, to be hybrid police and intelligence agencies. Seemingly for this reason, SUPO (the rebranded Finnish security intelligence service) has functions determined by police legislation.  Section 10 of the Act on Police Administration (p. 4) reads:

The Finish Security Intelligence Service is tasked with preventing undertakings and crimes that may endanger the governmental and political system and internal or external security, and with investigating such crimes.  The Finnish Security Intelligence Service shall also maintain and develop the overall preparedness for preventing activities endangering national security.  The matters investigated by the Finnish Security Intelligence Service shall be determined by the National Police Board.

Based on a review of SUPO's website, the Service does describe itself as carrying out "preventive security work" which "comprises security clearances as well as steering and counselling activities.  The objective of the operations is to develop preventive, comprehensive and methodological security activities among authorities, communities and enterprises.  Supervisory and counselling activities include general security education and expert assistance on, for example, facility and information security" (p.29)

SUPO also has powers granted by the Finnish Police Act for the prevention and detection of crimes, especially in terms of terrorist offences (p.16). 

We must, however, avoid a Category 2 error by claiming that because SUPO has police style preventive powers, so too must CSIS.  SUPO seems to be performing a classic police role in relation to codified crimes (and indeed, the Finnish terrorism crimes appear in some respect to be narrower than our own).  That classic police role tied to crimes is not justification for a much more expansive C-51 style function, to "threat reduce" tied to broadly defined national security preoccupations.

Citing SUPO as a precedent for C-51 only make senses if SUPO is filling a gap not currently covered by a combination of CSIS and our own police services.  Is there such a gap?

Having reviewed a translation of the Police Act (Finnish version here), it does not appear that SUPO, in performing its hybridized security police functions, has any greater powers than does the RCMP in Canada -- and indeed probably has less.  What the Police Act means by "prevent" lines up with the sort of peace office arrest powers Canadian police exercise every day.

But critically, there appears to be no suggestion that SUPO has some sort of special, unique disruption power above and beyond regular police powers or that it can by-pass the law, in the way C-51 anticipates being available for CSIS.

This assessment is consistent with that of Professor Iain Cameron, Uppsala University law school, Sweden.  Professor Cameron is familiar with the security law of the Nordic countries.  He writes in relation to Sweden, Finland and Norway:

The Swedish security police mandate is framed in similar ways to the Norwegian and Finnish mandates you quote. It certainly does NOT mean that the Swedish police is permitted to violate any (minor or major) laws, when it is trying to prevent security crime. Even the use of fake identities - when police go undercover and infiltrate - is subject to supervision by the oversight body in Sweden. I am almost certain that the same position is taken on disruption in Norway and Finland. Of course, there are occasional rumours that the security police use covert human sources (informants) not simply passively, but occasionally actively (to infiltrate). Infiltration as such is not explicitly regulated in the law - even though I wrote a report for the government commission of inquiry that the ECHR requires this, and this was the inquiry's conclusion too. Defences exist of necessity (nödvärn) in the criminal law, but there is no question of permission being given in advance to commit crimes/disrupt.

In sum, the Canadian government appears not to have a compelling case in pointing to Finland as justification for C-51.


Ms. Buzzetti concludes that France, in fact, of all the countries she investigates, may have an intelligence service with disruptive powers.  She spoke to the French embassy.  But that embassy was unable to confirm whether the services have disruptive powers because these services operate in a "legislative void".  In fact, a new French bill is trying to fill an aspect of this legislative void.  In the wake of that bill, France is now in the midst of a heated political debate, focusing particularly on surveillance and privacy but not on more kinetic powers. 

Our French colleague confirms the information from the French embassy and tells us that the legislative void for the intelligence services will not be fully filled by this bill because it focuses on information issues and not physical measures.  In the past the DGSE (the foreign intelligence service) has reportedly acted physically to reduce threats on direct orders from the executive, and in a manner not actually authorized by French law.

On the basis of this information, we think using France as a precedent for the CSIS powers would be unpersuasive (and indeed concerning) and amounts to a Category 6 error.

Norway and Sweden

As noted, Norway and Sweden have security police services with police powers, analogous to those exercised in Canada by the RCMP.  It is a Category 2 mistake to point to this police powers and say that CSIS needs them.  Moreover, per Professor Cameron's comments these powers do not appear to reach as far as the sort of law-violating conduct sanctioned by C-51. 

United Kingdom

Some Canadian government MPs now have statements on their websites claiming "In the United Kingdom, MI5 can, pursuant to section 1 of the Security Service Act conduct any activity to protect national security."  We have shared these comments with colleagues in the United Kingdom.  Their reaction is best described as "astonishment".  Section 1 of the Security Service Act specifies the mandate of MI5, but does not spell out its powers.  In other words, it does not authorize anything, it just describe the jurisdiction of MI5.  At any rate, s.1 does not say that MI5 can conduct "any activity".

As compared to CSIS, our interlocutors report that MI5 is more active and less passive. That is, MI5 is not strictly a passive watcher and collector.

That said, persons with experience in the UK security system do not believe that MI5 has the range of powers in play with C-51 for CSIS.  They also report that MI5 activity must be lawful. Here, the key piece of legislation governing the specifics is RIPA 2000. MI5 investigations must accord with the Human Rights Act 1998, and investigative activity is conducted under RIPA by warrant or via internal authorisation. RIPA is about surveillance activities.

Our United Kingdom interlocutors are consistent in stating that MI5 does not have the power to do threat reduction of the sort at issue in C-51 (that is, measures that may break laws or violate human rights protections).  It is not, in other words, a precedent for C-51, and to claim it as such is a Category 4 error.

MI6 -- the foreign intelligence service -- operates outside of the UK.  Here, under the Intelligence Services Act, conduct that would render someone lible civilly or criminally in the United Kingdom may be exonerated for international operations through prior authorization by the Secretary of State.  Pointing to this overseas activity power to justify CSIS domestic powers is a Category 3 error.

In sum, our interlocutors report that neither intelligence service has legal powers to carry out domestic disruption activities involving the breaking of law and the violating of human rights, with warrant or without.

United States

The FBI is a law enforcement body, with police powers.  It would, therefore, be a Category 2 error to point to the FBI police powers as justifying CSIS's C-51 expanded role.  Police roles in Canada are already amply exercised in Canada by our own police agencies.

The CIA is a foreign intelligence agency that does also perform covert, physical actions including overseas use of force that would be unlawful if committed in the United States.  It would be a Category 3 error to point to this fact to justify CSIS domestic powers. 

However, the one page speaking note document posted by some Canadian government MPs on their websites claims: "In the US the Central Intelligence Agency can, pursuant to the National Security Act, conduct domestic threat disruption with an executive order." 

CIA is generally prohibited from conflating its intelligence functions with other activities.  The National Security Act, in codified version at 50 USC Code § 3036(d)(1), specifies that the CIA shall collect intelligence, "except that the Director of the Central Intelligence Agency shall have no police, subpoena, or law enforcement powers or internal security functions" (emphasis added).

It would appear, though, that the Canadian government has simply looked to 50 USC Code § 3036(d)(4), specifying that the CIA director shall "perform such other functions and duties related to intelligence affecting the national security as the President or the Director of National Intelligence may direct".   

Executive Order 12223 (as amended) is the relevant executive instrument, and appears to spell out the CIA's powers in section 1.7.  Our interlocutors were persuaded that a domestic covert operation would never happen now in the United States, and that the National Security Act limitations on "internal security" should also constrain the exercise internal covert action activities. 

This may be opaque legal ground.  The Order does not appear emphatically to limit "covert action activities" to foreign activities.  It does, however, require presidential approval, and a presidential conclusion that no other agency is more likely to achieve the objective.  And more than that, in a statement that would be true even if not included in the Executive Order, section 2.8 specifies that: "Nothing in this Order shall be construed to authorize any activity in violation of the Constitution or statutes of the United States."

In other words, even if the CIA were authorized by a presidential order to conduct a domestic covert activity, it's conduct could not violate US law or the US Constitution. So pointing to the CIA is a variant on a Catagory 5 mistake.

We would add, again, that our US interlocutors would be astonished if the CIA were actually authorized by the president to conduct a domestic covert operation, and we suspect that such conduct would be subject to rapid legal challenge given the ambiguity of the National Security Act.

It follows, again, that the US experience is not adequate precedent for Bill C-51.



Bill C-51: What Did We Learn About the Government's Intentions from the Clause-by-Clause

Last night, the Commons national security committee conducted its lengthy clause-by-clause review of bill C-51.  The government MPs refused every opposition amendment and only passed small amendments of their own. These amendments were in fact those discussed (and analyzed) here.

I will not comment here on the nature of this process, and what it says about our current capacity for reasoned debate.  Colleagues have already commented and continue to comment (correctly) on the state of our democracy, as have members of the media.  I will only say that this process is night and day compared to the more important role Parliament played in both the enactment of the original CSIS Act in 1983/84 and that of the first Anti-terrorism Act in 2001.

Rather, in this post, I want to explore what more we learned about C-51 from the committee process, and especially the statements by the government lawyers in attendance and the prepared speaking notes of the government MPs, as they defeated every opposition amendment.  And I want to focus specifically on the controversial CSIS powers.

1. Powers will allow CSIS to breach law or the Charter, with warrant domestically

At issue is proposed s.12.1(3): "The Service shall not take measures to reduce a threat to the security of Canada if those measures will contravene a right or freedom guaranteed by the Canadian Charter
of Rights and Freedoms or will be contrary to other Canadian law, unless the Service
is authorized to take them by a warrant issued under section 21.1"
(emphasis added).

Basically, this has been read by almost every lawyer outside government whose opinion I have access to as "the Service can violate the Charter -- and potentially any Charter right -- if it has a warrant". 

A couple of key points:

  • This is very different from search and arrest warrants -- those are tied to Charter rights that have qualifying language in the right itself (Section 8 of the Charter only guards against unreasonable searches and seizures.  Section 9 only protects against arbitrary detention). A search or an arrest warrant satisfies this qualifying language, and therefore a government acting under such a warrant does not breach the Charter
  • Most other Charter rights are not imbued with built in qualifying language.  There is no concept of permissible free speech, or arbitrary cruel and unusual treatment, or appropriate mobility rights to enter or leave the country or limited habeas corpus. 
  • Such rights can only be trumped under s.1 of the Charter (or if the government uses the s.33 notwithstanding clause, which it is not doing in C-51).  Section 1 reads: "The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society."
  • But s.1 issues simply aren't dealt with through the peculiar mechanism of a warrant -- as noted below, procedurally to do so is a very concerning approach.

But, not surprisingly, the government now asserts that s.1 is exactly what they have in mind.  And because their proposed system supposedly allows an advance s.1 "get out of jail" blessing for a subsequent Charter right breach, there is no Charter violation:

Government witness: "The suggestion that the bill is designed to actually have a judge violate the charter or be co-opted into violating the charter... that is not what the bill does. What the bill does is precisely the opposite. It puts the judge in the position of deciding whether or not the charter would be violated by the proposed measure. If it would be violated, that is the end of the matter. No one, including the judge, can authorize the measure. ... The judge in fact is being put in precisely the position of looking at the facts of a particular case and determining whether or not the rights that are at issue are reasonably restricted. That is precisely one of the functions allowed a judge under the charter. Section 1 provides for that determination and that's what the bill in fact provides for. So it is not correct, in our submission, that in fact the bill is in any way co-opting the court or anyone else into sanctioning a charter violation. It goes to a judge precisely for that reason, to make sure the charter will not be violated, and the charter violation occurs when a particular right is restricted in a way that is not reasonable, and that is the inquiry that a judge makes under the statute. ... What it turns on is section 1 of the charter which provides that the rights referred to in the charter are guaranteed only to the extent that they are not restricted by reasonable limits prescribed by law in a free and democratic society. That what it turns on."


Our Response:

This legal theory is novel both on substance and radical in terms of procedure. (For footnotes, see p.23 and following here).

Substantive Objections

It is certainly true that s.1 has been used to justify administrative action in violation of the Charter under what is known as the Dore test, in after-the-fact judicial consideration of such actions.

But in every instance where the Dore approach to s.1 has been applied, the delegated power is much more closely anchored to a limited and specific range of possible government conduct, and the Charter rights potentially at play are not “the entire Charter”. 

In C-51, every possible Charter right may be at issue, if its infringement is viewed as necessary for threat reduction.  Put another way, the discretion to violate the Charter at issue here is much less bounded than any equivalent of which we are aware.  Moreover, the CSIS Act is not your typical administrative regime; it is one that by design skates very close to the edge of what the state should be doing in a democratic society.

Even more critically, we only reach s.1 in the first place where the rights limitation is “prescribed by law”. 

“Prescribed by law” is rarely an issue in constitutional disputes, since the government action in question is either expressly authorized in a stature or is sufficiently linked to it to meet the “prescribed by law” standard. In challenges to legislative action, the Supreme Court takes a “flexible approach to the ‘prescribed by law’ requirement as regards both the form (e.g., statute, regulation, municipal by-law, rule of a regulatory body or collective agreement provision) and articulation of a limit on a Charter right (i.e., a standard intelligible to the public and to those who apply the law).”

In bill C-51, neither this form requirement nor the intelligible standard expectation is met.  We cannot predict in advance which Charter right will be violated, or the specific circumstances or nature of the breach.  That is a matter that will be decided on a case-by-case basis. 

As a consequence, the bill constitutes an even more aggressive manifestation of the “prescribed by law” shortcomings identified by the Supreme Court in provisions that once governed court-authorized bail conditions, as in Morales.  It offers exactly the sort of vagueness and imprecision that disentitles the measure to a full s.1 inquiry. In the proposed new CSIS powers, the only statutory framework translates into: “you can do anything to ‘reduce’ broadly defined threats to the security of Canada, including violating every right in the Charter, so long as it doesn’t do bodily harm, violate sexual integrity or obstruct justice”. 

We are not aware of any circumstances in which the Supreme Court has concluded that such an open-textured invitation to violate the Charter is “prescribed by law”. That is probably because we have never before seen such an open-textured invitation. 

This would be bad enough, but we also believe that this system amounts to a drafting of judges into the legislative function of limiting Charter rights through s.1-style justifications. As suggested, it differs so significantly from the traditional search warrant process that we do not accept that approach as a plausible analogy.  Moreover, the present system is dramatically different from the “investigative hearings” process upheld by the Supreme Court the last time judicial independence was a live issue in a national security context. 

Most notably, a key ingredient saving the latter process from being unconstitutional was the fact that investigative hearings are held presumptively in open court.  As we discuss next, that safeguard does not exist in bill C-51.  Moreover, a strong minority of the Court concluded that even the relatively banal investigative hearing system did violence to the role of judges. The dissent concluded that judges were in effect being made into police investigators, even though investigative hearings are adversarial hearings held in open court.

Bill C-51 concentrates the legislative power to authorize and limit Charter rights on “section 1 reasonableness grounds” into the hands of those Federal Court judges who have been specially designated to sit in security cases, in secret.  This is dramatically different from (and much more concerning than) investigative hearings.

It is radically different from what judges do under s.1.  Judges adjudicate whether a Charter breach by another branch is prescribed by law and demonstrably reasonable in a free and democratic society.  They do not pre-authorize a Charter breach because they think it reasonable and necessary.  That is not what they have ever done, and it profoundly contorts their role.

Procedural complaints

And then we get to the procedural objection. The government's exotic C-51 s.1 analysis will be conducted in a warrant proceeding.  These are ex parte (only government party present) and in camera (closed) proceedings, done in advance of government Charter breaching action without necessarily knowing its full contours (as in, how it will play out on the ground) and without a full evidentiary record.  So the hearings will be done in secret, with only the judge and the government side represented, and no real full facts on what exactly has happened to the the target (or whether the target has been misidentified or otherwise constitute a "false positive"). The person affected by the illegal activity will not be represented — in fact they will often never know who visited the misfortune on them, depending on what is done to them. They cannot defend their rights. No civil rights group will be able to weigh in.  There is no appeal system.

This is unlike any s.1 analysis done by the courts in other circumstances -- usually a profoundly methodical process with a developed factual record. 

And so when you place the substantive concerns together with the procedural complaints we end up with a government's legal theory that goes from novel to "radical".  The government wishes to breach every and any right, with prior blessing by a court performing the dirty work of limiting indisputable Charter rights in a secret, closed process, without appeal and with the government the master of all facts (and the only party formally represented).

We are just two lawyers, and the government has its own lawyers.  And lawyers argue.  But we are hardly alone (to say the least) in saying this is all seriously unconstitutional.  And so since the government persists with this theory, this matter will now have to be litigated. Parliamentarians should appreciate, however, that in enacting this extreme measure, they place the burden of defending foundational concepts of our system of democracy and law on the shoulders of underfunded non-profits, in litigation where the government will challenge their right to do so at every turn.

2. What Will CSIS Do With its Powers?

A question many of us have asked is "what does CSIS need this vast new power to do".  In their speaking notes, the government has made persistent reference to “CSIS needs to be able to speak to parents of radicalizing children” – something that CSIS reportedly does already and at any rate does not justify the astonishing breadth of the new powers.

So it was both interesting and disturbing to see the government's response to Green Party and NDP efforts to limit CSIS's powers to exclude, e.g., detention. 

Violate Mobility Rights

First, one government witness used limitations on mobility rights as an example of what might be limited by CSIS measures (that is, entering and returning to Canada).


On detention, a key question is whether the government has really, truly wanted CSIS to be in the detention business.  The responses on this have either been non-responses or confusing.  The government tabled and the committee passed a "greater certainty" provision saying CSIS will have no "law enforcement powers".  But for the reasons we discuss here, this does nothing to clarify this question -- it still leaves open the prospect of "threat reduction detention".  And our views seem confirmed by what then happened in clause-by-clause.  And one government witness said this:

The concept of detention means different things in different contexts. In some cases it may give rise to treatment that would amount to bodily harm, but not necessarily. People are detained at the border for inspection purposes but they don't necessarily find themselves subjected to bodily harm or treatment that is referred to in the act.

If I might, I would just indicate that the reference to CSIS not having law enforcement powers, as the member indicated, was intended to address the concern that certain powers associated with a law enforcement agency were not being given to CSIS. The important point that was reflected in the drafting is that CSIS itself as an agency cannot take it upon itself to exercise those powers. It has no power and never has had a power to detain or arrest or imprison. Nothing in this bill changes that.

When the concept of detention is used for example, again, to repeat myself, the services never had a power to detain. That is a peace or police officer power that is either conferred by common law or by a statute. It doesn't find itself in CSIS. The point is that in the course of CSIS operations, they may in fact identify opportunities to take measures to interfere with a person's movement. What the act provides is, if CSIS wishes to do that, and if to do that would contravene the law, they have to obtain judicial authorization to do it. The important point in the legislation that we tried to reflect in the drafting was the it was never up to CSIS to make that decision on its own. It would always fall to a judge to make that determination.

The government MPs then defeated a Green amendment that would prohibit detention, saying "The amendment that is on the floor right now is not consistent with the intent of the bill. It would unduly narrow the range of possible measures that CSIS could take, and would so weaken CSIS's capacity to carry out its threat reduction mandate."

Short version: CSIS hasn't had detention in the past, but with judicial authorization it will be able to detain after C-51, and that is exactly what the government has in mind. 

I would add that nothing in the bill requires a warrant where the detention takes place overseas, at least for non-Canadians.  C-51 only requires a warrant where the conduct would violate Canadian law or the Charter.  These instruments rarely extend to overseas government conduct.  And so the trigger requiring the warrant will rarely be pulled.  Under current Federal Court jurisprudence, the Charter only applies overseas when, in essence, the Canadian government conduct would violate international human rights law and (a more doubtful requirement) it is directed at a Canadian.  We think the last requirement very doubtful, but it is what the Federal Court has said.

So the clear: C-51 combined with present understandings of the Charter would allow overseas detentions of foreign nationals, without warrant or judicial supervision.


Another concerning comment: According to government witnesses, the government amendment specifying CSIS does not have "law enforcement powers" will not bar "rendition".  In the words of the government witness: "just the reference to 'rendition' or 'removal to another state' is not necessarily a law enforcement power. So to the extent that the amendment refers to 'law enforcement', it may not be a like thing." 

In recent history, rendition is the process by which a person is kidnapped from one jurisdiction and taken to another, sometimes for trial and sometimes for abusive interrogation.

This is not to say CSIS would do this. But in truth, the Service has done forms of detention and removal of Canadians to foreign jurisdictions before.  With warrant, this treatment of Canadian would now be potentially legal.

If the treatment related to overseas treatment of a foreign national, again under current understandings of the Charter, a warrant would not be required.

We do not propose that CSIS's culture would lead it to the excess of other services who have practiced detention and rendition.  We also do not believe our law should ever be degraded to the point where that excess becomes legally permissible.

3. The Foreign Analogies

Last night, the government persists in the view that other Five Eyes countries (or at least the UK, US and Australia) have powers to the open ended degree C-51 proposes for CSIS.  Specifically: the power to breach laws and the constitution, in domestic operations. 

I have written on this before here and here.  Kent Roach has since spoken to his contacts. We have consulted with people with considerable experience in the UK. Since last night's repeated assertion, I have re-written to colleagues in the UK, Australia and the United States, some of whom are consulting with members of their security services.  I await full results, but again so far, no one has been able to identify close analogues. 

My concern right now is that the government is pointing to much more modest powers possessed by foreign domestic services, adding to those the sorts of powers exercised by other services only in their international operations, and then has proposed a package for CSIS that gives them sweeping powers both domestically and internationally that other like-countries just don't have.

Nothing stops the government from releasing its comparative analysis.  But it refuses to do so, making it very difficult to determine whether it is justifying apples with references to oranges.

And that is a symptom of how this debate has unfolded. This is lamentable on many different levels.