Not Such an Emergency: Canada (Attorney General) v. Canadian Civil Liberties Association, 2026 FCA 6

Last month, the Federal Court of Appeal handed down an important judgment in Canada (Attorney General) v. Canadian Civil Liberties Association, 2026 FCA 6. The Court of Appeal unanimously agreed with the Federal Court that the federal government’s invocation of the Emergencies Act during the so-called ‘Freedom Convoy’ protests in Ottawa in 2022 was unlawful.

Common law courts do not have a particularly proud history in addressing the lawfulness of declarations of emergency by the executive branch. Reflexive deference has been the norm — even though Lord Atkin’s dissent in Liversidge v. Anderson came to be (justly) celebrated, by the time the War on Terror rolled around, judges again found themselves drawn to deference. As such, the fact that four experienced judges have now concluded that the declaration here was unlawful is quite remarkable, albeit (I suppose) that the ’emergency’ in question has long since passed. Regular readers will know of my view that the invocation was, indeed, unlawful (and that I have assisted counsel for the Canadian Constitution Foundation in this case). In a nutshell, the intolerable situation in Ottawa (and, to be clear, it was intolerable for the residents most affected by the occupation) arose due to municipal incompetence and persisted due to provincial reticence to get involved and, although it inspired other protests elsewhere in Canada, these were dismantled using existing legal authority, without resort to the Act. It was a local affair, not a national emergency.

Here, I will make four general observations about where this significant decision sits in relation to important themes in contemporary Canadian public law.

First, the concept of legal and factual constraints, introduced by the Supreme Court in the Vavilov decision, was central to the outcome, with the Federal Court of Appeal finding that the invocation of the Act was unreasonable. These constraints are objective: they either exist and are relevant, or they do not and are not. Ultimately, the existence and relevance of a constraint is a question for a reviewing court. If the decision-maker has provided reasons on a particular constraint (or has explained why it believes a particular constraint is not relevant), then a court should ordinarily defer to the decision-maker. But if no reasons have been provided, then no deference can be given. So it was here. Legally, the federal government was tightly constrained by the text and scheme of the Act, designed, the legislative history makes clear, to limit recourse to the extraordinary powers Parliament had created. Whilst it had discretion about invoking the Act, it had to satisfy objective criteria to do so:

There is no disagreement between the parties that the decision to invoke the Act is highly discretionary and attracts a high degree of deference, not only because Cabinet sits at the apex of the executive and makes decisions on the basis of wide considerations of policy and public interest, but also because of the fluid, unpredictable, and fast-moving nature of an emergency situation. The same cannot be said, however, of the objective legal requirements that have to be met before the GIC can exercise its legitimate discretion…the wording of subsection 17(1) is quite circumscribed and cannot be interpreted as conferring unconstrained discretion. The “belief on reasonable grounds” concept has been the subject of numerous judicial pronouncements, and the definition of “public order emergency” as found in section 16 refers to objective standards and to a clear definition in the CSIS Act which makes this provision “more akin to the legal determinations courts make, governed by legal authorities, not policy” (at paras. 166, 171).

Here, the reasons given by the federal government simply did not satisfy those criteria, for example in relation to the requirement that the federal government have reasonable grounds to believe that serious violence is afoot:

What the AGC is really arguing before us is that rendering critical infrastructure unusable creates the same danger to Canadians’ safety and security as physical damage to that infrastructure and amounts to serious violence with respect to property. In the Proclamation, Cabinet invoked three reasons of an economic nature as justification to declare a public order emergency: the adverse effects on the Canadian economy and threats to its economic security resulting from the impacts of blockades of critical infrastructures, the adverse effects resulting from the impacts of the blockades on Canada’s relationship with its trading partners, and the breakdown in the distribution chain and availability of essential goods, services and resources caused by the existing (and potential additional) blockades. Interestingly, the Section 58 Explanation expanded on these risks in speculative language.

In light of the text, context and purpose … this expansive interpretation of “serious violence” to property is unwarranted and unreasonable. It could stifle all kinds of protests and demonstrations that blockade pipelines, nuclear plants, railway lines and other kinds of infrastructure to advance a cause. There is no indication, either in the definition itself of “threats to the security of Canada” in the CSIS Act or in the debates surrounding the adoption of that definition, that the kind of economic disruptions described in the Section 58 Explanation could be the basis for declaring a public order emergency. As previously mentioned, the adoption of the Act was clearly meant to curb the excesses and prevent the abuses that occurred under the [War Measures Act, the notorious predecessor to the Act]. The reference to the definition of “threats to the security of Canada” in the CSIS Act, which itself requires threat or use of acts of “serious violence”, was meant to assuage the concerns that pure economic considerations, especially those of a speculative or tentative nature, would not prevail over democratic values and fundamental freedoms of assembly and expression (at paras. 207-208).

The decision (set out in the so-called “section 58” reasons that had to be provided under the Act) did not fare any better on compliance with factual constraints:

Relying on the Section 58 Explanation, the AGC claims that there were considerable cumulative threats of serious violence to individuals, including the threat of lethal violence. Having carefully read that document, we agree with the Federal Court that there is very little hard evidence of any actual serious violence or threats of it, except at Coutts; otherwise, the Section 58 Explanation focuses mostly on the economic impact of the blockades, on speculation as to what might happen if the protests were not brought to an end, and on unsubstantiated and vague reports from unidentified sources (at para. 212).

At the end of the day, most of the blockades across Canada were dismantled without resort to the Act, making it difficult to satisfy this requirement (at para. 224, 263):

As for the threat or use of acts of serious violence against persons, when properly understood as requiring bodily harm, the evidence is quite simply lacking. Aside from the economic disturbance, the only incident of violence put forward by the AGC was the seizure of a cache of firearms and ammunition at Coutts, as well as vague reports of harassment, intimidation, and assault, and the fact that the police forces in Ottawa were overwhelmed. In our view, this is insufficient to satisfy the compelling and credible information requirement to justify the conclusion that there were reasonable grounds to believe that there was a threat or use of acts of serious violence. Declaring a public order emergency is a very serious matter, considering the extraordinary powers vested in the executive branch of the federal government once the Act is invoked, and for that reason the exacting requirements set out by Parliament must be strictly adhered to (at para. 231).

Second, the emphasis in Vavilov on responsiveness and justification may, in some instances, impose a positive duty on decision-makers to seek out additional information — or, at the very least, justify a negative inference when they could, but did not, do so. As the Federal Court of Appeal explained, “the proper question on judicial review is not whether there was actually a threat to the security of Canada when the GIC exercised its discretion to make a declaration, but whether the GIC acted reasonably in finding reasonable grounds to believe there was such a threat. But that belief will not be based on reasonable grounds if it is based on insufficient evidence or if that evidence is not sufficient to meet the legal standards prescribed in the Act” (at para. 228). Here, the security services had advised that the blockade in Ottawa was not a threat to the security of Canada (one of the pre-requisites for invoking the Act). The fact that the federal government did not seek a further threat assessment from the security services weighed against the reasonableness of its decision:

Of course, the CSIS assessment that there were no threats to the security of Canada was not determinative and could not bind Cabinet. The Federal Court and the respondents agree that the GIC was not limited to considering the intelligence collected by CSIS or by its analysis of that intelligence. Because of its expertise in investigating threats to the security of Canada, however, CSIS’s threat assessment should nevertheless have carried substantial weight; after all, it is one of CSIS’s principal activities to investigate, analyse, and retain information and intelligence on security threats: X(Re), 2016 FC 1105 at para. 159, citing the Pitfield Report at para. 28.

If Cabinet was not satisfied with CSIS’s threat assessment, it was always open to it to ask for further information and analysis from CSIS, the RCMP, or other relevant federal departments or agencies. In fact, it appears from the record that an alternative threat assessment was requested by the Clerk of the Privy Council on February 14, 2022. Without going into the details of what happened on that day, what is clear is that no alternative threat assessment was ever prepared before the invocation of the Act.

In the Invocation Memorandum which was prepared by the Clerk of the Privy Council and that ended up being the last piece of advice that went to the Prime Minister, we find the mention that “[a] more detailed threat assessment is being provided under separate cover” (Invocation Memorandum, p. 2; AB, Vol. 1, Tab 5, p. 189). Yet the Clerk testified that “there was no written detailed threat assessment provided under separate cover”: Commission Testimony of Clerk Charette and Deputy Clerk Drouin (excerpts) (November 18, 2022), Exhibit D, Zwibel Affidavit, AB, Vol. 1, Tab 11.8, p. 419 (at paras. 219-220).

Similarly, in relation to whether the alleged emergency was serious enough to overwhelm provincial authority and capacity and require the invocation of the Act rather than (say) the armed forces assisting the provincial authorities, it was “no answer for the AGC to argue now, ex post facto, that the military did not have either the training or the equipment to assist the police. It was incumbent on Cabinet to provide that evidence, since it is for the decision-maker to establish the reasonableness of its decision” (at para. 258; see also at para. 269 in relation to federal police). Again, further inquiries were possible and the failure to make them was held against the federal government.

Third, the consequences of invoking the Act, which upsets the ordinary division of powers between the federal and provincial governments and permits the use of sweeping emergency powers, were an important factor in the reasonableness analysis. In Vavilov, the Supreme Court stated that where a decision would visit “harsh consequences” on an individual, more by way of justification is required. The Federal Court of Appeal relied on this in two ways: to begin with, the impact the Act may have “on an individual’s life” was a “reason why the specific conditions underlying the existence of a national emergency must be strictly adhered to” (at para. 245); and, in addition, the “constitutional underpinning” of the Act as an exceptional measure of last resort was another reason “to require a robust justification for the finding of a national emergency” (at para. 247). The “harsh consequences” aspect of Vavilov fits well here, in my view, given the extraordinary nature of the Act — it is appropriate that respect for the division of powers can trigger a heightened burden of justification just as interference with personal interests will.

Fourth, the Federal Court of Appeal administered a welcome dose of constitutional realism in response to the federal government’s argument that the decision-maker targeted by judicial review was the Governor in Council rather than the federal cabinet. This was potentially significant for the content of the record before the Federal Court of Appeal. The basic principle is that only material that was before the decision-maker when it made its decision can be included in the record on judicial review. Here, the government argued, the decision-maker was the Governor in Council, not the cabinet. And, it followed, the record for judicial review purposes should only contain what was before the Governor in Council (a relatively limited amount of material, because the Governor in Council acts on cabinet’s advice: see para. 129) and not what was before cabinet. In this regard, it should be noted that under the Act, there is a mandatory public inquiry into its invocation: here, the parties had sought to introduce into the record material that came to light during the public inquiry, as part of the general background to the invocation of the Act. On the federal government’s approach, none of this should have been considered on judicial review: even though the material was before cabinet, it was never put before the Governor in Council at the point the Act was actually invoked. The Federal Court of Appeal roundly rejected the government’s argument, stating that “accepting this argument would turn back the clock of our constitutional history and undo many of the constitutional conventions that have evolved over the course of more than a century” (at para. 126):

To argue, as does the AGC, that the GIC is the “sole decision maker” under the Emergencies Act is antithetical to the principles of democracy and to the firmly established constitutional convention of responsible government. It is well understood since the early days of the Confederation that the GIC does not act on her own, but must act on the advice of Cabinet, which is the only active part of the Privy Council. Except for a few remaining “reserve powers” (such as the right to dismiss a Prime Minister or dissolve the House of Commons when the government remains in office after having lost the confidence of the House), the GIC has no choice but to confirm Cabinet’s advice. The notion that the GIC was convened separately from the Cabinet and exercised the powers Parliament gave to it under subsections 17(1) and 19(1) of the Emergencies Act is therefore a fiction, as no such meeting ever took place (at para. 130).

Accordingly, the Federal Court’s decision to permit the parties to introduce evidence from the public inquiry was correct in law (at para. 138). In addition, the government’s argument that cabinet is not an entity reviewable under the Federal Courts Act (on the rather technical basis that it exists by way of constitutional convention rather than under statute) was also rejected: “For all intents and purposes, therefore, Cabinet exercises the powers vested in the Privy Council. Accordingly, its decisions must be reviewable by the Federal Courts” (at para. 141). As I say, a welcome dose of constitutional realism and a decision that is likely to enhance transparency in the exercise of public power (albeit that the federal government still has expansive authority to claim privilege over cabinet-level documents under the Canada Evidence Act).

Evidently, I have strong priors about the lawfulness of the invocation of the Act, so I am content that the Federal Court of Appeal’s legal analysis of the relevant administrative law principles is correct. More than that, on these important themes in contemporary Canadian public law, the Federal Court of Appeal’s reasons make a significant contribution to the general principles of judicial review of administrative action.

This content has been updated on February 18, 2026 at 02:14.